In practical use, there is shrinkage in the width direction in existing overflow water film. This study introduces the ” flow deflectors ” to solve the problem of discontinuous water film. According to the principle of the experimental devices, the model was established by FLOW-3D software and the corresponding boundary conditions were set up, the minimum size of the grid was 0.25mm × 0.25mm × 0.25mm. The film model was investigated under ten working conditions with three factors. Through the analysis of the distribution curves of velocities and thickness, impacts of unit width flux, tank width and the distance between the flow deflectors on the velocity and thickness of the liquid film were obtained. The results indicated that the water film was able to keep continuous and airtight with flow deflectors. The variation law of water film with unit width flux was obtained and a correlation formula was proposed according to Nusselt correlation. And, it was found that the tank width has little influence on the water film itself. The “dry zones” were also founded on the film when the distance between flow deflectors increased.
References
Gong H, Jiang J, Qiu J, Jiang R and Zhang W 2016 Theoretical and Experimental Study on the Water Curtain Shape and Mechanical Relationship of the Smoke Control System Advances in Engineering Research 500-7 Gong H, Jiang J, Zhang W and Guan Y 2017 Theory and Simulation of the Relationship between Smoke-proof Water Curtain’s Tank Structure and Flow Non-uniformity DEStech Transactions on Environment, Energy and Earth Science iceepe 257-63 ZHENG Li, QIAN Zhongdong, CAO Zhixian and TAN Guangming 2009 Comparison of computed result for dam over-topping flow with two 3D models Engineering Journal of Wuhan University 06 758-63 Hartley DE and Murgatroyd W 1964 Criteria for the break-up of thin liquid layers flowing isothermally over solid surfaces International Journal of Heat and Mass Transfer 9 1003-15 QIN Wei, LIU Jianhua, WENG Zemin and JIANG Zhangyan 1997 Characteristics of Interfacial Surface Wave of Free-Falling Liquid Film JOURNAL OF JIMEI NAVIGATION INSTITUTE 02 3-8 Ye X, Yan W, Jiang Z and Li C 2002 Hydrodynamics of Free-Falling Turbulent Wavy Films and Implications for Enhanced Heat Transfer Heat Transfer Engineering 1 48-60
Microstructural defects in laser powder bed fusion (LPBF) metallic materials are correlated with processing parameters. A multi-physics model and a crystal plasticity framework are employed to predict microstructure growth in molten pools and assess the impact of manufacturing defects on plastic damage parameters. Criteria for optimising the LPBF process are identified, addressing microstructural defects and tensile properties of LPBF Hastelloy X at various volumetric energy densities (VED). The results show that higher VED levels foster a specific Goss texture {110} <001>, with irregular lack of fusion defects significantly affecting plastic damage, especially near the material surface. A critical threshold emerges between manufacturing defects and grain sizes in plastic strain accumulation. The optimal processing window for superior Hastelloy X mechanical properties ranges from 43 to 53 J/mm3 . This work accelerates the development of superior strengthductility alloys via LPBF, streamlining the trial-and-error process and reducing associated costs.
Figure 3. The simulated temperature distribution and single-layer multi-track isothermograms of LPBF Hastelloy X, located at the bottom of the powder bed, are presented for various laser energy densities. (a) depicts the single-point temperature distribution at the bottom of the powder bed, followed by the isothermograms corresponding to laser energy densities of (b) 31 J/mm3 , (c) 43 J/mm3 , (d) 53 J/mm3 , (e) 67 J/mm3 , and (f) 91 J/mm3 .
References [1] DebRoy T, Wei HL, Zuback JS, et al. Additive manufacturing of metallic components – process, structure and properties. Prog Mater Sci. 2018;92:112–224. doi:10. 1016/j.pmatsci.2017.10.001 [2] Mostafaei A, Ghiaasiaan R, Ho IT, et al. Additive manufacturing of nickel-based superalloys: A state-of-the-art review on process-structure-defect-property relationship. Prog Mater Sci. 2023;136:101108. doi:10.1016/j.pmatsci. 2023.101108 [3] Akande IG, Oluwole OO, Fayomi OSI, et al. Overview of mechanical, microstructural, oxidation properties and high-temperature applications of superalloys. Mater Today Proc. 2021;43:2222–2231. doi:10.1016/j.matpr. 2020.12.523 [4] Sanchez S, Smith P, Xu Z, et al. Powder bed fusion of nickel-based superalloys: a review. Int J Machine Tools Manuf. 2021;165:103729. doi:10.1016/j.ijmachtools.2021. 103729 [5] Xie Y, Teng Q, Shen M, et al. The role of overlap region width in multi-laser powder bed fusion of Hastelloy X superalloy. Virtual Phys Prototyp. 2023;18(1):e2142802. doi:10.1080/17452759.2022.2142802 [6] Yuan W, Chen H, Cheng T, et al. Effects of laser scanning speeds on different states of the molten pool during selective laser melting: simulation and experiment. Mater Des. 2020;189:108542. doi:10.1016/j.matdes.2020. 108542 [7] He X, Kong D, Zhou Y, et al. Powder recycling effects on porosity development and mechanical properties of Hastelloy X alloy during laser powder bed fusion process. Addit Manuf. 2022;55:102840. doi:10.1016/j. addma.2022.102840 [8] Sanaei N, Fatemi A. Defects in additive manufactured metals and their effect on fatigue performance: a stateof-the-art review. Prog Mater Sci. 2021;117:100724. doi:10.1016/j.pmatsci.2020.100724 [9] Pourbabak S, Montero-Sistiaga ML, Schryvers D, et al. Microscopic investigation of as built and hot isostatic pressed Hastelloy X processed by selective laser melting. Mater Charact. 2019;153:366–371. doi:10.1016/j. matchar.2019.05.024 [10] He X, Wang L, Kong D, et al. Recrystallization effect on surface passivation of Hastelloy X alloy fabricated by laser powder bed fusion. J Mater Sci Technol. 2023;163:245–258. doi:https://doi.org/10.1016j.jmst. 2023.06.003. [11] Sabzi HE, Maeng S, Liang X, et al. Controlling crack formation and porosity in laser powder bed fusion: alloy design and process optimisation. Addit Manuf. 2020;34:101360. doi:10.1016/j.addma.2020.101360 [12] Yu C, Chen N, Li R, et al. Selective laser melting of GH3536 superalloy: microstructure, mechanical properties, and hydrocyclone manufacturing. Adv Powder Mater. 2023:
doi:10.1016/j.apmate.2023.100134 [13] Ye C, Zhang C, Zhao J, et al. Effects of post-processing on the surface finish, porosity, residual stresses, and fatigue performance of additive manufactured metals: a review. J Mater Eng Perform. 2021;30(9):6407–6425. doi:10. 1007/s11665-021-06021-7 [14] Zhang W, Zheng Y, Liu F, et al. Effect of solution temperature on the microstructure and mechanical properties of Hastelloy X superalloy fabricated by laser directed energy deposition. Mater Sci Eng A. 2021;820:141537. doi:10. 1016/j.msea.2021.141537 [15] Lehmann T, Rose D, Ranjbar E, et al. Large-scale metal additive manufacturing: a holistic review of the state of the art and challenges. Int Mater Rev. 2021;67(4):410–459. doi:10.1080/09506608.2021.1971427
[16] Wu S, Hu Y, Yang B, et al. Review on defect characterization and structural integrity assessment method of additively manufactured materials. J Mech Eng. 2021;57 (22):3–34. doi:10.3901/JME.2021.22.003
[17] Keller C, Mokhtari M, Vieille B, et al. Influence of a rescanning strategy with different laser powers on the microstructure and mechanical properties of Hastelloy X elaborated by powder bed fusion. Mater Sci Eng A. 2021;803:140474. doi:10.1016/j.msea.2020.140474
[18] Keshavarzkermani A, Marzbanrad E, Esmaeilizadeh R,et al. An investigation into the effect of process parameters on melt pool geometry, cell spacing, and grain refinement during laser powder bed fusion. Optics & Laser Technol. 2019;116:83–91. doi:10.1016/j.optlastec. 2019.03.012
[19] Watring DS, Benzing JT, Hrabe N, et al. Effects of laserenergy density and build orientation on the structureproperty relationships in as-built Inconel 718 manufactured by laser powder bed fusion. Addit Manuf. 2020;36:101425. doi:10.1016/j.addma.2020.101425
[20] Xiao H, Liu X, Xiao W, et al. Influence of molten-pool cooling rate on solidification structure and mechanical property of laser additive manufactured Inconel 718. J Mater Res Technol. 2022;19:4404–4416. doi:10.1016/j. jmrt.2022.06.162
[21] Wang J, Zhu R, Liu Y, et al. Understanding melt pool characteristics in laser powder bed fusion: An overview of single- and multi-track melt pools for process optimization. Adv Powder Mater. 2023;2(4):100137. doi:10.1016/j. apmate.2023.100137
[22] Li Z, Deng Y, Yao B, et al. Effect of laser scan speed on pool size and densification of selective laser melted CoCr alloy under constant laser energy density. Laser Optoelectronics Progress. 2022;59(7):0736001. doi:10. 3788/LOP202259.0736001
[23] Zhang J, Yuan W, Song B, et al. Towards understanding metallurgical defect formation of selective laser melted wrought aluminum alloys. Adv Powder Mater. 2022;1 (4):100035. doi:10.1016/j.apmate.2022.100035
[24] Rui H, Meiping W, Chen C, et al. Effects of laser energy density on microstructure and corrosion resistance of FeCrNiMnAl high entropy alloy coating. Optics & Laser Technol. 2022;152:108188. doi:https://doi.org/10.1016j. optlastec.2022.108188.
[25] Zhao Y, Sun W, Wang Q, et al. Effect of beam energy density characteristics on microstructure and mechanical properties of nickel-based alloys manufactured by laser directed energy deposition. J Mater Process Technol. 2023;319:118074. doi:10.1016/j.jmatprotec.2023.118074
[26] Tan P, Zhou M, Tang C, et al. Multiphysics modelling of powder bed fusion for polymers. Virtual Phys Prototyp. 2023;18(1):e2257191. doi:10.1080/17452759.2023. 2257191
[27] Tan P, Shen F, Shian Tey W, et al. A numerical study on the packing quality of fibre/polymer composite powder for powder bed fusion additive manufacturing. Virtual Phys Prototyp. 2021;16(sup1):S1–S18. doi:10.1080/17452759. 2021.1922965
[28] Kusano M, Watanabe M. Microstructure control of Hastelloy X by geometry-induced elevation of sample temperature during a laser powder bed fusion process. Mater Des. 2022;222:111016. doi:10.1016/j.matdes.2022. 111016
[29] Lee YS, Zhang W. Modeling of heat transfer, fluid flow and solidification microstructure of nickel-base superalloy fabricated by laser powder bed fusion. Addit Manuf. 2016;12:178–188. doi:10.1016/j.addma.2016.05.003
[30] Lv F, Liang HX, Xie DQ, et al. On the role of laser in situ remelting into pore elimination of Ti-6Al-4V components fabricated by selective laser melting. J Alloys Compd. 2021;854:156866. doi:10.1016/j.jallcom.2020.156866
[31] Prithivirajan V, Sangid MD. The role of defects and critical pore size analysis in the fatigue response of additively manufactured IN718 via crystal plasticity. Mater Des. 2018;150:139–153. doi:10.1016/j.matdes.2018.04.022
[32] Huang Y. A user-material subroutine incroporating single crystal plasticity in the ABAQUS finite element program. Cambridge: Harvard University Press; 1991.
[33] Pilgar CM, Fernandez AM, Lucarini S, et al. Effect of printing direction and thickness on the mechanical behavior of SLM fabricated Hastelloy-X. Int J Plasticity. 2022;153:103250. doi:10.1016/j.ijplas.2022.103250
[34] Garlea E, Choo H, Sluss CC, et al. Variation of elastic mechanical properties with texture, porosity, and defect characteristics in laser powder bed fusion 316L stainless steel. Mater Sci Eng A. 2019;763:138032. doi:10.1016/j. msea.2019.138032
[35] Sanchez-Mata O, Wang X, Muñiz-Lerma JA, et al. Dependence of mechanical properties on crystallographic orientation in nickel-based superalloy Hastelloy X fabricated by laser powder bed fusion. J Alloys Compd. 2021;865:158868. doi:10.1016/j.jallcom.2021. 158868
[36] Gu H, Wei C, Li L, et al. Multi-physics modelling of molten
pool development and track formation in multi-track, multi-layer and multi-material selective laser melting. Int J Heat Mass Transf. 2020;151:119458. doi:10.1016/j. ijheatmasstransfer.2020.119458
[37] Johnson L, Mahmoudi M, Zhang B, et al. Assessing printability maps in additive manufacturing of metal alloys. Acta Mater. 2019;176:199–210. doi:10.1016/j.actamat. 2019.07.005
[38] Wang S, Ning J, Zhu L, et al. Role of porosity defects in metal 3D printing: formation mechanisms, impacts on properties and mitigation strategies. Mater Today. 2022;59:133–160. doi:10.1016/j.mattod.2022.08.014
[39] Guo Y, Collins DM, Tarleton E, et al. Measurements of stress fields near a grain boundary: exploring blocked arrays of dislocations in 3D. Acta Mater. 2015;96:229–doi:10.1016/j.actamat.2015.05.041 [40] Kong D, Dong C, Ni X, et al. Hetero-deformation-induced stress in additively manufactured 316L stainless steel. Mater Res Lett. 2020;8(10):390–397. doi:10.1080/ 21663831.2020.1775149
Unintended end-of-process depression (EOPD) commonly occurs in laser powder bed fusion (LPBF), leading to poor surface quality and lower fatigue strength, especially for many implants. In this study, a high-fidelity multi-physics meso-scale simulation model is developed to uncover the forming mechanism of this defect. A defect-process map of the EOPD phenomenon is obtained using this simulation model. It is found that the EOPD formation mechanisms are different under distinct regions of process parameters. At low scanning speeds in keyhole mode, the long-lasting recoil pressure and the large temperature gradient easily induce EOPD. While at high scanning speeds in keyhole mode, the shallow molten pool morphology and the large solidification rate allow the keyhole to evolve into an EOPD quickly. Nevertheless, in the conduction mode, the Marangoni effects along with a faster solidification rate induce EOPD. Finally, a ‘step’ variable power strategy is proposed to optimise the EOPD defects for the case with high volumetric energy density at low scanning speeds. This work provides a profound understanding and valuable insights into the quality control of LPBF fabrication.
의도하지 않은 공정 종료 후 함몰(EOPD)은 LPBF(레이저 분말층 융합)에서 흔히 발생하며, 특히 많은 임플란트의 경우 표면 품질이 떨어지고 피로 강도가 낮아집니다. 본 연구에서는 이 결함의 형성 메커니즘을 밝히기 위해 충실도가 높은 다중 물리학 메조 규모 시뮬레이션 모델을 개발했습니다.
이 시뮬레이션 모델을 사용하여 EOPD 현상의 결함 프로세스 맵을 얻습니다. EOPD 형성 메커니즘은 공정 매개변수의 별개 영역에서 서로 다른 것으로 밝혀졌습니다.
키홀 모드의 낮은 스캔 속도에서는 오래 지속되는 반동 압력과 큰 온도 구배로 인해 EOPD가 쉽게 유발됩니다. 키홀 모드에서 높은 스캐닝 속도를 유지하는 동안 얕은 용융 풀 형태와 큰 응고 속도로 인해 키홀이 EOPD로 빠르게 진화할 수 있습니다.
그럼에도 불구하고 전도 모드에서는 더 빠른 응고 속도와 함께 마랑고니 효과가 EOPD를 유발합니다. 마지막으로, 낮은 스캐닝 속도에서 높은 체적 에너지 밀도를 갖는 경우에 대해 EOPD 결함을 최적화하기 위한 ‘단계’ 가변 전력 전략이 제안되었습니다.
이 작업은 LPBF 제조의 품질 관리에 대한 심오한 이해와 귀중한 통찰력을 제공합니다.
Figure 5. Simulation of the molten pool under low-speed scanning (1.06 m/s). (a) Sequential solidification of the molten pool at the
end of the melt track for laser powers of 190 and 340 W, respectively. (b) Recoil pressure on the molten pool at the keyhole for laser
powers of 190 and 340 W, respectively. (c) The force diagram of the melt at the back of the keyhole at t = 750 μs in case B. (d) Temperature gradient at the solid–liquid interface of the molten pool at the moment the laser is deactivated in case A. (e) Temperature
gradient at the solid–liquid interface of the molten pool at the moment the laser is deactivated in case B.
References
[1] Zhang C, Li Z, Zhang J, et al. Additive manufacturing of magnesium matrix composites: comprehensive review of recent progress and research perspectives. J Mag Alloys. 2023. doi:10.1016/j.jma.2023.02.005 [2] Webster S, Lin H, Carter III FM, et al. Physical mechanisms in hybrid additive manufacturing: a process design framework. J Mater Process Technol. 2022;291:117048. doi:10. 1016/j.jmatprotec.2021.117048 [3] Wang S, Ning J, Zhu L, et al. Role of porosity defects in metal 3D printing: formation mechanisms, impacts on properties and mitigation strategies. Mater Today. 2022. doi:10.1016/j.mattod.2022.08.014 [4] Wei C, Li L. Recent progress and scientific challenges in multi-material additive manufacturing via laser-based powder bed fusion. Virtual Phys Prototyp. 2021;16 (3):347–371. doi:10.1080/17452759.2021.1928520 [5] Lin X, Wang Q, Fuh JYH, et al. Motion feature based melt pool monitoring for selective laser melting process. J Mater Process Technol. 2022;303:117523. doi:10.1016/j. jmatprotec.2022.117523 [6] Gockel J, Sheridan L, Koerper B, et al. The influence of additive manufacturing processing parameters on surface roughness and fatigue life. Int J Fatigue. 2019;124:380–388. doi:10.1016/j.ijfatigue.2019.03.025 [7] Nicoletto G. Influence of rough as-built surfaces on smooth and notched fatigue behavior of L-PBF AlSi10Mg. Addit Manuf. 2020;34:101251. doi:10.1016/j. addma.2020.101251 [8] Spece H, Yu T, Law AW, et al. 3D printed porous PEEK created via fused filament fabrication for osteoconductive orthopaedic surfaces. J Mech Behav Biomed Mater. 2020;109:103850. doi:10.1115/1.0004270v [9] Andrukhov O, Huber R, Shi B, et al. Proliferation, behavior, and differentiation of osteoblasts on surfaces of different microroughness. Dent Mater. 2016;32(11):1374–1384. doi:10.1016/j.dental.2016.08.217 [10] Dai N, Zhang LC, Zhang J, et al. Corrosion behavior of selective laser melted Ti-6Al-4 V alloy in NaCl solution. Corros Sci. 2016;102:484–489. doi:10.1016/j.corsci.2015. 10.041 [11] Li EL, Wang L, Yu AB, et al. A three-phase model for simulation of heat transfer and melt pool behaviour in laser powder bed fusion process. Powder Technol. 2021;381:298–312. doi:10.1016/j.powtec.2020.11.061 [12] Liao B, Xia RF, Li W, et al. 3D-printed ti6al4v scaffolds with graded triply periodic minimal surface structure for bone tissue engineering. J Mater Eng Perform. 2021;30:4993– 5004. doi:10.1007/s11665-021-05580-z [13] Li E, Zhou Z, Wang L, et al. Melt pool dynamics and pores formation in multi-track studies in laser powder bed fusion process. Powder Technol. 2022;405:117533. doi:10.1016/j.powtec.2022.117533 [14] Guo L, Geng S, Gao X, et al. Numerical simulation of heat transfer and fluid flow during nanosecond pulsed laser processing of Fe78Si9B13 amorphous alloys. Int J Heat Mass Transfer. 2021;170:121003. doi:10.1016/j.ijheatma sstransfer.2021.121003 [15] Guo L, Li Y, Geng S, et al. Numerical and experimental analysis for morphology evolution of 6061 aluminum alloy during nanosecond pulsed laser cleaning. Surf Coat Technol. 2022;432:128056. doi:10.1016/j.surfcoat. 2021.128056 [16] Li S, Liu D, Mi H, et al. Numerical simulation on evolution process of molten pool and solidification characteristics of melt track in selective laser melting of ceramic powder. Ceram Int. 2022;48(13):18302–18315. doi:10. 1016/j.ceramint.2022.03.089 [17] Aboulkhair NT, Maskery I, Tuck C, et al. On the formation of AlSi10Mg single tracks and layers in selective laser melting: microstructure and nano-mechanical properties. J Mater Process Technol. 2016;230:88–98. doi:10.1016/j. jmatprotec.2015.11.016 [18] Thijs L, Kempen K, Kruth JP, et al. Fine-structured aluminium products with controllable texture by selective laser melting of pre-alloyed AlSi10Mg powder. Acta Mater. 2013;61(5):1809–1819. doi:10.1016/j.actamat.2012.11.052 [19] Qiu C, Adkins NJE, Attallah MM. Microstructure and tensile properties of selectively laser-melted and of HIPed laser-melted Ti–6Al–4 V. Mater Sci Eng A. 2013;578:230–239. doi:10.1016/j.msea.2013.04.099 [20] Kazemi Z, Soleimani M, Rokhgireh H, et al. Melting pool simulation of 316L samples manufactured by selective laser melting method, comparison with experimental results. Int J Therm Sci. 2022;176:107538. doi:10.1016/j. ijthermalsci.2022.107538 [21] Cao L. Workpiece-scale numerical simulations of SLM molten pool dynamic behavior of 316L stainless steel. Comput Math Appl. 2021;96:209–228. doi:10.1016/j. camwa.2020.04.020 [22] Liu B, Fang G, Lei L, et al. Predicting the porosity defects in selective laser melting (SLM) by molten pool geometry. Int J Mech Sci. 2022;228:107478. doi:10.1016/j.ijmecsci. 2022.107478 [23] Ur Rehman A, Pitir F, Salamci MU. Full-field mapping and flow quantification of melt pool dynamics in laser powder bed fusion of SS316L. Materials. 2021;14(21):6264. doi:10. 3390/ma14216264 [24] Chia HY, Wang L, Yan W. Influence of oxygen content on melt pool dynamics in metal additive manufacturing: high-fidelity modeling with experimental validation. Acta Mater. 2023;249:118824. doi:10.1016/j.actamat. 2023.118824 [25] Cheng B, Loeber L, Willeck H, et al. Computational investigation of melt pool process dynamics and pore formation in laser powder bed fusion. J Mater Eng Perform. 2019;28:6565–6578. doi:10.1007/s11665-019- 04435-y [26] Li X, Guo Q, Chen L, et al. Quantitative investigation of gas flow, powder-gas interaction, and powder behavior under different ambient pressure levels in laser powder bed fusion. Int J Mach Tools Manuf. 2021;170:103797. doi:10.1016/j.ijmachtools.2021.103797 [27] Wu Y, Li M, Wang J, et al. Powder-bed-fusion additive manufacturing of molybdenum: process simulation, optimization, and property prediction. Addit Manuf. 2022;58:103069. doi:10.1016/j.addma.2022.103069 [28] Wu S, Yang Y, Huang Y, et al. Study on powder particle behavior in powder spreading with discrete element method and its critical implications for binder jetting additive manufacturing processes. Virtual Phys Prototyp. 2023;18(1):e2158877. doi:10.1080/17452759.2022.2158877 [29] Klassen A, Schakowsky T, Kerner C. Evaporation model for beam based additive manufacturing using free surface lattice Boltzmann methods. J Phys D Appl Phys. 2014;47 (27):275303. doi:10.1088/0022-3727/47/27/275303 [30] Cao L. Mesoscopic-scale numerical simulation including the influence of process parameters on slm single-layer multi-pass formation. Metall Mater Trans A. 2020;51:4130–4145. doi:10.1007/s11661-020-05831-z [31] Zhuang JR, Lee YT, Hsieh WH, et al. Determination of melt pool dimensions using DOE-FEM and RSM with process window during SLM of Ti6Al4V powder. Opt Laser Technol. 2018;103:59–76. doi:10.1016/j.optlastec.2018. 01.013 [32] Li Y, Gu D. Thermal behavior during selective laser melting of commercially pure titanium powder: numerical simulation and experimental study. Addit Manuf. 2014;1–4:99–109. doi:10.1016/j.addma.2014.09.001 [33] Dai D, Gu D. Thermal behavior and densification mechanism during selective laser melting of copper matrix composites: simulation and experiments. Mater Des. 2014;55 (0):482–491. doi:10.1016/j.matdes.2013.10.006 [34] Wang S, Zhu L, Dun Y, et al. Multi-physics modeling of direct energy deposition process of thin-walled structures: defect analysis. Comput Mech. 2021;67:c1229– c1242. doi:10.1007/s00466-021-01992-9 [35] Wu J, Zheng J, Zhou H, et al. Molten pool behavior and its mechanism during selective laser melting of polyamide 6 powder: single track simulation and experiments. Mater Res Express. 2019;6. doi:10.1088/2053-1591/ab2747 [36] Cho JH, Farson DF, Milewski JO, et al. Weld pool flows during initial stages of keyhole formation in laser welding. J Phys D Appl Phys. 2009;42. doi:10.1088/0022- 3727/42/17/175502 [37] Sinha KN. Identification of a suitable volumetric heat source for modelling of selective laser melting of Ti6Al4V powder using numerical and experimental validation approach. Int J Adv Manuf Technol. 2018;99:2257–2270. doi:10.1007/s00170-018-2631-4 [38] Fu CH, Guo YB. Three-dimensional temperature gradient mechanism in selective laser melting of Ti-6Al-4V. J Manuf Sci Eng. 2014;136(6):061004. doi:10.1115/1.4028539 [39] Ansari P, Rehman AU, Pitir F, et al. Selective laser melting of 316 l austenitic stainless steel: detailed process understanding using multiphysics simulation and experimentation. Metals. 2021;11(7):1076. doi:10.3390/met11071076 [40] Zhao C, Shi B, Chen S, et al. Laser melting modes in metal powder bed fusion additive manufacturing. Rev Mod Phys. 2022;94(4):045002. doi:10.1103/revmodphys.94. 045002 [41] Bertoli US, Wolfer AJ, Matthews MJ, et al. On the limitations of volumetric energy density as a design parameter for selective laser melting. Mater Des. 2017;113:331–340. doi:10.1016/j.matdes.2016.10.037 [42] Dash A, Kamaraj A. Prediction of the shift in melting mode during additive manufacturing of 316L stainless steel. Mater Today Commun. 2023: 107238. doi:10.1016/j. mtcomm.2023.107238 [43] Majeed M, Khan HM, Rasheed I. Finite element analysis of melt pool thermal characteristics with passing laser in SLM process. Optik. 2019;194:163068. doi:10.1016/j.ijleo. 2019.163068
In order to comprehensively reveal the evolutionary dynamics of the molten pool and the state of motion of the fluid during the high-precision laser powder bed fusion (HP-LPBF) process, this study aims to deeply investigate the specific manifestations of the multiphase flow, solidification phenomena, and heat transfer during the process by means of numerical simulation methods. Numerical simulation models of SS316L single-layer HP-LPBF formation with single and double tracks were constructed using the discrete element method and the computational fluid dynamics method. The effects of various factors such as Marangoni convection, surface tension, vapor recoil, gravity, thermal convection, thermal radiation, and evaporative heat dissipation on the heat and mass transfer in the molten pool have been paid attention to during the model construction process. The results show that the molten pool exhibits a “comet” shape, in which the temperature gradient at the front end of the pool is significantly larger than that at the tail end, with the highest temperature gradient up to 1.69 × 108 K/s. It is also found that the depth of the second track is larger than that of the first one, and the process parameter window has been determined preliminarily. In addition, the application of HP-LPBF technology helps to reduce the surface roughness and minimize the forming size.
Laser powder bed fusion (LPBF) has become a research hotspot in the field of additive manufacturing of metals due to its advantages of high-dimensional accuracy, good surface quality, high density, and high material utilization.1,2 With the rapid development of electronics, medical, automotive, biotechnology, energy, communication, and optics, the demand for microfabrication technology is increasing day by day.3 High-precision laser powder bed fusion (HP-LPBF) is one of the key manufacturing technologies for tiny parts in the fields of electronics, medical, automotive, biotechnology, energy, communication, and optics because of its process characteristics such as small focal spot diameter, small powder particle size, and thin powder layup layer thickness.4–13 Compared with LPBF, HP-LPBF has the significant advantages of smaller focal spot diameter, smaller powder particle size, and thinner layer thickness. These advantages make HP-LPBF perform better in producing micro-fine parts, high surface quality, and parts with excellent mechanical properties.
HP-LPBF is in the exploratory stage, and researchers have already done some exploratory studies on the focal spot diameter, the amount of defocusing, and the powder particle size. In order to explore the influence of changing the laser focal spot diameter on the LPBF process characteristics of the law, Wildman et al.14 studied five groups of different focal spot diameter LPBF forming 316L stainless steel (SS316L) processing effect, the smallest focal spot diameter of 26 μm, and the results confirm that changing the focal spot diameter can be achieved to achieve the energy control, so as to control the quality of forming. Subsequently, Mclouth et al.15 proposed the laser out-of-focus amount (focal spot diameter) parameter, which characterizes the distance between the forming plane and the laser focal plane. The laser energy density was controlled by varying the defocusing amount while keeping the laser parameters constant. Sample preparation at different focal positions was investigated, and their microstructures were characterized. The results show that the samples at the focal plane have finer microstructure than those away from the focal plane, which is the effect of higher power density and smaller focal spot diameter. In order to explore the influence of changing the powder particle size on the characteristics of the LPBF process, Qian et al.16 carried out single-track scanning simulations on powder beds with average powder particle sizes of 70 and 40 μm, respectively, and the results showed that the melt tracks sizes were close to each other under the same process parameters for the two particle-size distributions and that the molten pool of powder beds with small particles was more elongated and the edges of the melt tracks were relatively flat. In order to explore the superiority of HP-LPBF technology, Xu et al.17 conducted a comparative analysis of HP-LPBF and conventional LPBF of SS316L. The results showed that the average surface roughness of the top surface after forming by HP-LPBF could reach 3.40 μm. Once again, it was verified that HP-LPBF had higher forming quality than conventional LPBF. On this basis, Wei et al.6 comparatively analyzed the effects of different laser focal spot diameters on different powder particle sizes formed by LPBF. The results showed that the smaller the laser focal spot diameter, the fewer the defects on the top and side surfaces. The above research results confirm that reducing the laser focal spot diameter can obtain higher energy density and thus better forming quality.
LPBF involves a variety of complex systems and mechanisms, and the final quality of the part is influenced by a large number of process parameters.18–24 Some research results have shown that there are more than 50 factors affecting the quality of the specimen. The influencing factors are mainly categorized into three main groups: (1) laser parameters, (2) powder parameters, and (3) equipment parameters, which interact with each other to determine the final specimen quality. With the continuous development of technologies such as computational materials science and computational fluid dynamics (CFD), the method of studying the influence of different factors on the forming quality of LPBF forming process has been shifted from time-consuming and laborious experimental characterization to the use of numerical simulation methods. As a result, more and more researchers are adopting this approach for their studies. Currently, numerical simulation studies on LPBF are mainly focused on the exploration of molten pool, temperature distribution, and residual stresses.
Finite element simulation based on continuum mechanics and free surface fluid flow modeling based on fluid dynamics are two common approaches to study the behavior of LPBF molten pool.25–28 Finite element simulation focuses on the temperature and thermal stress fields, treats the powder bed as a continuum, and determines the molten pool size by plotting the elemental temperature above the melting point. In contrast, fluid dynamics modeling can simulate the 2D or 3D morphology of the metal powder pile and obtain the powder size and distribution by certain algorithms.29 The flow in the molten pool is mainly affected by recoil pressure and the Marangoni effect. By simulating the molten pool formation, it is possible to predict defects, molten pool shape, and flow characteristics, as well as the effect of process parameters on the molten pool geometry.30–34 In addition, other researchers have been conducted to optimize the laser processing parameters through different simulation methods and experimental data.35–46 Crystal growth during solidification is studied to further understand the effect of laser parameters on dendritic morphology and solute segregation.47–54 A multi-scale system has been developed to describe the fused deposition process during 3D printing, which is combined with the conductive heat transfer model and the dendritic solidification model.55,56
Relevant scholars have adopted various different methods for simulation, such as sequential coupling theory,57 Lagrangian and Eulerian thermal models,58 birth–death element method,25 and finite element method,59 in order to reveal the physical phenomena of the laser melting process and optimize the process parameters. Luo et al.60 compared the LPBF temperature field and molten pool under double ellipsoidal and Gaussian heat sources by ANSYS APDL and found that the diffusion of the laser energy in the powder significantly affects the molten pool size and the temperature field.
The thermal stresses obtained from the simulation correlate with the actual cracks,61 and local preheating can effectively reduce the residual stresses.62 A three-dimensional thermodynamic finite element model investigated the temperature and stress variations during laser-assisted fabrication and found that powder-to-solid conversion increases the temperature gradient, stresses, and warpage.63 Other scholars have predicted residual stresses and part deflection for LPBF specimens and investigated the effects of deposition pattern, heat, laser power, and scanning strategy on residual stresses, noting that high-temperature gradients lead to higher residual stresses.64–67
In short, the process of LPBF forming SS316L is extremely complex and usually involves drastic multi-scale physicochemical changes that will only take place on a very small scale. Existing literature employs DEM-based mesoscopic-scale numerical simulations to investigate the effects of process parameters on the molten pool dynamics of LPBF-formed SS316L. However, a few studies have been reported on the key mechanisms of heating and solidification, spatter, and convective behavior of the molten pool of HP-LPBF-formed SS316L with small laser focal spot diameters. In this paper, the geometrical properties of coarse and fine powder particles under three-dimensional conditions were first calculated using DEM. Then, numerical simulation models for single-track and double-track cases in the single-layer HP-LPBF forming SS316L process were developed at mesoscopic scale using the CFD method. The flow genesis of the melt in the single-track and double-track molten pools is discussed, and their 3D morphology and dimensional characteristics are discussed. In addition, the effects of laser process parameters, powder particle size, and laser focal spot diameter on the temperature field, characterization information, and defects in the molten pool are discussed.
II. MODELING
A. 3D powder bed modeling
HP-LPBF is an advanced processing technique for preparing target parts layer by layer stacking, the process of which involves repetitive spreading and melting of powders. In this process, both the powder spreading and the morphology of the powder bed are closely related to the results of the subsequent melting process, while the melted surface also affects the uniform distribution of the next layer of powder. For this reason, this chapter focuses on the modeling of the physical action during the powder spreading process and the theory of DEM to establish the numerical model of the powder bed, so as to lay a solid foundation for the accuracy of volume of fluid (VOF) and CFD.
1. DEM
DEM is a numerical technique for calculating the interaction of a large number of particles, which calculates the forces and motions of the spheres by considering each powder sphere as an independent unit. The motion of the powder particles follows the laws of classical Newtonian mechanics, including translational and rotational,38,68–70 which are expressed as follows:����¨=���+∑��ij,
(1)����¨=∑�(�ij×�ij),
(2)
where �� is the mass of unit particle i in kg, ��¨ is the advective acceleration in m/s2, And g is the gravitational acceleration in m/s2. �ij is the force in contact with the neighboring particle � in N. �� is the rotational inertia of the unit particle � in kg · m2. ��¨ is the unit particle � angular acceleration in rad/s2. �ij is the vector pointing from unit particle � to the contact point of neighboring particle �.
Equations (1) and (2) can be used to calculate the velocity and angular velocity variations of powder particles to determine their positions and velocities. A three-dimensional powder bed model of SS316L was developed using DEM. The powder particles are assumed to be perfect spheres, and the substrate and walls are assumed to be rigid. To describe the contact between the powder particles and between the particles and the substrate, a non-slip Hertz–Mindlin nonlinear spring-damping model71 was used with the following expression:�hz=��������+��[(�����ij−�eff����)−(�����+�eff����)],
(3)
where �hz is the force calculated using the Hertzian in M. �� and �� are the radius of unit particles � and � in m, respectively. �� is the overlap size of the two powder particles in m. ��, �� are the elastic constants in the normal and tangential directions, respectively. �ij is the unit vector connecting the centerlines of the two powder particles. �eff is the effective mass of the two powder particles in kg. �� and �� are the viscoelastic damping constants in the normal and tangential directions, respectively. �� and �� are the components of the relative velocities of the two powder particles. ��� is the displacement vector between two spherical particles. The schematic diagram of overlapping powder particles is shown in Fig. 1.
Schematic diagram of overlapping powder particles.
Because the particle size of the powder used for HP-LPBF is much smaller than 100 μm, the effect of van der Waals forces must be considered. Therefore, the cohesive force �jkr of the Hertz–Mindlin model was used instead of van der Waals forces,72 with the following expression:�jkr=−4��0�*�1.5+4�*3�*�3,
(4)1�*=(1−��2)��+(1−��2)��,
(5)1�*=1��+1��,
(6)
where �* is the equivalent Young’s modulus in GPa; �* is the equivalent particle radius in m; �0 is the surface energy of the powder particles in J/m2; α is the contact radius in m; �� and �� are the Young’s modulus of the unit particles � and �, respectively, in GPa; and �� and �� are the Poisson’s ratio of the unit particles � and �, respectively.
2. Model building
Figure 2 shows a 3D powder bed model generated using DEM with a coarse powder geometry of 1000 × 400 × 30 μm3. The powder layer thickness is 30 μm, and the powder bed porosity is 40%. The average particle size of this spherical powder is 31.7 μm and is normally distributed in the range of 15–53 μm. The geometry of the fine powder was 1000 × 400 × 20 μm3, with a layer thickness of 20 μm, and the powder bed porosity of 40%. The average particle size of this spherical powder is 11.5 μm and is normally distributed in the range of 5–25 μm. After the 3D powder bed model is generated, it needs to be imported into the CFD simulation software for calculation, and the imported geometric model is shown in Fig. 3. This geometric model is mainly composed of three parts: protective gas, powder bed, and substrate. Under the premise of ensuring the accuracy of the calculation, the mesh size is set to 3 μm, and the total number of coarse powder meshes is 1 704 940. The total number of fine powder meshes is 3 982 250.
Geometric modeling of the powder bed computational domain: (a) coarse powder, (b) fine powder.
B. Modeling of fluid mechanics simulation
In order to solve the flow, melting, and solidification problems involved in HP-LPBF molten pool, the study must follow the three governing equations of conservation of mass, conservation of energy, and conservation of momentum.73 The VOF method, which is the most widely used in fluid dynamics, is used to solve the molten pool dynamics model.
1. VOF
VOF is a method for tracking the free interface between the gas and liquid phases on the molten pool surface. The core idea of the method is to define a volume fraction function F within each grid, indicating the proportion of the grid space occupied by the material, 0 ≤ F ≤ 1 in Fig. 4. Specifically, when F = 0, the grid is empty and belongs to the gas-phase region; when F = 1, the grid is completely filled with material and belongs to the liquid-phase region; and when 0 < F < 1, the grid contains free surfaces and belongs to the mixed region. The direction normal to the free surface is the direction of the fastest change in the volume fraction F (the direction of the gradient of the volume fraction), and the direction of the gradient of the volume fraction can be calculated from the values of the volume fractions in the neighboring grids.74 The equations controlling the VOF are expressed as follows:𝛻����+�⋅(��→)=0,
(7)
where t is the time in s and �→ is the liquid velocity in m/s.
The material parameters of the mixing zone are altered due to the inclusion of both the gas and liquid phases. Therefore, in order to represent the density of the mixing zone, the average density �¯ is used, which is expressed as follows:72�¯=(1−�1)�gas+�1�metal,
(8)
where �1 is the proportion of liquid phase, �gas is the density of protective gas in kg/m3, and �metal is the density of metal in kg/m3.
2. Control equations and boundary conditions
Figure 5 is a schematic diagram of the HP-LPBF melting process. First, the laser light strikes a localized area of the material and rapidly heats up the area. Next, the energy absorbed in the region is diffused through a variety of pathways (heat conduction, heat convection, and surface radiation), and this process triggers complex phase transition phenomena (melting, evaporation, and solidification). In metals undergoing melting, the driving forces include surface tension and the Marangoni effect, recoil due to evaporation, and buoyancy due to gravity and uneven density. The above physical phenomena interact with each other and do not occur independently.
Laser heat sourceThe Gaussian surface heat source model is used as the laser heat source model with the following expression:�=2�0����2exp(−2�12��2),(9)where � is the heat flow density in W/m2, �0 is the absorption rate of SS316L, �� is the radius of the laser focal spot in m, and �1 is the radial distance from the center of the laser focal spot in m. The laser focal spot can be used for a wide range of applications.
Energy absorptionThe formula for calculating the laser absorption �0 of SS316L is as follows:�0=0.365(�0[1+�0(�−20)]/�)0.5,(10)where �0 is the direct current resistivity of SS316L at 20 °C in Ω m, �0 is the resistance temperature coefficient in ppm/°C, � is the temperature in °C, and � is the laser wavelength in m.
Heat transferThe basic principle of heat transfer is conservation of energy, which is expressed as follows:𝛻𝛻𝛻�(��)��+�·(��→�)=�·(�0����)+��,(11)where � is the density of liquid phase SS316L in kg/m3, �� is the specific heat capacity of SS316L in J/(kg K), 𝛻� is the gradient operator, t is the time in s, T is the temperature in K, 𝛻�� is the temperature gradient, �→ is the velocity vector, �0 is the coefficient of thermal conduction of SS316L in W/(m K), and �� is the thermal energy dissipation term in the molten pool.
Molten pool flowThe following three conditions need to be satisfied for the molten pool to flow:
Conservation of mass with the following expression:𝛻�·(��→)=0.(12)
Conservation of momentum (Navier–Stokes equation) with the following expression:𝛻𝛻𝛻𝛻���→��+�(�→·�)�→=�·[−pI+�(��→+(��→)�)]+�,(13)where � is the pressure in Pa exerted on the liquid phase SS316L microelement, � is the unit matrix, � is the fluid viscosity in N s/m2, and � is the volumetric force (gravity, atmospheric pressure, surface tension, vapor recoil, and the Marangoni effect).
Surface tension and the Marangoni effectThe effect of temperature on the surface tension coefficient is considered and set as a linear relationship with the following expression:�=�0−��dT(�−��),(14)where � is the surface tension of the molten pool at temperature T in N/m, �� is the melting temperature of SS316L in K, �0 is the surface tension of the molten pool at temperature �� in Pa, and σdσ/ dT is the surface tension temperature coefficient in N/(m K).In general, surface tension decreases with increasing temperature. A temperature gradient causes a gradient in surface tension that drives the liquid to flow, known as the Marangoni effect.
Metal vapor recoilAt higher input energy densities, the maximum temperature of the molten pool surface reaches the evaporation temperature of the material, and a gasification recoil pressure occurs vertically downward toward the molten pool surface, which will be the dominant driving force for the molten pool flow.75 The expression is as follows:��=0.54�� exp ���−���0���,(15)where �� is the gasification recoil pressure in Pa, �� is the ambient pressure in kPa, �� is the latent heat of evaporation in J/kg, �0 is the gas constant in J/(mol K), T is the surface temperature of the molten pool in K, and Te is the evaporation temperature in K.
Solid–liquid–gas phase transitionWhen the laser hits the powder layer, the powder goes through three stages: heating, melting, and solidification. During the solidification phase, mutual transformations between solid, liquid, and gaseous states occur. At this point, the latent heat of phase transition absorbed or released during the phase transition needs to be considered.68 The phase transition is represented based on the relationship between energy and temperature with the following expression:�=�����,(�<��),�(��)+�−����−����,(��<�<��)�(��)+(�−��)����,(��<�),,(16)where �� and �� are solid and liquid phase density, respectively, of SS316L in kg/m3. �� and �� unit volume of solid and liquid phase-specific heat capacity, respectively, of SS316L in J/(kg K). �� and ��, respectively, are the solidification temperature and melting temperature of SS316L in K. �� is the latent heat of the phase transition of SS316L melting in J/kg.
3. Assumptions
The CFD model was computed using the commercial software package FLOW-3D.76 In order to simplify the calculation and solution process while ensuring the accuracy of the results, the model makes the following assumptions:
It is assumed that the effects of thermal stress and material solid-phase thermal expansion on the calculation results are negligible.
The molten pool flow is assumed to be a Newtonian incompressible laminar flow, while the effects of liquid thermal expansion and density on the results are neglected.
It is assumed that the surface tension can be simplified to an equivalent pressure acting on the free surface of the molten pool, and the effect of chemical composition on the results is negligible.
Neglecting the effect of the gas flow field on the molten pool.
The mass loss due to evaporation of the liquid metal is not considered.
The influence of the plasma effect of the molten metal on the calculation results is neglected.
It is worth noting that the formulation of assumptions requires a trade-off between accuracy and computational efficiency. In the above models, some physical phenomena that have a small effect or high difficulty on the calculation results are simplified or ignored. Such simplifications make numerical simulations more efficient and computationally tractable, while still yielding accurate results.
4. Initial conditions
The preheating temperature of the substrate was set to 393 K, at which time all materials were in the solid state and the flow rate was zero.
5. Material parameters
The material used is SS316L and the relevant parameters required for numerical simulations are shown in Table I.46,77,78
TABLE I.
SS316L-related parameters.
Property
Symbol
Value
Density of solid metal (kg/m3)
�metal
7980
Solid phase line temperature (K)
��
1658
Liquid phase line temperature (K)
��
1723
Vaporization temperature (K)
��
3090
Latent heat of melting ( J/kg)
��
2.60×105
Latent heat of evaporation ( J/kg)
��
7.45×106
Surface tension of liquid phase (N /m)
�
1.60
Liquid metal viscosity (kg/m s)
��
6×10−3
Gaseous metal viscosity (kg/m s)
�gas
1.85×10−5
Temperature coefficient of surface tension (N/m K)
��/�T
0.80×10−3
Molar mass ( kg/mol)
M
0.05 593
Emissivity
�
0.26
Laser absorption
�0
0.35
Ambient pressure (kPa)
��
101 325
Ambient temperature (K)
�0
300
Stefan–Boltzmann constant (W/m2 K4)
�
5.67×10−8
Thermal conductivity of metals ( W/m K)
�
24.55
Density of protective gas (kg/m3)
�gas
1.25
Coefficient of thermal expansion (/K)
��
16×10−6
Generalized gas constant ( J/mol K)
R
8.314
III. RESULTS AND DISCUSSION
With the objective of studying in depth the evolutionary patterns of single-track and double-track molten pool development, detailed observations were made for certain specific locations in the model, as shown in Fig. 6. In this figure, P1 and P2 represent the longitudinal tangents to the centers of the two melt tracks in the XZ plane, while L1 is the transverse profile in the YZ plane. The scanning direction is positive and negative along the X axis. Points A and B are the locations of the centers of the molten pool of the first and second melt tracks, respectively (x = 1.995 × 10−4, y = 5 × 10−7, and z = −4.85 × 10−5).
A series of single-track molten pool simulation experiments were carried out in order to investigate the influence law of laser power as well as scanning speed on the HP-LPBF process. Figure 7 demonstrates the evolution of the 3D morphology and temperature field of the single-track molten pool in the time period of 50–500 μs under a laser power of 100 W and a scanning speed of 800 mm/s. The powder bed is in the natural cooling state. When t = 50 μs, the powder is heated by the laser heat and rapidly melts and settles to form the initial molten pool. This process is accompanied by partial melting of the substrate and solidification together with the melted powder. The molten pool rapidly expands with increasing width, depth, length, and temperature, as shown in Fig. 7(a). When t = 150 μs, the molten pool expands more obviously, and the temperature starts to transfer to the surrounding area, forming a heat-affected zone. At this point, the width of the molten pool tends to stabilize, and the temperature in the center of the molten pool has reached its peak and remains largely stable. However, the phenomenon of molten pool spatter was also observed in this process, as shown in Fig. 7(b). As time advances, when t = 300 μs, solidification begins to occur at the tail of the molten pool, and tiny ripples are produced on the solidified surface. This is due to the fact that the melt flows toward the region with large temperature gradient under the influence of Marangoni convection and solidifies together with the melt at the end of the bath. At this point, the temperature gradient at the front of the bath is significantly larger than at the end. While the width of the molten pool was gradually reduced, the shape of the molten pool was gradually changed to a “comet” shape. In addition, a slight depression was observed at the top of the bath because the peak temperature at the surface of the bath reached the evaporation temperature, which resulted in a recoil pressure perpendicular to the surface of the bath downward, creating a depressed region. As the laser focal spot moves and is paired with the Marangoni convection of the melt, these recessed areas will be filled in as shown in Fig. 7(c). It has been shown that the depressed regions are the result of the coupled effect of Marangoni convection, recoil pressure, and surface tension.79 By t = 500 μs, the width and height of the molten pool stabilize and show a “comet” shape in Fig. 7(d).
Single-track molten pool process: (a) t = 50 ��, (b) t = 150 ��, (c) t = 300 ��, (d) t = 500 ��.
Figure 8 depicts the velocity vector diagram of the P1 profile in a single-track molten pool, the length of the arrows represents the magnitude of the velocity, and the maximum velocity is about 2.36 m/s. When t = 50 μs, the molten pool takes shape, and the velocities at the two ends of the pool are the largest. The variation of the velocities at the front end is especially more significant in Fig. 8(a). As the time advances to t = 150 μs, the molten pool expands rapidly, in which the velocity at the tail increases and changes more significantly, while the velocity at the front is relatively small. At this stage, the melt moves backward from the center of the molten pool, which in turn expands the molten pool area. The melt at the back end of the molten pool center flows backward along the edge of the molten pool surface and then converges along the edge of the molten pool to the bottom center, rising to form a closed loop. Similarly, a similar closed loop is formed at the front end of the center of the bath, but with a shorter path. However, a large portion of the melt in the center of the closed loop formed at the front end of the bath is in a nearly stationary state. The main cause of this melt flow phenomenon is the effect of temperature gradient and surface tension (the Marangoni effect), as shown in Figs. 8(b) and 8(e). This dynamic behavior of the melt tends to form an “elliptical” pool. At t = 300 μs, the tendency of the above two melt flows to close the loop is more prominent and faster in Fig. 8(c). When t = 500 μs, the velocity vector of the molten pool shows a stable trend, and the closed loop of melt flow also remains stable. With the gradual laser focal spot movement, the melt is gradually solidified at its tail, and finally, a continuous and stable single track is formed in Fig. 8(d).
Vector plot of single-track molten pool velocity in XZ longitudinal section: (a) t = 50 ��, (b) t = 150 ��, (c) t = 300 ��, (d) t = 500 ��, (e) molten pool flow.
In order to explore in depth the transient evolution of the molten pool, the evolution of the single-track temperature field and the melt flow was monitored in the YZ cross section. Figure 9(a) shows the state of the powder bed at the initial moment. When t = 250 μs, the laser focal spot acts on the powder bed and the powder starts to melt and gradually collects in the molten pool. At this time, the substrate will also start to melt, and the melt flow mainly moves in the downward and outward directions and the velocity is maximum at the edges in Fig. 9(b). When t = 300 μs, the width and depth of the molten pool increase due to the recoil pressure. At this time, the melt flows more slowly at the center, but the direction of motion is still downward in Fig. 9(c). When t = 350 μs, the width and depth of the molten pool further increase, at which time the intensity of the melt flow reaches its peak and the direction of motion remains the same in Fig. 9(d). When t = 400 μs, the melt starts to move upward, and the surrounding powder or molten material gradually fills up, causing the surface of the molten pool to begin to flatten. At this time, the maximum velocity of the melt is at the center of the bath, while the velocity at the edge is close to zero, and the edge of the melt starts to solidify in Fig. 9(e). When t = 450 μs, the melt continues to move upward, forming a convex surface of the melt track. However, the melt movement slows down, as shown in Fig. 9(f). When t = 500 μs, the melt further moves upward and its speed gradually becomes smaller. At the same time, the melt solidifies further, as shown in Fig. 9(g). When t = 550 μs, the melt track is basically formed into a single track with a similar “mountain” shape. At this stage, the velocity is close to zero only at the center of the molten pool, and the flow behavior of the melt is poor in Fig. 9(h). At t = 600 μs, the melt stops moving and solidification is rapidly completed. Up to this point, a single track is formed in Fig. 9(i). During the laser action on the powder bed, the substrate melts and combines with the molten state powder. The powder-to-powder fusion is like the convergence of water droplets, which are rapidly fused by surface tension. However, the fusion between the molten state powder and the substrate occurs driven by surface tension, and the molten powder around the molten pool is pulled toward the substrate (a wetting effect occurs), which ultimately results in the formation of a monolithic whole.38,80,81
Evolution of single-track molten pool temperature and melt flow in the YZ cross section: (a) t = 0 ��, (b) t = 250 ��, (c) t = 300 ��, (d) t = 350 ��, (e) t = 400 ��, (f) t = 450 ��, (g) t = 500 ��, (h) t = 550 ��, (i) t = 600 ��.
The wetting ability between the liquid metal and the solid substrate in the molten pool directly affects the degree of balling of the melt,82,83 and the wetting ability can be measured by the contact angle of a single track in Fig. 10. A smaller value of contact angle represents better wettability. The contact angle α can be calculated by�=�1−�22,
(17)
where �1 and �2 are the contact angles of the left and right regions, respectively.
Relevant studies have confirmed that the wettability is better at a contact angle α around or below 40°.84 After measurement, a single-track contact angle α of about 33° was obtained under this process parameter, which further confirms the good wettability.
B. Double-track simulation
In order to deeply investigate the influence of hatch spacing on the characteristics of the HP-LPBF process, a series of double-track molten pool simulation experiments were systematically carried out. Figure 11 shows in detail the dynamic changes of the 3D morphology and temperature field of the double-track molten pool in the time period of 2050–2500 μs under the conditions of laser power of 100 W, scanning speed of 800 mm/s, and hatch spacing of 0.06 mm. By comparing the study with Fig. 7, it is observed that the basic characteristics of the 3D morphology and temperature field of the second track are similar to those of the first track. However, there are subtle differences between them. The first track exhibits a basically symmetric shape, but the second track morphology shows a slight deviation influenced by the difference in thermal diffusion rate between the solidified metal and the powder. Otherwise, the other characteristic information is almost the same as that of the first track. Figure 12 shows the velocity vector plot of the P2 profile in the double-track molten pool, with a maximum velocity of about 2.63 m/s. The melt dynamics at both ends of the pool are more stable at t = 2050 μs, where the maximum rate of the second track is only 1/3 of that of the first one. Other than that, the rest of the information is almost no significant difference from the characteristic information of the first track. Figure 13 demonstrates a detailed observation of the double-track temperature field and melts flow in the YZ cross section, and a comparative study with Fig. 9 reveals that the width of the second track is slightly wider. In addition, after the melt direction shifts from bottom to top, the first track undergoes four time periods (50 μs) to reach full solidification, while the second track takes five time periods. This is due to the presence of significant heat buildup in the powder bed after the forming of the first track, resulting in a longer dynamic time of the melt and an increased molten pool lifetime. In conclusion, the level of specimen forming can be significantly optimized by adjusting the laser power and hatch spacing.
Evolution of double-track molten pool temperature and melt flow in the YZ cross section: (a) t = 2250 ��, (b) t = 2300 ��, (c) t = 2350 ��, (d) t = 2400 ��, (e) t = 2450 ��, (f) t = 2500 ��, (g) t = 2550 ��, (h) t = 2600 ��, (i) t = 2650 ��.
In order to quantitatively detect the molten pool dimensions as well as the remolten region dimensions, the molten pool characterization information in Fig. 14 is constructed by drawing the boundary on the YZ cross section based on the isothermal surface of the liquid phase line. It can be observed that the heights of the first track and second track are basically the same, but the depth of the second track increases relative to the first track. The molten pool width is mainly positively correlated with the laser power as well as the scanning speed (the laser line energy density �). However, the remelted zone width is negatively correlated with the hatch spacing (the overlapping ratio). Overall, the forming quality of the specimens can be directly influenced by adjusting the laser power, scanning speed, and hatch spacing.
Double-track molten pool characterization information on YZ cross section.
In order to study the variation rule of the temperature in the center of the molten pool with time, Fig. 15 demonstrates the temperature variation curves with time for two reference points, A and B. Among them, the red dotted line indicates the liquid phase line temperature of SS316L. From the figure, it can be seen that the maximum temperature at the center of the molten pool in the first track is lower than that in the second track, which is mainly due to the heat accumulation generated after passing through the first track. The maximum temperature gradient was calculated to be 1.69 × 108 K/s. When the laser scanned the first track, the temperature in the center of the molten pool of the second track increased slightly. Similarly, when the laser scanned the second track, a similar situation existed in the first track. Since the temperature gradient in the second track is larger than that in the first track, the residence time of the liquid phase in the molten pool of the first track is longer than that of the second track.
Temperature profiles as a function of time for two reference points A and B.
C. Simulation analysis of molten pool under different process parameters
In order to deeply investigate the effects of various process parameters on the mesoscopic-scale temperature field, molten pool characteristic information and defects of HP-LPBF, numerical simulation experiments on mesoscopic-scale laser power, scanning speed, and hatch spacing of double-track molten pools were carried out.
1. Laser power
Figure 16 shows the effects of different laser power on the morphology and temperature field of the double-track molten pool at a scanning speed of 800 mm/s and a hatch spacing of 0.06 mm. When P = 50 W, a smaller molten pool is formed due to the lower heat generated by the Gaussian light source per unit time. This leads to a smaller track width, which results in adjacent track not lapping properly and the presence of a large number of unmelted powder particles, resulting in an increase in the number of defects, such as pores in the specimen. The surface of the track is relatively flat, and the depth is small. In addition, the temperature gradient before and after the molten pool was large, and the depression location appeared at the biased front end in Fig. 16(a). When P = 100 W, the surface of the track is flat and smooth with excellent lap. Due to the Marangoni effect, the velocity field of the molten pool is in the form of “vortex,” and the melt has good fluidity, and the maximum velocity reaches 2.15 m/s in Fig. 16(b). When P = 200 W, the heat generated by the Gaussian light source per unit time is too large, resulting in the melt rapidly reaching the evaporation temperature, generating a huge recoil pressure, forming a large molten pool, and the surface of the track is obviously raised. The melt movement is intense, especially the closed loop at the center end of the molten pool. At this time, the depth and width of the molten pool are large, leading to the expansion of the remolten region and the increased chance of the appearance of porosity defects in Fig. 16(c). The results show that at low laser power, the surface tension in the molten pool is dominant. At high laser power, recoil pressure is its main role.
Simulation results of double-track molten pool under different laser powers: (a) P = 50 W, (b) P = 100 W, (c) P = 200 W.
Table II shows the effect of different laser powers on the characteristic information of the double-track molten pool at a scanning speed of 800 mm/s and a hatch spacing of 0.06 mm. The negative overlapping ratio in the table indicates that the melt tracks are not lapped, and 26/29 indicates the melt depth of the first track/second track. It can be seen that with the increase in laser power, the melt depth, melt width, melt height, and remelted zone show a gradual increase. At the same time, the overlapping ratio also increases. Especially in the process of laser power from 50 to 200 W, the melting depth and melting width increased the most, which increased nearly 2 and 1.5 times, respectively. Meanwhile, the overlapping ratio also increases with the increase in laser power, which indicates that the melting and fusion of materials are better at high laser power. On the other hand, the dimensions of the molten pool did not change uniformly with the change of laser power. Specifically, the depth-to-width ratio of the molten pool increased from about 0.30 to 0.39 during the increase from 50 to 120 W, which further indicates that the effective heat transfer in the vertical direction is greater than that in the horizontal direction with the increase in laser power. This dimensional response to laser power is mainly affected by the recoil pressure and also by the difference in the densification degree between the powder layer and the metal substrate. In addition, according to the experimental results, the contact angle shows a tendency to increase and then decrease during the process of laser power increase, and always stays within the range of less than 33°. Therefore, in practical applications, it is necessary to select the appropriate laser power according to the specific needs in order to achieve the best processing results.
TABLE II.
Double-track molten pool characterization information at different laser powers.
Laser power (W)
Depth (μm)
Width (μm)
Height (μm)
Remolten region (μm)
Overlapping ratio (%)
Contact angle (°)
50
16
54
11
/
−10
23
100
26/29
74
14
18
23.33
33
200
37/45
116
21
52
93.33
28
2. Scanning speed
Figure 17 demonstrates the effect of different scanning speeds on the morphology and temperature field of the double-track molten pool at a laser power of 100 W and a hatch spacing of 0.06 mm. With the gradual increase in scanning speed, the surface morphology of the molten pool evolves from circular to elliptical. When � = 200 mm/s, the slow scanning speed causes the material to absorb too much heat, which is very easy to trigger the overburning phenomenon. At this point, the molten pool is larger and the surface morphology is uneven. This situation is consistent with the previously discussed scenario with high laser power in Fig. 17(a). However, when � = 1600 mm/s, the scanning speed is too fast, resulting in the material not being able to absorb sufficient heat, which triggers the powder particles that fail to melt completely to have a direct effect on the bonding of the melt to the substrate. At this time, the molten pool volume is relatively small and the neighboring melt track cannot lap properly. This result is consistent with the previously discussed case of low laser power in Fig. 17(b). Overall, the ratio of the laser power to the scanning speed (the line energy density �) has a direct effect on the temperature field and surface morphology of the molten pool.
Simulation results of double-track molten pool under different scanning speed: (a) � = 200 mm/s, (b) � = 1600 mm/s.
Table III shows the effects of different scanning speed on the characteristic information of the double-track molten pool under the condition of laser power of 100 W and hatch spacing of 0.06 mm. It can be seen that the scanning speed has a significant effect on the melt depth, melt width, melt height, remolten region, and overlapping ratio. With the increase in scanning speed, the melt depth, melt width, melt height, remelted zone, and overlapping ratio show a gradual decreasing trend. Among them, the melt depth and melt width decreased faster, while the melt height and remolten region decreased relatively slowly. In addition, when the scanning speed was increased from 200 to 800 mm/s, the decreasing speeds of melt depth and melt width were significantly accelerated, while the decreasing speeds of overlapping ratio were relatively slow. When the scanning speed was further increased to 1600 mm/s, the decreasing speeds of melt depth and melt width were further accelerated, and the un-lapped condition of the melt channel also appeared. In addition, the contact angle increases and then decreases with the scanning speed, and both are lower than 33°. Therefore, when selecting the scanning speed, it is necessary to make reasonable trade-offs according to the specific situation, and take into account the factors of melt depth, melt width, melt height, remolten region, and overlapping ratio, in order to achieve the best processing results.
TABLE III.
Double-track molten pool characterization information at different scanning speeds.
Scanning speed (mm/s)
Depth (μm)
Width (μm)
Height (μm)
Remolten region (μm)
Overlapping ratio (%)
Contact angle (°)
200
55/68
182
19/32
124
203.33
22
1600
13
50
11
/
−16.67
31
3. Hatch spacing
Figure 18 shows the effect of different hatch spacing on the morphology and temperature field of the double-track molten pool under the condition of laser power of 100 W and scanning speed of 800 mm/s. The surface morphology and temperature field of the first track and second track are basically the same, but slightly different. The first track shows a basically symmetric morphology along the scanning direction, while the second track shows a slight offset due to the difference in the heat transfer rate between the solidified material and the powder particles. When the hatch spacing is too small, the overlapping ratio increases and the probability of defects caused by remelting phenomenon grows. When the hatch spacing is too large, the neighboring melt track cannot overlap properly, and the powder particles are not completely melted, leading to an increase in the number of holes. In conclusion, the ratio of the line energy density � to the hatch spacing (the volume energy density E) has a significant effect on the temperature field and surface morphology of the molten pool.
Simulation results of double-track molten pool under different hatch spacings: (a) H = 0.03 mm, (b) H = 0.12 mm.
Table IV shows the effects of different hatch spacing on the characteristic information of the double-track molten pool under the condition of laser power of 100 W and scanning speed of 800 mm/s. It can be seen that the hatch spacing has little effect on the melt depth, melt width, and melt height, but has some effect on the remolten region. With the gradual expansion of hatch spacing, the remolten region shows a gradual decrease. At the same time, the overlapping ratio also decreased with the increase in hatch spacing. In addition, it is observed that the contact angle shows a tendency to increase and then remain stable when the hatch spacing increases, which has a more limited effect on it. Therefore, trade-offs and decisions need to be made on a case-by-case basis when selecting the hatch spacing.
TABLE IV.
Double-track molten pool characterization information at different hatch spacings.
Hatch spacing (mm)
Depth (μm)
Width (μm)
Height (μm)
Remolten region (μm)
Overlapping ratio (%)
Contact angle (°)
0.03
25/27
82
14
59
173.33
30
0.12
26
78
14
/
−35
33
In summary, the laser power, scanning speed, and hatch spacing have a significant effect on the formation of the molten pool, and the correct selection of these three process parameters is crucial to ensure the forming quality. In addition, the melt depth of the second track is slightly larger than that of the first track at higher line energy density � and volume energy density E. This is mainly due to the fact that a large amount of heat accumulation is generated after the first track, forming a larger molten pool volume, which leads to an increase in the melt depth.
D. Simulation analysis of molten pool with powder particle size and laser focal spot diameter
Figure 19 demonstrates the effect of different powder particle sizes and laser focal spot diameters on the morphology and temperature field of the double-track molten pool under a laser power of 100 W, a scanning speed of 800 mm/s, and a hatch spacing of 0.06 mm. In the process of melting coarse powder with small laser focal spot diameter, the laser energy cannot completely melt the larger powder particles, resulting in their partial melting and further generating excessive pore defects. The larger powder particles tend to generate zigzag molten pool edges, which cause an increase in the roughness of the melt track surface. In addition, the molten pool is also prone to generate the present spatter phenomenon, which can directly affect the quality of forming. The volume of the formed molten pool is relatively small, while the melt depth, melt width, and melt height are all smaller relative to the fine powder in Fig. 19(a). In the process of melting fine powders with a large laser focal spot diameter, the laser energy is able to melt the fine powder particles sufficiently, even to the point of overmelting. This results in a large number of fine spatters being generated at the edge of the molten pool, which causes porosity defects in the melt track in Fig. 19(b). In addition, the maximum velocity of the molten pool is larger for large powder particle sizes compared to small powder particle sizes, which indicates that the temperature gradient in the molten pool is larger for large powder particle sizes and the melt motion is more intense. However, the size of the laser focal spot diameter has a relatively small effect on the melt motion. However, a larger focal spot diameter induces a larger melt volume with greater depth, width, and height. In conclusion, a small powder size helps to reduce the surface roughness of the specimen, and a small laser spot diameter reduces the minimum forming size of a single track.
Simulation results of double-track molten pool with different powder particle size and laser focal spot diameter: (a) focal spot = 25 μm, coarse powder, (b) focal spot = 80 μm, fine powder.
Table V shows the maximum temperature gradient at the reference point for different powder sizes and laser focal spot diameters. As can be seen from the table, the maximum temperature gradient is lower than that of HP-LPBF for both coarse powders with a small laser spot diameter and fine powders with a large spot diameter, a phenomenon that leads to an increase in the heat transfer rate of HP-LPBF, which in turn leads to a corresponding increase in the cooling rate and, ultimately, to the formation of finer microstructures.
TABLE V.
Maximum temperature gradient at the reference point for different powder particle sizes and laser focal spot diameters.
Laser power (W)
Scanning speed (mm/s)
Hatch spacing (mm)
Average powder size (μm)
Laser focal spot diameter (μm)
Maximum temperature gradient (×107 K/s)
100
800
0.06
31.7
25
7.89
11.5
80
7.11
IV. CONCLUSIONS
In this study, the geometrical characteristics of 3D coarse and fine powder particles were first calculated using DEM and then numerical simulations of single track and double track in the process of forming SS316L from monolayer HP-LPBF at mesoscopic scale were developed using CFD method. The effects of Marangoni convection, surface tension, recoil pressure, gravity, thermal convection, thermal radiation, and evaporative heat dissipation on the heat and mass transfer in the molten pool were considered in this model. The effects of laser power, scanning speed, and hatch spacing on the dynamics of the single-track and double-track molten pools, as well as on other characteristic information, were investigated. The effects of the powder particle size on the molten pool were investigated comparatively with the laser focal spot diameter. The main conclusions are as follows:
The results show that the temperature gradient at the front of the molten pool is significantly larger than that at the tail, and the molten pool exhibits a “comet” morphology. At the top of the molten pool, there is a slightly concave region, which is the result of the coupling of Marangoni convection, recoil pressure, and surface tension. The melt flow forms two closed loops, which are mainly influenced by temperature gradients and surface tension. This special dynamic behavior of the melt tends to form an “elliptical” molten pool and an almost “mountain” shape in single-track forming.
The basic characteristics of the three-dimensional morphology and temperature field of the second track are similar to those of the first track, but there are subtle differences. The first track exhibits a basically symmetrical shape; however, due to the difference in thermal diffusion rates between the solidified metal and the powder, a slight asymmetry in the molten pool morphology of the second track occurs. After forming through the first track, there is a significant heat buildup in the powder bed, resulting in a longer dynamic time of the melt, which increases the life of the molten pool. The heights of the first track and second track remained essentially the same, but the depth of the second track was greater relative to the first track. In addition, the maximum temperature gradient was 1.69 × 108 K/s during HP-LPBF forming.
At low laser power, the surface tension in the molten pool plays a dominant role. At high laser power, recoil pressure becomes the main influencing factor. With the increase of laser power, the effective heat transfer in the vertical direction is superior to that in the horizontal direction. With the gradual increase of scanning speed, the surface morphology of the molten pool evolves from circular to elliptical. In addition, the scanning speed has a significant effect on the melt depth, melt width, melt height, remolten region, and overlapping ratio. Too large or too small hatch spacing will lead to remelting or non-lap phenomenon, which in turn causes the formation of defects.
When using a small laser focal spot diameter, it is difficult to completely melt large powder particle sizes, resulting in partial melting and excessive porosity generation. At the same time, large powder particles produce curved edges of the molten pool, resulting in increased surface roughness of the melt track. In addition, spatter occurs, which directly affects the forming quality. At small focal spot diameters, the molten pool volume is relatively small, and the melt depth, the melt width, and the melt height are correspondingly small. Taken together, the small powder particle size helps to reduce surface roughness, while the small spot diameter reduces the forming size.
REFERENCES
S. L. Sing and W. Y. Yeong , “ Laser powder bed fusion for metal additive manufacturing: Perspectives on recent developments,” Virtual Phys. Prototyping. 15, 359–370 (2020).https://doi.org/10.1080/17452759.2020.1779999 Google ScholarCrossref
A. M. Khorasani , I. G. Jithin , J. K. Veetil , and A. H. Ghasemi , “ A review of technological improvements in laser-based powder bed fusion of metal printers,” Int. J. Adv. Manuf. Technol. 108, 191–209 (2020).https://doi.org/10.1007/s00170-020-05361-3 Google ScholarCrossref
Y. Qin , A. Brockett , Y. Ma , A. Razali , J. Zhao , C. Harrison , W. Pan , X. Dai , and D. Loziak , “ Micro-manufacturing: Research, technology outcomes and development issues,” Int. J. Adv. Manuf. Technol. 47, 821–837 (2010).https://doi.org/10.1007/s00170-009-2411-2 Google ScholarCrossref
B. Nagarajan , Z. Hu , X. Song , W. Zhai , and J. Wei , “ Development of micro selective laser melting: The state of the art and future perspectives,” Engineering. 5, 702–720 (2019).https://doi.org/10.1016/j.eng.2019.07.002 Google ScholarCrossref
Y. Wei , G. Chen , W. Li , Y. Zhou , Z. Nie , J. Xu , and W. Zhou , “ Micro selective laser melting of SS316L: Single tracks, defects, microstructures and thermal/mechanical properties,” Opt. Laser Technol. 145, 107469 (2022).https://doi.org/10.1016/j.optlastec.2021.107469 Google ScholarCrossref
Y. Wei , G. Chen , W. Li , M. Li , Y. Zhou , Z. Nie , and J. Xu , “ Process optimization of micro selective laser melting and comparison of different laser diameter for forming different powder,” Opt. Laser Technol. 150, 107953 (2022).https://doi.org/10.1016/j.optlastec.2022.107953 Google ScholarCrossref
H. Zhiheng , B. Nagarajan , X. Song , R. Huang , W. Zhai , and J. Wei , “ Formation of SS316L single tracks in micro selective laser melting: Surface, geometry, and defects,” Adv. Mater. Sci. Eng. 2019, 9451406.https://doi.org/10.1155/2019/9451406 Crossref
B. Nagarajan , Z. Hu , S. Gao , X. Song , R. Huang , M. Seita , and J. Wei , “ Effect of in-situ laser remelting on the microstructure of SS316L fabricated by micro selective laser melting,” in Advanced Surface Enhancement, edited by Sho Itoh and Shashwat Shukla , Lecture Notes in Mechanical Engineering ( Springer Singapore, Singapore, 2020), pp. 330–336. Google ScholarCrossref
H. Zhiheng , B. Nagarajan , X. Song , R. Huang , W. Zhai , and J. Wei , “ Tailoring surface roughness of micro selective laser melted SS316L by in-situ laser remelting,” in Advanced Surface Enhancement, edited by Sho Itoh and Shashwat Shukla , Lecture Notes in Mechanical Engineering ( Springer Singapore, Singapore, 2020), pp. 337–343. Google Scholar
J. Fu , Z. Hu , X. Song , W. Zhai , Y. Long , H. Li , and M. Fu , “ Micro selective laser melting of NiTi shape memory alloy: Defects, microstructures and thermal/mechanical properties,” Opt. Laser Technol. 131, 106374 (2020).https://doi.org/10.1016/j.optlastec.2020.106374 Google ScholarCrossref
E. Abele and M. Kniepkamp , “ Analysis and optimisation of vertical surface roughness in micro selective laser melting,” Surf. Topogr.: Metrol. Prop. 3, 034007 (2015).https://doi.org/10.1088/2051-672X/3/3/034007 Google ScholarCrossref
S. Qu , J. Ding , J. Fu , M. Fu , B. Zhang , and X. Song , “ High-precision laser powder bed fusion processing of pure copper,” Addit. Manuf. 48, 102417 (2021).https://doi.org/10.1016/j.addma.2021.102417 Google ScholarCrossref
Y. Wei , G. Chen , M. Li , W. Li , Y. Zhou , J. Xu , and Z. wei , “ High-precision laser powder bed fusion of 18Ni300 maraging steel and its SiC reinforcement composite materials,” J. Manuf. Process. 84, 750–763 (2022).https://doi.org/10.1016/j.jmapro.2022.10.049 Google ScholarCrossref
B. Liu , R. Wildman , T. Christopher , I. Ashcroft , and H. Richard , “ Investigation the effect of particle size distribution on processing parameters optimisation in selective laser melting process,” in 2011 International Solid Freeform Fabrication Symposium ( University of Texas at Austin, 2011). Google Scholar
T. D. McLouth , G. E. Bean , D. B. Witkin , S. D. Sitzman , P. M. Adams , D. N. Patel , W. Park , J.-M. Yang , and R. J. Zaldivar , “ The effect of laser focus shift on microstructural variation of Inconel 718 produced by selective laser melting,” Mater. Des. 149, 205–213 (2018).https://doi.org/10.1016/j.matdes.2018.04.019 Google ScholarCrossref
Y. Qian , Y. Wentao , and L. Feng , “ Mesoscopic simulations of powder bed fusion: Research progresses and conditions,” Electromachining Mould 06, 46–52 (2017).https://doi.org/10.3969/j.issn.1009-279X.2017.06.012 Google Scholar
J. Fu , S. Qu , J. Ding , X. Song , and M. W. Fu , “ Comparison of the microstructure, mechanical properties and distortion of stainless Steel 316L fabricated by micro and conventional laser powder bed fusion,” Addit. Manuf. 44, 102067 (2021).https://doi.org/10.1016/j.addma.2021.102067 Google ScholarCrossref
N. T. Aboulkhair , I. Maskery , C. Tuck , I. Ashcroft , and N. M. Everitt , “ The microstructure and mechanical properties of selectively laser Melted AlSi10Mg: The effect of a conventional T6-like heat treatment,” Mater. Sci. Eng. A 667, 139–146 (2016).https://doi.org/10.1016/j.msea.2016.04.092 Google ScholarCrossref
S. Y. Chen , J. C. Huang , C. T. Pan , C. H. Lin , T. L. Yang , Y. S. Huang , C. H. Ou , L. Y. Chen , D. Y. Lin , H. K. Lin , T. H. Li , J. S. C. Jang , and C. C. Yang , “ Microstructure and mechanical properties of open-cell porous Ti-6Al-4V fabricated by selective laser melting,” J. Alloys Compd. 713, 248–254 (2017).https://doi.org/10.1016/j.jallcom.2017.04.190 Google ScholarCrossref
Y. Bai , Y. Yang , D. Wang , and M. Zhang , “ Influence mechanism of parameters process and mechanical properties evolution mechanism of Maraging steel 300 by selective laser melting,” Mater. Sci. Eng. A 703, 116–123 (2017).https://doi.org/10.1016/j.msea.2017.06.033 Google ScholarCrossref
Y. Bai , Y. Yang , Z. Xiao , M. Zhang , and D. Wang , “ Process optimization and mechanical property evolution of AlSiMg0.75 by selective laser melting,” Mater. Des. 140, 257–266 (2018).https://doi.org/10.1016/j.matdes.2017.11.045 Google ScholarCrossref
Y. Liu , M. Zhang , W. Shi , Y. Ma , and J. Yang , “ Study on performance optimization of 316L stainless steel parts by high-efficiency selective laser melting,” Opt. Laser Technol. 138, 106872 (2021).https://doi.org/10.1016/j.optlastec.2020.106872 Google ScholarCrossref
D. Gu , Y.-C. Hagedorn , W. Meiners , G. Meng , R. J. S. Batista , K. Wissenbach , and R. Poprawe , “ Densification behavior, microstructure evolution, and wear performance of selective laser melting processed commercially pure titanium,” Acta Mater. 60, 3849–3860 (2012).https://doi.org/10.1016/j.actamat.2012.04.006 Google ScholarCrossref
N. Read , W. Wang , K. Essa , and M. M. Attallah , “ Selective laser melting of AlSi10Mg alloy: Process optimisation and mechanical properties development,” Mater. Des. 65, 417–424 (2015).https://doi.org/10.1016/j.matdes.2014.09.044 Google ScholarCrossref
I. A. Roberts , C. J. Wang , R. Esterlein , M. Stanford , and D. J. Mynors , “ A three-dimensional finite element analysis of the temperature field during laser melting of metal powders in additive layer manufacturing,” Int. J. Mach. Tools Manuf. 49(12–13), 916–923 (2009).https://doi.org/10.1016/j.ijmachtools.2009.07.004 Google ScholarCrossref
K. Dai and L. Shaw , “ Finite element analysis of the effect of volume shrinkage during laser densification,” Acta Mater. 53(18), 4743–4754 (2005).https://doi.org/10.1016/j.actamat.2005.06.014 Google ScholarCrossref
K. Carolin , E. Attar , and P. Heinl , “ Mesoscopic simulation of selective beam melting processes,” J. Mater. Process. Technol. 211(6), 978–987 (2011).https://doi.org/10.1016/j.jmatprotec.2010.12.016 Google ScholarCrossref
F.-J. Gürtler , M. Karg , K.-H. Leitz , and M. Schmidt , “ Simulation of laser beam melting of steel powders using the three-dimensional volume of fluid method,” Phys. Procedia 41, 881–886 (2013).https://doi.org/10.1016/j.phpro.2013.03.162 Google ScholarCrossref
P. Meakin and R. Jullien , “ Restructuring effects in the rain model for random deposition,” J. Phys. France 48(10), 1651–1662 (1987).https://doi.org/10.1051/jphys:0198700480100165100 Google ScholarCrossref
J-m Wang , G-h Liu , Y-l Fang , and W-k Li , “ Marangoni effect in nonequilibrium multiphase system of material processing,” Rev. Chem. Eng. 32(5), 551–585 (2016).https://doi.org/10.1515/revce-2015-0067 Google ScholarCrossref
W. Ye , S. Zhang , L. L. Mendez , M. Farias , J. Li , B. Xu , P. Li , and Y. Zhang , “ Numerical simulation of the melting and alloying processes of elemental titanium and boron powders using selective laser alloying,” J. Manuf. Process. 64, 1235–1247 (2021).https://doi.org/10.1016/j.jmapro.2021.02.044 Google ScholarCrossref
U. S. Bertoli , A. J. Wolfer , M. J. Matthews , J.-P. R. Delplanque , and J. M. Schoenung , “ On the limitations of volumetric energy density as a design parameter for selective laser melting,” Mater. Des. 113, 331–340 (2017).https://doi.org/10.1016/j.matdes.2016.10.037 Google ScholarCrossref
W. E. King , H. D. Barth , V. M. Castillo , G. F. Gallegos , J. W. Gibbs , D. E. Hahn , C. Kamath , and A. M. Rubenchik , “ Observation of keyhole-mode laser melting in laser powder-bed fusion additive manufacturing,” J. Mater. Process. Technol. 214(12), 2915–2925 (2014).https://doi.org/10.1016/j.jmatprotec.2014.06.005 Google ScholarCrossref
L. Cao , “ Numerical simulation of the impact of laying powder on selective laser melting single-pass formation,” Int. J. Heat Mass Transfer 141, 1036–1048 (2019).https://doi.org/10.1016/j.ijheatmasstransfer.2019.07.053 Google ScholarCrossref
L. Huang , X. Hua , D. Wu , and F. Li , “ Numerical study of keyhole instability and porosity formation mechanism in laser welding of aluminum alloy and steel,” J. Mater. Process. Technol. 252, 421–431 (2018).https://doi.org/10.1016/j.jmatprotec.2017.10.011 Google ScholarCrossref
K. Q. Le , C. Tang , and C. H. Wong , “ On the study of keyhole-mode melting in selective laser melting process,” Int. J. Therm. Sci. 145, 105992 (2019).https://doi.org/10.1016/j.ijthermalsci.2019.105992 Google ScholarCrossref
J.-H. Cho and S.-J. Na , “ Theoretical analysis of keyhole dynamics in polarized laser drilling,” J. Phys. D: Appl. Phys. 40(24), 7638 (2007).https://doi.org/10.1088/0022-3727/40/24/007 Google ScholarCrossref
W. Ye , “ Mechanism analysis of selective laser melting and metallurgy process based on base element powder of titanium and boron,” Ph.D. dissertation ( Nanchang University, 2021). Google Scholar
R. Ammer , M. Markl , U. Ljungblad , C. Körner , and U. Rüde , “ Simulating fast electron beam melting with a parallel thermal free surface lattice Boltzmann method,” Comput. Math. Appl. 67(2), 318–330 (2014).https://doi.org/10.1016/j.camwa.2013.10.001 Google ScholarCrossref
H. Chen , Q. Wei , S. Wen , Z. Li , and Y. Shi , “ Flow behavior of powder particles in layering process of selective laser melting: Numerical modeling and experimental verification based on discrete element method,” Int. J. Mach. Tools Manuf. 123, 146–159 (2017).https://doi.org/10.1016/j.ijmachtools.2017.08.004 Google ScholarCrossref
F. Verhaeghe , T. Craeghs , J. Heulens , and L. Pandelaers , “ A pragmatic model for selective laser melting with evaporation,” Acta Mater. 57(20), 6006–6012 (2009).https://doi.org/10.1016/j.actamat.2009.08.027 Google ScholarCrossref
C. H. Fu and Y. B. Guo , “ Three-dimensional temperature gradient mechanism in selective laser melting of Ti-6Al-4V,” J. Manuf. Sci. Eng. 136(6), 061004 (2014).https://doi.org/10.1115/1.4028539 Google ScholarCrossref
Y. Xiang , Z. Shuzhe , L. Junfeng , W. Zhengying , Y. Lixiang , and J. Lihao , “ Numerical simulation and experimental verification for selective laser single track melting forming of Ti6Al4V,” J. Zhejiang Univ. (Eng. Sci.) 53(11), 2102–2109 + 2117 (2019).https://doi.org/10.3785/j.issn.1008-973X.2019.11.007 Google Scholar
Q. He , H. Xia , J. Liu , X. Ao , and S. Lin , “ Modeling and numerical studies of selective laser melting: Multiphase flow, solidification and heat transfer,” Mater. Des. 196, 109115 (2020).https://doi.org/10.1016/j.matdes.2020.109115 Google ScholarCrossref
L. Cao , “ Mesoscopic-scale numerical simulation including the influence of process parameters on SLM single-layer multi-pass formation,” Metall. Mater. Trans. A 51, 4130–4145 (2020).https://doi.org/10.1007/s11661-020-05831-z Google ScholarCrossref
L. Cao , “ Mesoscopic-scale numerical investigation including the influence of process parameters on LPBF multi-layer multi-path formation,” Comput. Model. Eng. Sci. 126(1), 5–23 (2021).https://doi.org/10.32604/cmes.2021.014693 Google ScholarCrossref
H. Yin and S. D. Felicelli , “ Dendrite growth simulation during solidification in the LENS process,” Acta Mater. 58(4), 1455–1465 (2010).https://doi.org/10.1016/j.actamat.2009.10.053 Google ScholarCrossref
P. Nie , O. A. Ojo , and Z. Li , “ Numerical modeling of microstructure evolution during laser additive manufacturing of a nickel-based superalloy,” Acta Mater. 77, 85–95 (2014).https://doi.org/10.1016/j.actamat.2014.05.039 Google ScholarCrossref
Z. Liu and H. Qi , “ Effects of substrate crystallographic orientations on crystal growth and microstructure formation in laser powder deposition of nickel-based superalloy,” Acta Mater. 87, 248–258 (2015).https://doi.org/10.1016/j.actamat.2014.12.046 Google ScholarCrossref
L. Wei , L. Xin , W. Meng , and H. Weidong , “ Cellular automaton simulation of the molten pool of laser solid forming process,” Acta Phys. Sin. 64(01), 018103–018363 (2015).https://doi.org/10.7498/aps.64.018103 Google ScholarCrossref
R. Acharya , J. A. Sharon , and A. Staroselsky , “ Prediction of microstructure in laser powder bed fusion process,” Acta Mater. 124, 360–371 (2017).https://doi.org/10.1016/j.actamat.2016.11.018 Google ScholarCrossref
M. R. Rolchigo and R. LeSar , “ Modeling of binary alloy solidification under conditions representative of additive manufacturing,” Comput. Mater. Sci. 150, 535–545 (2018).https://doi.org/10.1016/j.commatsci.2018.04.004 Google ScholarCrossref
S. Geng , P. Jiang , L. Guo , X. Gao , and G. Mi , “ Multi-scale simulation of grain/sub-grain structure evolution during solidification in laser welding of aluminum alloys,” Int. J. Heat Mass Transfer 149, 119252 (2020).https://doi.org/10.1016/j.ijheatmasstransfer.2019.119252 Google ScholarCrossref
W. L. Wang , W. Q. Liu , X. Yang , R. R. Xu , and Q. Y. Dai , “ Multi-scale simulation of columnar-to-equiaxed transition during laser selective melting of rare earth magnesium alloy,” J. Mater. Sci. Technol. 119, 11–24 (2022).https://doi.org/10.1016/j.jmst.2021.12.029 Google ScholarCrossref
Q. Xia , J. Yang , and Y. Li , “ On the conservative phase-field method with the N-component incompressible flows,” Phys. Fluids 35, 012120 (2023).https://doi.org/10.1063/5.0135490 Google ScholarCrossref
Q. Xia , G. Sun , J. Kim , and Y. Li , “ Multi-scale modeling and simulation of additive manufacturing based on fused deposition technique,” Phys. Fluids 35, 034116 (2023).https://doi.org/10.1063/5.0141316 Google ScholarCrossref
A. Hussein , L. Hao , C. Yan , and R. Everson , “ Finite element simulation of the temperature and stress fields in single layers built without-support in selective laser melting,” Mater. Des. 52, 638–647 (2013).https://doi.org/10.1016/j.matdes.2013.05.070 Google ScholarCrossref
J. Ding , P. Colegrove , J. Mehnen , S. Ganguly , P. M. Sequeira Almeida , F. Wang , and S. Williams , “ Thermo-mechanical analysis of wire and arc additive layer manufacturing process on large multi-layer parts,” Comput. Mater. Sci. 50(12), 3315–3322 (2011).https://doi.org/10.1016/j.commatsci.2011.06.023 Google ScholarCrossref
Y. Du , X. You , F. Qiao , L. Guo , and Z. Liu , “ A model for predicting the temperature field during selective laser melting,” Results Phys. 12, 52–60 (2019).https://doi.org/10.1016/j.rinp.2018.11.031 Google ScholarCrossref
X. Luo , M. Liu , L. Zhenhua , H. Li , and J. Shen , “ Effect of different heat-source models on calculated temperature field of selective laser melted 18Ni300,” Chin. J. Lasers 48(14), 1402005–1402062 (2021).https://doi.org/10.3788/CJL202148.1402005 Google ScholarCrossref
J. F. Li , L. Li , and F. H. Stott , “ Thermal stresses and their implication on cracking during laser melting of ceramic materials,” Acta Mater. 52(14), 4385–4398 (2004).https://doi.org/10.1016/j.actamat.2004.06.005 Google ScholarCrossref
P. Aggarangsi and J. L. Beuth , “ Localized preheating approaches for reducing residual stress in additive manufacturing,” paper presented at the 2006 International Solid Freeform Fabrication Symposium, The University of Texas in Austin on August 14–16, 2006.
K. Dai and L. Shaw , “ Thermal and mechanical finite element modeling of laser forming from metal and ceramic powders,” Acta Mater. 52(1), 69–80 (2004).https://doi.org/10.1016/j.actamat.2003.08.028 Google ScholarCrossref
A. H. Nickel , D. M. Barnett , and F. B. Prinz , “ Thermal stresses and deposition patterns in layered manufacturing,” Mater. Sci. Eng. A 317(1–2), 59–64 (2001).https://doi.org/10.1016/S0921-5093(01)01179-0 Google ScholarCrossref
M. F. Zaeh and G. Branner , “ Investigations on residual stresses and deformations in selective laser melting,” Prod. Eng. 4(1), 35–45 (2010).https://doi.org/10.1007/s11740-009-0192-y Google ScholarCrossref
P. Bian , J. Shi , Y. Liu , and Y. Xie , “ Influence of laser power and scanning strategy on residual stress distribution in additively manufactured 316L steel,” Opt. Laser Technol. 132, 106477 (2020).https://doi.org/10.1016/j.optlastec.2020.106477 Google ScholarCrossref
B. M. Marques , C. M. Andrade , D. M. Neto , M. C. Oliveira , J. L. Alves , and L. F. Menezes , “ Numerical analysis of residual stresses in parts produced by selective laser melting process,” Procedia Manuf. 47, 1170–1177 (2020).https://doi.org/10.1016/j.promfg.2020.04.167 Google ScholarCrossref
W. Mu , “ Numerical simulation of SLM forming process and research and prediction of forming properties,” MA thesis ( Anhui Jianzhu University, 2022). Google Scholar
Y. Zhang , “ Multi-scale multi-physics modeling of laser powder bed fusion process of metallic materials with experiment validation,” Ph.D. dissertation ( Purdue University, 2018). Google Scholar
Y. Qian , “ Mesoscopic simulation studies of key processing issues for powder bed fusion technology,” Ph.D. dissertation ( Tsinghua University, 2019). Google Scholar
N. V. Brilliantov , S. Frank , J.-M. Hertzsch , and T. Pöschel , “ Model for collisions in granular gases,” Phys. Rev. E 53(5), 5382–5392 (1996).https://doi.org/10.1103/PhysRevE.53.5382 Google ScholarCrossref
Z. Xiao , “ Research on microscale selective laser melting process of high strength pure copper specimens,” MA thesis ( Hunan University, 2022). Google Scholar
Z. Li , K. Mukai , M. Zeze , and K. C. Mills , “ Determination of the surface tension of liquid stainless steel,” J. Mater. Sci. 40(9–10), 2191–2195 (2005).https://doi.org/10.1007/s10853-005-1931-x Google ScholarCrossref
R. Scardovelli and S. Zaleski , “ Analytical relations connecting linear interfaces and volume fractions in rectangular grids,” J. Comput. Phys. 164(1), 228–237 (2000).https://doi.org/10.1006/jcph.2000.6567 Google ScholarCrossref
D.-W. Cho , W.-I. Cho , and S.-J. Na , “ Modeling and simulation of arc: Laser and hybrid welding process,” J. Manuf. Process. 16(1), 26–55 (2014).https://doi.org/10.1016/j.jmapro.2013.06.012 Google ScholarCrossref 76.Flow3D. Version 11.1.0: User Manual ( FlowScience, Santa Fe, NM, USA, 2015).
Y. Tian , L. Yang , D. Zhao , Y. Huang , and J. Pan , “ Numerical analysis of powder bed generation and single track forming for selective laser melting of ss316l stainless steel,” J. Manuf. Process. 58, 964–974 (2020).https://doi.org/10.1016/j.jmapro.2020.09.002 Google ScholarCrossref
C. Tang , K. Q. Le , and C. H. Wong , “ Physics of humping formation in laser powder bed fusion,” Int. J. Heat Mass Transfer 149, 119172 (2020).https://doi.org/10.1016/j.ijheatmasstransfer.2019.119172 Google ScholarCrossref
L. Cao , “ Mesoscopic-scale simulation of pore evolution during laser powder bed fusion process,” Comput. Mater. Sci. 179, 109686 (2020).https://doi.org/10.1016/j.commatsci.2020.109686 Google ScholarCrossref
R. Li , J. Liu , Y. Shi , W. Li , and W. Jiang , “ Balling behavior of stainless steel and nickel powder during selective laser melting process,” Int. J. Adv. Manuf. Technol. 59(9–12), 1025–1035 (2012).https://doi.org/10.1007/s00170-011-3566-1 Google ScholarCrossref
S. A. Khairallah and A. Anderson , “ Mesoscopic simulation model of selective laser melting of stainless steel powder,” J. Mater. Process. Technol. 214(11), 2627–2636 (2014).https://doi.org/10.1016/j.jmatprotec.2014.06.001 Google ScholarCrossref
J. Liu , D. Gu , H. Chen , D. Dai , and H. Zhang , “ Influence of substrate surface morphology on wetting behavior of tracks during selective laser melting of aluminum-based alloys,” J. Zhejiang Univ. Sci. A 19(2), 111–121 (2018).https://doi.org/10.1631/jzus.A1700599 Google ScholarCrossref
L. Li , J. Li , and T. Fan , “ Phase-field modeling of wetting and balling dynamics in powder bed fusion process,” Phys. Fluids 33, 042116 (2021).https://doi.org/10.1063/5.0046771 Google ScholarCrossref
X. Nie , Z. Hu , H. Zhu , Z. Hu , L. Ke , and X. Zeng , “ Analysis of processing parameters and characteristics of selective laser melted high strength Al-Cu-Mg alloys: from single tracks to cubic samples,” J. Mater. Process. Technol. 256, 69–77 (2018).https://doi.org/10.1016/j.jmatprotec.2018.01.030 Google ScholarCrossref
금속 적층 제조 중 고체 상 변형 예측: Inconel-738의 전자빔 분말층 융합에 대한 사례 연구
Nana Kwabena Adomako a, Nima Haghdadi a, James F.L. Dingle bc, Ernst Kozeschnik d, Xiaozhou Liao bc, Simon P. Ringer bc, Sophie Primig a
Abstract
Metal additive manufacturing (AM) has now become the perhaps most desirable technique for producing complex shaped engineering parts. However, to truly take advantage of its capabilities, advanced control of AM microstructures and properties is required, and this is often enabled via modeling. The current work presents a computational modeling approach to studying the solid-state phase transformation kinetics and the microstructural evolution during AM. Our approach combines thermal and thermo-kinetic modelling. A semi-analytical heat transfer model is employed to simulate the thermal history throughout AM builds. Thermal profiles of individual layers are then used as input for the MatCalc thermo-kinetic software. The microstructural evolution (e.g., fractions, morphology, and composition of individual phases) for any region of interest throughout the build is predicted by MatCalc. The simulation is applied to an IN738 part produced by electron beam powder bed fusion to provide insights into how γ′ precipitates evolve during thermal cycling. Our simulations show qualitative agreement with our experimental results in predicting the size distribution of γ′ along the build height, its multimodal size character, as well as the volume fraction of MC carbides. Our findings indicate that our method is suitable for a range of AM processes and alloys, to predict and engineer their microstructures and properties.
Additive manufacturing (AM) is an advanced manufacturing method that enables engineering parts with intricate shapes to be fabricated with high efficiency and minimal materials waste. AM involves building up 3D components layer-by-layer from feedstocks such as powder [1]. Various alloys, including steel, Ti, Al, and Ni-based superalloys, have been produced using different AM techniques. These techniques include directed energy deposition (DED), electron- and laser powder bed fusion (E-PBF and L-PBF), and have found applications in a variety of industries such as aerospace and power generation[2], [3], [4]. Despite the growing interest, certain challenges limit broader applications of AM fabricated components in these industries and others. One of such limitations is obtaining a suitable and reproducible microstructure that offers the desired mechanical properties consistently. In fact, the AM as-built microstructure is highly complex and considerably distinctive from its conventionally processed counterparts owing to the complicated thermal cycles arising from the deposition of several layers upon each other [5], [6].
Several studies have reported that the solid-state phases and solidification microstructure of AM processed alloys such as CMSX-4, CoCr [7], [8], Ti-6Al-4V [9], [10], [11], IN738[6], 304L stainless steel[12], and IN718 [13], [14] exhibit considerable variations along the build direction. For instance, references [9], [10] have reported that there is a variation in the distribution of α and β phases along the build direction in Ti-alloys. Similarly, the microstructure of an L-PBF fabricated martensitic steel exhibits variations in the fraction of martensite [15]. Furthermore, some of the present authors and others [6], [16], [17], [18], [19], [20] have recently reviewed and reported that there is a difference in the morphology and fraction of nanoscale precipitates as a function of build height in Ni-based superalloys. These non-uniformities in the as-built microstructure result in an undesired heterogeneity in mechanical and other important properties such as corrosion and oxidation[19], [21], [22], [23]. To obtain the desired microstructure and properties, additional processing treatments are utilized, but this incurs extra costs and may lead to precipitation of detrimental phases and grain coarsening. Therefore, a through-process understanding of the microstructure evolution under repeated heating and cooling is now needed to further advance 3D printed microstructure and property control.
It is now commonly understood that the microstructure evolution during printing is complex, and most AM studies concentrate on the microstructure and mechanical properties of the final build only. Post-printing studies of microstructure characteristics at room temperature miss crucial information on how they evolve. In-situ measurements and modelling approaches are required to better understand the complex microstructural evolution under repeated heating and cooling. Most in-situ measurements in AM focus on monitoring the microstructural changes, such as phase transformations and melt pool dynamics during fabrication using X-ray scattering and high-speed X-ray imaging [24], [25], [26], [27]. For example, Zhao et al. [25] measured the rate of solidification and described the α/β phase transformation during L-PBF of Ti-6Al-4V in-situ. Also, Wahlmann et al. [21] recently used an L-PBF machine coupled with X-ray scattering to investigate the changes in CMSX-4 phase during successive melting processes. Although these techniques provide significant understanding of the basic principles of AM, they are not widely accessible. This is due to the great cost of the instrument, competitive application process, and complexities in terms of the experimental set-up, data collection, and analysis [26], [28].
Computational modeling techniques are promising and more widely accessible tools that enable advanced understanding, prediction, and engineering of microstructures and properties during AM. So far, the majority of computational studies have concentrated on physics based process models for metal AM, with the goal of predicting the temperature profile, heat transfer, powder dynamics, and defect formation (e.g., porosity) [29], [30]. In recent times, there have been efforts in modeling of the AM microstructure evolution using approaches such as phase-field [31], Monte Carlo (MC) [32], and cellular automata (CA) [33], coupled with finite element simulations for temperature profiles. However, these techniques are often restricted to simulating the evolution of solidification microstructures (e.g., grain and dendrite structure) and defects (e.g., porosity). For example, Zinovieva et al. [33] predicted the grain structure of L-PBF Ti-6Al-4V using finite difference and cellular automata methods. However, studies on the computational modelling of the solid-state phase transformations, which largely determine the resulting properties, remain limited. This can be attributed to the multi-component and multi-phase nature of most engineering alloys in AM, along with the complex transformation kinetics during thermal cycling. This kind of research involves predictions of the thermal cycle in AM builds, and connecting it to essential thermodynamic and kinetic data as inputs for the model. Based on the information provided, the thermokinetic model predicts the history of solid-state phase microstructure evolution during deposition as output. For example, a multi-phase, multi-component mean-field model has been developed to simulate the intermetallic precipitation kinetics in IN718 [34] and IN625 [35] during AM. Also, Basoalto et al. [36] employed a computational framework to examine the contrasting distributions of process-induced microvoids and precipitates in two Ni-based superalloys, namely IN718 and CM247LC. Furthermore, McNamara et al. [37] established a computational model based on the Johnson-Mehl-Avrami model for non-isothermal conditions to predict solid-state phase transformation kinetics in L-PBF IN718 and DED Ti-6Al-4V. These models successfully predicted the size and volume fraction of individual phases and captured the repeated nucleation and dissolution of precipitates that occur during AM.
In the current study, we propose a modeling approach with appreciably short computational time to investigate the detailed microstructural evolution during metal AM. This may include obtaining more detailed information on the morphologies of phases, such as size distribution, phase fraction, dissolution and nucleation kinetics, as well as chemistry during thermal cycling and final cooling to room temperature. We utilize the combination of the MatCalc thermo-kinetic simulator and a semi-analytical heat conduction model. MatCalc is a software suite for simulation of phase transformations, microstructure evolution and certain mechanical properties in engineering alloys. It has successfully been employed to simulate solid-state phase transformations in Ni-based superalloys [38], [39], steels [40], and Al alloys[41] during complex thermo-mechanical processes. MatCalc uses the classical nucleation theory as well as the so-called Svoboda-Fischer-Fratzl-Kozeschnik (SFFK) growth model as the basis for simulating precipitation kinetics [42]. Although MatCalc was originally developed for conventional thermo-mechanical processes, we will show that it is also applicable for AM if the detailed time-temperature profile of the AM build is known. The semi-analytical heat transfer code developed by Stump and Plotkowski [43] is used to simulate these profile throughout the AM build.
1.1. Application to IN738
Inconel-738 (IN738) is a precipitation hardening Ni-based superalloy mainly employed in high-temperature components, e.g. in gas turbines and aero-engines owing to its exceptional mechanical properties at temperatures up to 980 °C, coupled with high resistance to oxidation and corrosion [44]. Its superior high-temperature strength (∼1090 MPa tensile strength) is provided by the L12 ordered Ni3(Al,Ti) γ′ phase that precipitates in a face-centered cubic (FCC) γ matrix [45], [46]. Despite offering great properties, IN738, like most superalloys with high γ′ fractions, is challenging to process owing to its propensity to hot cracking [47], [48]. Further, machining of such alloys is challenging because of their high strength and work-hardening rates. It is therefore difficult to fabricate complex INC738 parts using traditional manufacturing techniques like casting, welding, and forging.
The emergence of AM has now made it possible to fabricate such parts from IN738 and other superalloys. Some of the current authors’ recent research successfully applied E-PBF to fabricate defect-free IN738 containing γ′ throughout the build [16], [17]. The precipitated γ′ were heterogeneously distributed. In particular, Haghdadi et al. [16] studied the origin of the multimodal size distribution of γ′, while Lim et al. [17] investigated the gradient in γ′ character with build height and its correlation to mechanical properties. Based on these results, the present study aims to extend the understanding of the complex and site-specific microstructural evolution in E-PBF IN738 by using a computational modelling approach. New experimental evidence (e.g., micrographs not published previously) is presented here to support the computational results.
2. Materials and Methods
2.1. Materials preparation
IN738 Ni-based superalloy (59.61Ni-8.48Co-7.00Al-17.47Cr-3.96Ti-1.01Mo-0.81W-0.56Ta-0.49Nb-0.47C-0.09Zr-0.05B, at%) gas-atomized powder was used as feedstock. The powders, with average size of 60 ± 7 µm, were manufactured by Praxair and distributed by Astro Alloys Inc. An Arcam Q10 machine by GE Additive with an acceleration voltage of 60 kV was used to fabricate a 15 × 15 × 25 mm3 block (XYZ, Z: build direction) on a 316 stainless steel substrate. The block was 3D-printed using a ‘random’ spot melt pattern. The random spot melt pattern involves randomly selecting points in any given layer, with an equal chance of each point being melted. Each spot melt experienced a dwell time of 0.3 ms, and the layer thickness was 50 µm. Some of the current authors have previously characterized the microstructure of the very same and similar builds in more detail [16], [17]. A preheat temperature of ∼1000 °C was set and kept during printing to reduce temperature gradients and, in turn, thermal stresses [49], [50], [51]. Following printing, the build was separated from the substrate through electrical discharge machining. It should be noted that this sample was simultaneously printed with the one used in [17] during the same build process and on the same build plate, under identical conditions.
2.2. Microstructural characterization
The printed sample was longitudinally cut in the direction of the build using a Struers Accutom-50, ground, and then polished to 0.25 µm suspension via standard techniques. The polished x-z surface was electropolished and etched using Struers A2 solution (perchloric acid in ethanol). Specimens for image analysis were polished using a 0.06 µm colloidal silica. Microstructure analyses were carried out across the height of the build using optical microscopy (OM) and scanning electron microscopy (SEM) with focus on the microstructure evolution (γ′ precipitates) in individual layers. The position of each layer being analyzed was determined by multiplying the layer number by the layer thickness (50 µm). It should be noted that the position of the first layer starts where the thermal profile is tracked (in this case, 2 mm from the bottom). SEM images were acquired using a JEOL 7001 field emission microscope. The brightness and contrast settings, acceleration voltage of 15 kV, working distance of 10 mm, and other SEM imaging parameters were all held constant for analysis of the entire build. The ImageJ software was used for automated image analysis to determine the phase fraction and size of γ′ precipitates and carbides. A 2-pixel radius Gaussian blur, following a greyscale thresholding and watershed segmentation was used [52]. Primary γ′ sizes (>50 nm), were measured using equivalent spherical diameters. The phase fractions were considered equal to the measured area fraction. Secondary γ′ particles (<50 nm) were not considered here. The γ′ size in the following refers to the diameter of a precipitate.
2.3. Hardness testing
A Struers DuraScan tester was utilized for Vickers hardness mapping on a polished x-z surface, from top to bottom under a maximum load of 100 mN and 10 s dwell time. 30 micro-indentations were performed per row. According to the ASTM standard [53], the indentations were sufficiently distant (∼500 µm) to assure that strain-hardened areas did not interfere with one another.
2.4. Computational simulation of E-PBF IN738 build
2.4.1. Thermal profile modeling
The thermal history was generated using the semi-analytical heat transfer code (also known as the 3DThesis code) developed by Stump and Plotkowski [43]. This code is an open-source C++ program which provides a way to quickly simulate the conductive heat transfer found in welding and AM. The key use case for the code is the simulation of larger domains than is practicable with Computational Fluid Dynamics/Finite Element Analysis programs like FLOW-3D AM. Although simulating conductive heat transfer will not be an appropriate simplification for some investigations (for example the modelling of keyholding or pore formation), the 3DThesis code does provide fast estimates of temperature, thermal gradient, and solidification rate which can be useful for elucidating microstructure formation across entire layers of an AM build. The mathematics involved in the code is as follows:
In transient thermal conduction during welding and AM, with uniform and constant thermophysical properties and without considering fluid convection and latent heat effects, energy conservation can be expressed as:(1)��∂�∂�=�∇2�+�̇where � is density, � specific heat, � temperature, � time, � thermal conductivity, and �̇ a volumetric heat source. By assuming a semi-infinite domain, Eq. 1 can be analytically solved. The solution for temperature at a given time (t) using a volumetric Gaussian heat source is presented as:(2)��,�,�,�−�0=33�����32∫0�1������exp−3�′�′2��+�′�′2��+�′�′2����′(3)and��=12��−�′+��2for�=�,�,�(4)and�′�′=�−���′Where � is the vector �,�,� and �� is the location of the heat source.
The numerical integration scheme used is an adaptive Gaussian quadrature method based on the following nondimensionalization:(5)�=��xy2�,�′=��xy2�′,�=��xy,�=��xy,�=��xy,�=���xy
A more detailed explanation of the mathematics can be found in reference [43].
The main source of the thermal cycling present within a powder-bed fusion process is the fusion of subsequent layers. Therefore, regions near the top of a build are expected to undergo fewer thermal cycles than those closer to the bottom. For this purpose, data from the single scan’s thermal influence on multiple layers was spliced to represent the thermal cycles experienced at a single location caused by multiple subsequent layers being fused.
The cross-sectional area simulated by this model was kept constant at 1 × 1 mm2, and the depth was dependent on the build location modelled with MatCalc. For a build location 2 mm from the bottom, the maximum number of layers to simulate is 460. Fig. 1a shows a stitched overview OM image of the entire build indicating the region where this thermal cycle is simulated and tracked. To increase similarity with the conditions of the physical build, each thermal history was constructed from the results of two simulations generated with different versions of a random scan path. The parameters used for these thermal simulations can be found in Table 1. It should be noted that the main purpose of the thermal profile modelling was to demonstrate how the conditions at different locations of the build change relative to each other. Accurately predicting the absolute temperature during the build would require validation via a temperature sensor measurement during the build process which is beyond the scope of the study. Nonetheless, to establish the viability of the heat source as a suitable approximation for this study, an additional sensitivity analysis was conducted. This analysis focused on the influence of energy input on γ′ precipitation behavior, the central aim of this paper. This was achieved by employing varying beam absorption energies (0.76, 0.82 – the values utilized in the simulation, and 0.9). The direct impact of beam absorption efficiency on energy input into the material was investigated. Specifically, the initial 20 layers of the build were simulated and subsequently compared to experimental data derived from SEM. While phase fractions were found to be consistent across all conditions, disparities emerged in the mean size of γ′ precipitates. An absorption efficiency of 0.76 yielded a mean size of approximately 70 nm. Conversely, absorption efficiencies of 0.82 and 0.9 exhibited remarkably similar mean sizes of around 130 nm, aligning closely with the outcomes of the experiments.
The numerical analyses of the evolution of precipitates was performed using MatCalc version 6.04 (rel 0.011). The thermodynamic (‘mc_ni.tdb’, version 2.034) and diffusion (‘mc_ni.ddb’, version 2.007) databases were used. MatCalc’s basic principles are elaborated as follows:
The nucleation kinetics of precipitates are computed using a computational technique based on a classical nucleation theory[54] that has been modified for systems with multiple components [42], [55]. Accordingly, the transient nucleation rate (�), which expresses the rate at which nuclei are formed per unit volume and time, is calculated as:(6)�=�0��*∙�xp−�*�∙�∙exp−��where �0 denotes the number of active nucleation sites, �* the rate of atomic attachment, � the Boltzmann constant, � the temperature, �* the critical energy for nucleus formation, τ the incubation time, and t the time. � (Zeldovich factor) takes into consideration that thermal excitation destabilizes the nucleus as opposed to its inactive state [54]. Z is defined as follows:(7)�=−12�kT∂2∆�∂�2�*12where ∆� is the overall change in free energy due to the formation of a nucleus and n is the nucleus’ number of atoms. ∆�’s derivative is evaluated at n* (critical nucleus size). �* accounts for the long-range diffusion of atoms required for nucleation, provided that the matrix’ and precipitates’ composition differ. Svoboda et al. [42] developed an appropriate multi-component equation for �*, which is given by:(8)�*=4��*2�4�∑�=1��ki−�0�2�0��0�−1where �* denotes the critical radius for nucleation, � represents atomic distance, and � is the molar volume. �ki and �0� represent the concentration of elements in the precipitate and matrix, respectively. The parameter �0� denotes the rate of diffusion of the ith element within the matrix. The expression for the incubation time � is expressed as [54]:(9)�=12�*�2
and �*, which represents the critical energy for nucleation:(10)�*=16�3�3∆�vol2where � is the interfacial energy, and ∆Gvol the change in the volume free energy. The critical nucleus’ composition is similar to the γ′ phase’s equilibrium composition at the same temperature. � is computed based on the precipitate and matrix compositions, using a generalized nearest neighbor broken bond model, with the assumption of interfaces being planar, sharp, and coherent [56], [57], [58].
In Eq. 7, it is worth noting that �* represents the fundamental variable in the nucleation theory. It contains �3/∆�vol2 and is in the exponent of the nucleation rate. Therefore, even small variations in γ and/or ∆�vol can result in notable changes in �, especially if �* is in the order of �∙�. This is demonstrated in [38] for UDIMET 720 Li during continuous cooling, where these quantities change steadily during precipitation due to their dependence on matrix’ and precipitate’s temperature and composition. In the current work, these changes will be even more significant as the system is exposed to multiple cycles of rapid cooling and heating.
Once nucleated, the growth of a precipitate is assessed using the radius and composition evolution equations developed by Svoboda et al. [42] with a mean-field method that employs the thermodynamic extremal principle. The expression for the total Gibbs free energy of a thermodynamic system G, which consists of n components and m precipitates, is given as follows:(11)�=∑���0��0�+∑�=1�4���33��+∑�=1��ki�ki+∑�=1�4���2��.
The chemical potential of component � in the matrix is denoted as �0�(�=1,…,�), while the chemical potential of component � in the precipitate is represented by �ki(�=1,…,�,�=1,…,�). These chemical potentials are defined as functions of the concentrations �ki(�=1,…,�,�=1,…,�). The interface energy density is denoted as �, and �� incorporates the effects of elastic energy and plastic work resulting from the volume change of each precipitate.
Eq. (12) establishes that the total free energy of the system in its current state relies on the independent state variables: the sizes (radii) of the precipitates �� and the concentrations of each component �ki. The remaining variables can be determined by applying the law of mass conservation to each component �. This can be represented by the equation:(12)��=�0�+∑�=1�4���33�ki,
Furthermore, the global mass conservation can be expressed by equation:(13)�=∑�=1���When a thermodynamic system transitions to a more stable state, the energy difference between the initial and final stages is dissipated. This model considers three distinct forms of dissipation effects [42]. These include dissipations caused by the movement of interfaces, diffusion within the precipitate and diffusion within the matrix.
Consequently, �̇� (growth rate) and �̇ki (chemical composition’s rate of change) of the precipitate with index � are derived from the linear system of equation system:(14)�ij��=��where �� symbolizes the rates �̇� and �̇ki [42]. Index i contains variables for precipitate radius, chemical composition, and stoichiometric boundary conditions suggested by the precipitate’s crystal structure. Eq. (10) is computed separately for every precipitate �. For a more detailed description of the formulae for the coefficients �ij and �� employed in this work please refer to [59].
The MatCalc software was used to perform the numerical time integration of �̇� and �̇ki of precipitates based on the classical numerical method by Kampmann and Wagner [60]. Detailed information on this method can be found in [61]. Using this computational method, calculations for E-PBF thermal cycles (cyclic heating and cooling) were computed and compared to experimental data. The simulation took approximately 2–4 hrs to complete on a standard laptop.
3. Results
3.1. Microstructure
Fig. 1 displays a stitched overview image and selected SEM micrographs of various γ′ morphologies and carbides after observations of the X-Z surface of the build from the top to 2 mm above the bottom. Fig. 2 depicts a graph that charts the average size and phase fraction of the primary γ′, as it changes with distance from the top to the bottom of the build. The SEM micrographs show widespread primary γ′ precipitation throughout the entire build, with the size increasing in the top to bottom direction. Particularly, at the topmost height, representing the 460th layer (Z = 22.95 mm), as seen in Fig. 1b, the average size of γ′ is 110 ± 4 nm, exhibiting spherical shapes. This is representative of the microstructure after it solidifies and cools to room temperature, without experiencing additional thermal cycles. The γ′ size slightly increases to 147 ± 6 nm below this layer and remains constant until 0.4 mm (∼453rd layer) from the top. At this position, the microstructure still closely resembles that of the 460th layer. After the 453rd layer, the γ′ size grows rapidly to ∼503 ± 19 nm until reaching the 437th layer (1.2 mm from top). The γ′ particles here have a cuboidal shape, and a small fraction is coarser than 600 nm. γ′ continue to grow steadily from this position to the bottom (23 mm from the top). A small fraction of γ′ is > 800 nm.
Besides primary γ′, secondary γ′ with sizes ranging from 5 to 50 nm were also found. These secondary γ′ precipitates, as seen in Fig. 1f, were present only in the bottom and middle regions. A detailed analysis of the multimodal size distribution of γ′ can be found in [16]. There is no significant variation in the phase fraction of the γ′ along the build. The phase fraction is ∼ 52%, as displayed in Fig. 2. It is worth mentioning that the total phase fraction of γ′ was estimated based on the primary γ′ phase fraction because of the small size of secondary γ′. Spherical MC carbides with sizes ranging from 50 to 400 nm and a phase fraction of 0.8% were also observed throughout the build. The carbides are the light grey precipitates in Fig. 1g. The light grey shade of carbides in the SEM images is due to their composition and crystal structure [52]. These carbides are not visible in Fig. 1b-e because they were dissolved during electro-etching carried out after electropolishing. In Fig. 1g, however, the sample was examined directly after electropolishing, without electro-etching.
Table 2 shows the nominal and measured composition of γ′ precipitates throughout the build by atom probe microscopy as determined in our previous study [17]. No build height-dependent composition difference was observed in either of the γ′ precipitate populations. However, there was a slight disparity between the composition of primary and secondary γ′. Among the main γ′ forming elements, the primary γ′ has a high Ti concentration while secondary γ′ has a high Al concentration. A detailed description of the atom distribution maps and the proxigrams of the constituent elements of γ′ throughout the build can be found in [17].
Table 2. Bulk IN738 composition determined using inductively coupled plasma atomic emission spectroscopy (ICP-AES). Compositions of γ, primary γ′, and secondary γ′ at various locations in the build measured by APT. This information is reproduced from data in Ref. [17] with permission.
at%
Ni
Cr
Co
Al
Mo
W
Ti
Nb
C
B
Zr
Ta
Others
Bulk
59.12
17.47
8.48
7.00
1.01
0.81
3.96
0.49
0.47
0.05
0.09
0.56
0.46
γ matrix
Top
50.48
32.91
11.59
1.94
1.39
0.82
0.44
0.8
0.03
0.03
0.02
–
0.24
Mid
50.37
32.61
11.93
1.79
1.54
0.89
0.44
0.1
0.03
0.02
0.02
0.01
0.23
Bot
48.10
34.57
12.08
2.14
1.43
0.88
0.48
0.08
0.04
0.03
0.01
–
0.12
Primary γ′
Top
72.17
2.51
3.44
12.71
0.25
0.39
7.78
0.56
–
0.03
0.02
0.05
0.08
Mid
71.60
2.57
3.28
13.55
0.42
0.68
7.04
0.73
–
0.01
0.03
0.04
0.04
Bot
72.34
2.47
3.86
12.50
0.26
0.44
7.46
0.50
0.05
0.02
0.02
0.03
0.04
Secondary γ′
Mid
70.42
4.20
3.23
14.19
0.63
1.03
5.34
0.79
0.03
–
0.04
0.04
0.05
Bot
69.91
4.06
3.68
14.32
0.81
1.04
5.22
0.65
0.05
–
0.10
0.02
0.11
3.2. Hardness
Fig. 3a shows the Vickers hardness mapping performed along the entire X-Z surface, while Fig. 3b shows the plot of average hardness at different build heights. This hardness distribution is consistent with the γ′ precipitate size gradient across the build direction in Fig. 1, Fig. 2. The maximum hardness of ∼530 HV1 is found at ∼0.5 mm away from the top surface (Z = 22.5), where γ′ particles exhibit the smallest observed size in Fig. 2b. Further down the build (∼ 2 mm from the top), the hardness drops to the 440–490 HV1 range. This represents the region where γ′ begins to coarsen. The hardness drops further to 380–430 HV1 at the bottom of the build.
3.3. Modeling of the microstructural evolution during E-PBF
3.3.1. Thermal profile modeling
Fig. 4 shows the simulated thermal profile of the E-PBF build at a location of 23 mm from the top of the build, using a semi-analytical heat conduction model. This profile consists of the time taken to deposit 460 layers until final cooling, as shown in Fig. 4a. Fig. 4b-d show the magnified regions of Fig. 4a and reveal the first 20 layers from the top, a single layer (first layer from the top), and the time taken for the build to cool after the last layer deposition, respectively.
The peak temperatures experienced by previous layers decrease progressively as the number of layers increases but never fall below the build preheat temperature (1000 °C). Our simulated thermal cycle may not completely capture the complexity of the actual thermal cycle utilized in the E-PBF build. For instance, the top layer (Fig. 4c), also representing the first deposit’s thermal profile without additional cycles (from powder heating, melting, to solidification), recorded the highest peak temperature of 1390 °C. Although this temperature is above the melting range of the alloy (1230–1360 °C) [62], we believe a much higher temperature was produced by the electron beam to melt the powder. Nevertheless, the solidification temperature and dynamics are outside the scope of this study as our focus is on the solid-state phase transformations during deposition. It takes ∼25 s for each layer to be deposited and cooled to the build temperature. The interlayer dwell time is 125 s. The time taken for the build to cool to room temperature (RT) after final layer deposition is ∼4.7 hrs (17,000 s).
3.3.2. MatCalc simulation
During the MatCalc simulation, the matrix phase is defined as γ. γ′, and MC carbide are included as possible precipitates. The domain of these precipitates is set to be the matrix (γ), and nucleation is assumed to be homogenous. In homogeneous nucleation, all atoms of the unit volume are assumed to be potential nucleation sites. Table 3 shows the computational parameters used in the simulation. All other parameters were set at default values as recommended in the version 6.04.0011 of MatCalc. The values for the interfacial energies are automatically calculated according to the generalized nearest neighbor broken bond model and is one of the most outstanding features in MatCalc [56], [57], [58]. It should be noted that the elastic misfit strain was not included in the calculation. The output of MatCalc includes phase fraction, size, nucleation rate, and composition of the precipitates. The phase fraction in MatCalc is the volume fraction. Although the experimental phase fraction is the measured area fraction, it is relatively similar to the volume fraction. This is because of the generally larger precipitate size and similar morphology at the various locations along the build [63]. A reliable phase fraction comparison between experiment and simulation can therefore be made.
Table 3. Computational parameters used in the simulation.
γ′ = 0.080–0.140 J/m2 and MC carbide = 0.410–0.430 J/m2
3.3.2.1. Precipitate phase fraction
Fig. 5a shows the simulated phase fraction of γ′ and MC carbide during thermal cycling. Fig. 5b is a magnified view of 5a showing the simulated phase fraction at the center points of the top 70 layers, whereas Fig. 5c corresponds to the first two layers from the top. As mentioned earlier, the top layer (460th layer) represents the microstructure after solidification. The microstructure of the layers below is determined by the number of thermal cycles, which increases with distance to the top. For example, layers 459, 458, 457, up to layer 1 (region of interest) experience 1, 2, 3 and 459 thermal cycles, respectively. In the top layer in Fig. 5c, the volume fraction of γ′ and carbides increases with temperature. For γ′, it decreases to zero when the temperature is above the solvus temperature after a few seconds. Carbides, however, remain constant in their volume fraction reaching equilibrium (phase fraction ∼ 0.9%) in a short time. The topmost layer can be compared to the first deposit, and the peak in temperature symbolizes the stage where the electron beam heats the powder until melting. This means γ′ and carbide precipitation might have started in the powder particles during heating from the build temperature and electron beam until the onset of melting, where γ′ dissolves, but carbides remain stable [28].
During cooling after deposition, γ′ reprecipitates at a temperature of 1085 °C, which is below its solvus temperature. As cooling progresses, the phase fraction increases steadily to ∼27% and remains constant at 1000 °C (elevated build temperature). The calculated equilibrium fraction of phases by MatCalc is used to show the complex precipitation characteristics in this alloy. Fig. 6 shows that MC carbides form during solidification at 1320 °C, followed by γ′, which precipitate when the solidified layer cools to 1140 °C. This indicates that all deposited layers might contain a negligible amount of these precipitates before subsequent layer deposition, while being at the 1000 °C build temperature or during cooling to RT. The phase diagram also shows that the equilibrium fraction of the γ′ increases as temperature decreases. For instance, at 1000, 900, and 800 °C, the phase fractions are ∼30%, 38%, and 42%, respectively.
Deposition of subsequent layers causes previous layers to undergo phase transformations as they are exposed to several thermal cycles with different peak temperatures. In Fig. 5c, as the subsequent layer is being deposited, γ′ in the previous layer (459th layer) begins to dissolve as the temperature crosses the solvus temperature. This is witnessed by the reduction of the γ′ phase fraction. This graph also shows how this phase dissolves during heating. However, the phase fraction of MC carbide remains stable at high temperatures and no dissolution is seen during thermal cycling. Upon cooling, the γ′ that was dissolved during heating reprecipitates with a surge in the phase fraction until 1000 °C, after which it remains constant. This microstructure is similar to the solidification microstructure (layer 460), with a similar γ′ phase fraction (∼27%).
The complete dissolution and reprecipitation of γ′ continue for several cycles until the 50th layer from the top (layer 411), where the phase fraction does not reach zero during heating to the peak temperature (see Fig. 5d). This indicates the ‘partial’ dissolution of γ′, which continues progressively with additional layers. It should be noted that the peak temperatures for layers that underwent complete dissolution were much higher (1170–1300 °C) than the γ′ solvus.
The dissolution and reprecipitation of γ′ during thermal cycling are further confirmed in Fig. 7, which summarizes the nucleation rate, phase fraction, and concentration of major elements that form γ′ in the matrix. Fig. 7b magnifies a single layer (3rd layer from top) within the full dissolution region in Fig. 7a to help identify the nucleation and growth mechanisms. From Fig. 7b, γ′ nucleation begins during cooling whereby the nucleation rate increases to reach a maximum value of approximately 1 × 1020 m−3s−1. This fast kinetics implies that some rearrangement of atoms is required for γ′ precipitates to form in the matrix [65], [66]. The matrix at this stage is in a non-equilibrium condition. Its composition is similar to the nominal composition and remains unchanged. The phase fraction remains insignificant at this stage although nucleation has started. The nucleation rate starts declining upon reaching the peak value. Simultaneously, diffusion-controlled growth of existing nuclei occurs, depleting the matrix of γ′ forming elements (Al and Ti). Thus, from (7), (11), ∆�vol continuously decreases until nucleation ceases. The growth of nuclei is witnessed by the increase in phase fraction until a constant level is reached at 27% upon cooling to and holding at build temperature. This nucleation event is repeated several times.
At the onset of partial dissolution, the nucleation rate jumps to 1 × 1021 m−3s−1, and then reduces sharply at the middle stage of partial dissolution. The nucleation rate reaches 0 at a later stage. Supplementary Fig. S1 shows a magnified view of the nucleation rate, phase fraction, and thermal profile, underpinning this trend. The jump in nucleation rate at the onset is followed by a progressive reduction in the solute content of the matrix. The peak temperatures (∼1130–1160 °C) are lower than those in complete dissolution regions but still above or close to the γ′ solvus. The maximum phase fraction (∼27%) is similar to that of the complete dissolution regions. At the middle stage, the reduction in nucleation rate is accompanied by a sharp drop in the matrix composition. The γ′ fraction drops to ∼24%, where the peak temperatures of the layers are just below or at γ′ solvus. The phase fraction then increases progressively through the later stage of partial dissolution to ∼30% towards the end of thermal cycling. The matrix solute content continues to drop although no nucleation event is seen. The peak temperatures are then far below the γ′ solvus. It should be noted that the matrix concentration after complete dissolution remains constant. Upon cooling to RT after final layer deposition, the nucleation rate increases again, indicating new nucleation events. The phase fraction reaches ∼40%, with a further depletion of the matrix in major γ′ forming elements.
3.3.2.2. γ′ size distribution
Fig. 8 shows histograms of the γ′ precipitate size distributions (PSD) along the build height during deposition. These PSDs are predicted at the end of each layer of interest just before final cooling to room temperature, to separate the role of thermal cycles from final cooling on the evolution of γ′. The PSD for the top layer (layer 460) is shown in Fig. 8a (last solidified region with solidification microstructure). The γ′ size ranges from 120 to 230 nm and is similar to the 44 layers below (2.2 mm from the top).
Further down the build, γ′ begins to coarsen after layer 417 (44th layer from top). Fig. 8c shows the PSD after the 44th layer, where the γ′ size exhibits two peaks at ∼120–230 and ∼300 nm, with most of the population being in the former range. This is the onset of partial dissolution where simultaneously with the reprecipitation and growth of fresh γ′, the undissolved γ′ grows rapidly through diffusive transport of atoms to the precipitates. This is shown in Fig. 8c, where the precipitate class sizes between 250 and 350 represent the growth of undissolved γ′. Although this continues in the 416th layer, the phase fractions plot indicates that the onset of partial dissolution begins after the 411th layer. This implies that partial dissolution started early, but the fraction of undissolved γ′ was too low to impact the phase fraction. The reprecipitated γ′ are mostly in the 100–220 nm class range and similar to those observed during full dissolution.
As the number of layers increases, coarsening intensifies with continued growth of more undissolved γ′, and reprecipitation and growth of partially dissolved ones. Fig. 8d, e, and f show this sequence. Further down the build, coarsening progresses rapidly, as shown in Figs. 8d, 8e, and 8f. The γ′ size ranges from 120 to 1100 nm, with the peaks at 160, 180, and 220 nm in Figs. 8d, 8e, and 8f, respectively. Coarsening continues until nucleation ends during dissolution, where only the already formed γ′ precipitates continue to grow during further thermal cycling. The γ′ size at this point is much larger, as observed in layers 361 and 261, and continues to increase steadily towards the bottom (layer 1). Two populations in the ranges of ∼380–700 and ∼750–1100 nm, respectively, can be seen. The steady growth of γ′ towards the bottom is confirmed by the gradual decrease in the concentration of solute elements in the matrix (Fig. 7a). It should be noted that for each layer, the γ′ class with the largest size originates from continuous growth of the earliest set of the undissolved precipitates.
Fig. 9, Fig. 10 and supplementary Figs. S2 and S3 show the γ′ size evolution during heating and cooling of a single layer in the full dissolution region, and early, middle stages, and later stages of partial dissolution, respectively. In all, the size of γ′ reduces during layer heating. Depending on the peak temperature of the layer which varies with build height, γ′ are either fully or partially dissolved as mentioned earlier. Upon cooling, the dissolved γ′ reprecipitate.
In Fig. 9, those layers that underwent complete dissolution (top layers) were held above γ′ solvus temperature for longer. In Fig. 10, layers at the early stage of partial dissolution spend less time in the γ′ solvus temperature region during heating, leading to incomplete dissolution. In such conditions, smaller precipitates are fully dissolved while larger ones shrink [67]. Layers in the middle stages of partial dissolution have peak temperatures just below or at γ′ solvus, not sufficient to achieve significant γ′ dissolution. As seen in supplementary Fig. S2, only a few smaller γ′ are dissolved back into the matrix during heating, i.e., growth of precipitates is more significant than dissolution. This explains the sharp decrease in concentration of Al and Ti in the matrix in this layer.
The previous sections indicate various phenomena such as an increase in phase fraction, further depletion of matrix composition, and new nucleation bursts during cooling. Analysis of the PSD after the final cooling of the build to room temperature allows a direct comparison to post-printing microstructural characterization. Fig. 11 shows the γ′ size distribution of layer 1 (460th layer from the top) after final cooling to room temperature. Precipitation of secondary γ′ is observed, leading to the multimodal size distribution of secondary and primary γ′. The secondary γ′ size falls within the 10–80 nm range. As expected, a further growth of the existing primary γ′ is also observed during cooling.
3.3.2.3. γ′ chemistry after deposition
Fig. 12 shows the concentration of the major elements that form γ′ (Al, Ti, and Ni) in the primary and secondary γ′ at the bottom of the build, as calculated by MatCalc. The secondary γ′ has a higher Al content (13.5–14.5 at% Al), compared to 13 at% Al in the primary γ′. Additionally, within the secondary γ′, the smallest particles (∼10 nm) have higher Al contents than larger ones (∼70 nm). In contrast, for the primary γ′, there is no significant variation in the Al content as a function of their size. The Ni concentration in secondary γ′ (71.1–72 at%) is also higher in comparison to the primary γ′ (70 at%). The smallest secondary γ′ (∼10 nm) have higher Ni contents than larger ones (∼70 nm), whereas there is no substantial change in the Ni content of primary γ′, based on their size. As expected, Ti shows an opposite size-dependent variation. It ranges from ∼ 7.7–8.7 at% Ti in secondary γ′ to ∼9.2 at% in primary γ′. Similarly, within the secondary γ′, the smallest (∼10 nm) have lower Al contents than the larger ones (∼70 nm). No significant variation is observed for Ti content in primary γ′.
4. Discussion
A combined modelling method is utilized to study the microstructural evolution during E-PBF of IN738. The presented results are discussed by examining the precipitation and dissolution mechanism of γ′ during thermal cycling. This is followed by a discussion on the phase fraction and size evolution of γ′ during thermal cycling and after final cooling. A brief discussion on carbide morphology is also made. Finally, a comparison is made between the simulation and experimental results to assess their agreement.
4.1. γ′ morphology as a function of build height
4.1.1. Nucleation of γ′
The fast precipitation kinetics of the γ′ phase enables formation of γ′ upon quenching from higher temperatures (above solvus) during thermal cycling [66]. In Fig. 7b, for a single layer in the full dissolution region, during cooling, the initial increase in nucleation rate signifies the first formation of nuclei. The slight increase in nucleation rate during partial dissolution, despite a decrease in the concentration of γ′ forming elements, may be explained by the nucleation kinetics. During partial dissolution and as the precipitates shrink, it is assumed that the regions at the vicinity of partially dissolved precipitates are enriched in γ′ forming elements [68], [69]. This differs from the full dissolution region, in which case the chemical composition is evenly distributed in the matrix. Several authors have attributed the solute supersaturation of the matrix around primary γ′ to partial dissolution during isothermal ageing [69], [70], [71], [72]. The enhanced supersaturation in the regions close to the precipitates results in a much higher driving force for nucleation, leading to a higher nucleation rate upon cooling. This phenomenon can be closely related to the several nucleation bursts upon continuous cooling of Ni-based superalloys, where second nucleation bursts exhibit higher nucleation rates [38], [68], [73], [74].
At middle stages of partial dissolution, the reduction in the nucleation rate indicates that the existing composition and low supersaturation did not trigger nucleation as the matrix was closer to the equilibrium state. The end of a nucleation burst means that the supersaturation of Al and Ti has reached a low level, incapable of providing sufficient driving force during cooling to or holding at 1000 °C for further nucleation [73]. Earlier studies on Ni-based superalloys have reported the same phenomenon during ageing or continuous cooling from the solvus temperature to RT [38], [73], [74].
4.1.2. Dissolution of γ′ during thermal cycling
γ′ dissolution kinetics during heating are fast when compared to nucleation due to exponential increase in phase transformation and diffusion activities with temperature [65]. As shown in Fig. 9, Fig. 10, and supplementary Figs. S2 and S3, the reduction in γ′ phase fraction and size during heating indicates γ′ dissolution. This is also revealed in Fig. 5 where phase fraction decreases upon heating. The extent of γ′ dissolution mostly depends on the temperature, time spent above γ′ solvus, and precipitate size[75], [76], [77]. Smaller γ′ precipitates are first to be dissolved [67], [77], [78]. This is mainly because more solute elements need to be transported away from large γ′ precipitates than from smaller ones [79]. Also, a high temperature above γ′ solvus temperature leads to a faster dissolution rate[80]. The equilibrium solvus temperature of γ′ in IN738 in our MatCalc simulation (Fig. 6) and as reported by Ojo et al. [47] is 1140 °C and 1130–1180 °C, respectively. This means the peak temperature experienced by previous layers decreases progressively from γ′ supersolvus to subsolvus, near-solvus, and far from solvus as the number of subsequent layers increases. Based on the above, it can be inferred that the degree of dissolution of γ′ contributes to the gradient in precipitate distribution.
Although the peak temperatures during later stages of partial dissolution are much lower than the equilibrium γ′ solvus, γ′ dissolution still occurs but at a significantly lower rate (supplementary Fig. S3). Wahlmann et al. [28] also reported a similar case where they observed the rapid dissolution of γ′ in CMSX-4 during fast heating and cooling cycles at temperatures below the γ′ solvus. They attributed this to the γ′ phase transformation process taking place in conditions far from the equilibrium. While the same reasoning may be valid for our study, we further believe that the greater surface area to volume ratio of the small γ′ precipitates contributed to this. This ratio means a larger area is available for solute atoms to diffuse into the matrix even at temperatures much below the solvus [81].
4.2. γ′ phase fraction and size evolution
4.2.1. During thermal cycling
In the first layer, the steep increase in γ′ phase fraction during heating (Fig. 5), which also represents γ′ precipitation in the powder before melting, has qualitatively been validated in [28]. The maximum phase fraction of 27% during the first few layers of thermal cycling indicates that IN738 theoretically could reach the equilibrium state (∼30%), but the short interlayer time at the build temperature counteracts this. The drop in phase fraction at middle stages of partial dissolution is due to the low number of γ′ nucleation sites [73]. It has been reported that a reduction of γ′ nucleation sites leads to a delay in obtaining the final volume fraction as more time is required for γ′ precipitates to grow and reach equilibrium [82]. This explains why even upon holding for 150 s before subsequent layer deposition, the phase fraction does not increase to those values that were observed in the previous full γ′ dissolution regions. Towards the end of deposition, the increase in phase fraction to the equilibrium value of 30% is as a result of the longer holding at build temperature or close to it [83].
During thermal cycling, γ′ particles begin to grow immediately after they first precipitate upon cooling. This is reflected in the rapid increase in phase fraction and size during cooling in Fig. 5 and supplementary Fig. S2, respectively. The rapid growth is due to the fast diffusion of solute elements at high temperatures [84]. The similar size of γ′ for the first 44 layers from the top can be attributed to the fact that all layers underwent complete dissolution and hence, experienced the same nucleation event and growth during deposition. This corresponds with the findings by Balikci et al. [85], who reported that the degree of γ′ precipitation in IN738LC does not change when a solution heat treatment is conducted above a certain critical temperature.
The increase in coarsening rate (Fig. 8) during thermal cycling can first be ascribed to the high peak temperature of the layers [86]. The coarsening rate of γ′ is known to increase rapidly with temperature due to the exponential growth of diffusion activity. Also, the simultaneous dissolution with coarsening could be another reason for the high coarsening rate, as γ′ coarsening is a diffusion-driven process where large particles grow by consuming smaller ones [78], [84], [86], [87]. The steady growth of γ′ towards the bottom of the build is due to the much lower layer peak temperature, which is almost close to the build temperature, and reduced dissolution activity, as is seen in the much lower solute concentration in γ′ compared to those in the full and partial dissolution regions.
4.2.2. During cooling
The much higher phase fraction of ∼40% upon cooling signifies the tendency of γ′ to reach equilibrium at lower temperatures (Fig. 4). This is due to the precipitation of secondary γ′ and a further increase in the size of existing primary γ′, which leads to a multimodal size distribution of γ′ after cooling [38], [73], [88], [89], [90]. The reason for secondary γ′ formation during cooling is as follows: As cooling progresses, it becomes increasingly challenging to redistribute solute elements in the matrix owing to their lower mobility [38], [73]. A higher supersaturation level in regions away from or free of the existing γ′ precipitates is achieved, making them suitable sites for additional nucleation bursts. More cooling leads to the growth of these secondary γ′ precipitates, but as the temperature and in turn, the solute diffusivity is low, growth remains slow.
4.3. Carbides
MC carbides in IN738 are known to have a significant impact on the high-temperature strength. They can also act as effective hardening particles and improve the creep resistance [91]. Precipitation of MC carbides in IN738 and several other superalloys is known to occur during solidification or thermal treatments (e.g., hot isostatic pressing) [92]. In our case, this means that the MC carbides within the E-PBF build formed because of the thermal exposure from the E-PBF thermal cycle in addition to initial solidification. Our simulation confirms this as MC carbides appear during layer heating (Fig. 5). The constant and stable phase fraction of MC carbides during thermal cycling can be attributed to their high melting point (∼1360 °C) and the short holding time at peak temperatures [75], [93], [94]. The solvus temperature for most MC carbides exceeds most of the peak temperatures observed in our simulation, and carbide dissolution kinetics at temperatures above the solvus are known to be comparably slow [95]. The stable phase fraction and random distribution of MC carbides signifies the slight influence on the gradient in hardness.
4.4. Comparison of simulations and experiments
4.4.1. Precipitate phase fraction and morphology as a function of build height
A qualitative agreement is observed for the phase fraction of carbides, i.e. ∼0.8% in the experiment and ∼0.9% in the simulation. The phase fraction of γ′ differs, with the experiment reporting a value of ∼51% and the simulation, 40%. Despite this, the size distribution of primary γ′ along the build shows remarkable consistency between experimental and computational analyses. It is worth noting that the primary γ′ morphology in the experimental analysis is observed in the as-fabricated state, whereas the simulation (Fig. 8) captures it during deposition process. The primary γ′ size in the experiment is expected to experience additional growth during the cooling phase. Regardless, both show similar trends in primary γ′ size increments from the top to the bottom of the build. The larger primary γ’ size in the simulation versus the experiment can be attributed to the fact that experimental and simulation results are based on 2D and 3D data, respectively. The absence of stereological considerations [96] in our analysis could have led to an underestimation of the precipitate sizes from SEM measurements. The early starts of coarsening (8th layer) in the experiment compared to the simulation (45th layer) can be attributed to a higher actual γ′ solvus temperature than considered in our simulation [47]. The solvus temperature of γ′ in a Ni-based superalloy is mainly determined by the detailed composition. A high amount of Cr and Co are known to reduce the solvus temperature, whereas Ta and Mo will increase it [97], [98], [99]. The elemental composition from our experimental work was used for the simulation except for Ta. It should be noted that Ta is not included in the thermodynamic database in MatCalc used, and this may have reduced the solvus temperature. This could also explain the relatively higher γ′ phase fraction in the experiment than in simulation, as a higher γ′ solvus temperature will cause more γ′ to precipitate and grow early during cooling [99], [100].
Another possible cause of this deviation can be attributed to the extent of γ′ dissolution, which is mainly determined by the peak temperature. It can be speculated that individual peak temperatures at different layers in the simulation may have been over-predicted. However, one needs to consider that the true thermal profile is likely more complicated in the actual E-PBF process [101]. For example, the current model assumes that the thermophysical properties of the material are temperature-independent, which is not realistic. Many materials, including IN738, exhibit temperature-dependent properties such as thermal conductivity, specific heat capacity, and density [102]. This means that heat transfer simulations may underestimate or overestimate the temperature gradients and cooling rates within the powder bed and the solidified part. Additionally, the model does not account for the reduced thermal diffusivity through unmelted powder, where gas separating the powder acts as insulation, impeding the heat flow [1]. In E-PBF, the unmelted powder regions with trapped gas have lower thermal diffusivity compared to the fully melted regions, leading to localized temperature variations, and altered solidification behavior. These limitations can impact the predictions, particularly in relation to the carbide dissolution, as the peak temperatures may be underestimated.
While acknowledging these limitations, it is worth emphasizing that achieving a detailed and accurate representation of each layer’s heat source would impose tough computational challenges. Given the substantial layer count in E-PBF, our decision to employ a semi-analytical approximation strikes a balance between computational feasibility and the capture of essential trends in thermal profiles across diverse build layers. In future work, a dual-calibration strategy is proposed to further reduce simulation-experiment disparities. By refining temperature-independent thermophysical property approximations and absorptivity in the heat source model, and by optimizing interfacial energy descriptions in the kinetic model, the predictive precision could be enhanced. Further refining the simulation controls, such as adjusting the precipitate class size may enhance quantitative comparisons between modeling outcomes and experimental data in future work.
4.4.2. Multimodal size distribution of γ′ and concentration
Another interesting feature that sees qualitative agreement between the simulation and the experiment is the multimodal size distribution of γ′. The formation of secondary γ′ particles in the experiment and most E-PBF Ni-based superalloys is suggested to occur at low temperatures, during final cooling to RT [16], [73], [90]. However, so far, this conclusion has been based on findings from various continuous cooling experiments, as the study of the evolution during AM would require an in-situ approach. Our simulation unambiguously confirms this in an AM context by providing evidence for secondary γ′ precipitation during slow cooling to RT. Additionally, it is possible to speculate that the chemical segregation occurring during solidification, due to the preferential partitioning of certain elements between the solid and liquid phases, can contribute to the multimodal size distribution during deposition [51]. This is because chemical segregation can result in variations in the local composition of superalloys, which subsequently affects the nucleation and growth of γ′. Regions with higher concentrations of alloying elements will encourage the formation of larger γ′ particles, while regions with lower concentrations may favor the nucleation of smaller precipitates. However, it is important to acknowledge that the elevated temperature during the E-PBF process will largely homogenize these compositional differences [103], [104].
A good correlation is also shown in the composition of major γ′ forming elements (Al and Ti) in primary and secondary γ′. Both experiment and simulation show an increasing trend for Al content and a decreasing trend for Ti content from primary to secondary γ′. The slight composition differences between primary and secondary γ′ particles are due to the different diffusivity of γ′ stabilizers at different thermal conditions [105], [106]. As the formation of multimodal γ′ particles with different sizes occurs over a broad temperature range, the phase chemistry of γ′ will be highly size dependent. The changes in the chemistry of various γ′ (primary, secondary, and tertiary) have received significant attention since they have a direct influence on the performance [68], [105], [107], [108], [109]. Chen et al. [108], [109], reported a high Al content in the smallest γ′ precipitates compared to the largest, while Ti showed an opposite trend during continuous cooling in a RR1000 Ni-based superalloy. This was attributed to the temperature and cooling rate at which the γ′ precipitates were formed. The smallest precipitates formed last, at the lowest temperature and cooling rate. A comparable observation is evident in the present investigation, where the secondary γ′ forms at a low temperature and cooling rate in comparison to the primary. The temperature dependence of γ′ chemical composition is further evidenced in supplementary Fig. S4, which shows the equilibrium chemical composition of γ′ as a function of temperature.
5. Conclusions
A correlative modelling approach capable of predicting solid-state phase transformations kinetics in metal AM was developed. This approach involves computational simulations with a semi-analytical heat transfer model and the MatCalc thermo-kinetic software. The method was used to predict the phase transformation kinetics and detailed morphology and chemistry of γ′ and MC during E-PBF of IN738 Ni-based superalloy. The main conclusions are:
1.The computational simulations are in qualitative agreement with the experimental observations. This is particularly true for the γ′ size distribution along the build height, the multimodal size distribution of particles, and the phase fraction of MC carbides.
2.The deviations between simulation and experiment in terms of γ′ phase fraction and location in the build are most likely attributed to a higher γ′ solvus temperature during the experiment than in the simulation, which is argued to be related to the absence of Ta in the MatCalc database.
3.The dissolution and precipitation of γ′ occur fast and under non-equilibrium conditions. The level of γ′ dissolution determines the gradient in γ′ size distribution along the build. After thermal cycling, the final cooling to room temperature has further significant impacts on the final γ′ size, morphology, and distribution.
4.A negligible amount of γ′ forms in the first deposited layer before subsequent layer deposition, and a small amount of γ′ may also form in the powder induced by the 1000 °C elevated build temperature before melting.
Our findings confirm the suitability of MatCalc to predict the microstructural evolution at various positions throughout a build in a Ni-based superalloy during E-PBF. It also showcases the suitability of a tool which was originally developed for traditional thermo-mechanical processing of alloys to the new additive manufacturing context. Our simulation capabilities are likely extendable to other alloy systems that undergo solid-state phase transformations implemented in MatCalc (various steels, Ni-based superalloys, and Al-alloys amongst others) as well as other AM processes such as L-DED and L-PBF which have different thermal cycle characteristics. New tools to predict the microstructural evolution and properties during metal AM are important as they provide new insights into the complexities of AM. This will enable control and design of AM microstructures towards advanced materials properties and performances.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This research was sponsored by the Department of Industry, Innovation, and Science under the auspices of the AUSMURI program – which is a part of the Commonwealth’s Next Generation Technologies Fund. The authors acknowledge the facilities and the scientific and technical assistance at the Electron Microscope Unit (EMU) within the Mark Wainwright Analytical Centre (MWAC) at UNSW Sydney and Microscopy Australia. Nana Adomako is supported by a UNSW Scientia PhD scholarship. Michael Haines’ (UNSW Sydney) contribution to the revised version of the original manuscript is thankfully acknowledged.
[1]T. Debroy, H.L. Wei, J.S. Zuback, T. Mukherjee, J.W. Elmer, J.O. Milewski, A.M. Beese, A. Wilson-heid, A. De, W. ZhangAdditive manufacturing of metallic components – process, structure and propertiesProg. Mater. Sci., 92 (2018), pp. 112-224, 10.1016/j.pmatsci.2017.10.001View PDFView articleView in ScopusGoogle Scholar
[4]N.K. Adomako, J.J. Lewandowski, B.M. Arkhurst, H. Choi, H.J. Chang, J.H. KimMicrostructures and mechanical properties of multi-layered materials composed of Ti-6Al-4V, vanadium, and 17–4PH stainless steel produced by directed energy depositionAddit. Manuf., 59 (2022), Article 103174, 10.1016/j.addma.2022.103174View PDFView articleView in ScopusGoogle Scholar
[5]H. Wang, Z.G. Zhu, H. Chen, A.G. Wang, J.Q. Liu, H.W. Liu, R.K. Zheng, S.M.L. Nai, S. Primig, S.S. Babu, S.P. Ringer, X.Z. LiaoEffect of cyclic rapid thermal loadings on the microstructural evolution of a CrMnFeCoNi high-entropy alloy manufactured by selective laser meltingActa Mater., 196 (2020), pp. 609-625, 10.1016/J.ACTAMAT.2020.07.006View PDFView articleView in ScopusGoogle Scholar
[10]S.S. Al-Bermani, M.L. Blackmore, W. Zhang, I. ToddThe origin of microstructural diversity, texture, and mechanical properties in electron beam melted Ti-6Al-4VMetall. Mater. Trans. A Phys. Metall. Mater. Sci., 41 (2010), pp. 3422-3434, 10.1007/s11661-010-0397-xView article View in ScopusGoogle Scholar
[13]H. Helmer, A. Bauereiß, R.F. Singer, C. KörnerErratum to: ‘Grain structure evolution in Inconel 718 during selective electron beam melting’ (Materials Science & Engineering A (2016) 668 (180–187 (S0921509316305536) (10.1016/j.msea.2016.05.046))Mater. Sci. Eng. A., 676 (2016), p. 546, 10.1016/j.msea.2016.09.016View PDFView articleView in ScopusGoogle Scholar
[16]N. Haghdadi, E. Whitelock, B. Lim, H. Chen, X. Liao, S.S. Babu, S.P. Ringer, S. PrimigMultimodal γ′ precipitation in Inconel-738 Ni-based superalloy during electron-beam powder bed fusion additive manufacturingJ. Mater. Sci., 55 (2020), pp. 13342-13350, 10.1007/s10853-020-04915-wView article View in ScopusGoogle Scholar
[17]B. Lim, H. Chen, Z. Chen, N. Haghdadi, X. Liao, S. Primig, S.S. Babu, A. Breen, S.P. RingerMicrostructure–property gradients in Ni-based superalloy (Inconel 738) additively manufactured via electron beam powder bed fusionAddit. Manuf. (2021), Article 102121, 10.1016/j.addma.2021.102121View PDFView articleView in ScopusGoogle Scholar
[18]P. Karimi, E. Sadeghi, P. Åkerfeldt, J. Ålgårdh, J. AnderssonInfluence of successive thermal cycling on microstructure evolution of EBM-manufactured alloy 718 in track-by-track and layer-by-layer designMater. Des., 160 (2018), pp. 427-441, 10.1016/j.matdes.2018.09.038View PDFView articleView in ScopusGoogle Scholar
[19]E. Chauvet, P. Kontis, E.A. Jägle, B. Gault, D. Raabe, C. Tassin, J.J. Blandin, R. Dendievel, B. Vayre, S. Abed, G. MartinHot cracking mechanism affecting a non-weldable Ni-based superalloy produced by selective electron Beam MeltingActa Mater., 142 (2018), pp. 82-94, 10.1016/j.actamat.2017.09.047View PDFView articleView in ScopusGoogle Scholar
[20]M. Ramsperger, R.F. Singer, C. KörnerMicrostructure of the nickel-base superalloy CMSX-4 fabricated by selective electron beam meltingMetall. Mater. Trans. A Phys. Metall. Mater. Sci., 47 (2016), pp. 1469-1480, 10.1007/s11661-015-3300-y View PDF This article is free to access.View in ScopusGoogle Scholar
[21]B. Zhang, P. Wang, Y. Chew, Y. Wen, M. Zhang, P. Wang, G. Bi, J. WeiMechanical properties and microstructure evolution of selective laser melting Inconel 718 along building direction and sectional dimensionMater. Sci. Eng. A, 794 (2020), Article 139941, 10.1016/j.msea.2020.139941View PDFView articleView in ScopusGoogle Scholar
[22]C. Körner, M. Ramsperger, C. Meid, D. Bürger, P. Wollgramm, M. Bartsch, G. EggelerMicrostructure and mechanical properties of CMSX-4 single crystals prepared by additive manufacturingMetall. Mater. Trans. A Phys. Metall. Mater. Sci., 49 (2018), pp. 3781-3792, 10.1007/s11661-018-4762-5 View PDF This article is free to access.View in ScopusGoogle Scholar
[23]B. Lim, H. Chen, K. Nomoto, Z. Chen, A.I. Saville, S. Vogel, A.J. Clarke, A. Paradowska, M. Reid, S. Primig, X. Liao, S.S. Babu, A.J. Breen, S.P. RingerAdditively manufactured Haynes-282 monoliths containing thin wall struts of varying thicknessesAddit. Manuf., 59 (2022), Article 103120, 10.1016/j.addma.2022.103120View PDFView articleView in ScopusGoogle Scholar
[24]C.L.A. Leung, S. Marussi, R.C. Atwood, M. Towrie, P.J. Withers, P.D. LeeIn situ X-ray imaging of defect and molten pool dynamics in laser additive manufacturingNat. Commun., 9 (2018), pp. 1-9, 10.1038/s41467-018-03734-7 View PDF This article is free to access.View in ScopusGoogle Scholar
[25]C. Zhao, K. Fezzaa, R.W. Cunningham, H. Wen, F. De Carlo, L. Chen, A.D. Rollett, T. SunReal-time monitoring of laser powder bed fusion process using high-speed X-ray imaging and diffractionSci. Rep., 7 (2017), pp. 1-11, 10.1038/s41598-017-03761-2 View PDF This article is free to access.View in ScopusGoogle Scholar
[26]C. Kenel, D. Grolimund, X. Li, E. Panepucci, V.A. Samson, D.F. Sanchez, F. Marone, C. LeinenbachIn situ investigation of phase transformations in Ti-6Al-4V under additive manufacturing conditions combining laser melting and high-speed micro-X-ray diffractionSci. Rep., 7 (2017), pp. 1-10, 10.1038/s41598-017-16760-0 View PDF This article is free to access.Google Scholar
[27]W.L. Bevilaqua, J. Epp, H. Meyer, J. Dong, H. Roelofs, A. da, S. Rocha, A. RegulyRevealing the dynamic transformation of austenite to bainite during uniaxial warm compression through in-situ synchrotron X-ray diffractionMetals, 11 (2021), pp. 1-14, 10.3390/met11030467View article View in ScopusGoogle Scholar
[28]B. Wahlmann, E. Krohmer, C. Breuning, N. Schell, P. Staron, E. Uhlmann, C. KörnerIn situ observation of γ′ phase transformation dynamics during selective laser melting of CMSX-4Adv. Eng. Mater., 23 (2021), 10.1002/adem.202100112 View PDF This article is free to access.Google Scholar
[35]M.J. Anderson, J. Benson, J.W. Brooks, B. Saunders, H.C. BasoaltoPredicting precipitation kinetics during the annealing of additive manufactured Inconel 625 componentsIntegr. Mater. Manuf. Innov., 8 (2019), pp. 154-166, 10.1007/S40192-019-00134-7/FIGURES/11 View PDFThis article is free to access.View in ScopusGoogle Scholar
[36]H.C. Basoalto, C. Panwisawas, Y. Sovani, M.J. Anderson, R.P. Turner, B. Saunders, J.W. BrooksA computational study on the three-dimensional printability of precipitate-strengthened nickel-based superalloysProc. R. Soc. A, 474 (2018), 10.1098/RSPA.2018.0295View article Google Scholar
[37]K. McNamara, Y. Ji, F. Lia, P. Promoppatum, S.C. Yao, H. Zhou, Y. Wang, L.Q. Chen, R.P. MartukanitzPredicting phase transformation kinetics during metal additive manufacturing using non-isothermal Johnson-Mehl-Avrami models: application to Inconel 718 and Ti-6Al-4VAddit. Manuf., 49 (2022), Article 102478, 10.1016/J.ADDMA.2021.102478View PDFView articleView in ScopusGoogle Scholar
[39]A. Drexler, B. Oberwinkler, S. Primig, C. Turk, E. Povoden-karadeniz, A. Heinemann, W. Ecker, M. StockingerMaterials Science & Engineering A Experimental and numerical investigations of the γ ″ and γ ′ precipitation kinetics in Alloy 718Mater. Sci. Eng. A., 723 (2018), pp. 314-323, 10.1016/j.msea.2018.03.013View PDFView articleView in ScopusGoogle Scholar
[44]A.V. Sotov, A.V. Agapovichev, V.G. Smelov, V.V. Kokareva, M.O. Dmitrieva, A.A. Melnikov, S.P. Golanov, Y.M. AnurovInvestigation of the IN-738 superalloy microstructure and mechanical properties for the manufacturing of gas turbine engine nozzle guide vane by selective laser meltingInt. J. Adv. Manuf. Technol., 107 (2020), pp. 2525-2535, 10.1007/s00170-020-05197-xView article View in ScopusGoogle Scholar
[49]S. Sanchez, P. Smith, Z. Xu, G. Gaspard, C.J. Hyde, W.W. Wits, I.A. Ashcroft, H. Chen, A.T. ClarePowder Bed Fusion of nickel-based superalloys: a reviewInt. J. Mach. Tools Manuf., 165 (2021), 10.1016/j.ijmachtools.2021.103729 View PDF This article is free to access.Google Scholar
[50]C.L.A. Leung, R. Tosi, E. Muzangaza, S. Nonni, P.J. Withers, P.D. LeeEffect of preheating on the thermal, microstructural and mechanical properties of selective electron beam melted Ti-6Al-4V componentsMater. Des., 174 (2019), Article 107792, 10.1016/j.matdes.2019.107792View PDFView articleView in ScopusGoogle Scholar
[51]S. Griffiths, H. Ghasemi Tabasi, T. Ivas, X. Maeder, A. De Luca, K. Zweiacker, R. Wróbel, J. Jhabvala, R.E. Logé, C. LeinenbachCombining alloy and process modification for micro-crack mitigation in an additively manufactured Ni-base superalloyAddit. Manuf., 36 (2020), 10.1016/j.addma.2020.101443View article Google Scholar
[52]P. Soille, L. Vincent Pierre Soille, L.M. Vincent, Determining watersheds in digital pictures via flooding simulations, Https://Doi.Org/10.1117/12.24211. 1360 (1990) 240–250. https://doi.org/10.1117/12.24211.Google Scholar
[53]ASTM Standard Test Method for Microindentation Hardness of MaterialsKnoop and Vickers Hardness of Materials 1, Annu. B ASTM Stand. i 2010 1 42.Google Scholar
[56]B. Sonderegger, E. KozeschnikGeneralized nearest-neighbor broken-bond analysis of randomly oriented coherent interfaces in multicomponent Fcc and Bcc structuresMetall. Mater. Trans. A Phys. Metall. Mater. Sci., 40 (2009), pp. 499-510, 10.1007/S11661-008-9752-6/FIGURES/8View articleView in ScopusGoogle Scholar
[58]B. Sonderegger, E. KozeschnikInterfacial energy of diffuse phase boundaries in the generalized broken-bond approachMetall. Mater. Trans. A Phys. Metall. Mater. Sci., 41 (2010), pp. 3262-3269, 10.1007/S11661-010-0370-8/FIGURES/4View articleView in ScopusGoogle Scholar
[67]P. Strunz, M. Petrenec, J. Polák, U. Gasser, G. FarkasFormation and dissolution of’ precipitates in IN792 superalloy at elevated temperaturesMetals, 6 (2016), 10.3390/met6020037View article Google Scholar
[68]A.R.P. Singh, S. Nag, J.Y. Hwang, G.B. Viswanathan, J. Tiley, R. Srinivasan, H.L. Fraser, R. BanerjeeInfluence of cooling rate on the development of multiple generations of γ′ precipitates in a commercial nickel base superalloyMater. Charact., 62 (2011), pp. 878-886, 10.1016/j.matchar.2011.06.002View PDFView articleView in ScopusGoogle Scholar
[69]E. Balikci, A. Raman, R. MirshamsMicrostructure evolution in polycrystalline IN738LC in the range 1120 to 1250C, Zeitschrift FuerMet, 90 (1999), pp. 132-140View in ScopusGoogle Scholar
[71]Ł. Rakoczy, M. Grudzień-Rakoczy, F. Hanning, G. Cempura, R. Cygan, J. Andersson, A. Zielińska-LipiecInvestigation of the γ′ precipitates dissolution in a Ni-based superalloy during stress-free short-term annealing at high homologous temperaturesMetall. Mater. Trans. A Phys. Metall. Mater. Sci., 52 (2021), pp. 4767-4784, 10.1007/s11661-021-06420-4 View PDF This article is free to access.View in ScopusGoogle Scholar
[73]F. Masoumi, D. Shahriari, M. Jahazi, J. Cormier, A. DevauxKinetics and Mechanisms of γ′ Reprecipitation in a Ni-based SuperalloySci. Rep., 6 (2016), pp. 1-16, 10.1038/srep28650View articleGoogle Scholar
[74]A.R.P. Singh, S. Nag, S. Chattopadhyay, Y. Ren, J. Tiley, G.B. Viswanathan, H.L. Fraser, R. BanerjeeMechanisms related to different generations of γ′ precipitation during continuous cooling of a nickel base superalloyActa Mater., 61 (2013), pp. 280-293, 10.1016/j.actamat.2012.09.058View PDFView articleView in ScopusGoogle Scholar
[76]N.D. Souza, M.C. Hardy, B. Roebuck, W.E.I. Li, G.D. West, D.M. Collins, On the Rate Dependence of Precipitate Formation and Dissolution in a Nickel-Base Superalloy, Metall. Mater. Trans. A. (n.d.). https://doi.org/10.1007/s11661–022-06680–8.Google Scholar
[79]H. Huang, G. Liu, H. Wang, A. Ullah, B. HuDissolution behavior and kinetics of γ′ phase during solution treatment in powder metallurgy nickel-based superalloyMetall. Mater. Trans. A Phys. Metall. Mater. Sci., 51 (2020), pp. 1075-1084, 10.1007/s11661-019-05581-7View article View in ScopusGoogle Scholar
[80]A.J. Goodfellow, E.I. Galindo-Nava, K.A. Christofidou, N.G. Jones, T. Martin, P.A.J. Bagot, C.D. Boyer, M.C. Hardy, H.J. StoneGamma prime precipitate evolution during aging of a model nickel-based superalloyMetall. Mater. Trans. A Phys. Metall. Mater. Sci., 49 (2018), pp. 718-728, 10.1007/s11661-017-4336-y View PDF This article is free to access.View in ScopusGoogle Scholar
[81]T.P. Gabb, D.G. Backman, D.Y. Wei, D.P. Mourer, D. Furrer, A. Garg, D.L. Ellis, #947;’ Form. a Nickel-Base Disk Superalloy 2012 405 414 doi: 10.7449/2000/superalloys_2000_405_414.Google Scholar
[82]A. PlatiModelling of γ precipitation in superalloys University of CambridgeMater. Sci. (2003), p. 73Google Scholar
[91]F. Theska, W.F. Tse, B. Schulz, R. Buerstmayr, S.R. Street, M. Lison-Pick, S. PrimigReview of microstructure–mechanical property relationships in cast and wrought ni-based superalloys with boron, carbon, and zirconium microalloying additionsAdv. Eng. Mater. (2022), p. 2201514, 10.1002/ADEM.202201514 View PDF This article is free to access.Google Scholar
[93]L. Zhang, Y. Li, Q. Zhang, S. ZhangMicrostructure evolution, phase transformation and mechanical properties of IN738 superalloy fabricated by selective laser melting under different heat treatmentsMater. Sci. Eng. A, 844 (2022), Article 142947, 10.1016/j.msea.2022.142947View PDFView articleView in ScopusGoogle Scholar
[94]J.C. Franco-Correa, E. Martínez-Franco, J.M. Alvarado-Orozco, L.A. Cáceres-Díaz, D.G. Espinosa-Arbelaez, J.A. VilladaEffect of conventional heat treatments on the microstructure and microhardness of IN718 obtained by wrought and additive manufacturingJ. Mater. Eng. Perform., 30 (2021), pp. 7035-7045, 10.1007/s11665-021-06138-9View article View in ScopusGoogle Scholar
[96]N. Li, M.J. Anderson, H.C. BasoaltoAutomated stereology and uncertainty quantification considering spherical non-penetrating dispersionsPage 464.Cryst 2023, Vol. 13 (13) (2023), p. 464, 10.3390/CRYST13030464View article Google Scholar
[97]W.T. Loomis, J.W. Freeman, D.L. SponsellerInfluence of molybdenum on the γ′- phase in experimental nickelbase superalloysMet. Trans., 3 (1972), pp. 989-1000Google Scholar
[98]A.S. Shaikh Development of a γ’ Precipitation Hardening Ni-Base Superalloy for Additive Manufacturing Thesis 2018 102.〈https://odr.chalmers.se/handle/20.500.12380/255645%0Ahttps://www.researchgate.net/profile/Abdul_Shaafi_Shaikh2/publication/326226200%0Ahttps://drive.google.com/open?id=1BIez-aJyBTjnSgazzzTvv3jlrYpFC0N-〉.Google Scholar
[102]P.N. Quested, R.F. Brooks, L. Chapman, R. Morrell, Y. Youssef, K.C. MillsMeasurement and estimation of thermophysical properties of nickel based superalloysMater. Sci. Technol., 25 (2009), pp. 154-162, 10.1179/174328408×361454View articleView in ScopusGoogle Scholar
액체-증기 상 변화 모델은 밀폐된 용기의 자체 가압 프로세스 시뮬레이션에 매우 큰 영향을 미칩니다. Hertz-Knudsen 관계, 에너지 점프 모델 및 그 파생물과 같은 널리 사용되는 액체-증기 상 변화 모델은 실온 유체를 기반으로 개발되었습니다. 액체-증기 전이를 통한 극저온 시뮬레이션에 널리 적용되었지만 각 모델의 성능은 극저온 조건에서 명시적으로 조사 및 비교되지 않았습니다. 본 연구에서는 171가지 일반적인 액체-증기 상 변화 모델을 통합한 통합 다상 솔버가 제안되었으며, 이를 통해 이러한 모델을 실험 데이터와 직접 비교할 수 있습니다. 증발 및 응축 모델의 예측 정확도와 계산 속도를 평가하기 위해 총 <>개의 자체 가압 시뮬레이션이 수행되었습니다. 압력 예측은 최적화 전략이 서로 다른 모델 계수에 크게 의존하는 것으로 나타났습니다. 에너지 점프 모델은 극저온 자체 가압 시뮬레이션에 적합하지 않은 것으로 나타났습니다. 평균 편차와 CPU 소비량에 따르면 Lee 모델과 Tanasawa 모델은 다른 모델보다 안정적이고 효율적인 것으로 입증되었습니다.
Liquid-vapor phase change models vitally influence the simulation of self-pressurization processes in closed containers. Popular liquid-vapor phase change models, such as the Hertz-Knudsen relation, energy jump model, and their derivations were developed based on room-temperature fluids. Although they had widely been applied in cryogenic simulations with liquid-vapor transitions, the performance of each model was not explicitly investigated and compared yet under cryogenic conditions. A unified multi-phase solver incorporating four typical liquid-vapor phase change models has been proposed in the present study, which enables direct comparison among those models against experimental data. A total number of 171 self-pressurization simulations were conducted to evaluate the evaporation and condensation models’ prediction accuracy and calculation speed. It was found that the pressure prediction highly depended on the model coefficients, whose optimization strategies differed from each other. The energy jump model was found inadequate for cryogenic self-pressurization simulations. According to the average deviation and CPU consumption, the Lee model and the Tanasawa model were proven to be more stable and more efficient than the others.
Introduction
The liquid-vapor phase change of cryogenic fluids is widely involved in industrial applications, such as the hydrogen transport vehicles [1], shipborne liquid natural gas (LNG) containers [2] and on-orbit cryogenic propellant tanks [3]. These applications require cryogenic fluids to be stored for weeks to months. Although high-performance insulation measures are adopted, heat inevitably enters the tank via radiation and conduction. The self-pressurization in the tank induced by the heat leakage eventually causes the venting loss of the cryogenic fluids and threatens the safety of the craft in long-term missions. To reduce the boil-off loss and extend the cryogenic storage duration, a more comprehensive understanding of the self-pressurization mechanism is needed.
Due to the difficulties and limitations in implementing cryogenic experiments, numerical modeling is a convenient and powerful way to study the self-pressurization process of cryogenic fluids. However, how the phase change models influence the mass and heat transfer under cryogenic conditions is still unsettled [4]. As concluded by Persad and Ward [5], a seemingly slight variation in the liquid-vapor phase change models can lead to erroneous predictions.
Among the liquid-vapor phase change models, the kinetic theory gas (KTG) based models and the energy jump model are the most popular ones used in recent self-pressurization simulations [6]. The KTG based models, also known as the Hertz-Knudsen relation models, were developed on the concept of the Maxwell-Boltzmann distribution of the gas molecular [7]. The Hertz-Knudsen relation has evolved to several models, including the Schrage model [8], the Tanasawa model [9], the Lee model [10] and the statistical rate theory (SRT) [11], which will be described in Section 2.2. Since the Schrage model and the Lee model are embedded and configured as the default ones in the commercial CFD solvers Flow-3D® and Ansys Fluent® respectively, they have been widely used in self-pressurization simulations for liquid nitrogen [12], [13] and liquid hydrogen [14], [15]. The major drawback of the KTG models lies in the difficulty of selecting model coefficients, which were reported in a considerably wide range spanning three magnitudes even for the same working fluid [16], [17], [18], [19], [20], [21]. Studies showed that the liquid level, pressure and mass transfer rate are directly influenced by the model coefficients [16], [22], [23], [24], [25]. Wrong coefficients will lead to deviation or even divergence of the results. The energy jump model is also known as the thermal limitation model. It assumes that the evaporation and condensation at the liquid-vapor interface are induced only by heat conduction. The model is widely adopted in lumped node simulations due to its simplicity [6], [26], [27]. To improve the accuracy of mass flux prediction, the energy jump model was modified by including the convection heat transfer [28], [29]. However, the convection correlations are empirical and developed mainly for room-temperature fluids. Whether the correlation itself can be precisely applied in cryogenic simulations still needs further investigation.
Fig. 1 summarizes the cryogenic simulations involving the modeling of evaporation and condensation processes in recent years. The publication has been increasing rapidly. However, the characteristics of each evaporation and condensation model are not explicitly revealed when simulating self-pressurization. A comparative study of the phase change models is highly needed for cryogenic fluids for a better simulation of the self-pressurization processes.
In the present paper, a unified multi-phase solver incorporating four typical liquid-vapor phase change models, namely the Tanasawa model, the Lee model, the energy jump model, and the modified energy jump model has been proposed, which enables direct comparison among different models. The models are used to simulate the pressure and temperature evolutions in an experimental liquid nitrogen tank in normal gravity, which helps to evaluate themselves in the aspects of accuracy, calculation speed and robustness.
Section snippets
Governing equations for the self-pressurization tank
In the present study, both the fluid domain and the solid wall of the tank are modeled and discretized. The heat transportation at the solid boundaries is considered to be irrelevant with the nearby fluid velocity. Consequently, two sets of the solid and the fluid governing equations can be decoupled and solved separately. The pressures in the cryogenic container are usually from 100 kPa to 300 kPa. Under these conditions, the Knudsen number is far smaller than 0.01, and the fluids are
Self-pressurization results and phase change model comparison
This section compares the simulation results by different phase change models. Section 3.1 compares the pressure and temperature outputs from two KTG based models, namely the Lee model and the Tanasawa model. Section 3.2 presents the pressure predictions from the energy transport models, namely the energy jump model and the modified energy jump model, and compares pressure prediction performances between the KTG based models and the energy transport models. Section 3.3 evaluates the four models
Conclusion
A unified vapor-liquid-solid multi-phase numerical solver has been accomplished for the self pressurization simulation in cryogenic containers. Compared to the early fluid-only solver, the temperature prediction in the vicinity of the tank wall improves significantly. Four liquid-vapor phase change models were integrated into the solver, which enables fair and effective comparison for performances between each other. The pressure and temperature prediction accuracies, and the calculation speed
Authors declare that they have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled, “Validity evaluation of popular liquid-vapor phase change models for cryogenic self-pressurization process”.
Acknowledgement
This project is supported by the National Natural Science Foundation of China (No. 51936006).
Progress in physical modelling and numerical simulation of phase transitions in cryogenic pool boiling and cavitation2023, Applied Mathematical ModellingCitation Excerpt :We will not delve into cryogenic evaporation phenomena, that predominantly drive the phase-change in well-insulated storage facilities, and thus are less relevant to spill scenarios. We instead refer the reader to the works of Zuo et al. [31–33]. If the static pressure at any location in a turbomachine drops below a fluid’s saturation pressure, localized evaporation events may occur, followed by rapid collapse of the vapour cavities in a process termed “cavitation” [34].Show abstract
Thermodynamic performance in a liquid oxygen tank during active-pressurization under different gas injection temperatures2023, International Communications in Heat and Mass TransferCitation Excerpt :The volume of fluid method is adopted to predict the tank pressurization performance. The associated governing equations could refer to previous published investigations [33–39,41,45,46]. Subjected to external heat input and gas injection, the phase change occurs at the interface and within the tank.Show abstract
Interfacial mass and energy transport during steady-state evaporation in liquid oxygen storage tanks2022, Applied EnergyCitation Excerpt :However, most of them simply used the Lee model for mass transport as did for regular fluids, and seldom focus on the evaporation itself related to the interfacial temperature distribution or were unable to validate their results against credible experimental data. A recent study proposed an optimized evaporation model for the cryogenic self-pressurization with a thorough comparison between popular phase change models [8], but still lacked of experimental data to validate the results. A series of experiments have been conducted on the heat and mass transport in a thin liquid layer in the vicinity of the liquid–vapor interface of room-temperature fluids [9–14].Show abstract
Thermal destratification of cryogenic liquid storage tanks by continuous bubbling of gases2022, International Journal of Hydrogen EnergyCitation Excerpt :It was concluded that a single injector with a larger diameter configuration showed a higher chance of developing a vertical temperature gradient. Zuo et al. [48] carried out a numerical analysis to investigate the temperature distribution within the LH2 storage tank with a self-pinning spraying bar. They used the SST turbulence model coupled with the 6-DOF model.Show abstract
Analyses of Cryogenic Propellant Tank Pressurization based upon Experiments and Numerical Simulations Carina Ludwig? and Michael Dreyer** *DLR – German Aerospace Center, Space Launcher Systems Analysis (SART), Institute of Space Systems, 28359 Bremen, Germany, Carina.Ludwig@dlr.de **ZARM – Center for Applied Space Technology and Microgravity, University of Bremen, 28359 Bremen, Germany
Abstract
본 연구에서는 발사대 적용을 위한 극저온 추진제 탱크의 능동 가압을 분석하였다. 따라서 지상 실험, 수치 시뮬레이션 및 분석 연구를 수행하여 다음과 같은 중요한 결과를 얻었습니다.
필요한 가압 기체 질량을 최소화하기 위해 더 높은 가압 기체 온도가 유리하거나 헬륨을 가압 기체로 적용하는 것이 좋습니다.
Flow-3D를 사용한 가압 가스 질량의 수치 시뮬레이션은 실험 결과와 잘 일치함을 보여줍니다. 가압 중 지배적인 열 전달은 주입된 가압 가스에서 축방향 탱크 벽으로 나타나고 능동 가압 단계 동안 상 변화의 주된 방식은 가압 가스의 유형에 따라 다릅니다.
가압 단계가 끝나면 상당한 압력 강하가 발생합니다. 이 압력 강하의 분석적 결정을 위해 이론적 모델이 제공됩니다.
The active-pressurization of cryogenic propellant tanks for the launcher application was analyzed in this study. Therefore, ground experiments, numerical simulations and analytical studies were performed with the following important results: In order to minimize the required pressurant gas mass, a higher pressurant gas temperature is advantageous or the application of helium as pressurant gas. Numerical simulations of the pressurant gas mass using Flow-3D show good agreement to the experimental results. The dominating heat transfer during pressurization appears from the injected pressurant gas to the axial tank walls and the predominant way of phase change during the active-pressurization phase depends on the type of the pressurant gas. After the end of the pressurization phase, a significant pressure drop occurs. A theoretical model is presented for the analytical determination of this pressure drop.
Figure 1: Drawing of the experimental set-up, Figure 2: Experimental tank with locations of temperature sensorsFigure 3: Non-dimensional (a) tank pressure, (b) liquid temperatures, (c) vapor temperatures, (d) wall and lid temperatures during pressurization and relaxation of the N300h experiment (for details see Table 2). T14 is the pressurant
gas temperature at the diffuser. Pressurization starts at tp,0 (t
∗ = 0.06·10−4
) and ends at tp, f (t
∗ = 0.84·10−4
). Relaxation
takes place until tp,T (t
∗ = 2.79·10−4
) and ∆p is the characteristic pressure dropFigure 5: Nondimensional vapor mass at pressurization start (m
∗
v,0
), pressurant gas mass (m
∗
pg), condensed vapor mass
from pressurization start to pressurization end (m
∗
cond,0,f
) and condensed vapor mass from pressurization end to relaxation end (m
∗
cond, f,T
) for all GN2 (a) and the GHe (b) pressurized experiments with the relating errors.Figure 6: Schematical propellant tank with vapor and liquid phase, pressurant gas and condensation mass flow as well as the applied control volumes. ., Figure 7: N300h experiment: wall to fluid heat flux at pressurization end (tp, f) over the tank height.
References
[1] M.E. Nein and R.R. Head. Experiences with pressurized discharge of liquid oxygen from large flight vehicle propellant tanks. In Advances in Cryogenig Engineering, vol. 7, New York, Plenum Press, 244–250. [2] M.E. Nein and J.F. Thompson. Experimental and analytical studies of cryogenic propellant tank pressurant requirements: NASA TN D-3177, 1966. [3] R.J. Stochl, J.E. Maloy, P.A. Masters and R.L. DeWitt. Gaseous-helium requirements for the discharge of liquid hydrogen from a 1.52-meter- (5-ft-) diameter spherical tank: NASA TN D-5621, 1970. [4] R.J. Stochl, J.E. Maloy, P.A. Masters and R.L. DeWitt. Gaseous-helium requirements for the discharge of liquid hydrogen from a 3.96-meter- (13-ft-) diameter spherical tank: NASA TN D-7019, 1970. [5] R.J. Stochl, P.A. Masters, R.L. DeWitt and J.E. Maloy. Gaseous-hydrogen requirements for the discharge of liquid hydrogen from a 1.52-meter- (5-ft-) diameter spherical tank: NASA TN D-5336, 1969. [6] R.J. Stochl, P.A. Masters, R.L. DeWitt and J.E. Maloy. Gaseous-hydrogen requirements for the discharge of liquid hydrogen from a 3.96-meter- (13-ft-) diameter spherical tank: NASA TN D-5387, 1969. [7] R.F. Lacovic. Comparison of experimental and calculated helium requirements for pressurization of a Centaur liquid oxygen tank: NASA TM X-2013, 1970. [8] N.T. van Dresar and R.J. Stochl. Pressurization and expulsion of a flightweight liquid hydrogen tank: AIAA-93- 1966, 1993. [9] T.L. Hardy and T.M. Tomsik. Prediction of the ullage gas thermal stratification in a NASP vehicle propellant tank experimental simulation using Flow-3D: Nasa technical memorandum 103217, 1990. [10] G.P. Samsal, J.I. Hochstein, M.C. Wendl and T.L. Hardy. Computational modeling of the pressurization process in a NASP vehicle propellant tank experimental simulation: AIAA 91-2407. AIAA Joint Propulsion Conference and Exhibit, 1991. [11] P. Adnani and R.W. Jennings. Pressurization analysis of cryogenic propulsion systems: AIAA 2000-3788. In 36th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit, Huntsville, Alabama, USA, 2000. [12] C. Ludwig and M. Dreyer. Analyses of cryogenic propellant tank pressurization based upon ground experiments: AIAA 2012-5199. In AIAA Space 2012 Conference & Exhibit, Pasadena, California, USA, 2012. [13] Flow Science Inc. Flow-3D User Manual – Version 10.0, 2011. [14] R.F. Barron. Cryogenic heat transfer, 3. ed., Taylor & Francis, Philadelphia, 1999, p. 23 [15] E.W. Lemmon, M.L. Huber and M.O. McLinden. NIST Standard Reference Database 23: Reference Fluid Thermodynamic and Transport Properties-REFPROP, Version 9.0, National Institute of Standards and Technology, Standard Reference Data Program, Gaithersburg, 2010. [16] E.J. Hopfinger and S.P. Das. Mass transfer enhancement by capillary waves at a liquid–vapour interface. Experiments in Fluids, Vol. 46, No.4: 597-605, 2009. [17] S.P. Das and E.J. Hopfinger. Mass transfer enhancement by gravity waves at a liquid–vapour interface. International Journal of Heat and Mass Transfer, Vol. 52, No. 5-6: 1400-1411, 2009. [18] H.D. Baehr and K. Stephan. Wärme- und Stoffübertragung, 6. ed., Springer, Berlin, 2008, p.491, p.302.
Pan Lu1 , Zhang Cheng-Lin2,6,Wang Liang3, Liu Tong4 and Liu Jiang-lin5 1 Aviation and Materials College, Anhui Technical College of Mechanical and Electrical Engineering, Wuhu Anhui 241000, People’s Republic of China 2 School of Engineering Science, University of Science and Technology of China, Hefei Anhui 230026, People’s Republic of China 3 Anhui Top Additive Manufacturing Technology Co., Ltd., Wuhu Anhui 241300, People’s Republic of China 4 Anhui Chungu 3D Printing Institute of Intelligent Equipment and Industrial Technology, Anhui 241300, People’s Republic of China 5 School of Mechanical and Transportation Engineering, Taiyuan University of Technology, Taiyuan Shanxi 030024, People’s Republic of China 6 Author to whom any correspondence should be addressed. E-mail: ahjdpanlu@126.com, jiao__zg@126.com, ahjdjxx001@126.com,tongliu1988@126.com and liujianglin@tyut.edu.cn
선택적 레이저 용융(SLM)은 열 전달, 용융, 상전이, 기화 및 물질 전달을 포함하는 복잡한 동적 비평형 프로세스인 금속 적층 제조(MAM)에서 가장 유망한 기술 중 하나가 되었습니다. 용융 풀의 특성(구조, 온도 흐름 및 속도 흐름)은 SLM의 최종 성형 품질에 결정적인 영향을 미칩니다. 이 연구에서는 선택적 레이저 용융 AlCu5MnCdVA 합금의 용융 풀 구조, 온도 흐름 및 속도장을 연구하기 위해 수치 시뮬레이션과 실험을 모두 사용했습니다.
그 결과 용융풀의 구조는 다양한 형태(깊은 오목 구조, 이중 오목 구조, 평면 구조, 돌출 구조 및 이상적인 평면 구조)를 나타냈으며, 용융 풀의 크기는 약 132 μm × 107 μm × 50 μm였습니다. : 용융풀은 초기에는 여러 구동력에 의해 깊이 15μm의 깊은 오목형상이었으나, 성형 후기에는 장력구배에 의해 높이 10μm의 돌출형상이 되었다. 용융 풀 내부의 금속 흐름은 주로 레이저 충격력, 금속 액체 중력, 표면 장력 및 반동 압력에 의해 구동되었습니다.
AlCu5MnCdVA 합금의 경우, 금속 액체 응고 속도가 매우 빠르며(3.5 × 10-4 S), 가열 속도 및 냉각 속도는 각각 6.5 × 107 K S-1 및 1.6 × 106 K S-1 에 도달했습니다. 시각적 표준으로 표면 거칠기를 선택하고, 낮은 레이저 에너지 AlCu5MnCdVA 합금 최적 공정 매개변수 창을 수치 시뮬레이션으로 얻었습니다: 레이저 출력 250W, 부화 공간 0.11mm, 층 두께 0.03mm, 레이저 스캔 속도 1.5m s-1 .
또한, 실험 프린팅과 수치 시뮬레이션과 비교할 때, 용융 풀의 폭은 각각 약 205um 및 약 210um이었고, 인접한 두 용융 트랙 사이의 중첩은 모두 약 65um이었다. 결과는 수치 시뮬레이션 결과가 실험 인쇄 결과와 기본적으로 일치함을 보여 수치 시뮬레이션 모델의 정확성을 입증했습니다.
Selective Laser Melting (SLM) has become one of the most promising technologies in Metal Additive Manufacturing (MAM), which is a complex dynamic non-equilibrium process involving heat transfer, melting, phase transition, vaporization and mass transfer. The characteristics of the molten pool (structure, temperature flow and velocity flow) have a decisive influence on the final forming quality of SLM. In this study, both numerical simulation and experiments were employed to study molten pool structure, temperature flow and velocity field in Selective Laser Melting AlCu5MnCdVA alloy. The results showed the structure of molten pool showed different forms(deep-concave structure, double-concave structure, plane structure, protruding structure and ideal planar structure), and the size of the molten pool was approximately 132 μm × 107 μm × 50 μm: in the early stage, molten pool was in a state of deep-concave shape with a depth of 15 μm due to multiple driving forces, while a protruding shape with a height of 10 μm duo to tension gradient in the later stages of forming. The metal flow inside the molten pool was mainly driven by laser impact force, metal liquid gravity, surface tension and recoil pressure. For AlCu5MnCdVA alloy, metal liquid solidification speed was extremely fast(3.5 × 10−4 S), the heating rate and cooling rate reached 6.5 × 107 K S−1 and 1.6 × 106 K S−1 , respectively. Choosing surface roughness as a visual standard, low-laser energy AlCu5MnCdVA alloy optimum process parameters window was obtained by numerical simulation: laser power 250 W, hatching space 0.11 mm, layer thickness 0.03 mm, laser scanning velocity 1.5 m s−1 . In addition, compared with experimental printing and numerical simulation, the width of the molten pool was about 205 um and about 210 um, respectively, and overlapping between two adjacent molten tracks was all about 65 um. The results showed that the numerical simulation results were basically consistent with the experimental print results, which proved the correctness of the numerical simulation model.
Figure 1. AlCu5MnCdVA powder particle size distribution.Figure 2. AlCu5MnCdVA powderFigure 3. Finite element model and calculation domains of SLM.Figure 4. SLM heat transfer process.Figure 17. Two-pass molten tracks overlapping for Scheme NO.2.
References
[1] Cuiyun H 2008 Phase diagram determination and thermodynamic study of Al–Cu–Mn, Al–Cu–Si, Al–Mg–Ni and Ni–Ti–Si systems Central South University [2] Zhanfei Z 2017 Study on theta phase segregation and room temperature properties of high strength cast Al–Cu–Mn alloy Lanzhou University of Technology [3] Nie X et al 2018 Analysis of processing parameters and characteristics of selective laser melted high strength Al–Cu–Mg alloys: from single tracks to cubic samplesJ. Mater. Process. Technol. 256 69–77 [4] Shenping Y et al 2017 Laser absorptance measurement of commonly used metal materials in laser additive manufacturing technology Aviation Manufacturing Technology 12 23–9 [5] Wenqing W 2007 Relationship between cooling rate and grain size of AlCu5MnCdVA alloy Harbin University of Technology [6] Majeed M, Vural M, Raja S and Bilal Naim Shaikh M 2019 Finite element analysis of thermal behavior in maraging steel during SLM process Optik 208 113–24 [7] Khairallah S A, Anderson A T, Rubenchik A and King W E 2016 Laser powder-bed fusion additive manufacturing: physics of complex melt flow and formation mechanisms of pores, spatter, and denudation zones Acta Mater. 108 36–45 [8] Bo C, Zhiyu X, Quanquan Z, Yuanbiao W, Liping W and Jin C 2020 Process optimization and microstructure and properties of SLM forming Cu6AlNiSnInCe imitation gold alloy Chin. J. Nonferr. Met. 30 372–82 [9] Li W 2012 Research on performance of metal parts formed by selective laser melting Huazhong University of Science and Technology [10] Yu Q 2013 The influence of different laser heat sources on the surface shape of the molten pool in laser cladding Surf. Technol. 42 40–3
[11] Xianfeng J, Xiangchen M, Rongwei S, Xigen Y and Ming Y 2015 Research on the influence of material state change on temperature field in SLM processing Applied Laser 35 155–9 [12] Körner C, Attar E and Heinl P 2011 Mesoscopic simulation of selective beam melting processesJ. Mater. Process. Technol. 211 978–87 [13] Yadroitsev I, Gusarov A, Yadroitsava I and Smurov I 2010 Single track formation in selective laser melting of metal powdersJ. Mater. Process. Technol. 210 1624–31 [14] King W, Anderson A T, Ferencz R M, Hodge N E, Kamath C and Khairallah S A 2014 Overview of modelling and simulation of metal powder bed fusion process at Lawrence Livermore National Laboratory Mater. Sci. Technol. 31 957–68 [15] Hussein A, Hao L, Yan C and Everson R 2013 Finite element simulation of the temperature and stress fields in single layers built without-support in selective laser melting Materials & Design (1980–2015) 52 638–47 [16] Qiu C, Panwisawas C, Ward M, Basoalto H C, Brooks J W and Attallah M M 2015 On the role of melt flow into the surface structure and porosity development during selective laser melting Acta Mater. 96 72–9 [17] Weihao Y, Hui C and Qingsong W 2020 Thermodynamic behavior of laser selective melting molten pool under the action of recoil pressure Journal of Mechanical Engineering 56 213–9 [18] Weijuan Y 2019 Numerical simulation of melt pool temperature field and morphology evolution during laser selective melting process Xi’an University of Technology [19] Genwang W 2017 Research on the establishment of laser heat source model based on energy distribution and its simulation application Harbin Institute of Technology [20] FLOW-3D 2017 User Manual (USA: FLOW SCIENCE) [21] Hirt C and Nichols B 1981 Volume of fluid (VOF) method for the dynamics of free boundariesJ. Comput. Phys. 39 201–25 [22] Hu Z, Zhang H, Zhu H, Xiao Z, Nie X and Zeng X 2019 Microstructure, mechanical properties and strengthening mechanisms of AlCu5MnCdVA aluminum alloy fabricated by selective laser melting Materials Science and Engineering: A 759 154–66 [23] Ketai H, Liu Z and Lechang Y 2020 Simulation of temperature field, microstructure and mechanical properties of 316L stainless steel in selected laser melting Progress in Laser and Optoelectronics 9 1–18 [24] Cao L 2020 Workpiece-scale numerical simulations of SLM molten pool dynamic behavior of 316L stainless steel Comput. Math. Appl. 4 22–34 [25] Dening Z, Yongping L, Tinglu H and Junyi S 2000 Numerical study of fluid flow and heat transfer in molten pool under the condition of moving heat source J. Met. 4 387–90 [26] Chengyun C, Cui F and Wenlong Z 2018 The effect of Marangoni flow on the thermal behavior and melt flow behavior of laser cladding Applied Laser 38 409–16 [27] Peiying B and Enhuai Y 2020 The effect of laser power on the morphology and residual stress of the molten pool of metal laser selective melting Progress in Laser and Optoelectronics 7 1–12 http://kns.cnki.net/kcms/detail/31.1690.TN.20190717.0933.032.html [28] Zhen L, Dongyun Z, Zhe F and Chengjie W 2017 Numerical simulation of the influence of overlap rate on the forming quality of Inconel 718 alloy by selective laser melting processing Applied Laser 37 187–93 [29] Wei W, Qi L, Guang Y, Lanyun Q and Xiong X 2015 Numerical simulation of electromagnetic field, temperature field and flowfield of laser melting pool under the action of electromagnetic stirring China Laser 42 48–55 [30] Hu Y, He X, Yu G and Zhao S 2016 Capillary convection in pulsed—butt welding of miscible dissimilar couple Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci. 231 2429–40 [31] Li R 2010 Research on the key basic problems of selective laser melting forming of metal powder Huazhong University of Science and Technology [32] Zijue T, Weiwei L, Zhaorui Y, Hao W and Hongchao Z 2019 Study on the shape evolution behavior of metal laser melting deposition based on molten pool dynamic characteristicsJournal of Mechanical Engineering 55 39–47 [33] Pan L, Cheng-Lin Z, Hai-Yi L, Liang W and Tong L 2020 A new two-step selective laser remelting of 316L stainless steel: process, density, surface roughness, mechanical properties, microstructure Mater. Res. Express 7 056503 [34] Pan L, Cheng-Lin Z, Hai-Yi L, Jiang H, Tong L and Liang W 2019 The influence and optimization of forming process parameters of 316L stainless steel prepared by laser melting on the density Forging Technology 44 103–9
Due to their high efficiency, low heat loss and associated sustainability advantages, impinging jets have been used extensively in marine engineering, geotechnical engineering and other engineering practices. In this paper, the flow structure and impact characteristics of impinging jets with different Reynolds numbers and impact distances are systematically studied by Flow-3D based on PIV experiments. In the study, the relevant state parameters of the jets are dimensionlessly treated, obtaining not only the linear relationship between the length of the potential nucleation zone and the impinging distance, but also the linear relationship between the axial velocity and the axial distance in the impinging zone. In addition, after the jet impinges on the flat plate, the vortex action range caused by the wall-attached flow of the jet gradually decreases inward with the increase of the impinging distance. By examining the effect of Reynolds number Re on the hydraulic characteristics of the submerged impact jet, it can be found that the structure of the continuous submerged impact jet is relatively independent of the Reynolds number. At the same time, the final simulation results demonstrate the applicability of the linear relationship between the length of the potential core region and the impact distance. This study provides methodological guidance and theoretical support for relevant engineering practice and subsequent research on impinging jets, which has strong theoretical and practical significance.
Figure 3. (a) Schematic diagram of the experimental setup; (b) PIV images of vertical impinging jets with velocity fields.
Figure 4. (a) Velocity distribution verification at the outlet of the jet pipe; (b) Distribution of flow angle in the mid-axis of the jet [39].
Figure 5. Along-range distribution of the dimensionless axial velocity of the jet at different impact distances.Figure 6 shows the variation of H
Figure 6. Relationship between the distribution of potential core region and the impact height H/D.
Figure 7. The relationship between the potential core length
Figure 8. Along-range distribution of the flow angle φ of the jet at different impact distances.
Figure 9. Velocity distribution along the axis of the jet at different impinging regions.
Figure 10. The absolute value distribution of slope under different impact distances.
Figure 11. Velocity distribution of impinging jet on wall under different impinging distances.
Figure 12. Along-range distribution of the dimensionless axial velocity of the jet at different Reynolds numbers.
Figure 13. Along-range distribution of the flow angle φ of the jet at different Reynolds numbers.
Figure 14. Velocity distribution along the jet axis at different Reynolds numbers.
Figure 15. Velocity distribution of impinging jet on a wall under different Reynolds numbers.
References
Zhang, J.; Li, Y.; Zhang, Y.; Yang, F.; Liang, C.; Tan, S. Using a high-pressure water jet-assisted tunnel boring machine to break rock. Adv. Mech. Eng.2020, 12, 1687814020962290. [Google Scholar] [CrossRef]
Shi, X.; Zhang, G.; Xu, G.; Ma, Y.; Wu, X. Inactivating Microorganism on Medical Instrument Using Plasma Jet. High Volt. Eng.2009, 35, 632–635. [Google Scholar]
Gao, Y.; Han, P.; Wang, F.; Cao, J.; Zhang, S. Study on the Characteristics of Water Jet Breaking Coal Rock in a Drilling Hole. Sustainability2022, 14, 8258. [Google Scholar] [CrossRef]
Xu, W.; Wang, C.; Zhang, L.; Ge, J.; Zhang, D.; Gao, Z. Numerical study of continuous jet impinging on a rotating wall based on Wray—Agarwal turbulence model. J. Braz. Soc. Mech. Sci. Eng.2022, 44, 433. [Google Scholar] [CrossRef]
Hu, B.; Wang, C.; Wang, H.; Yu, Q.; Liu, J.; Zhu, Y.; Ge, J.; Chen, X.; Yang, Y. Numerical Simulation Study of the Horizontal Submerged Jet Based on the Wray—Agarwal Turbulence Model. J. Mar. Sci. Eng.2022, 10, 1217. [Google Scholar] [CrossRef]
Dahiya, A.K.; Bhuyan, B.K.; Kumar, S. Perspective study of abrasive water jet machining of composites—A review. J. Mech. Sci. Technol.2022, 36, 213–224. [Google Scholar] [CrossRef]
Abushanab, W.S.; Moustafa, E.B.; Harish, M.; Shanmugan, S.; Elsheikh, A.H. Experimental investigation on surface characteristics of Ti6Al4V alloy during abrasive water jet machining process. Alex. Eng. J.2022, 61, 7529–7539. [Google Scholar] [CrossRef]
Hu, B.; Wang, H.; Liu, J.; Zhu, Y.; Wang, C.; Ge, J.; Zhang, Y. A numerical study of a submerged water jet impinging on a stationary wall. J. Mar. Sci. Eng.2022, 10, 228. [Google Scholar] [CrossRef]
Peng, J.; Shen, H.; Xie, W.; Zhai, S.; Xi, G. Influence of flow fluctuation characteristics on flow and heat transfer in different regions. J. Drain. Irrig. Mach. Eng.2022, 40, 826–833. [Google Scholar]
Zhai, S.; Xie, F.; Yin, G.; Xi, G. Effect of gap ratio on vortex-induced vibration characteristics of different blunt bodies near-wall. J. Drain. Irrig. Mach. Eng.2021, 39, 1132–1138. [Google Scholar]
Lin, W.; Zhou, Y.; Wang, L.; Tao, L. PIV experiment and numerical simulation of trailing vortex structure of improved INTER-MIG impeller. J. Drain. Irrig. Mach. Eng.2021, 39, 158–164. [Google Scholar]
Han, B.; Yao, Z.; Tang, R.; Xu, H. On the supersonic impinging jet by laser Doppler velocimetry. Exp. Meas. Fluid Mech.2002, 16, 99–103. [Google Scholar]
Darisse, A.; Lemay, J.; Benaissa, A. LDV measurements of well converged third order moments in the far field of a free turbulent round jet. Exp. Therm. Fluid Sci.2013, 44, 825–833. [Google Scholar] [CrossRef]
Kumar, S.; Kumar, A. Effect of initial conditions on mean flow characteristics of a three dimensional turbulent wall jet. Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci.2021, 235, 6177–6190. [Google Scholar] [CrossRef]
Tao, D.; Zhang, R.; Ying, C. Development and application of the pollutant diffusion testing apparatus based on the image analysis. J. Saf. Environ.2016, 16, 247–251. [Google Scholar]
Seo, H.; Kim, K.C. Experimental study on flow and turbulence characteristics of bubbly jet with low void fraction. Int. J. Multiph. Flow2021, 142, 103738. [Google Scholar] [CrossRef]
Wen, Q.; Sha, J.; Liu, Y. TR-PIV measurement of the turbulent submerged jet and POB analysis of the dynamic structure. J. Exp. Fluid Mech.2014, 4, 16–24. [Google Scholar]
Yang, Y.; Zhou, L.; Shi, W.; He, Z.; Han, Y.; Xiao, Y. Interstage difference of pressure pulsation in a three-stage electrical submersible pump. J. Petrol. Sci. Eng.2021, 196, 107653. [Google Scholar] [CrossRef]
Tang, S.; Zhu, Y.; Yuan, S. An improved convolutional neural network with an adaptable learning rate towards multi-signal fault diagnosis of hydraulic piston pump. Adv. Eng. Inform.2021, 50, 101406. [Google Scholar] [CrossRef]
Han, Y.; Song, X.; Li, K.; Yan, X. Hybrid modeling for submergence depth of the pumping well using stochastic configuration networks with random sampling. J. Petrol. Sci. Eng.2022, 208, 109423. [Google Scholar] [CrossRef]
Tang, S.; Zhu, Y.; Yuan, S. A novel adaptive convolutional neural network for fault diagnosis of hydraulic piston pump with acoustic images. Adv. Eng. Inform.2022, 52, 101554. [Google Scholar] [CrossRef]
Long, J.; Song, X.; Shi, J.; Chen, J. Optimization and CFD Analysis on Nozzle Exit Position of Two-phase Ejector. J. Refrig.2022, 43, 39–45. [Google Scholar]
Ni, Q.; Ruan, W. Optimization design of desilting jet pump parameters based on response surface model. J. Ship Mech.2022, 26, 365–374. [Google Scholar]
Zhang, K.; Zhu, X.; Ren, X.; Qiu, Q.; Shen, S. Numerical investigation on the effect of nozzle position for design of high performance ejector. Appl. Therm. Eng.2017, 126, 594–601. [Google Scholar] [CrossRef]
Fu, W.; Liu, Z.; Li, Y.; Wu, H.; Tang, Y. Numerical study for the influences of primary steam nozzle distance and mixing chamber throat diameter on steam ejector performance. Int. J. Therm. Sci.2018, 132, 509–516. [Google Scholar] [CrossRef]
Lucas, C.; Rusche, H.; Schroeder, A.; Koehler, J. Numerical investigation of a two-phase CO2 ejector. Int. J. Refrigeration2014, 43, 154–166. [Google Scholar] [CrossRef]
Ma, X.; Zhu, T.; Fu, Y.; Yan, Y.; Chen, W. Numerical simulation of rock breaking by abrasive water jet. J. Coast. Res.2019, 93, 274–283. [Google Scholar] [CrossRef]
He, L.; Liu, Y.; Shen, K.; Yang, X.; Ba, Q.; Xiong, W. Numerical research on the dynamic rock-breaking process of impact drilling with multi-nozzle water jets. J. Pet. Sci. Eng.2021, 207, 109145. [Google Scholar] [CrossRef]
Yu, Z.; Wang, Z.; Lei, C.; Zhou, Y.; Qiu, X. Numerical Simulation on Internal Flow Field of a Self-excited Oscillation Pulsed Jet Nozzle with Back-flow. Mech. Sci. Technol. Aerosp. Eng.2022, 41, 998–1002. [Google Scholar]
Huang, J.; Ni, F.; Gu, L. Numerical method of FLOW-3D for sediment erosion simulation. China Harb. Eng.2019, 39, 6–11. [Google Scholar]
Al Shaikhli, H.I.; Khassaf, S.I. Using of flow 3d as CFD materials approach in waves generation. Mater. Today Proc.2022, 49, 2907–2911. [Google Scholar] [CrossRef]
Kosaj, R.; Alboresha, R.S.; Sulaiman, S.O. Comparison Between Numerical Flow3d Software and Laboratory Data, For Sediment Incipient Motion. IOP Conf. Ser. Earth Environ. Sci.2022, 961, 012031. [Google Scholar] [CrossRef]
Du, C.; Liu, X.; Zhang, J.; Wang, B.; Chen, X.; Yu, X. Long-distance water hammer protection of pipeline after pump being first lowered and then rasied. J. Drain. Irrig. Mach. Eng.2022, 40, 1248–1253, 1267. [Google Scholar]
Gao, F.; Li, X.; Gao, Q. Experiment and numerical simulation on hydraulic characteristics of novel trapezoidal measuring weir. J. Drain. Irrig. Mach. Eng.2022, 40, 1104–1111. [Google Scholar]
Tu, A.; Nie, X.; Li, Y.; Li, H. Experimental and simulation study on water infiltration characteristics of layered red soil. J. Drain. Irrig. Mach. Eng.2021, 39, 1243–1249. [Google Scholar]
Chen, J.; Zeng, B.; Liu, L.; Tao, K.; Zhao, H.; Zhang, C.; Zhang, J.; Li, D. Investigating the anchorage performance of full-grouted anchor bolts with a modified numerical simulation method. Eng. Fail. Anal.2022, 141, 106640. [Google Scholar] [CrossRef]
Hu, B.; Yao, Y.; Wang, M.; Wang, C.; Liu, Y. Flow and Performance of the Disk Cavity of a Marine Gas Turbine at Varying Nozzle Pressure and Low Rotation Speeds: A Numerical Investigation. Machines2023, 11, 68. [Google Scholar] [CrossRef]
Yao, J.; Wang, X.; Zhang, S.; Xu, S.; Jin, B.; Ding, S. Orthogonal test of important parameters affecting hydraulic performance of negative pressure feedback jet sprinkler. J. Drain. Irrig. Mach. Eng.2021, 39, 966–972. [Google Scholar]
Wang, C.; Wang, X.; Shi, W.; Lu, W.; Tan, S.K.; Zhou, L. Experimental investigation on impingement of a submerged circular water jet at varying impinging angles and Reynolds numbers. Exp. Therm. Fluid Sci.2017, 89, 189–198. [Google Scholar] [CrossRef]
Speziale, C.G.; Thangam, S. Analysis of an RNG based turbulence model for separated flows. Int. J. Eng. Sci.1992, 30, 1379–1388. [Google Scholar] [CrossRef]
El Hassan, M.; Assoum, H.H.; Sobolik, V.; Vétel, J.; Abed-Meraim, K.; Garon, A.; Sakout, A. Experimental investigation of the wall shear stress and the vortex dynamics in a circular impinging jet. Exp. Fluids2012, 52, 1475–1489. [Google Scholar] [CrossRef]
Fairweather, M.; Hargrave, G. Experimental investigation of an axisymmetric, impinging turbulent jet. 1. Velocity field. Exp. Fluids2002, 33, 464–471. [Google Scholar] [CrossRef]
Ashforth-Frost, S.; Jambunathan, K. Effect of nozzle geometry and semi-confinement on the potential core of a turbulent axisymmetric free jet. Int. Commun. Heat Mass Transf.1996, 23, 155–162. [Google Scholar] [CrossRef]
Chen, M.; Huang, H.; Wang, D.; Lv, S.; Chen, Y. PIV tests for flow characteristics of impinging jet in a semi-closed circular pipe. J. Vib. Shock2021, 40, 90–97, 113. [Google Scholar]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Mi, H.; Wang, C.; Jia, X.; Hu, B.; Wang, H.; Wang, H.; Zhu, Y. Hydraulic Characteristics of Continuous Submerged Jet Impinging on a Wall by Using Numerical Simulation and PIV Experiment. Sustainability2023, 15, 5159. https://doi.org/10.3390/su15065159
AMA Style
Mi H, Wang C, Jia X, Hu B, Wang H, Wang H, Zhu Y. Hydraulic Characteristics of Continuous Submerged Jet Impinging on a Wall by Using Numerical Simulation and PIV Experiment. Sustainability. 2023; 15(6):5159. https://doi.org/10.3390/su15065159Chicago/Turabian Style
Mi, Hongbo, Chuan Wang, Xuanwen Jia, Bo Hu, Hongliang Wang, Hui Wang, and Yong Zhu. 2023. “Hydraulic Characteristics of Continuous Submerged Jet Impinging on a Wall by Using Numerical Simulation and PIV Experiment” Sustainability 15, no. 6: 5159. https://doi.org/10.3390/su15065159
화성 미션 애플리케이션을 위한 NERVA 파생 원자로 냉각수 채널 모델은 1.3m NERVA 파생 원자로(NDR) 냉각수 채널의 전산유체역학(CFD) 연구 결과를 제시합니다. CFD 코드 FLOW-3D는 NDR 코어를 통과하는 기체 수소의 흐름을 모델링하는 데 사용되었습니다. 수소는 냉각제 채널을 통해 노심을 통과하여 원자로의 냉각제 및 로켓의 추진제 역할을 합니다. 수소는 고밀도/저온 상태로 채널에 들어가고 저밀도/고온 상태로 빠져나오므로 압축성 모델을 사용해야 합니다. 기술 문서의 설계 사양이 모델에 사용되었습니다. 채널 길이에 걸친 압력 강하가 이전에 추정한 것(0.9MPa)보다 높은 것으로 확인되었으며, 이는 더 강력한 냉각수 펌프가 필요하고 설계 사양을 재평가해야 함을 나타냅니다.
NERVA-Derived Reactor Coolant Channel Model for Mars Mission Applications presents the results of a computational fluid dynamics (CFD) study of a 1.3m NERVA-Derived Reactor (NDR) coolant channel; The CFD code FLOW-3D was used to model the flow of gaseous hydrogen through the core of a NDR. Hydrogen passes through the core by way of coolant channels, acting as the coolant for the reactor as well as the propellant for the rocket. Hydrogen enters the channel in a high density/low temperature state and exits in a low density/high temperature state necessitating the use of a compressible model. Design specifications from a technical paper were used for the model; It was determined that the pressure drop across the length of the channel was higher than previously estimated (0.9 MPa), indicating the possible need for more powerful coolant pumps and a re-evaluation of the design specifications.
Figure 1 Nuclear Rocket Schematic DiagramFigure 2 Fuel Element – Tip ViewFigure 3 Fuel Element – Tie-Tube Structure (Tie-tubes are black)Figure 5 Three-Dimensional Coolant Channel ModelFigure 6 Two-Dimensional Coolant Channel Model
REFERENCES
Anderson, J. D., Jr., (1990) Modern Compressible Flow, 2d ed., McGraw-Hill, New York. Avallone E. A. and T. Baumeister III, eds., (1987) Mark’s Standard Handbookfor Mechanical Engineers, 9th ed., McGraw-Hill, New York. Bennett, G. L. and T. J. Miller (1992) “Nuclear Propulsion: A Key Transportation Technology for the Exploration of Mars,” Proceedings o f the 9th Symposium on Space Nuclear Power Systems, CONF-920104, M. S. El-Genk and M. D. Hoover, eds., American Institute of Physics, New York, AIP Conference Proceedings No. 246, 2: 383-388. Black, D. L., and S. V. Gunn (1991) “A Technical Summary of Engine and Reactor Subsystem Design Performance during the NERVA Program,” AIAA-91-3450, American Institute of Aeronautics and Astronautics, Washington, D. C. Borowski, S. K., et al. (1992) “Nuclear Thermal Rockets: Key to Moon-Mars Exploration,” Aerospace America, July 1992, pp. 34(5). Borowski, S. K., et al. (1993) “ Nuclear Thermal Rocket/Vehicle Design Options for Future NASA Missions to the Moon and Mars,” AIAA-93-4170, American Institute of Aeronautics and Astronautics, Washington, D. C. Borowski, S. K., et al. (1994) “Nuclear Thermal Rocket/Stage Technology Options for NASA’s Future Human Exploration Missions to the Moon and Mars,” Proceedings o f the 11th Symposium on Space Nuclear Power and Propulsion, CONF-940101, M. S. El-Genk and M. D. Hoover, eds., American Institute of Physics, New York, NY, AIP Conference Proceedings No. 301, 2: 745 – 758. Burmeister, L. C. (1993) Convective Heat Transfer, 2d ed., John Wiley & Sons, New York. Chi, J., R. Holman, and B. Pierce (1989) “Nerva Derivative Reactors for Thermal and Electrical Propulsion,” AIAA-89-2770, American Institute of Aeronautics and Astronautics, Washington, D. C. FIDAP (1993) FIDAP 7.0 User’s Manual, Fluid Dynamics International, Inc. FL0W-3D (1994) FL0W-3D Version 6.0 Quick Reference Guide, Flow Science, Inc., Los Alamos, NM. Hill, P. G. and C. R. Peterson (1970) Mechanics and Thermodynamics o f Propulsion, Addison-Wesley, Reading, MA. Lamarsh, J. R. (1983) Introduction to Nuclear Engineering, 2d ed., Addison-Wesley, Reading, MA. Nassersharif, B. (1991) Notes from a Nuclear Propulsion Short Course, 3-5 January 1991, American Institute of Physics. Nassersharif, B., E. Porta, and D. Hailes (1994) “A Proposal Entitled: Scenario Based Design of Nuclear Propulsion for Manned Mars Mission,” NSCEE, Las Vegas, NV. Shepard, K., et al. (1992) “A Split Sprint Mission to Mars,” Proceedings o f the 9th Symposium on Space Nuclear Power Systems, CONF-920104, M. S. El-Genk and M. D. Hoover, eds., American Institute of Physics, New York, AIP Conference Proceedings No. 246, 1: 58 – 63. Sutton, G. P. (1986) Rocket Propulsion Elements: An Introduction to the Engineering o f Rockets, 5th ed., John Wiley & Sons, New York. U.S. President (1989) “Remarks on the 20th Anniversary of the Apollo 11 Moon Landing July 20, 1989,” Administration o f George Bush, Office of the Federal Register. National Archives and Records Service, 1989, Washington D. C., George Bush, 1989, p. 992. VSAERO (1994) VSAERO User’s Manual E.5, Analytical Methods, Inc., Redmond, WA. White, F. M. (1991) Viscous Fluid Flow, 2d ed., McGraw-Hill, Inc., New York. Zweig, H. R. and M. H. Cooper (1993) “NERVA-Derived Rocket Module for Solar System Exploration,” AIAA-93-2110, American Institute of Aeronautics and Astronautics, Washington, D. C.
Dissertação de Mestrado Ciclo de Estudos Integrados Conducentes ao Grau de Mestre em Engenharia Mecânica Trabalho efectuado sob a orientação do Doutor Hélder de Jesus Fernades Puga Professor Doutor José Joaquim Carneiro Barbosa
ABSTRACT
논문의 일부로 튜터 선택 가능성과 해결해야 할 주제가 설정되는 매개변수를 염두에 두고 개발 주제 ‘Flow- 3D ®에 의한 저압 충전 시스템 최적화’가 선택되었습니다. 이를 위해서는 달성해야 할 목표와 이를 달성하기 위한 방법을 정의하는 것이 필요했습니다.
충전 시스템을 시뮬레이션하고 검증할 수 있는 광범위한 소프트웨어에도 불구하고 Flow-3D®는 시장에서 최고의 도구 중 하나로 표시되어 전체 충전 프로세스 및 행동 표현과 관련하여 탁월한 정확도로 시뮬레이션하는 능력을 입증했습니다.
이를 위해 관련 프로세스를 더 잘 이해하고 충진 시스템 시뮬레이션을 위한 탐색적 기반 역할을 하기 위해 이 도구를 탐색하는 것이 중요합니다. 지연 및 재료 낭비에 반영되는 실제적인 측면에서 충전 장치의 치수를 완벽하게 만드는 비용 및 시간 낭비. 이러한 방식으로 저압 주조 공정에서 충진 시스템을 설계하고 물리적 모델을 탐색하여 특성화하는 방법론을 검증하기 위한 것입니다.
이를 위해 다음 주요 단계를 고려하십시오.
시뮬레이션 소프트웨어 Flow 3D® 탐색; 충전 시스템 모델링; 모델의 매개변수를 탐색하여 모델링된 시스템의 시뮬레이션, 검증 및 최적화.
따라서 연구 중인 압력 곡선과 주조 분석에서 가장 관련성이 높은 정보의 최종 마이닝을 검증하기 위한 것입니다.
사용된 압력 곡선은 수집된 문헌과 이전에 수행된 실제 작업을 통해 얻었습니다. 결과를 통해 3단계 압력 곡선이 층류 충진 체계의 의도된 목적과 관련 속도가 0.5 𝑚/𝑠를 초과하지 않는다는 결론을 내릴 수 있었습니다.
충전 수준이 2인 압력 곡선은 0.5 𝑚/𝑠 이상의 속도로 영역을 채우는 더 난류 시스템을 갖습니다. 열전달 매개변수는 이전에 얻은 값이 주물에 대한 소산 거동을 확증하지 않았기 때문에 연구되었습니다.
이러한 방식으로 주조 공정에 더 부합하는 새로운 가치를 얻었습니다. 달성된 결과는 유사한 것으로 나타난 NovaFlow & Solid®에 의해 생성된 결과와 비교되어 시뮬레이션에서 설정된 매개변수를 검증했습니다. Flow 3D®는 주조 부품 시뮬레이션을 위한 강력한 도구로 입증되었습니다.
As part of the dissertation and bearing in mind the parameters in which the possibility of a choice of tutor and the subject to be addressed is established, the subject for development ’Optimization of filling systems for low pressure by Flow 3D ®’ was chosen. For this it was necessary to define the objectives to achieve and the methods to attain them. Despite the wide range of software able to simulate and validate filling systems, Flow 3D® has been shown as one of the best tools in the market, demonstrating its ability to simulate with distinctive accuracy with respect to the entire process of filling and the behavioral representation of the fluid obtained. To this end, it is important to explore this tool for a better understanding of the processes involved and to serve as an exploratory basis for the simulation of filling systems, simulation being one of the great strengths of the current industry due to the need to reduce costs and time waste, in practical terms, that lead to the perfecting of the dimensioning of filling devices, which are reflected in delays and wasted material. In this way it is intended to validate the methodology to design a filling system in lowpressure casting process, exploring their physical models and thus allowing for its characterization. For this, consider the following main phases: The exploration of the simulation software Flow 3D®; modeling of filling systems; simulation, validation and optimization of systems modeled by exploring the parameters of the models. Therefore, it is intended to validate the pressure curves under study and the eventual mining of the most relevant information in a casting analysis. The pressure curves that were used were obtained through the gathered literature and the practical work previously performed. Through the results it was possible to conclude that the pressure curve with 3 levels meets the intended purpose of a laminar filling regime and associated speeds never exceeding 0.5 𝑚/𝑠. The pressure curve with 2 filling levels has a more turbulent system, having filling areas with velocities above 0.5 𝑚/𝑠. The heat transfer parameter was studied due to the values previously obtained didn’t corroborate the behavior of dissipation regarding to the casting. In this way, new values, more in tune with the casting process, were obtained. The achieved results were compared with those generated by NovaFlow & Solid®, which were shown to be similar, validating the parameters established in the simulations. Flow 3D® was proven a powerful tool for the simulation of casting parts.
Figure 4.24 – Model with virtual valves in the extremities of the geometries to simulate the permeability of the mold promoting a more uniformed fillingFigure 4.39 – Values of temperature contours using full energy heat transfer parameter for simulaFigure 4.40 – Comparison between software simulations (a) Flow 3D® simulation,
(b) NovaFlow & Solid® simulation
BIBLIOGRAPHY
[1] E. Stanley and D. B. Sc, “Fluid Flow Aspects of Solidification Modelling : Simulation of Low Pressure Die Casting .” [2] Y. Sahin, “Computer aided foundry die-design,” Metallography, vol. 24, no. 8, pp. 671–679, 2003. [3] F. Bonollo, J. Urban, B. Bonatto, and M. Botter, “Gravity and low pressure die casting of aluminium alloys : a technical and economical benchmark,” La Metall. Ital., vol. 97, no. 6, pp. 23–32, 2005. [4] P. a and R. R, “Study of the effect of process parameters on the production of a nonsimmetric low pressure die casting part,” La Metall. Ital., pp. 57–63, 2009. [5] “Fundição em baixa pressão | Aluinfo.” [Online]. Available: http://www.aluinfo.com.br/novo/materiais/fundicao-em-baixa-pressao. [Accessed: 18- Sep-2015]. [6] “Low Pressure Sand Casting by Wolverine Bronze.” [Online]. Available: http://www.wolverinebronze.com/low-pressure-sand-casting.php. [Accessed: 18-Sep2015]. [7] A. Reikher, “Numerical Analysis of Die-Casting Process in Thin Cavities Using Lubrication Approximation,” no. December, 2012. [8] P. Fu, A. a. Luo, H. Jiang, L. Peng, Y. Yu, C. Zhai, and A. K. Sachdev, “Low-pressure die casting of magnesium alloy AM50: Response to process parameters,” J. Mater. Process. Technol., vol. 205, no. 1–3, pp. 224–234, 2008. [9] X. Li, Q. Hao, W. Jie, and Y. Zhou, “Development of pressure control system in counter gravity casting for large thin-walled A357 aluminum alloy components,” Trans. Nonferrous Met. Soc. China, vol. 18, no. 4, pp. 847–851, 2008. [10] J. a. Hines, “Determination of interfacial heat-transfer boundary conditions in an aluminum low-pressure permanent mold test casting,” Metall. Mater. Trans. B, vol. 35, no. 2, pp. 299–311, 2004. [11] A. Lima, A. Freitas, and P. Magalhães, “Processos de vazamento em moldações permanentes,” pp. 40–49, 2003. [12] Y. B. Choi, K. Matsugi, G. Sasaki, K. Arita, and O. Yanagisawa, “Analysis of Manufacturing Processes for Metal Fiber Reinforced Aluminum Alloy Composite Fabricated by Low-Pressure Casting,” Mater. Trans., vol. 47, no. 4, pp. 1227–1231, 68 2006. [13] G. Mi, X. Liu, K. Wang, and H. Fu, “Numerical simulation of low pressure die-casting aluminum wheel,” China Foundry, vol. 6, no. 1, pp. 48–52, 2009. [14] J. Kuo, F. Hsu, and W. Hwang, “ADVANCED Development of an interactive simulation system for the determination of the pressure ± time relationship during the ® lling in a low pressure casting process,” vol. 2, pp. 131–145, 2001. [15] S.-G. Liu, F.-Y. Cao, X.-Y. Zhao, Y.-D. Jia, Z.-L. Ning, and J.-F. Sun, “Characteristics of mold filling and entrainment of oxide film in low pressure casting of A356 alloy,” Mater. Sci. Eng. A, vol. 626, pp. 159–164, 2015. [16] “Casting Training Class – Lecture 10 – Solidification and Shrinkage-Casting.” FLOW3D®. [17] “UAB Casting Engineering Laboratory.” [Online]. Available: file:///C:/Users/Jos%C3%A9 Belo/Desktop/Artigo_Software/UAB Casting Engineering Laboratory.htm. [Accessed: 09-Nov-2015]. [18] A. Louvo, “Casting Simulation as a Tool in Concurrent Engineering,” pp. 1–12, 1997. [19] T. R. Vijayaram and P. Piccardo, “Computers in Foundries,” vol. 30, 2012. [20] M. Sadaiah, D. R. Yadav, P. V. Mohanram, and P. Radhakrishnan, “A generative computer-aided process planning system for prismatic components,” Int. J. Adv. Manuf. Technol., vol. 20, no. 10, pp. 709–719, 2002. [21] Ministry_of_Planning, “Digital Data,” vol. 67, pp. 1–6, 2004. [22] S. Shamasundar, D. Ramachandran, and N. S. Shrinivasan, “COMPUTER SIMULATION AND ANALYSIS OF INVESTMENTCASTING PROCESS.” [23] J. M. Siqueira and G. Motors, “Simulation applied to Aluminum High Pressure Die Casting,” pp. 1–5, 1998. [24] C. Fluid, COMPUTATIONAL FLUID DYNAMICS. Abdulnaser Sayma & Ventus Publishing ApS, 2009. [25] C. a. Felippa, “1 – Overview,” Adv. Finite Elem. Methods, pp. 1–9. [26] a. Meena and M. El Mansori, “Correlative thermal methodology for castability simulation of ductile iron in ADI production,” J. Mater. Process. Technol., vol. 212, no. 11, pp. 2484–2495, 2012. [27] T. R. Vijayaram, S. Sulaiman, a. M. S. Hamouda, and M. H. M. Ahmad, “Numerical simulation of casting solidification in permanent metallic molds,” J. Mater. Process. 69 Technol., vol. 178, pp. 29–33, 2006. [28] “General CFD FAQ — CFD-Wiki, the free CFD reference.” [Online]. Available: http://www.cfd-online.com/Wiki/General_CFD_FAQ. [Accessed: 10-Nov-2015]. [29] “FEM | FEA | CFD.” [Online]. Available: http://fem4analyze.blogspot.pt/. [Accessed: 09-Nov-2015]. [30] “Fundição; revista da Associação portuguesa de fundição,” Fundição, vol. N o 227. [31] “Casting Training Class – Lecture 1 – Introduction_to_FLOW-3D – Casting.” FLOW3D®. [32] F. Science, “FLOW-3D Cast Documentation,” no. 3.5, p. 80, 2012. [33] “Casting Training Class – Lecture 4 – Geometry Building – General.” FLOW-3D®. [34] F. Science, “FLOW-3D v11.0.3 User Manual,” pp. 1–132, 2015. [35] “Casting Training Class – Lecture 5 Meshing Concept – General.” FLOW-3D®. [36] “Casting Training Class – Lecture 6 – Boundary_Conditions – Casting.” FLOW-3D®. [37] “Casting Training Class – Lecture 9 – Physical Models-castings.” FLOW-3D®. [38] P. A. D. Jácome, M. C. Landim, A. Garcia, A. F. Furtado, and I. L. Ferreira, “The application of computational thermodynamics and a numerical model for the determination of surface tension and Gibbs–Thomson coefficient of aluminum based alloys,” Thermochim. Acta, vol. 523, no. 1–2, pp. 142–149, 2011. [39] J. P. Anson, R. A. L. Drew, and J. E. Gruzleski, “The surface tension of molten aluminum and Al-Si-Mg alloy under vacuum and hydrogen atmospheres,” Metall. Mater. Trans. B Process Metall. Mater. Process. Sci., vol. 30, no. 6, pp. XVI–1032, 1999.
Publication Date:2013-07-24 Research Org.: Los Alamos National Lab. (LANL), Los Alamos, NM (United States) Sponsoring Org.: DOE/LANL OSTI Identifier: 1088904 Report Number(s): LA-UR-13-25537 DOE Contract Number: AC52-06NA25396 Resource Type: Technical Report Country of Publication: United States Language: English Subject: Engineering(42); Materials Science(36); Radiation Chemistry, Radiochemistry, & Nuclear Chemistry(38)
Introduction
The plutonium foundry at Los Alamos National Laboratory casts products for various special nuclear applications. However, plutonium’s radioactivity, material properties, and security constraints complicate the ability to perform experimental analysis of mold behavior. The Manufacturing Engineering and Technologies (MET-2) group previously developed a graphite mold to vacuum cast small plutonium disks to be used by the Department of Homeland Security as point sources for radiation sensor testing.
A two-stage pouring basin consisting of a funnel and an angled cavity directs the liquid into a vertical runner. A stack of ten disk castings connect to the runner by horizontal gates. Volumetric flow rates were implemented to limit overflow into the funnel and minimize foundry returns. Models using Flow-3D computational fluid dynamics software are employed here to determine liquid Pu flow paths, optimal pour regimes, temperature changes, and pressure variations.
Setup
Hardcopy drawings provided necessary information to create 3D .stl models for import into Flow-3D (Figs. 1 and 2). The mesh was refined over several iterations to isolate the disk cavities, runner, angled cavity, funnel, and input pour. The final flow and mold-filling simulation utilizes a fine mesh with ~5.5 million total cells. For the temperature study, the mesh contained 1/8 as many cells to reduce computational time and set temperatures to 850 °C for the molten plutonium and 500 °C for the solid graphite mold components (Fig. 3).
Flow-3D solves mass continuity and Navier-Stokes momentum equations over the structured rectangular grid model using finite difference and finite volume numerical algorithms. The solver includes terms in the momentum equation for body and viscous accelerations and uses convective heat transfer.
Simulation settings enabled Flow-3D physics calculations for gravity at 980.665 cm/s 2 in the negative Z direction (top of mold to bottom); viscous, turbulent, incompressible flow using dynamically-computed Renormalized Group Model turbulence calculations and no-slip/partial slip wall shear, and; first order, full energy equation heat transfer.
Mesh boundaries were all set to symmetric boundary conditions except for the Zmin boundary set to outflow and the Zmax boundary set to a volume flow. Vacuum casting conditions and the high reactivity of remaining air molecules with Pu validate the assumption of an initially fluidless void.
Results
The flow follows a unique three-dimensional path. The mold fills upwards with two to three disks receiving fluid in a staggered sequence. Figures 5-9 show how the fluid fills the cavity, and Figure 7 includes the color scale for pressure levels in these four figures. The narrow gate causes a high pressure region which forces the fluid to flow down the cavity centerline.
It proceeds to splash against the far wall and then wrap around the circumference back to the gate (Figs. 5 and 6). Flow in the angled region of the pouring basin cascades over the bottom ledge and attaches to the far wall of the runner, as seen in Figure 7.
This channeling becomes less pronounced as fluid volume levels increase. Finally, two similar but non-uniform depressed regions form about the centerline. These regions fill from their perimeter and bottom until completion (Fig. 8). Such a pattern is counter, for example, to a steady scenario in which a circle of molten Pu encompassing the entire bottom surface rises as a growing cylinder.
Cavity pressure becomes uniform when the cavity is full. Pressure levels build in the rising well section of the runner, where impurities were found to settle in actual casting. Early test simulations optimized the flow as three pours so that the fluid would never overflow to the funnel, the cavities would all fill completely, and small amounts of fluid would remain as foundry returns in the angled cavity.
These rates and durations were translated to the single 2.7s pour at 100 cm 3 per second used here. Figure 9 shows anomalous pressure fluctuations which occurred as the cavities became completely filled. Multiple simulations exhibited a rapid change in pressure from positive to negative and back within the newly-full disk and surrounding, already-full disks.
The time required to completely fill each cavity is plotted in Figure 10. Results show negligible temperature change within the molten Pu during mold filling and, as seen in Figure 11, at fill completion.
Figure 1: Mold drawingsFigure 2: Mold AssemblyFigure 4: Actual mold and cast PuFigure 5: Bottom cavity filling
from runnerFigure 6: Pouring and fillingFigure 8: Edge detection of cavity fill geometry. Two similar depressed areas form
about the centerline. Top cavity shown; same pressure scale as other figuresFigure 10: Cavity fill times,from first fluid contact with pouring basin, Figure 11:Fluid temperature remains essentially constant
Conclusions
Non-uniform cavity filling could cause crystal microstructure irregularities during solidification. However, the small temperature changes seen – due to large differences in specific heat between Pu and graphite – over a relatively short time make such problems unlikely in this case.
In the actual casting, cooling required approximately ten minutes. This large difference in time scales further reduces the chance for temperature effects in such a superheated scenario. Pouring basin emptying decreases pressure at the gate which extends fill time of the top two cavities.
The bottom cavity takes longer to fill because fluid must first enter the runner and fill the well. Fill times continue linearly until the top two cavities. The anomalous pressure fluctuations may be due to physical attempts by the system to reach equilibrium, but they are more likely due to numerical errors in the Flow3D solver.
Unsuccessful tests were performed to remove them by halving fluid viscosity. The fine mesh reduced, but did not eliminate, the extent of the fluctuations. Future work is planned to study induction and heat transfer in the full Pu furnace system, including quantifying temporal lag of the cavity void temperature to the mold wall temperature during pre-heat and comparing heat flux levels between furnace components during cool-down.
Thanks to Doug Kautz for the opportunity to work with MET-2 and for assigning an interesting unclassified project. Additional thanks to Mike Bange for CFD guidance, insight of the project’s history, and draft review.
As part of the National Aero-Space Plane (NASP) project, the multi-dimensional effects of gravitational force, initial tank pressure, initial ullage temperature, and heat transfer rate on the 2-D temperature profiles were studied. FLOW-3D, a commercial finite difference fluid flow model, was used for the evaluation. These effects were examined on the basis of previous liquid hydrogen experimental data with gaseous hydrogen pressurant. FLOW-3D results were compared against an existing 1-D model. In addition, the effects of mesh size and convergence criteria on the analytical results were investigated. Suggestions for future modifications and uses of FLOW-3D for modeling of a NASP tank are also presented.
Document ID : 19900016844
Document Type : Technical Memorandum (TM)
AuthorsHardy, Terry L.
(NASA Lewis Research Center Cleveland, OH, United States)Tomsik, Thomas M. (NASA Lewis Research Center Cleveland, OH, United States)
Date Acquired : September 6, 2013
Publication Date : July 1, 1990
Subject CategoryPropellants And Fuels : Report/Patent NumberNASA-TM-103217E-5629NAS 1.15:103217
Funding Number(s)PROJECT: RTOP 763-01-21
Distribution Limits : Public
Copyright : Work of the US Gov. Public Use Permitted.
Prediction of the ullage gas thermal stratification in a NASP vehicle propellant tank experimental simulation using FLOW-3DPrediction of the ullage gas thermal stratification in a NASP vehicle propellant tank experimental simulation using FLOW-3D
Beom-Jin Kim 1, Jae-Hong Hwang 2 and Byunghyun Kim 3,* 1 Advanced Structures and Seismic Safety Research Division, Korea Atomic Energy Research Institute, Daejeon 34057, Korea 2 Korea Water Resources Corporation (K-Water), Daejeon 34350, Korea 3 Department of Civil Engineering, Kyungpook National University, Daegu 41566, Korea
Hydraulic structures installed in rivers inevitably create a water level difference between upstream and downstream regions. The potential energy due to this difference in water level is converted into kinetic energy, causing high-velocity flow and hydraulic jumps in the river. As a result, problems such as scouring and sloping downstream may occur around the hydraulic structures. In this study, a FLOW-3D model was constructed to perform a numerical analysis of the ChangnyeongHaman weir in the Republic of Korea. The constructed model was verified based on surface velocity measurements from a field gate operation experiment. In the simulation results, the flow discharge differed from the measured value by 9–15 m3/s, from which the accuracy was evaluated to be 82–87%. The flow velocity was evaluated with an accuracy of 92% from a difference of 0.01 to 0.16 m/s. Following this verification, a flow analysis of the hydraulic structures was performed according to boundary conditions and operation conditions for numerous scenarios. Since 2018, the ChangnyeongHaman weir gate has been fully opened due to the implementation of Korea’s eco-environmental policy; therefore, in this study, the actual gate operation history data prior to 2018 was applied and evaluated. The evaluation conditions were a 50% open gate condition and the flow discharge of two cases with a large difference in water level. As a result of the analysis, the actual operating conditions showed that the velocity and the Froude number were lower than the optimal conditions, confirming that the selected design was appropriate. It was also found that in the bed protection section, the average flow velocity was high when the water level difference was large, whereas the bottom velocity was high when the gate opening was large. Ultimately, through the reviewed status survey data in this study, the downstream flow characteristics of hydraulic structures along with adequacy verification techniques, optimal design techniques such as procedures for design, and important considerations were derived. Based on the current results, the constructed FLOW-3D-based model can be applied to creating or updating flow analysis guidelines for future repair and reinforcement measures as well as hydraulic structure design.
하천에 설치되는 수력구조물은 필연적으로 상류와 하류의 수위차를 발생시킨다. 이러한 수위차로 인한 위치에너지는 운동에너지로 변환되어 하천의 고속유동과 수압점프를 일으킨다. 그 결과 수력구조물 주변에서 하류의 세굴, 경사 등의 문제가 발생할 수 있다.
본 연구에서는 대한민국 창녕함안보의 수치해석을 위해 FLOW-3D 모델을 구축하였다. 구축된 모델은 현장 게이트 작동 실험에서 표면 속도 측정을 기반으로 검증되었습니다.
시뮬레이션 결과에서 유량은 측정값과 9~15 m3/s 차이가 나고 정확도는 82~87%로 평가되었다. 유속은 0.01~0.16m/s의 차이에서 92%의 정확도로 평가되었습니다.
검증 후 다양한 시나리오에 대한 경계조건 및 운전조건에 따른 수리구조물의 유동해석을 수행하였다. 2018년부터 창녕함안보 문은 한국의 친환경 정책 시행으로 전면 개방되었습니다.
따라서 본 연구에서는 2018년 이전의 실제 게이트 운영 이력 데이터를 적용하여 평가하였다. 평가조건은 50% open gate 조건과 수위차가 큰 2가지 경우의 유수방류로 하였다. 해석 결과 실제 운전조건은 속도와 Froude수가 최적조건보다 낮아 선정된 설계가 적합함을 확인하였다.
또한 베드보호구간에서는 수위차가 크면 평균유속이 높고, 수문개구가 크면 저저유속이 높은 것으로 나타났다. 최종적으로 본 연구에서 검토한 실태조사 자료를 통해 적정성 검증기법과 함께 수력구조물의 하류 유동특성, 설계절차 등 최적 설계기법 및 중요 고려사항을 도출하였다.
현재의 결과를 바탕으로 구축된 FLOW-3D 기반 모델은 수력구조 설계뿐만 아니라 향후 보수 및 보강 조치를 위한 유동해석 가이드라인 생성 또는 업데이트에 적용할 수 있습니다.
Figure 1. Effect of downstream riverbed erosion according to the type of weir foundation.Figure 2. Changnyeong-Haman weir depth survey results (June 2015)Figure 4. Field gate discharge experiment.Figure 16. Analysis results for Case 7 and Case 8
References
Wanoschek, R.; Hager, W.H. Hydraulic jump in trapezoidal channel. J. Hydraul. Res. 1989, 27, 429–446. [CrossRef]
Bohr, T.; Dimon, P.; Putkaradze, V. Shallow-water approach to the circular hydraulic jump. J. Fluid Mech. 1993, 254, 635–648. [CrossRef]
Chanson, H.; Brattberg, T. Experimental study of the air–water shear flow in a hydraulic jump. Int. J. Multiph. Flow 2000, 26, 583–607. [CrossRef]
Dhamotharan, S.; Gulliver, J.S.; Stefan, H.G. Unsteady one-dimensional settling of suspended sediment. Water Resour. Res. 1981, 17, 1125–1132. [CrossRef]
Ziegler, C.K.; Nisbet, B.S. Long-term simulation of fine-grained sediment transport in large reservoir. J. Hydraul. Eng. 1995, 121, 773–781. [CrossRef]
Olsen, N.R.B. Two-dimensional numerical modelling of flushing processes in water reservoirs. J. Hydraul. Res. 1999, 37, 3–16. [CrossRef]
Saad, N.Y.; Fattouh, E.M. Hydraulic characteristics of flow over weirs with circular openings. Ain Shams Eng. J. 2017, 8, 515–522. [CrossRef]
Bagheri, S.; Kabiri-Samani, A.R. Hydraulic Characteristics of flow over the streamlined weirs. Modares Civ. Eng. J. 2018, 17, 29–42.
Hussain, Z.; Khan, S.; Ullah, A.; Ayaz, M.; Ahmad, I.; Mashwani, W.K.; Chu, Y.-M. Extension of optimal homotopy asymptotic method with use of Daftardar–Jeffery polynomials to Hirota–Satsuma coupled system of Korteweg–de Vries equations. Open Phys. 2020, 18, 916–924. [CrossRef]
Arifeen, S.U.; Haq, S.; Ghafoor, A.; Ullah, A.; Kumam, P.; Chaipanya, P. Numerical solutions of higher order boundary value problems via wavelet approach. Adv. Differ. Equ. 2021, 2021, 347. [CrossRef]
Sharafati, A.; Haghbin, M.; Motta, D.; Yaseen, Z.M. The application of soft computing models and empirical formulations for hydraulic structure scouring depth simulation: A comprehensive review, assessment and possible future research direction. Arch. Comput. Methods Eng. 2021, 28, 423–447. [CrossRef]
Khan, S.; Selim, M.M.; Khan, A.; Ullah, A.; Abdeljawad, T.; Ayaz, M.; Mashwani, W.K. On the analysis of the non-Newtonian fluid flow past a stretching/shrinking permeable surface with heat and mass transfer. Coatings 2021, 11, 566. [CrossRef]
Khan, S.; Selim, M.M.; Gepreel, K.A.; Ullah, A.; Ayaz, M.; Mashwani, W.K.; Khan, E. An analytical investigation of the mixed convective Casson fluid flow past a yawed cylinder with heat transfer analysis. Open Phys. 2021, 19, 341–351. [CrossRef]
Ullah, A.; Selim, M.M.; Abdeljawad, T.; Ayaz, M.; Mlaiki, N.; Ghafoor, A. A Magnetite–Water-Based Nanofluid Three-Dimensional Thin Film Flow on an Inclined Rotating Surface with Non-Linear Thermal Radiations and Couple Stress Effects. Energies 2021, 14, 5531. [CrossRef]
Aamir, M.; Ahmad, Z.; Pandey, M.; Khan, M.A.; Aldrees, A.; Mohamed, A. The Effect of Rough Rigid Apron on Scour Downstream of Sluice Gates. Water 2022, 14, 2223. [CrossRef]
Gharebagh, B.A.; Bazargan, J.; Mohammadi, M. Experimental Investigation of Bed Scour Rate in Flood Conditions. Environ. Water Eng. 2022, in press. [CrossRef]
Laishram, K.; Devi, T.T.; Singh, N.B. Experimental Comparison of Hydraulic Jump Characteristics and Energy Dissipation Between Sluice Gate and Radial Gate. In Innovative Trends in Hydrological and Environmental Systems; Springer: Berlin/Heidelberg, Germany, 2022; pp. 207–218.
Varaki, M.E.; Sedaghati, M.; Sabet, B.S. Effect of apron length on local scour at the downstream of grade control structures with labyrinth planform. Arab. J. Geosci. 2022, 15, 1240. [CrossRef]
Rizk, D.; Ullah, A.; Elattar, S.; Alharbi, K.A.M.; Sohail, M.; Khan, R.; Khan, A.; Mlaiki, N. Impact of the KKL Correlation Model on the Activation of Thermal Energy for the Hybrid Nanofluid (GO+ ZnO+ Water) Flow through Permeable Vertically Rotating Surface. Energies 2022, 15, 2872. [CrossRef]
Kim, K.H.; Choi, G.W.; Jo, J.B. An Experimental Study on the Stream Flow by Discharge Ratio. Korea Water Resour. Assoc. Acad. Conf. 2005, 05b, 377–382.
Lee, D.S.; Yeo, H.G. An Experimental Study for Determination of the Material Diameter of Riprap Bed Protection Structure. Korea Water Resour. Assoc. Acad. Conf. 2005, 05b, 1036–1039.
Choi, G.W.; Byeon, S.J.; Kim, Y.G.; Cho, S.U. The Flow Characteristic Variation by Installing a Movable Weir having Water Drainage Equipment on the Bottom. J. Korean Soc. Hazard Mitig. 2008, 8, 117–122.
Jung, J.G. An Experimental Study for Estimation of Bed Protection Length. J. Korean Wetl. Soc. 2011, 13, 677–686.
Kim, S.H.; Kim, W.; Lee, E.R.; Choi, G.H. Analysis of Hydraulic Effects of Singok Submerged Weir in the Lower Han River. J. Korean Water Resour. Assoc. 2005, 38, 401–413. [CrossRef]
Kim, J.H.; Sim, M.P.; Choi, G.W.; Oh, J.M. Hydraulic Analysis of Air Entrainment by Weir Types. J. Korean Water Resour. Assoc. 2003, 36, 971–984. [CrossRef]
Jeong, S.; Yeo, C.G.; Yun, G.S.; Lee, S.O. Analysis of Characteristics for Bank Scour around Low Dam using 3D Numerical Simulation. Korean Soc. Hazard Mitig. Acad. Conf. 2011, 02a, 102.
Son, A.R.; Kim, B.H.; Moon, B.R.; Han, G.Y. An Analysis of Bed Change Characteristics by Bed Protection Work. J. Korean Soc. Civ. Eng. 2015, 35, 821–834.
French, R.H.; French, R.H. Open-Channel Hydraulics; McGraw-Hill: New York, NY, USA, 1985; ISBN 0070221340.
린 첸 가오 양 미시 옹 장 춘밍 왕 Lin Chen , Gaoyang Mi , Xiong Zhang , Chunming Wang * 중국 우한시 화중과학기술대학 재료공학부, 430074
Effects of sinusoidal oscillating laser beam on weld formation, melt flow and grain structure during aluminum alloys lap welding
Abstract
A numerical model of 1.5 mm 6061/5182 aluminum alloys thin sheets lap joints under laser sinusoidal oscillation (sine) welding and laser welding (SLW) weld was developed to simulate temperature distribution and melt flow. Unlike the common energy distribution of SLW, the sinusoidal oscillation of laser beam greatly homogenized the energy distribution and reduced the energy peak. The energy peaks were located at both sides of the sine weld, resulting in the tooth-shaped sectional formation. This paper illustrated the effect of the temperature gradient (G) and solidification rate (R) on the solidification microstructure by simulation. Results indicated that the center of the sine weld had a wider area with low G/R, promoting the formation of a wider equiaxed grain zone, and the columnar grains were slenderer because of greater GR. The porosity-free and non-penetration welds were obtained by the laser sinusoidal oscillation. The reasons were that the molten pool volume was enlarged, the volume proportion of keyhole was reduced and the turbulence in the molten pool was gentled, which was observed by the high-speed imaging and simulation results of melt flow. The tensile test of both welds showed a tensile fracture form along the fusion line, and the tensile strength of sine weld was significantly better than that of the SLW weld. This was because that the wider equiaxed grain area reduced the tendency of cracks and the finer grain size close to the fracture location. Defect-free and excellent welds are of great significance to the new energy vehicles industry.
온도 분포 및 용융 흐름을 시뮬레이션하기 위해 레이저 사인파 진동 (사인) 용접 및 레이저 용접 (SLW) 용접에서 1.5mm 6061/5182 알루미늄 합금 박판 랩 조인트 의 수치 모델이 개발되었습니다. SLW의 일반적인 에너지 분포와 달리 레이저 빔의 사인파 진동은 에너지 분포를 크게 균질화하고 에너지 피크를 줄였습니다. 에너지 피크는 사인 용접의 양쪽에 위치하여 톱니 모양의 단면이 형성되었습니다. 이 논문은 온도 구배(G)와 응고 속도 의 영향을 설명했습니다.(R) 시뮬레이션에 의한 응고 미세 구조. 결과는 사인 용접의 중심이 낮은 G/R로 더 넓은 영역을 가짐으로써 더 넓은 등축 결정립 영역의 형성을 촉진하고 더 큰 GR로 인해 주상 결정립 이 더 가늘다는 것을 나타냅니다. 다공성 및 비관통 용접은 레이저 사인파 진동에 의해 얻어졌습니다. 그 이유는 용융 풀의 부피가 확대되고 열쇠 구멍의 부피 비율이 감소하며 용융 풀의 난류가 완만해졌기 때문이며, 이는 용융 흐름의 고속 이미징 및 시뮬레이션 결과에서 관찰되었습니다. 두 용접부 의 인장시험 은 융착선을 따라 인장파괴형태를인장강도사인 용접의 경우 SLW 용접보다 훨씬 우수했습니다. 이는 등축 결정립 영역이 넓을수록 균열 경향이 감소하고 파단 위치에 근접한 입자 크기가 미세 하기 때문입니다. 결함이 없고 우수한 용접은 신에너지 자동차 산업에 매우 중요합니다.
Fig. 1. Schematic of lap welding for 6061/5182 aluminum alloys.Fig. 2. Finite element mesh.Fig. 3. Weld morphologies of cross-section and upper surface for the two welds: (a) sine pattern weld; (b) SLW weld.Fig. 4. Calculation of laser energy distribution: (a)-(c) sine pattern weld; (d)-(f) SLW weld.Fig. 5. The partially melted region of zone A.Fig. 6. The simulated profiles of melted region for the two welds: (a) SLW weld; (b) sine pattern weld.Fig. 7. The temperature field simulation results of cross section for sine pattern weld.Fig. 8. Dynamic behavior of the molten pool at the same time interval of 0.004 s within one oscillating period: (a) SLW weld; (b) sine pattern weld.Fig. 9. The temperature field and flow field of the molten pool for the SLW weld: (a)~(f) t = 80 ms~100 ms.Fig. 10. The temperature field and flow field of the molten pool for the sine pattern weld: (a)~(f) t = 151 ms~171 ms.Fig. 11. The evolution of the molten pool volume and keyhole depth within one period.Fig. 12. The X-ray inspection results for the two welds: (a) SLW weld, (b) sine pattern weld.Fig. 13. Comparison of the solidification parameters for sine and SLW patterns: (a) the temperature field simulated results of the molten pool upper surfaces; (b)
temperature gradient G and solidification rate R along the molten pool boundary isotherm from weld centerline to the fusion boundary; (c) G/R; (d) GR.Fig. 14. The EBSD results of equiaxed grain zone in the weld center of: (a) sine pattern weld; (b) SLW weld; (c) grain size.Fig. 15. (a) EBSD results of horizontal sections of SLW weld and sine pattern weld; (b) The columnar crystal widths of SLW weld and sine pattern weld.Fig. 16. (a) The tensile test results of the two welds; (b) Fracture location of SLW weld; (b) Fracture location of sine pattern weld.
Keywords
Laser welding, Sinusoidal oscillating, Energy distribution, Numerical simulation, Molten pool flow, Grain structure
References
Chen, X., 2014. Study on laser-MAG Hybrid Weaving Welding Charateristics. Master thesis. Harbin Institute of Technology, China. Chen, G., Wang, B., Mao, S., Zhong, P., He, J., 2019. Research on the “∞”-shaped laser scanning welding process for aluminum alloy. Opt. Laser Technol. 115, 32–41. Cho, W.-I., Na, S.-J., Cho, M.-H., Lee, J.-S., 2010. Numerical study of alloying element distribution in CO2 laser–GMA hybrid welding. Comput. Mater. Sci. 49, 792–800. Cho, W.-I., Na, S.-J., Thomy, C., Vollertsen, F., 2012. Numerical simulation of molten pool dynamics in high power disk laser welding. J. Mater. Process. Technol. 212, 262–275. Das, A., Butterworth, I., Masters, I., Williams, D., 2018. Microstructure and mechanical properties of gap-bridged remote laser welded (RLW) automotive grade AA 5182 joints. Mater. Charact. 145, 697–712. Fetzer, F., Sommer, M., Weber, R., Weberpals, J.-P., Graf, T., 2018. Reduction of pores by means of laser beam oscillation during remote welding of AlMgSi. Opt. Lasers Eng. 108, 68–77. Geng, S., Jiang, P., Shao, X., Guo, L., Gao, X., 2020. Heat transfer and fluid flow and their effects on the solidification microstructure in full-penetration laser welding of aluminum sheet. J. Mater. Sci. Technol. 46, 50–63. Hagenlocher, C., Sommer, M., Fetzer, F., Weber, R., Graf, T., 2018a. Optimization of the solidification conditions by means of beam oscillation during laser beam welding of aluminum. Mater. Des. 160, 1178–1185. Hagenlocher, C., Weller, D., Weber, R., Graf, T., 2018b. Reduction of the hot cracking susceptibility of laser beam welds in AlMgSi alloys by increasing the number of grain boundaries. Sci. Technol. Weld. Join. 24, 313–319. Hagenlocher, C., Fetzer, F., Weller, D., Weber, R., Graf, T., 2019. Explicit analytical expressions for the influence of welding parameters on the grain structure of laser beam welds in aluminium alloys. Mater. Des. 174, 107791. Han, X., Tang, X., Wang, T., Shao, C., Lu, F., Cui, H., 2018. Role of ambient pressure in keyhole dynamics based on beam transmission path method for laser welding on Al alloy. Int. J. Adv. Manuf. Technol. 99, 1639–1651. Hao, K., Li, G., Gao, M., Zeng, X., 2015. Weld formation mechanism of fiber laser oscillating welding of austenitic stainless steel. J. Mater. Process. Technol. 225, 77–83. Hirt, C.W., Nichols, B.D., 1981. Volume of fluid (VOF) method for the dynamics of free boundaries. J. Comput. Phys. 39, 201–225. Jiang, Z., Chen, X., Li, H., Lei, Z., Chen, Y., Wu, S., Wang, Y., 2020. Grain refinement and laser energy distribution during laser oscillating welding of Invar alloy. Mater. Des. 186, 108195. Kaplan, A., 1994. A model of deep penetration laser welding based on calculation of the keyhole profile. J. Phys. D Appl. Phys. 27, 1805–1814. Kou, S., 2002. Welding Metallurgy, 2nd ed. Wiley-Interscience, New Jersey, USA. Kuryntsev, S.V., Gilmutdinov, A.K., 2015. The effect of laser beam wobbling mode in welding process for structural steels. Int. J. Adv. Manuf. Technol. 81, 1683–1691. Li, P., Nie, F., Dong, H., Li, S., Yang, G., Zhang, H., 2018. Pulse MIG welding of 6061-T6/ A356-T6 aluminum alloy dissimilar T-joint. J. Mater. Eng. Perform. 27, 4760–4769. Liu, T., Mu, Z., Hu, R., Pang, S., 2019. Sinusoidal oscillating laser welding of 7075 aluminum alloy: hydrodynamics, porosity formation and optimization. Int. J. Heat Mass Transf. 140, 346–358. Seto, N., Katayama, S., Matsunawa, A., 2000. High-speed simultaneous observation of plasma and keyhole behavior during high power CO2 laser welding: effect of shielding gas on porosity formation. J. Laser Appl. 12, 245–250. Tang, Z., Vollertsen, F., 2014. Influence of grain refinement on hot cracking in laser welding of aluminum. Weld. World 58, 355–366. Wang, L., Gao, M., Zhang, C., Zeng, X., 2016. Effect of beam oscillating pattern on weld characterization of laser welding of AA6061-T6 aluminum alloy. Mater. Des. 108, 707–717. Wang, L., Gao, M., Zeng, X., 2018. Experiment and prediction of weld morphology for laser oscillating welding of AA6061 aluminium alloy. Sci. Technol. Weld. Join. 24, 334–341. Yamazaki, Y., Abe, Y., Hioki, Y., Nakatani, M., Kitagawa, A., Nakata, K., 2016. Fundamental study of narrow-gap welding with oscillation laser beam. Weld. Int. 30, 699–707. Yuan, Z., Tu, Y., Yuan, T., Zhang, Y., Huang, Y., 2021. Size effects on mechanical properties of pure industrial aluminum sheet for micro/meso scale plastic deformation: experiment and modeling. J. Alloys. Compd. 859, 157752. Zou, J., 2016. Characteristics of laser scanning welding process for 5A06 aluminum alloy thick plate with narrow gap. Materials Processing Engineering. Harbin Welding Institute, China. Master thesis.
NadhiraKarimaaIkhaMagdalenaabIndrianaMarcelaaMohammadFaridbaFaculty of Mathematics and Natural Sciences, Bandung Institute of Technology, 40132, IndonesiabCenter for Coastal and Marine Development, Bandung Institute of Technology, Indonesia
Highlights
•A new three-layer model for n-block submerged porous breakwaters is developed.
•New analytical approach in finding the wave transmission coefficient is presented.
•A finite volume method successfully simulates the wave attenuation process.
•Porous media blocks characteristics and configuration can optimize wave reduction.
Abstract
높은 파도 진폭은 해안선에 위험한 영향을 미치고 해안 복원력을 약화시킬 수 있습니다. 그러나 다중 다공성 매체는 해양 생태계의 환경 친화적인 해안 보호 역할을 할 수 있습니다.
이 논문에서 우리는 n개의 잠긴 다공성 미디어 블록이 있는 영역에서 파동 진폭 감소를 계산하기 위해 3층 깊이 통합 방정식을 사용합니다. 수학적 모델은 파동 전달 계수를 얻기 위해 여러 행렬 방정식을 포함하는 변수 분리 방법을 사용하여 해석적으로 해결됩니다.
이 계수는 진폭 감소의 크기에 대한 정보를 제공합니다. 또한 모델을 수치적으로 풀기 위해 지그재그 유한 체적 방법이 적용됩니다.
수치 시뮬레이션을 통해 다공성 매질 블록의 구성과 특성이 투과파 진폭을 줄이는 데 중요하다는 결론을 내렸습니다.
High wave amplitudes may cause dangerous effects on the shoreline and weaken coastal resilience. However, multiple porous media can act as environmental friendly coastal protectors of the marine ecosystem. In this paper, we use three-layer depth-integrated equations to calculate wave amplitude reduction in a domain with n submerged porous media blocks. The mathematical model is solved analytically using the separation of variables method involving several matrix equations to obtain the wave transmission coefficient. This coefficient provides information about the magnitude of amplitude reduction. Additionally, a staggered finite volume method is applied to solve the model numerically. By conducting numerical simulations, we conclude that porous media blocks’ configuration and characteristics are crucial in reducing transmitted wave amplitude.
Fig. 1. Sketch of the problem configuration.Fig. 6. Experiment of waves passing through a single block of porous medium.
References
[1]M. Beck, G. Lange, Managing Coasts with Natural Solutions: Guidelines for Measuring and Valuing the Coastal Protection Services of Mangroves and Coral Reefs.
ArticleDownload PDFView Record in ScopusGoogle Scholar[9]F. Hajivalie, S. M. Mahmoudof, Experimental study of energy dissipation at rectangular submerged breakwater, Proceedings of the 8th International Conference on Fluid Mechanics.
Google Scholar[10]G. T. Klonaris, A. S. Metallinos, C. D. Memos, K. A. Galani, Experimental and numerical investigation of bed morphology in the lee of porous submerged breakwaters, Coast. Eng. 155.
Finite element simulation on the convective double diffusive water-based copper oxide nanofluid flow in a square cavity having vertical wavy surfaces in presence of hydro-magnetic field
Liril D.SilviaDinesh K.ChandrakercSumanaGhoshaArup KDasb aDepartment of Chemical Engineering, Indian Institute of Technology, Roorkee, India bDepartment of Mechanical Engineering, Indian Institute of Technology, Roorkee, India cReactor Engineering Division, Bhabha Atomic Research Centre, Mumbai, India
Abstract
Present work reports numerical understanding of interfacial dynamics during co-flow of vapor and liquid phases of water inside a typical Boiling Water Reactor (BWR), consisting of a nuclear fuel rod bundle assembly of 7 pins in a circular array. Two representative spacings between rods in a circular array are used to carry out the simulation. In literature, flow boiling in a nuclear reactor is dealt with mechanistic models or averaged equations. Hence, in the present study using the Volume of Fluid (VOF) based multiphase model, a detailed numerical understanding of breaking and making in interfaces during flow boiling in BWR is targeted. Our work will portray near realistic vapor bubble and liquid flow dynamics in rod bundle scenario. Constant wall heat flux for fuel rod and uniform velocity of the liquid at the inlet patch is applied as a boundary condition. The saturation properties of water are taken at 30 bar pressure. Flow boiling stages involving bubble nucleation, growth, merging, local dry-out, rewetting with liquid patches, and complete dry-out are illustrated. The dry-out phenomenon with no liquid presence is numerically observed with phase fraction contours at various axial cut-sections. The quantification of the liquid phase fraction at different axial planes is plotted over time, emphasizing the progressive dry-out mechanism. A comparison of liquid-vapor distribution for inner and outer rods reveals that the inner rod’s dry-out occurs sooner than that of the outer rod. The heat transfer coefficient to identify the heat dissipation capacity of each case is also reported.
현재 작업은 원형 배열에 있는 7개의 핀으로 구성된 핵연료봉 다발 어셈블리로 구성된 일반적인 끓는 물 원자로(BWR) 내부의 물의 증기 및 액체상의 동시 흐름 동안 계면 역학에 대한 수치적 이해를 보고합니다.
원형 배열의 막대 사이에 두 개의 대표적인 간격이 시뮬레이션을 수행하는 데 사용됩니다. 문헌에서 원자로의 유동 비등은 기계론적 모델 또는 평균 방정식으로 처리됩니다.
따라서 VOF(Volume of Fluid) 기반 다상 모델을 사용하는 본 연구에서는 BWR에서 유동 비등 동안 계면의 파괴 및 생성에 대한 자세한 수치적 이해를 목표로 합니다.
우리의 작업은 막대 번들 시나리오에서 거의 사실적인 증기 기포 및 액체 흐름 역학을 묘사합니다. 연료봉에 대한 일정한 벽 열유속과 입구 패치에서 액체의 균일한 속도가 경계 조건으로 적용됩니다. 물의 포화 특성은 30bar 압력에서 취합니다.
기포 핵 생성, 성장, 병합, 국소 건조, 액체 패치로 재습윤 및 완전한 건조를 포함하는 유동 비등 단계가 설명됩니다. 액체가 존재하지 않는 건조 현상은 다양한 축 단면에서 위상 분율 윤곽으로 수치적으로 관찰됩니다.
다른 축 평면에서 액상 분율의 정량화는 점진적인 건조 메커니즘을 강조하면서 시간이 지남에 따라 표시됩니다. 내부 막대와 외부 막대의 액-증기 분포를 비교하면 내부 막대의 건조가 외부 막대보다 더 빨리 발생함을 알 수 있습니다. 각 경우의 방열 용량을 식별하기 위한 열 전달 계수도 보고됩니다.
Fig. 1. A typical Boiling Water Reactor (BWR) and selected segment of study for simulationFig. 2. (a-c) dimensions and mesh configuration for G = 6 mm; (d-f) dimensions and mesh configuration for G = 0.6 mmFig. 3. Simulating the effect of spacer (a) Spacer configuration around rod bundle (b) Mesh structure in spacer zone (c) Distribution of vapor bubbles in a rod bundle with
spacer (d) Liquid phase fraction comparison for geometry with and without spacer (e,f,g) Wall temperature comparison for geometry with and without spacer; WS: With
Spacer, WOS: Without Spacer; Temperature in the y-axis is in (f) and (g) is same as (e).Fig. 4. Validation of the present numerical model with crossflow boiling over a heated cylindrical rod [40]Fig. 5. Grid-Independent study in terms of vapor volume in 1/4th of computational domainFig. 6. Interface contour for G = 6 mm; ul = 1.2 m/s; q˙ w = 396 kW/m2; they are showing nucleation, growth, merging, and pseudo-steady-state condition.Fig. 7. Interface contours for G = 0.6 mm; ul = 1.2 m/s; q˙ w = 396 kW/m2; It shows dry-out at pseudo-steady-state near the exitFig. 8. Vapor-liquid distribution across various distant cross-sections (Black color indicates liquid; Gray color indicates vapor); Magnification factor: 1 × (for a and b),
1.5 × (for c and d)Fig. 21. Two-phase flow mixture velocity (u¯z); for G = 6 mm, r = 5 means location at inner heated wall and r = 25 means location at outer adiabatic wall; for G = 0.66
mm, r = 5 means location at inner heated wall and r = 16.6 mm means location at outer adiabatic wall.
References
[1] J. Würtz, An Experimental and Theoretical Investigation of Annular Steam-Water Flow in Tubes and Annuli at 30 to 90 Bar, Risø National Laboratory, Roskilde, 1978. [2] W. Tian, A. Myint, Z. Li, S. Qiu, G.H. Su, D. Jia, Experimental study on dryout point in vertical narrow annulus under low flow conditions, in: International Conference on Nuclear Engineering, 4689, 2004, pp. 643–648. Jan 1Vol. [3] K.M. Becker, C.H. Ling, S. Hedberg, G. Strand, An experimental investigation of post dryout heat transfer, R. Inst. Technol. (1983). [4] K.M. Becker, A Burnout Correlation for Flow of Boiling Water in Vertical Rod Bundles, AB Atomenergi, 1967. [5] Jr J.R. Barbosa, G.F. Hewitt, S.M. Richardson, High-speed visualisation of nucleate boiling in vertical annular flow, Int. J. Heat Mass Transf. 46 (26) (2003) 5153–5160 1, doi:10.1016/S0017-9310(03)00255-2. [6] Y. Mizutani, A. Tomiyama, S. Hosokawa, A. Sou, Y. Kudo, K. Mishima, Twophase flow patterns in a four by four rod bundle, J. Nucl. Sci. Technol. 44 (6) (2007) 894–901 1, doi:10.1080/18811248.2007.9711327. [7] S.S. Paranjape, Two-Phase Flow Interfacial Structures in a Rod Bundle Geometry, Purdue University, 2009. [8] D. Lavicka, J. Polansky, Model of the cooling of a nuclear reactor fuel rod, Multiph. Sci. Technol. 25 (2-4) (2013), doi:10.1615/MultScienTechn.v25.i2-4.90. [9] M. Thurgood, J. Kelly, T. Guidotti, R. Kohrt, K. Crowell, Tech. rep., Pacific Northwest National Laboratory, 1983. [10] S. Sugawara, Droplet deposition and entrainment modeling based on the three-fluid model, Nucl. Eng. Des. 122 (1-3) (1990) 67–84, doi:10.1016/ 0029-5493(90)90197-6. [11] C. Adamsson, J.M. Le Corre, Modeling and validation of a mechanistic tool (MEFISTO) for the prediction of critical power in BWR fuel assemblies, Nucl. Eng. Des. 241 (8) (2011) 2843–2858, doi:10.1016/j.nucengdes.2011.01.033. [12] S. Talebi, H. Kazeminejad, A mathematical approach to predict dryout in a rod bundle, Nucl. Eng. Des. 249 (2012) 348–356, doi:10.1016/j.nucengdes.2012.04. 016. [13] H. Anglart, O. Nylund, N. Kurul, M.Z. Podowski, CFD prediction of flow and phase distribution in fuel assemblies with spacers, Nucl. Eng. Des. 177 (1-3) (1997) 215–228, doi:10.1016/S0029-5493(97)00195-7. [14] H. Li, H. Anglart, CFD model of diabatic annular two-phase flow using the Eulerian–Lagrangian approach, Ann. Nucl. Energy 77 (2015) 415–424, doi:10. 1016/j.anucene.2014.12.002. [15] G. Sorokin, A. Sorokin, Experimental and numerical investigation of liquid metal boiling in fuel subassemblies under natural circulation conditions, Prog. Nucl. Energy 47 (1-4) (2005) 656–663, doi:10.1016/j.pnucene.2005. 05.069. [16] W.D. Pointer, A. Tentner, T. Sofu, D. Weber, S. Lo, A. Splawski, Eulerian two-phase computational fluid dynamics for boiling water reactor core analysis, Joint International Topical Meeting on Mathematics and Computation and Supercomputing in Nuclear Applications (M and C± SNA), 2007. [17] K. Podila, Y. Rao, CFD modelling of supercritical water flow and heat transfer in a 2 × 2 fuel rod bundle, Nucl. Eng. Des. 301 (2016) 279–289, doi:10.1016/j. nucengdes.2016.03.019. [18] H. Pothukuchi, S. Kelm, B.S. Patnaik, B.V. Prasad, H.J. Allelein, Numerical investigation of subcooled flow boiling in an annulus under the influence of eccentricity, Appl. Therm. Eng. 129 (2018) 1604–1617, doi:10.1016/j.applthermaleng. 2017.10.105. [19] H. Pothukuchi, S. Kelm, B.S. Patnaik, B.V. Prasad, H.J. Allelein, CFD modeling of critical heat flux in flow boiling: validation and assessment of closure models, Appl. Therm. Eng. 150 (2019) 651–665, doi:10.1016/j.applthermaleng.2019.01. 030. [20] W. Fan, H. Li, H. Anglart, A study of rewetting and conjugate heat transfer influence on dryout and post-dryout phenomena with a multi-domain coupled CFD approach, Int. J. Heat Mass Transf. 163 (2020) 120503, doi:10.1016/j. ijheatmasstransfer.2020.120503. [21] R. Zhang, T. Cong, G. Su, J. Wang, S. Qiu, Investigation on the critical heat flux in typical 5 by 5 rod bundle at conditions prototypical of PWR based on CFD methodology, Appl. Therm. Eng. 179 (2020) 115582, doi:10.1016/j. applthermaleng.2020.115582.
[22] L.D. Silvi, A. Saha, D.K. Chandraker, S. Ghosh, A.K. Das, Numerical analysis of pre-dryout sequences through the route of interfacial evolution in annular gasliquid two-phase flow with phase change, Chem. Eng. Sci. 212 (2020) 115356, doi:10.1016/j.ces.2019.115356. [23] L.D. Silvi, D.K. Chandraker, S. Ghosh, A.K. Das, On-route to dryout through sequential interfacial dynamics in annular flow boiling around temperature and heat flux controlled heater rod, Chem. Eng. Sci. 229 (2021) 116014, doi:10.1016/ j.ces.2020.116014. [24] J.U. Brackbill, D.B. Kothe, C. Zemach, A continuum method for modeling surface tension, J. Comput. Phys. 100 (2) (1992) 335–354, doi:10.1016/0021-9991(92) 90240-Y. [25] B. Lafaurie, C. Nardone, R. Scardovelli, S. Zaleski, G. Zanetti, Modelling merging and fragmentation in multiphase flows with SURFER, J. Comput. Phys. 113 (1) (1994) 134–147, doi:10.1006/jcph.1994.1123. [26] I. Tanasawa, Advances in condensation heat transfer, Ad. Heat Transf. 21 (1991) 55–139 Vol, doi:10.1016/S0065-2717(08)70334-4. [27] V.H. Del Valle, D.B. Kenning, Subcooled flow boiling at high heat flux, Int. J. Heat Mass Transf. 28 (10) (1985) 1907–1920, doi:10.1016/0017-9310(85) 90213-3. [28] B. Matzner, G.M. Latter, Reduced pressure drop space for boiling water reactor fuel bundles, US Patent US5375154A, (1993) [29] C. Unal, O. Badr, K. Tuzla, J.C. Chen, S. Neti, Pressure drop at rod-bundle spacers in the post-CHF dispersed flow regime, Int. J. Multiphase Flow 20 (3) (1994) 515–522, doi:10.1016/0301-9322(94)90025-6. [30] D.K. Chandraker, A.K. Nayak, V.P. Krishnan, Effect of spacer on the dryout of BWR fuel rod assemblies, Nucl. Eng. Des. 294 (2015), doi:10.1016/j.nucengdes. 2015.09.004. [31] S.K Verma, S.L. Sinha, D.K. Chandraker, A comprehensive review of the spacer effect on performance of nuclear fuel bundle using computational fluid dynamics methodology, Mater. Today: Proc. 4 (2017) 100030–110034, doi:10. 1016/j.matpr.2017.06.315. [32] S.K Verma, S.L. Sinha, D.K. Chandraker, Experimental investigation on the effect of space on the turbulent mixing in vertical pressure tube-type boiling water reactor, Nucl. Sci. Eng. 190 (2) (2018), doi:10.1080/00295639.2017.1413874. [33] T. Zhang, Y. Liu, Numerical investigation of flow and heat transfer characteristics of subcooled boiling in a single rod channel with/without spacer grid, Case Stud. Therm. Eng. 20 (2020) 100644, doi:10.1016/j.csite.2020.100644. [34] K.M. Becker, G. Hernborg, M. Bode, O. Eriksson, Burnout data for flow of boiling water in vertical round ducts, annuli and rod clusters, AB Atomenergi (1965). [35] A. Saha, A.K. Das, Numerical study of boiling around wires and influence of active or passive neighbours on vapour film dynamics, Int. J. Heat Mass Transf. 130 (2019) 440–454, doi:10.1016/j.ijheatmasstransfer.2018.10.117. [36] M. Reimann, U. Grigull, Heat transfer with free convection and film boiling in the critical area of water and carbon dioxide, Heat Mass Transf. 8 (1975) 229– 239, doi:10.1007/BF01002151. [37] M.S. Plesset, S.A. Zwick, The growth of vapor bubbles in superheated liquids, J. Appl. Phys. 25 (4) (1954) 493–500, doi:10.1063/1.1721668. [38] N. Samkhaniani, M.R. Ansari, Numerical simulation of superheated vapor bubble rising in stagnant liquid, Heat Mass Transf. 53 (9) (2017) 2885–2899, doi:10.1007/S00231-017-2031-6. [39] N. Samkhaniani, M.R. Ansari, The evaluation of the diffuse interface method for phase change simulations using OpenFOAM, Heat Transf. Asian Res. 46 (8) (2017) 1173–1203, doi:10.1002/htj.21268. [40] P. Goel, A.K. Nayak, M.K. Das, J.B. Joshi, Bubble departure characteristics in a horizontal tube bundle under cross flow conditions, Int. J. Multiph. Flow 100 (2018) 143–154, doi:10.1016/j.ijmultiphaseflow.2017.12.013. [41] K.M. Becker, J. Engstorm, B.Scholin Nylund, B. Sodequist, Analysis of the dryout incident in the Oskarshamn 2 boiling water reactor, Int. J. Multiph. Flow 16 (6) (1990) 959–974, doi:10.1016/0301-9322(90)90101-N. [42] H.G. Weller, A New Approach to VOF-Based Interface Capturing Methods for Incompressible and Compressible Flow, A New Approach to VOF-Based Interface Capturing Methods for Incompressible and Compressible Flow, 4, OpenCFD Ltd., 2008 Report TR/HGW. [43] G. Boeing, Visual analysis of nonlinear dynamical systems: chaos, fractals, selfsimilarity and the limits of prediction, Systems 4 (4) (2016) 37, doi:10.3390/ systems4040037.
Investigating the breach outflow hydrograph is an essential task to conduct mitigation plans and flood warnings. In the present study, the spatial dam breach is simulated by using a three-dimensional computational fluid dynamics model, FLOW-3D. The model parameters were adjusted by making a comparison with a previous experimental model. The different parameters (initial breach shape, dimensions, location, and dam slopes) are studied to investigate their effects on dam breaching. The results indicate that these parameters have a significant impact. The maximum erosion rate and peak outflow for the rectangular shape are higher than those for the V-notch by 8.85% and 5%, respectively. Increasing breach width or decreasing depth by 5% leads to increasing maximum erosion rate by 11% and 15%, respectively. Increasing the downstream slope angle by 4° leads to an increase in both peak outflow and maximum erosion rate by 2.0% and 6.0%, respectively.
유출 유출 수문곡선을 조사하는 것은 완화 계획 및 홍수 경보를 수행하는 데 필수적인 작업입니다. 본 연구에서는 3차원 전산유체역학 모델인 FLOW-3D를 사용하여 공간 댐 붕괴를 시뮬레이션합니다. 이전 실험 모델과 비교하여 모델 매개변수를 조정했습니다.
다양한 매개변수(초기 붕괴 형태, 치수, 위치 및 댐 경사)가 댐 붕괴에 미치는 영향을 조사하기 위해 연구됩니다. 결과는 이러한 매개변수가 상당한 영향을 미친다는 것을 나타냅니다. 직사각형 형태의 최대 침식율과 최대 유출량은 V-notch보다 각각 8.85%, 5% 높게 나타났습니다.
위반 폭을 늘리거나 깊이를 5% 줄이면 최대 침식률이 각각 11% 및 15% 증가합니다. 하류 경사각을 4° 증가시키면 최대 유출량과 최대 침식률이 각각 2.0% 및 6.0% 증가합니다.
Keywords
Spatial dam breach; FLOW-3D; Overtopping erosion; Computational fluid dynamics (CFD)
1. Introduction
There are many purposes for dam construction, such as protection from flood disasters, water storage, and power generation. Embankment failures may have a catastrophic impact on lives and infrastructure in the downstream regions. One of the most common causes of embankment dam failure is overtopping. Once the overtopping of the dam begins, the breach formation will start in the dam body then end with the dam failure. This failure occurs within a very short time, which threatens to be very dangerous. Therefore, understanding and modeling the embankment breaching processes is essential for conducting mitigation plans, flood warnings, and forecasting flood damage.
The analysis of the dam breaching process is implemented by different techniques: comparative methods, empirical models with dimensional and dimensionless solutions, physical-based models, and parametric models. These models were described in detail [1]. Parametric modeling is commonly used to simulate breach growth as a time-dependent linear process and calculate outflow discharge from the breach using hydraulics principles [2]. Alhasan et al. [3] presented a simple one-dimensional mathematical model and a computer code to simulate the dam breaching process. These models were validated by small dams breaching during the floods in 2002 in the Czech Republic. Fread [4] developed an erosion model (BREACH) based on hydraulics principles, sediment transport, and soil mechanics to estimate breach size, time of formation, and outflow discharge. Říha et al. [5] investigated the dam break process for a cascade of small dams using a simple parametric model for piping and overtopping erosion, as well as a 2D shallow-water flow model for the flood in downstream areas. Goodarzi et al. [6] implemented mathematical and statistical methods to assess the effect of inflows and wind speeds on the dam’s overtopping failure.
Dam breaching studies can be divided into two main modes of erosion. The first mode is called “planar dam breach” where the flow overtops the whole dam width. While the second mode is called “spatial dam breach” where the flow overtops through the initial pilot channel (i.e., a channel created in the dam body). Therefore, the erosion will be in both vertical and horizontal directions [7].
The erosion process through the embankment dams occurs due to the shear stress applied by water flows. The dam breaching evolution can be divided into three stages [8], [9], but Y. Yang et al. [10] divided the breach development into five stages: Stage I, the seepage erosion; Stage II, the initial breach formation; Stage III, the head erosion; Stage IV, the breach expansion; and Stage V, the re-equilibrium of the river channel through the breach. Many experimental tests have been carried out on non-cohesive embankment dams with an initial breach to examine the effect of upstream inflow discharges on the longitudinal profile evolution and the time to inflection point[11].
Zhang et al. [12] studied the effect of changing downstream slope angle, sediment grain size, and dam crest length on erosion rates. They noticed that increasing dam crest length and decreasing downstream slope angle lead to decreasing sediment transport rate. While the increase in sediment grain size leads to an increased sediment transport rate at the initial stages. Höeg et al. [13] presented a series of field tests to investigate the stability of embankment dams made of various materials. Overtopping and piping were among the failure tests carried out for the dams composed of homogeneous rock-fill, clay, or gravel with a height of up to 6.0 m. Hakimzadeh et al. [14] constructed 40 homogeneous cohesive and non-cohesive embankment dams to study the effect of changing sediment diameter and dam height on the breaching process. They also used genetic programming (GP) to estimate the breach outflow. Refaiy et al. [15] studied different scenarios for the downstream drain geometry, such as length, height, and angle, to minimize the effect of piping phenomena and therefore increase dam safety.
Zhu et al. [16] examined the effect of headcut erosion on dam breach growth, especially in the case of cohesive dams. They found that the breach growth in non-cohesive embankments is slower than cohesive embankments due to the little effect of headcut. Schmocker and Hager [7] proposed a relationship for estimating peak outflow from the dam breach process.(1)QpQin-1=1.7exp-20hc23d5013H0
where: Qp = peak outflow discharge.
Qin = inflow discharge.
hc = critical flow depth.
d50 = mean sediment diameter.
Ho = initial dam height.
Yu et al. [17] carried out an experimental study for homogeneous non-cohesive embankment dams in a 180° bending rectangular flume to determine the effect of overtopping flows on breaching formation. They found that the main factors influencing breach formation are water level, river discharge, and embankment material diameter.
Wu et al. [18] carried out a series of experiments to investigate the effect of breaching geometry on both non-cohesive and cohesive embankment dams in a U-bend flume due to overtopping flows. In the case of non-cohesive embankments, the non-symmetrical lateral expansion was noticed during the breach formation. This expansion was described by a coefficient ranging from 2.7 to 3.3.
The numerical models of the dam breach can be categorized according to different parameters, such as flow dimensions (1D, 2D, or 3D), flow governing equations, and solution methods. The 1D models are mainly used to predict the outflow hydrograph from the dam breach. Saberi et al. [19] applied the 1D Saint-Venant equation, which is solved by the finite difference method to investigate the outflow hydrograph during dam overtopping failure. Because of the ability to study dam profile evolution and breach formation, 2D models are more applicable than 1D models. Guan et al. [20] and Wu et al. [21] employed both 2D shallow water equations (SWEs) and sediment erosion equations, which are solved by the finite volume method to study the effect of the dam’s geometry parameters on outflow hydrograph and dam profile evolution. Wang et al. [22] also proposed a second-order hybrid-type of total variation diminishing (TVD) finite-difference to estimate the breach outflow by solving the 2D (SWEs). The accuracy of (SWEs) for both vertical flow contraction and surface roughness has been assessed [23]. They noted that the accuracy of (SWEs) is acceptable for milder slopes, but in the case of steeper slopes, modelers should be more careful. Generally, the accuracy of 2D models is still low, especially with velocity distribution over the flow depth, lateral momentum exchange, density-driven flows, and bottom friction[24]. Therefore, 3D models are preferred. Larocque et al. [25] and Yang et al. [26] started to use three-dimensional (3D) models that depend on the Reynolds-averaged Navier-Stokes (RANS) equations.
Previous experimental studies concluded that there is no clear relationship between the peak outflow from the dam breach and the initial breach characteristics. Some of these studies depend on the sharp-crested weir fixed at the end of the flume to determine the peak outflow from the breach, which leads to a decrease in the accuracy of outflow calculations at the microscale. The main goals of this study are to carry out a numerical simulation for a spatial dam breach due to overtopping flows by using (FLOW-3D) software to find an empirical equation for the peak outflow discharge from the breach and determine the worst-case that leads to accelerating the dam breaching process.
2. Numerical simulation
The current study for spatial dam breach is simulated by using (FLOW-3D) software [27], which is a powerful computational fluid dynamics (CFD) program.
2.1. Geometric presentations
A stereolithographic (STL) file is prepared for each change in the initial breach geometry and dimensions. The CAD program is useful for creating solid objects and converting them to STL format, as shown in Fig. 1.
2.2. Governing equations
The governing equations for water flow are three-dimensional Reynolds Averaged Navier-Stokes equations (RANS).
The momentum equation:(3)∂ui∂t+1VFuj∂ui∂xj=1ρ∂∂xj-pδij+ν∂ui∂xj+∂uj∂xi-ρu`iu`j¯
where u is time-averaged velocity,ν is kinematic viscosity, VF is fractional volume open to flow, p is averaged pressure and -u`iu`j¯ are components of Reynold’s stress. The Volume of Fluid (VOF) technique is used to simulate the free surface profile. Hirt et al. [28] presented the VOF algorithm, which employs the function (F) to express the occupancy of each grid cell with fluid. The value of (F) varies from zero to unity. Zero value refers to no fluid in the grid cell, while the unity value refers to the grid cell being fully occupied with fluid. The free surface is formed in the grid cells having (F) values between zero and unity.(4)∂F∂t+1VF∂∂xFAxu+∂∂yFAyv+∂∂zFAzw=0
where (u, v, w) are the velocity components in (x, y, z) coordinates, respectively, and (Ax, Ay, Az) are the area fractions.
2.3. Boundary and initial conditions
To improve the accuracy of the results, the boundary conditions should be carefully determined. In this study, two mesh blocks are used to minimize the time consumed in the simulation. The boundary conditions for mesh block 1 are as follows: The inlet and sides boundaries are defined as a wall boundary condition (wall boundary condition is usually used for bound fluid by solid regions. In the case of viscous flows, no-slip means that the tangential velocity is equal to the wall velocity and the normal velocity is zero), the outlet is defined as a symmetry boundary condition (symmetry boundary condition is usually used to reduce computational effort during CFD simulation. This condition allows the flow to be transferred from one mesh block to another. No inputs are required for this boundary condition except that its location should be defined accurately), the bottom boundary is defined as a uniform flow rate boundary condition, and the top boundary is defined as a specific pressure boundary condition with assigned atmospheric pressure. The boundary conditions for mesh block 2 are as follows: The inlet is defined as a symmetry boundary condition, the outlet is defined as a free flow boundary condition, the bottom and sides boundaries are defined as a wall boundary condition, and the top boundary is defined as a specific pressure boundary condition with assigned atmospheric pressure as shown in Fig. 2. The initial conditions required to be set for the fluid (i.e., water) inside of the domain include configuration, temperature, velocities, and pressure distribution. The configuration of water depends on the dimensions and shape of the dam reservoir. While the other conditions have been assigned as follows: temperature is normal water temperature (25 °c) and pressure distribution is hydrostatic with no initial velocity.
2.4. Numerical method
FLOW-3D uses the finite volume method (FVM) to solve the governing equation (Reynolds-averaged Navier-Stokes) over the computational domain. A finite-volume method is an Eulerian approach for representing and evaluating partial differential equations in algebraic equations form [29]. At discrete points on the mesh geometry, values are determined. Finite volume expresses a small volume surrounding each node point on a mesh. In this method, the divergence theorem is used to convert volume integrals with a divergence term to surface integrals. After that, these terms are evaluated as fluxes at each finite volume’s surfaces.
2.5. Turbulent models
Turbulence is the chaotic, unstable motion of fluids that occurs when there are insufficient stabilizing viscous forces. In FLOW-3D, there are six turbulence models available: the Prandtl mixing length model, the one-equation turbulent energy model, the two-equation (k – ε) model, the Renormalization-Group (RNG) model, the two-equation (k – ω) models, and a large eddy simulation (LES) model. For simulating flow motion, the RNG model is adopted to simulate the motion behavior better than the k – ε and k – ω.
models [30]. The RNG model consists of two main equations for the turbulent kinetic energy KT and its dissipation.εT(5)∂kT∂t+1VFuAx∂kT∂x+vAy∂kT∂y+wAz∂kT∂z=PT+GT+DiffKT-εT(6)∂εT∂t+1VFuAx∂εT∂x+vAy∂εT∂y+wAz∂εT∂z=C1.εTKTPT+c3.GT+Diffε-c2εT2kT
where KT is the turbulent kinetic energy, PT is the turbulent kinetic energy production, GT is the buoyancy turbulence energy, εT is the turbulent energy dissipation rate, DiffKT and Diffε are terms of diffusion, c1, c2 and c3 are dimensionless parameters, in which c1 and c3 have a constant value of 1.42 and 0.2, respectively, c2 is computed from the turbulent kinetic energy (KT) and turbulent production (PT) terms.
2.6. Sediment scour model
The sediment scour model available in FLOW-3D can calculate all the sediment transport processes including Entrainment transport, Bedload transport, Suspended transport, and Deposition. The erosion process starts once the water flows remove the grains from the packed bed and carry them into suspension. It happens when the applied shear stress by water flows exceeds critical shear stress. This process is represented by entrainment transport in the numerical model. After entrained, the grains carried by water flow are represented by suspended load transport. After that, some suspended grains resort to settling because of the combined effect of gravity, buoyancy, and friction. This process is described through a deposition. Finally, the grains sliding motions are represented by bedload transport in the model. For the entrainment process, the shear stress applied by the fluid motion on the packed bed surface is calculated using the standard wall function as shown in Eq.7.(7)ks,i=Cs,i∗d50
where ks,i is the Nikuradse roughness and Cs,i is a user-defined coefficient. The critical bed shear stress is defined by a dimensionless parameter called the critical shields number as expressed in Eq.8.(8)θcr,i=τcr,i‖g‖diρi-ρf
where θcr,i is the critical shields number, τcr,i is the critical bed shear stress, g is the absolute value of gravity acceleration, di is the diameter of the sediment grain, ρi is the density of the sediment species (i) and ρf is the density of the fluid. The value of the critical shields number is determined according to the Soulsby-Whitehouse equation.(9)θcr,i=0.31+1.2d∗,i+0.0551-exp-0.02d∗,i
where d∗,i is the dimensionless diameter of the sediment, given by Eq.10.(10)d∗,i=diρfρi-ρf‖g‖μf213
where μf is the fluid dynamic viscosity. For the sloping bed interface, the value of the critical shields number is modified according to Eq.11.(11)θ`cr,i=θcr,icosψsinβ+cos2βtan2φi-sin2ψsin2βtanφi
where θ`cr,i is the modified critical shields number, φi is the angle of repose for the sediment, β is the angle of bed slope and ψ is the angle between the flow and the upslope direction. The effects of the rolling, hopping, and sliding motions of grains along the packed bed surface are taken by the bedload transport process. The volumetric bedload transport rate (qb,i) per width of the bed is expressed in Eq.12.(12)qb,i=Φi‖g‖ρi-ρfρfdi312
where Φi is the dimensionless bedload transport rate is calculated by using Meyer Peter and Müller equation.(13)Φi=βMPM,iθi-θ`cr,i1.5cb,i
where βMPM,i is the Meyer Peter and Müller user-defined coefficient and cb,i is the volume fraction of species i in the bed material. The suspended load transport is calculated as shown in Eq.14.(14)∂Cs,i∂t+∇∙Cs,ius,i=∇∙∇DCs,i
where Cs,i is the suspended sediment mass concentration, D is the diffusivity, and us,i is the grain velocity of species i. Entrainment and deposition are two opposing processes that take place at the same time. The lifting and settling velocities for both entrainment and deposition processes are calculated according to Eq.15 and Eq.16, respectively.(15)ulifting,i=αid∗,i0.3θi-θ`cr,igdiρiρf-1(16)usettling,i=υfdi10.362+1.049d∗,i3-10.36
where αi is the entrainment coefficient of species i and υf is the kinematic viscosity of the fluid.
2.7. Grid type
Using simple rectangular orthogonal elements in planes and hexahedral in volumes in the (FLOW-3D) program makes the mesh generation process easier, decreases the required memory, and improves numerical accuracy. Two mesh blocks were used in a joined form with a size ratio of 2:1. The first mesh block is coarser, which contains the reservoir water, and the second mesh block is finer, which contains the dam. For achieving accuracy and efficiency in results, the mesh size is determined by using a grid convergence test. The optimum uniform cell size for the first mesh block is 0.012 m and for the second mesh block is 0.006 m.
2.8. Time step
The maximum time step size is determined by using a Courant number, which controls the distance that the flow will travel during the simulation time step. In this study, the Courant number was taken equal to 0.25 to prevent the flow from traveling through more than one cell in the time step. Based on the Courant number, a maximum time step value of 0.00075 s was determined.
2.9. Numerical model validation
The numerical model accuracy was achieved by comparing the numerical model results with previous experimental results. The experimental study of Schmocker and Hager [7] was based on 31 tests with changes in six parameters (d50, Ho, Bo, Lk, XD, and Qin). All experimental tests were conducted in a straight open glass-sided flume. The horizontal flume has a rectangular cross-section with a width of 0.4 m and a height of 0.7 m. The flume was provided with a flow straightener and an intake with a length of 0.66 m. All tested dams were inserted at various distances (XD) from the intake. Test No.1 from this experimental program was chosen to validate the numerical model. The different parameters used in test No.1 are as follows:
(1) uniform sediment with a mean diameter (d50 = 0.31 mm), (2) Ho = 0.2 m, (3) Bo = 0.2 m, (4) Lk = 0.1 m,
(5) XD = 1.0 m, (6) Qin = 6.0 lit/s, (7) Su and Sd = 2:1, (8) mass density (ρs = 2650 kg/m3) (9) Homogenous and non-cohesive embankment dam. As shown in Fig. 2, the simulation is contained within a rectangular grid with dimensions: 3.56 m in the x-direction (where 0.66 m is used as inlet, 0.9 m as dam base width, and 1.0 m as outlet), in y-direction 0.2 m (dam length), and in the z-direction 0.3 m, which represents the dam height (0.2 m) with a free distance (0.1 m) above the dam. There are two main reasons that this experimental program is preferred for the validation process. The first reason is that this program deals with homogenous, non-cohesive soil, which is available in FLOW-3D. The second reason is that this program deals with small-scale models which saves time for numerical simulation. Finally, some important assumptions were considered during the validation process. The flow is assumed to be incompressible, viscous, turbulent, and three-dimensional.
By comparing dam profiles at different time instants for the experimental test with the current numerical model, it appears that the numerical model gives good agreement as shown in Fig. 3 and Fig. 4, with an average error percentage of 9% between the experimental results and the numerical model.
3. Analysis and discussions
The current model is used to study the effects of different parameters such as (initial breach shapes, dimensions, locations, upstream and downstream dam slopes) on the peak outflow discharge, QP, time of peak outflow, tP, and rate of erosion, E.
This study consists of a group of scenarios. The first scenario is changing the shapes of the initial breach according to Singh [1], the most predicted shapes are rectangular and V-notch as shown in Fig. 5. The second scenario is changing the initial breach dimensions (i.e., width and depth). While the third scenario is changing the location of the initial breach. Eventually, the last scenario is changing the upstream and downstream dam slopes.
All scenarios of this study were carried out under the same conditions such as inflow discharge value (Qin=1.0lit/s), dimensions of the tested dam, where dam height (Ho=0.20m), crest width.
(Lk=0.1m), dam length (Bo=0.20m), and homogenous & non-cohesive soil with a mean diameter (d50=0.31mm).
3.1. Dam breaching process evolution
The dam breaching process is a very complex process due to the quick changes in hydrodynamic conditions during dam failure. The dam breaching process starts once water flows reach the downstream face of the dam. During the initial stage of dam breaching, the erosion process is relatively quiet due to low velocities of flow. As water flows continuously, erosion rates increase, especially in two main zones: the crest and the downstream face. As soon as the dam crest is totally eroded, the water levels in the dam reservoir decrease rapidly, accompanied by excessive erosion in the dam body. The erosion process continues until the water levels in the dam reservoir equal the remaining height of the dam.
According to Zhou et al. [11], the breaching process consists of three main stages. The first stage starts with beginning overtopping flow, then ends when the erosion point directed upstream and reached the inflection point at the inflection time (ti). The second stage starts from the end of the stage1 until the occurrence of peak outflow discharge at the peak outflow time (tP). The third stage starts from the end of the stage2 until the value of outflow discharge becomes the same as the value of inflow discharge at the final time (tf). The outflow discharge from the dam breach increases rapidly during stage1 and stage2 because of the large dam storage capacity (i.e., the dam reservoir is totally full of water) and excessive erosion. While at stage3, the outflow values start to decrease slowly because most of the dam’s storage capacity was run out. The end of stage3 indicates that the dam storage capacity was totally run out, so the outflow equalized with the inflow discharge as shown in Fig. 6 and Fig. 7.
3.2. The effect of initial breach shape
To identify the effect of the initial breach shape on the evolution of the dam breaching process. Three tests were carried out with different cross-section areas for each shape. The initial breach is created at the center of the dam crest. Each test had an ID to make the process of arranging data easier. The rectangular shape had an ID (Rec5h & 5b), which means that its depth and width are equal to 5% of the dam height, and the V-notch shape had an ID (V-noch5h & 1:1) which means that its depth is equal to 5% of the dam height and its side slope is equal to 1:1. The comparison between rectangular and V-notch shapes is done by calculating the ratio between maximum dam height at different times (ZMax) to the initial dam height (Ho), rate of erosion, and hydrograph of outflow discharge for each test. The rectangular shape achieves maximum erosion rate and minimum inflection time, in addition to a rapid decrease in the dam reservoir levels. Therefore, the dam breaching is faster in the case of a rectangular shape than in a V-notch shape, which has the same cross-section area as shown in Fig. 8.
Also, by comparing the hydrograph for each test, the peak outflow discharge value in the case of a rectangular shape is higher than the V-notch shape by 5% and the time of peak outflow for the rectangular shape is shorter than the V-notch shape by 9% as shown in Fig. 9.
3.3. The effect of initial breach dimensions
The results of the comparison between the different initial breach shapes indicate that the worst initial breach shape is rectangular, so the second scenario from this study concentrated on studying the effect of a change in the initial rectangular breach dimensions. Groups of tests were carried out with different depths and widths for the rectangular initial breach. The first group had a depth of 5% from the dam height and with three different widths of 5,10, and 15% from the dam height, the second group had a depth of 10% with three different widths of 5,10, and 15%, the third group had a depth of 15% with three different widths of 5,10, and 15% and the final group had a width of 15% with three different heights of 5, 10, and 15% for a rectangular breach shape. The comparison was made as in the previous section to determine the worst case that leads to the quick dam failure as shown in Fig. 10.
The results show that the (Rec 5 h&15b) test achieves a maximum erosion rate for a shorter period of time and a minimum ratio for (Zmax / Ho) as shown in Fig. 10, which leads to accelerating the dam failure process. The dam breaching process is faster with the minimum initial breach depth and maximum initial breach width. In the case of a minimum initial breach depth, the retained head of water in the dam reservoir is high and the crest width at the bottom of the initial breach (L`K) is small, so the erosion point reaches the inflection point rapidly. While in the case of the maximum initial breach width, the erosion perimeter is large.
3.4. The effect of initial breach location
The results of the comparison between the different initial rectangular breach dimensions indicate that the worst initial breach dimension is (Rec 5 h&15b), so the third scenario from this study concentrated on studying the effect of a change in the initial breach location. Three locations were checked to determine the worst case for the dam failure process. The first location is at the center of the dam crest, which was named “Center”, the second location is at mid-distance between the dam center and dam edge, which was named “Mid”, and the third location is at the dam edge, which was named “Edge” as shown in Fig. 11. According to this scenario, the results indicate that the time of peak outflow discharge (tP) is the same in the three cases, but the maximum value of the peak outflow discharge occurs at the center location. The difference in the peak outflow values between the three cases is relatively small as shown in Fig. 12.
The rates of erosion were also studied for the three cases. The results show that the maximum erosion rate occurs at the center location as shown in Fig. 13. By making a comparison between the three cases for the dam storage volume. The results show that the center location had the minimum values for the dam storage volume, which means that a large amount of water has passed to the downstream area as shown in Fig. 14. According to these results, the center location leads to increased erosion rate and accelerated dam failure process compared with the two other cases. Because the erosion occurs on both sides, but in the case of edge location, the erosion occurs on one side.
3.5. The effect of upstream and downstream dam slopes
The results of the comparison between the different initial rectangular breach locations indicate that the worst initial breach location is the center location, so the fourth scenario from this study concentrated on studying the effect of a change in the upstream (Su) and downstream (Sd) dam slopes. Three slopes were checked individually for both upstream and downstream slopes to determine the worst case for the dam failure process. The first slope value is (2H:1V), the second slope value is (2.5H:1V), and the third slope value is (3H:1V). According to this scenario, the results show that the decreasing downstream slope angle leads to increasing time of peak outflow discharge (tP) and decreasing value of peak outflow discharge. The difference in the peak outflow values between the three cases for the downstream slope is 2%, as shown in Fig. 15, but changing the upstream slope has a negligible impact on the peak outflow discharge and its time as shown in Fig. 16.
The rates of erosion were also studied in the three cases for both upstream and downstream slopes. The results show that the maximum erosion rate increases by 6.0% with an increasing downstream slope angle by 4°, as shown in Fig. 17. The results also indicate that the erosion rates aren’t affected by increasing or decreasing the upstream slope angle, as shown in Fig. 18. According to these results, increasing the downstream slope angle leads to increased erosion rate and accelerated dam failure process compared with the upstream slope angle. Because of increasing shear stress applied by water flows in case of increasing downstream slope.
According to all previous scenarios, the dimensionless peak outflow discharge QPQin is presented for a fixed dam height (Ho) and inflow discharge (Qin). Fig. 19 illustrates the relationship between QP∗=QPQin and.
Lr=ho2/3∗bo2/3Ho. The deduced relationship achieves R2=0.96.(17)QP∗=2.2807exp-2.804∗Lr
4. Conclusions
A spatial dam breaching process was simulated by using FLOW-3D Software. The validation process was performed by making a comparison between the simulated results of dam profiles and the dam profiles obtained by Schmocker and Hager [7] in their experimental study. And also, the peak outflow value recorded an error percentage of 12% between the numerical model and the experimental study. This model was used to study the effect of initial breach shape, dimensions, location, and dam slopes on peak outflow discharge, time of peak outflow, and the erosion process. By using the parameters obtained from the validation process, the results of this study can be summarized in eight points as follows.1.
The rectangular initial breach shape leads to an accelerating dam failure process compared with the V-notch.2.
The value of peak outflow discharge in the case of a rectangular initial breach is higher than the V-notch shape by 5%.3.
The time of peak outflow discharge for a rectangular initial breach is shorter than the V-notch shape by 9%.4.
The minimum depth and maximum width for the initial breach achieve maximum erosion rates (increasing breach width, b0, or decreasing breach depth, h0, by 5% from the dam height leads to an increase in the maximum rate of erosion by 11% and 15%, respectively), so the dam failure is rapid.5.
The center location of the initial breach leads to an accelerating dam failure compared with the edge location.6.
The initial breach location has a negligible effect on the peak outflow discharge value and its time.7.
Increasing the downstream slope angle by 4° leads to an increase in both peak outflow discharge and maximum rate of erosion by 2.0% and 6.0%, respectively.8.
The upstream slope has a negligible effect on the dam breaching process.
References
[1]V. SinghDam breach modeling technologySpringer Science & Business Media (1996)Google Scholar[2]Wahl TL. Prediction of embankment dam breach parameters: a literature review and needs assessment. 1998.Google Scholar[3]Z. Alhasan, J. Jandora, J. ŘíhaStudy of dam-break due to overtopping of four small dams in the Czech RepublicActa Universitatis Agriculturae et Silviculturae Mendelianae Brunensis, 63 (3) (2015), pp. 717-729 View PDFCrossRefView Record in ScopusGoogle Scholar[4]D. FreadBREACH, an erosion model for earthen dam failures: Hydrologic Research LaboratoryNOAA, National Weather Service (1988)Google Scholar[5]J. Říha, S. Kotaška, L. PetrulaDam Break Modeling in a Cascade of Small Earthen Dams: Case Study of the Čižina River in the Czech RepublicWater, 12 (8) (2020), p. 2309, 10.3390/w12082309 View PDFView Record in ScopusGoogle Scholar[6]E. Goodarzi, L. Teang Shui, M. ZiaeiDam overtopping risk using probabilistic concepts–Case study: The Meijaran DamIran Ain Shams Eng J, 4 (2) (2013), pp. 185-197ArticleDownload PDFView Record in ScopusGoogle Scholar[7]L. Schmocker, W.H. HagerPlane dike-breach due to overtopping: effects of sediment, dike height and dischargeJ Hydraul Res, 50 (6) (2012), pp. 576-586 View PDFCrossRefView Record in ScopusGoogle Scholar[8]J.S. Walder, R.M. Iverson, J.W. Godt, M. Logan, S.A. SolovitzControls on the breach geometry and flood hydrograph during overtopping of noncohesive earthen damsWater Resour Res, 51 (8) (2015), pp. 6701-6724View Record in ScopusGoogle Scholar[9]H. Wei, M. Yu, D. Wang, Y. LiOvertopping breaching of river levees constructed with cohesive sedimentsNat Hazards Earth Syst Sci, 16 (7) (2016), pp. 1541-1551 View PDFCrossRefView Record in ScopusGoogle Scholar[10]Y. Yang, S.-Y. Cao, K.-J. Yang, W.-P. LiYang K-j, Li W-p. Experimental study of breach process of landslide dams by overtopping and its initiation mechanismsJ Hydrodynamics, 27 (6) (2015), pp. 872-883ArticleDownload PDFCrossRefView Record in ScopusGoogle Scholar[11]G.G.D. Zhou, M. Zhou, M.S. Shrestha, D. Song, C.E. Choi, K.F.E. Cui, et al.Experimental investigation on the longitudinal evolution of landslide dam breaching and outburst floodsGeomorphology, 334 (2019), pp. 29-43ArticleDownload PDFView Record in ScopusGoogle Scholar[12]J. Zhang, Z.-x. Guo, S.-y. CaoYang F-g. Experimental study on scour and erosion of blocked damWater Sci Eng, 5 (2012), pp. 219-229ArticleDownload PDFView Record in ScopusGoogle Scholar[13]K. Höeg, A. Løvoll, K. VaskinnStability and breaching of embankment dams: Field tests on 6 m high damsInt J Hydropower Dams, 11 (2004), pp. 88-92View Record in ScopusGoogle Scholar[14]H. Hakimzadeh, V. Nourani, A.B. AminiGenetic programming simulation of dam breach hydrograph and peak outflow dischargeJ Hydrol Eng, 19 (4) (2014), pp. 757-768View Record in ScopusGoogle Scholar[15]A.R. Refaiy, N.M. AboulAtta, N.Y. Saad, D.A. El-MollaModeling the effect of downstream drain geometry on seepage through earth damsAin Shams Eng J, 12 (3) (2021), pp. 2511-2531ArticleDownload PDFView Record in ScopusGoogle Scholar[16]Y. Zhu, P.J. Visser, J.K. Vrijling, G. WangExperimental investigation on breaching of embankmentsScience China Technological Sci, 54 (1) (2011), pp. 148-155 View PDFCrossRefView Record in ScopusGoogle Scholar[17]M.-H. Yu, H.-Y. Wei, Y.-J. Liang, Y. ZhaoInvestigation of non-cohesive levee breach by overtopping flowJ Hydrodyn, 25 (4) (2013), pp. 572-579ArticleDownload PDFCrossRefView Record in ScopusGoogle Scholar[18]S. Wu, M. Yu, H. Wei, Y. Liang, J. ZengNon-symmetrical levee breaching processes in a channel bend due to overtoppingInt J Sedim Res, 33 (2) (2018), pp. 208-215ArticleDownload PDFView Record in ScopusGoogle Scholar[19]O. Saberi, G. ZenzNumerical investigation on 1D and 2D embankment dams failure due to overtopping flowInt J Hydraulic Engineering, 5 (2016), pp. 9-18View Record in ScopusGoogle Scholar[20]M. Guan, N.G. Wright, P.A. Sleigh2D Process-Based Morphodynamic Model for Flooding by Noncohesive Dyke BreachJ Hydraul Eng, 140 (7) (2014), p. 04014022, 10.1061/(ASCE)HY.1943-7900.0000861 View PDFView Record in ScopusGoogle Scholar[21]W. Wu, R. Marsooli, Z. HeDepth-Averaged Two-Dimensional Model of Unsteady Flow and Sediment Transport due to Noncohesive Embankment Break/BreachingJ Hydraul Eng, 138 (6) (2012), pp. 503-516View Record in ScopusGoogle Scholar[22]Z. Wang, D.S. BowlesThree-dimensional non-cohesive earthen dam breach model. Part 1: Theory and methodologyAdv Water Resour, 29 (10) (2006), pp. 1528-1545ArticleDownload PDFView Record in ScopusGoogle Scholar[23]Říha J, Duchan D, Zachoval Z, Erpicum S, Archambeau P, Pirotton M, et al. Performance of a shallow-water model for simulating flow over trapezoidal broad-crested weirs. J Hydrology Hydromechanics. 2019;67:322-8.Google Scholar[24]C.B. VreugdenhilNumerical methods for shallow-water flowSpringer Science & Business Media (1994)Google Scholar[25]L.A. Larocque, J. Imran, M.H. Chaudhry3D numerical simulation of partial breach dam-break flow using the LES and k–∊ turbulence modelsJ Hydraul Res, 51 (2) (2013), pp. 145-157 View PDFCrossRefView Record in ScopusGoogle Scholar[26]C. Yang, B. Lin, C. Jiang, Y. LiuPredicting near-field dam-break flow and impact force using a 3D modelJ Hydraul Res, 48 (6) (2010), pp. 784-792 View PDFCrossRefView Record in ScopusGoogle Scholar[27]FLOW-3D. Version 11.1.1 Flow Science, Inc., Santa Fe, NM. https://wwwflow3dcom.Google Scholar[28]C.W. Hirt, B.D. NicholsVolume of fluid (VOF) method for the dynamics of free boundariesJ Comput Phys, 39 (1) (1981), pp. 201-225ArticleDownload PDFGoogle Scholar[29]S.V. PatankarNumerical heat transfer and fluid flow, Hemisphere PublCorp, New York, 58 (1980), p. 288View Record in ScopusGoogle Scholar[30]M. Alemi, R. MaiaNumerical simulation of the flow and local scour process around single and complex bridge piersInt J Civil Eng, 16 (5) (2018), pp. 475-487 View PDFCrossRefView Record in ScopusGoogle Scholar
W.E. Alphonso1, M.Bayat1,*, M. Baier 2, S. Carmignato2, J.H. Hattel1 1Department of Mechanical Engineering, Technical University of Denmark (DTU), Lyngby, Denmark 2Department of Management and Engineering – University of Padova, Padova, Italy
ABSTRACT
L-PBF(Laser Powder Bed Fusion)는 레이저 열원을 사용하여 선택적으로 통합되는 분말 층으로 복잡한 3D 금속 부품을 만드는 금속 적층 제조(MAM) 기술입니다. 처리 영역은 수십 마이크로미터 정도이므로 L-PBF를 다중 규모 제조 공정으로 만듭니다.
기체 기공의 형성 및 성장 및 용융되지 않은 분말 영역의 생성은 다중물리 모델에 의해 예측할 수 있습니다. 또한 이러한 모델을 사용하여 용융 풀 모양 및 크기, 온도 분포, 용융 풀 유체 흐름 및 입자 크기 및 형태와 같은 미세 구조 특성을 계산할 수 있습니다.
이 작업에서는 용융, 응고, 유체 흐름, 표면 장력, 열 모세관, 증발 및 광선 추적을 통한 다중 반사를 포함하는 스테인리스 스틸 316-L에 대한 충실도 다중 물리학 중간 규모 수치 모델이 개발되었습니다. 완전한 실험 설계(DoE) 방법을 사용하는 통계 연구가 수행되었으며, 여기서 불확실한 재료 특성 및 공정 매개변수, 즉 흡수율, 반동 압력(기화) 및 레이저 빔 크기가 용융수지 모양 및 크기에 미치는 영향을 분석했습니다.
또한 용융 풀 역학에 대한 위에서 언급한 불확실한 입력 매개변수의 중요성을 강조하기 위해 흡수율이 가장 큰 영향을 미치고 레이저 빔 크기가 그 뒤를 잇는 주요 효과 플롯이 생성되었습니다. 용융 풀 크기에 대한 반동 압력의 중요성은 흡수율에 따라 달라지는 용융 풀 부피와 함께 증가합니다.
모델의 예측 정확도는 유사한 공정 매개변수로 생성된 단일 트랙 실험과 시뮬레이션의 용융 풀 모양 및 크기를 비교하여 검증됩니다.
더욱이, 열 렌즈 효과는 레이저 빔 크기를 증가시켜 수치 모델에서 고려되었으며 나중에 결과적인 용융 풀 프로파일은 모델의 견고성을 보여주기 위한 실험과 비교되었습니다.
Laser Powder Bed Fusion (L-PBF) is a Metal Additive Manufacturing (MAM) technology where a complex 3D metal part is built from powder layers, which are selectively consolidated using a laser heat source. The processing zone is in the order of a few tenths of micrometer, making L-PBF a multi-scale manufacturing process. The formation and growth of gas pores and the creation of un-melted powder zones can be predicted by multiphysics models. Also, with these models, the melt pool shape and size, temperature distribution, melt pool fluid flow and its microstructural features like grain size and morphology can be calculated. In this work, a high fidelity multi-physics meso-scale numerical model is developed for stainless steel 316-L which includes melting, solidification, fluid flow, surface tension, thermo-capillarity, evaporation and multiple reflection with ray-tracing. A statistical study using a full Design of Experiments (DoE) method was conducted, wherein the impact of uncertain material properties and process parameters namely absorptivity, recoil pressure (vaporization) and laser beam size on the melt pool shape and size was analysed. Furthermore, to emphasize on the significance of the above mentioned uncertain input parameters on the melt pool dynamics, a main effects plot was created which showed that absorptivity had the highest impact followed by laser beam size. The significance of recoil pressure on the melt pool size increases with melt pool volume which is dependent on absorptivity. The prediction accuracy of the model is validated by comparing the melt pool shape and size from the simulation with single track experiments that were produced with similar process parameters. Moreover, the effect of thermal lensing was considered in the numerical model by increasing the laser beam size and later on the resultant melt pool profile was compared with experiments to show the robustness of the model.
Figure 1: a) Computational domain for single track L-PBF which includes a 200 μm thick substrate and 45 μm powder layer
thickness b) 3D temperature contour plot after scanning a single track with melt pool contours at two locations along the
scanning direction where the green region indicates the melted regions.Figure 2: Main effects plot of uncertain parameters: absorptivity, recoil pressure coefficient and laser beam radius on the melt
pool dimensions (width and depth)Figure 3: 3D temperature contours and 2D melt pool cross-sections where the melt pool is stabilized at x=500 µm from the
start of the laser initial location for cases where (a) absorptivity = 0.1, Recoil pressure coefficient B = 1 and laser beam radius
= 12 µm, (b) absorptivity = 0.1, Recoil pressure coefficient B = 20 and laser beam radius = 12 µm, (c) absorptivity = 0.1, Recoil
pressure coefficient B = 1 and laser beam radius = 18 µm, (d) absorptivity = 0.45, Recoil pressure coefficient B = 1 and laser
beam radius = 18 µm, (e) absorptivity = 0.45, Recoil pressure coefficient B = 20 and laser beam radius = 12 µm, (f) absorptivity
= 0.45, Recoil pressure coefficient B = 20 and laser beam radius = 18 µm.Figure 4: Validation of Numerical model with Recoil pressure coefficient B= 20, absorptivity = 0.45 and a) laser beam radius
= 15 µm b) laser beam radius = 20 µm
CONCLUSION
In this work, a high-fidelity multi-physics numerical model was developed for L-PBF using the FVM method in Flow-3D. The impact of uncertainty in the input parameters including absorptivity, recoil pressure and laser beam size on the melt pool is addressed using a DoE method. The DoE analysis shows that absorptivity has the highest impact on the melt pool. The recoil pressure and laser beam size only become significant once absorptivity is 0.45. Furthermore, the numerical model is validated by comparing the predicted melt pool shape and size with experiments conducted with similar process parameters wherein a high prediction accuracy is achieved by the model. In addition, the impact of thermal lensing on the melt pool dimensions by increasing the laser beam spot size is considered in the validated numerical model and the resultant melt pool is compared with experiments.
REFERENCES
[1] T. Bonhoff, M. Schniedenharn, J. Stollenwerk, P. Loosen, Experimental and theoretical analysis of thermooptical effects in protective window for selective laser melting, Proc. Int. Conf. Lasers Manuf. LiM. (2017) 26–29. https://www.wlt.de/lim/Proceedings2017/Data/PDF/Contribution31_final.pdf. [2] L.R. Goossens, Y. Kinds, J.P. Kruth, B. van Hooreweder, On the influence of thermal lensing during selective laser melting, Solid Free. Fabr. 2018 Proc. 29th Annu. Int. Solid Free. Fabr. Symp. – An Addit. Manuf. Conf. SFF 2018. (2020) 2267–2274. [3] J. Shinjo, C. Panwisawas, Digital materials design by thermal-fluid science for multi-metal additive manufacturing, Acta Mater. 210 (2021) 116825. https://doi.org/10.1016/j.actamat.2021.116825. [4] Z. Zhang, Y. Huang, A. Rani Kasinathan, S. Imani Shahabad, U. Ali, Y. Mahmoodkhani, E. Toyserkani, 3- Dimensional heat transfer modeling for laser powder-bed fusion additive manufacturing with volumetric heat sources based on varied thermal conductivity and absorptivity, Opt. Laser Technol. 109 (2019) 297–312. https://doi.org/10.1016/j.optlastec.2018.08.012. [5] M. Bayat, A. Thanki, S. Mohanty, A. Witvrouw, S. Yang, J. Thorborg, N.S. Tiedje, J.H. Hattel, Keyholeinduced porosities in Laser-based Powder Bed Fusion (L-PBF) of Ti6Al4V: High-fidelity modelling and experimental validation, Addit. Manuf. 30 (2019) 100835. https://doi.org/10.1016/j.addma.2019.100835. [6] M. Bayat, S. Mohanty, J.H. Hattel, Multiphysics modelling of lack-of-fusion voids formation and evolution in IN718 made by multi-track/multi-layer L-PBF, Int. J. Heat Mass Transf. 139 (2019) 95–114. https://doi.org/10.1016/j.ijheatmasstransfer.2019.05.003. [7] J. Metelkova, Y. Kinds, K. Kempen, C. de Formanoir, A. Witvrouw, B. Van Hooreweder, On the influence of laser defocusing in Selective Laser Melting of 316L, Addit. Manuf. 23 (2018) 161–169. https://doi.org/10.1016/j.addma.2018.08.006.
M. Bayat* , V. K. Nadimpalli, J. H. Hattel 1Department of Mechanical Engineering, Technical University of Denmark (DTU), Produktionstorvet 425, Kgs. 2800, Lyngby, Denmark
ABSTRACT
L-PBF(Laser Powder Bed Fusion)는 다양한 산업 분야에서 많은 관심을 받았으며, 주로 기존 제조 기술을 사용하여 만들 수 없었던 복잡한 토폴로지 최적화 구성 요소를 구현하는 잘 알려진 능력 덕분입니다. . 펄스 L-PBF(PL-PBF)에서 레이저의 시간적 프로파일은 주기 지속 시간과 듀티 주기 중 하나 또는 둘 다를 수정하여 변조할 수 있습니다. 따라서 레이저의 시간적 프로파일은 향후 적용을 위해 이 프로세스를 더 잘 제어할 수 있는 길을 열어주는 새로운 프로세스 매개변수로 간주될 수 있습니다. 따라서 이 작업에서 우리는 레이저의 시간적 프로파일을 변경하는 것이 PL-PBF 공정에서 용융 풀 조건과 트랙의 최종 모양 및 형상에 어떻게 영향을 미칠 수 있는지 조사하는 것을 목표로 합니다. 이와 관련하여 본 논문에서는 CFD(Computational Fluid Dynamics) 소프트웨어 패키지인 Flow-3D를 기반으로 하는 316-L 스테인리스강 PL-PBF 공정의 다중물리 수치 모델을 개발하고 이 모델을 사용하여 열과 유체를 시뮬레이션합니다. 다양한 펄스 모드에서 공정 과정 중 용융 풀 내부에서 발생하는 유동 조건. 따라서 고정된 레이저 듀티 사이클(50%)이 있는 레이저 주기 지속 시간이 용융 풀의 모양과 크기 및 최종 트랙 형태에 미치는 영향을 연구하기 위해 매개변수 연구가 수행됩니다. 더 긴 주기 기간에서 더 많은 재료가 더 큰 용융 풀 내에서 변위됨에 따라 용융 풀의 후류에 더 눈에 띄는 혹이 형성되며, 동시에 더 심각한 반동 압력을 받습니다. 또한 시뮬레이션에서 50% 듀티 사이클에서 1000μs에서 형성된 보다 대칭적인 용융 풀과 비교하여 400μs 사이클 주기에서 더 긴 용융 풀이 형성된다는 것이 관찰되었습니다. 풀 볼륨은 1000μs의 경우 더 큽니다. 매개변수 연구는 연속 트랙과 파손된 트랙 PL-PBF 사이의 경계를 설명하며, 여기서 연속 트랙은 항상 소량의 용융 재료를 유지함으로써 유지됩니다.
English Abstract
Laser Powder Bed Fusion (L-PBF) has attracted a lot of attention from various industrial sectors and mainly thanks to its well-proven well-known capacity of realizing complex topology-optimized components that have so far been impossible to make using conventional manufacturing techniques. In Pulsed L-PBF (PL-PBF), the laser’s temporal profile can be modulated via modifying either or both the cycle duration and the duty cycle. Thus, the laser’s temporal profile could be considered as a new process parameter that paves the way for a better control of this process for future applications. Therefore, in this work we aim to investigate how changing the laser’s temporal profile can affect the melt pool conditions and the final shape and geometry of a track in the PL-PBF process. In this respect, in this paper a multiphysics numerical model of the PL-PBF process of 316-L stainless steel is developed based on the computational fluid dynamics (CFD) software package Flow-3D and the model is used to simulate the heat and fluid flow conditions occurring inside the melt pool during the course of the process at different pulsing modes. Thus, a parametric study is carried out to study the influence of the laser’s cycle duration with a fixed laser duty cycle (50 %) on the shape and size of the melt pool and the final track morphology. It is noticed that at longer cycle periods, more noticeable humps form at the wake of the melt pool as more material is displaced within bigger melt pools, which are at the same time subjected to more significant recoil pressures. It is also observed in the simulations that at 50 % duty cycle, longer melt pools form at 400 μs cycle period compared to the more symmetrical melt pools formed at 1000 μs, primarily because of shorter laser off-times in the former, even though melt pool volume is bigger for the 1000 μs case. The parameteric study illustrates the boundary between a continuous track and a broken track PL-PBF wherein the continuous track is retained by always maintaining a small volume of molten material.
Figure 1: Front and side views of the computational domain. Note that the region along z and from -100 μm to +50 μm is void.Figure 2: Temperature contours and melt pool border lines at different times for the 50 % duty cycle case: (a) – (c) Δtcycle =
400 μs, (d) – (f) Δtcycle = 1000 μs and (g) – (i) Δtcycle = 3000 μs.Figure 3: Plot of melt pool volume versus time for four cases including continuous wave laser as well as 50 % duty cycle at
400 μs, 1000 μs and 3000 μs.
CONCLUSIONS
In this work a CFD model of the modulated PL-PBF process of stainless steel 316-L is developed in the commercial software package Flow-3D. The model involves physics such as solidification, melting, evaporation, convection, laser-material interaction, capillarity, Marangoni effect and the recoil pressure effect. In the current study, a parametric study is carried out to understand how the change in the cycle period duration affects the melt pool’s thermo-fluid conditions during the modulated PL-PBF process. It is observed that at the pulse mode with 50 % duty cycle and 400 μs cycle period, an overlapped chain of humps form at the wake of the melt pool and at a spatial frequency of occurrence of about 78 μm. Furthermore and as expected, it is noted that the melt pool volume, the size of the hump as well as the crater size at the end of the track, increase with increase in the cycle period duration, as more material is re-deposited at the back of the melt pool and that itself is caused by more pronounced recoil pressures. Moreover, it is noticed that due to the short off-time period of the laser in the 400 μs cycle period case, there is always an amount of liquid metal left from the previous cycle, at the time the new cycle starts. This is found to be the main reason why longer and elongated melt pools form at 400 μs cycle period, compared to the bigger, shorter and more symmetrical-like melt pools forming at the 1000 μs case. In this study PL-PBF single tracks including the broken track and the continuous track examples were studied to illustrate the boundary of this transition at a given laser scan parameter setting. At higher scan speeds, it is expected that the Plateau–Rayleigh instability will compete with the pulsing behavior to change the transition boundary between a broken and continuous track, which is suggested as future work from this study.
REFERENCES
[1] T. Craeghs, L. Thijs, F. Verhaeghe, J.-P. Kruth, J. Van Humbeeck, A study of the microstructural evolution during selective laser melting of Ti–6Al–4V, Acta Mater. 58 (2010) 3303–3312. https://doi.org/10.1016/j.actamat.2010.02.004. [2] J. Liu, A.T. Gaynor, S. Chen, Z. Kang, K. Suresh, A. Takezawa, L. Li, J. Kato, J. Tang, C.C.L. Wang, L. Cheng, X. Liang, A.C. To, Current and future trends in topology optimization for additive manufacturing, (2018) 2457–2483. [3] M. Bayat, W. Dong, J. Thorborg, A.C. To, J.H. Hattel, A review of multi-scale and multi-physics simulations of metal additive manufacturing processes with focus on modeling strategies, Addit. Manuf. 47 (2021). https://doi.org/10.1016/j.addma.2021.102278. [4] A. Foroozmehr, M. Badrossamay, E. Foroozmehr, S. Golabi, Finite Element Simulation of Selective Laser Melting process considering Optical Penetration Depth of laser in powder bed, Mater. Des. 89 (2016) 255–263. https://doi.org/10.1016/j.matdes.2015.10.002. [5] Y.S. Lee, W. Zhang, Modeling of heat transfer, fluid flow and solidification microstructure of nickel-base superalloy fabricated by laser powder bed fusion, Addit. Manuf. 12 (2016) 178–188. https://doi.org/10.1016/j.addma.2016.05.003. [6] S.A. Khairallah, A.T. Anderson, A. Rubenchik, W.E. King, Laser powder-bed fusion additive manufacturing: Physics of complex melt flow and formation mechanisms of pores, spatter, and denudation zones, Acta Mater. 108 (2016) 36–45. https://doi.org/10.1016/j.actamat.2016.02.014. [7] M. Bayat, A. Thanki, S. Mohanty, A. Witvrouw, S. Yang, J. Thorborg, N.S. Tiedje, J.H. Hattel, Keyholeinduced porosities in Laser-based Powder Bed Fusion (L-PBF) of Ti6Al4V: High-fidelity modelling and experimental validation, Addit. Manuf. 30 (2019). https://doi.org/10.1016/j.addma.2019.100835. [8] A. Charles, M. Bayat, A. Elkaseer, L. Thijs, J.H. Hattel, S. Scholz, Elucidation of dross formation in laser powder bed fusion at down-facing surfaces: phenomenon-oriented multiphysics simulation and experimental validation, Addit. Manuf. Under revi (2021). [9] M. Bayat, V.K. Nadimpalli, D.B. Pedersen, J.H. Hattel, A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys, Int. J. Heat Mass Transf. 166 (2021) 120766. https://doi.org/10.1016/j.ijheatmasstransfer.2020.120766. [10] J.D. Roehling, S.A. Khairallah, Y. Shen, A. Bayramian, C.D. Boley, A.M. Rubenchik, J. Demuth, N. Duanmu, M.J. Matthews, Physics of large-area pulsed laser powder bed fusion, Addit. Manuf. 46 (2021) https://doi.org/10.1016/j.addma.2021.102186. [11] M. Zheng, L. Wei, J. Chen, Q. Zhang, J. Li, S. Sui, G. Wang, W. Huang, Surface morphology evolution during pulsed selective laser melting: Numerical and experimental investigations, Appl. Surf. Sci. 496 (2019) 143649. https://doi.org/10.1016/j.apsusc.2019.143649. [12] M. Bayat, V.K. Nadimpalli, D.B. Pedersen, J.H. Hattel, A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys, Int. J. Heat Mass Transf. 166 (2021). https://doi.org/10.1016/j.ijheatmasstransfer.2020.120766.
A 3D numerical model of heat transfer and fluid flow of molten pool in the process of laser wire deposition was presented by computational fluid dynamics technique. The simulation results of the deposition morphology were also compared with the experimental results under the condition of liquid bridge transfer mode. Moreover, they showed a good agreement. Considering the effect of recoil pressure, the morphology of the deposit metal obtained by the simulation was similar to the experiment result. Molten metal at the wire tip was peeled off and flowed into the molten pool, and then spread to both sides of the deposition layer under the recoil pressure. In addition, the results of simulation and high-speed charge-coupled device presented that a wedge transition zone, with a length of ∼6 mm, was formed behind the keyhole in the liquid bridge transfer process, where the height of deposited metal decreased gradually. After solidification, metal in the transition zone retained the original melt morphology, resulting in a decrease in the height of the tail of the deposition layer.
Keywords
LWD, CFD, liquid bridge transfer, fluid dynamics, wedge transition zone
Fluid Thermodynamic Simulation of Ti-6Al-4V Alloy in Laser Wire DepositionFluid Thermodynamic Simulation of Ti-6Al-4V Alloy in Laser Wire Deposition
References
1. Matthews MJ, Guss G, Khairallah SA, et al. Denudation of metal powder layers in laser powder bed fusion processes. Acta Mater 2016;114:33–42. Crossref, Google Scholar
2. Ge WJ, Han SW, Fang YC, et al. Mechanism of surface morphology in electron beam melting of Ti6Al4V based on computational flow patterns. Appl Surf Sci 2017;419:150–158. Crossref, Google Scholar
3. Bai XW, Colegrove P, Ding JL, et al. Numerical analyswas of heat transfer and fluid flow in multilayer deposition of PAW-based wire and arc additive manufacturing. Int J Heat Mass Transf 2018;124:504–516. Crossref, Google Scholar
4. Torkamany MJ, Kaplan AFH, Ghaini FM. Wire deposition by a laser-induced boiling front. Opt Laser Technol 2015;69:104–112. Crossref, Google Scholar
5. Yu Y, Huang W, Wang G. Investigation of melting dynamics of filler wire during wire feed laser welding. J Mec Sci Technol 2013;27:1097–1108. Crossref, Google Scholar
6. Ma G, Li L, Chen Y. Effects of beam confgurations on wire melting and transfer behaviors in dual beam laser welding with fller wire. Opt Laser Technol 2017;91:138–148. Crossref, Google Scholar
7. Abioye TE, Folkes J, Clare AT. A parametric study of Inconel 625 wire laser deposition. J Mater Process Tech 2013;213:2145–2151. Crossref, Google Scholar
8. Wei S, Wang G, Shin YC, et al. Comprehensive modeling of transport phenomena in laser hot-wire deposition process. Int J Heat Mass Transf 2018;125:1356–1368. Crossref, Google Scholar
9. Gu H, Li L. Computational fluid dynamic simulation of gravity and pressure effects in laser metal deposition for potential additive manufacturing in space. Int J Heat Mass Transf 2019;140:51–65. Crossref, Google Scholar
10. Hu R, Luo M, Liu T, et al. Thermal fluid dynamics of liquid bridge transfer in laser wire deposition 3D printing. Sci Technolf Weld Join 2019;24:1–11. Google Scholar
11. Chatterjee D, Chakraborty S. A hybrid lattice Boltzmann model for solid–liquid phase transition in presence of fluid flow. Phys Lett A 2006;351:359–367. Crossref, Google Scholar
12. Wu L, Cheon J, Kiran DV, et al. CFD simulations of GMA welding of horizontal fillet joints based on coordinate rotation of arc models. J Mater Process Tech 2016;231:221–238. Crossref, Google Scholar
13. Gerhard W, Boyer RR, Collings EW. Materials Properties Handbook: Titanium Alloys. ASM International: Almere, The Netherlands, 1994. Google Scholar
14. Colegrove P, Simiand PE, Varughese A, et al. Evaluation of a drilling model approach to represent laser spot microwelding. In: ASM Proceedings of the international conference: trends in welding research; 2009. Google Scholar
15. Boivineau M, Cagran C, Doytier D, et al. Thermophysical properties of solid and liquid Ti-6Al-4V (TA6V) alloy. Int J Thermophys 2006;27:507–529. Crossref, Google Scholar
16. Shejndlin AE, Kenisarin MM, Chekhovskoj VY. Melting point of yttrium oxide. AN SSSR 1974;216:582–584. Google Scholar
17. Cho JH, Na SJ. Teflection and Fresnel absorption of laser beam in keyhole. J Phys D Appl Phys 2006;39:5372–5378. Crossref, Google Scholar
18. Han SW, Ahn J, Na SJ. A study on ray tracing method for CFD simulations of laser keyhole welding: Progressive search method. Weld World 2016;60:247–258. Crossref, Google Scholar
19. Allmen MV. Laser-Beam Interactions with Materials. Springer, Berlin-Heidelberg, 1995. Google Scholar
20. Dobson PJ. Absorption and scattering of light by small particles. Phys Bull 1984;35:104. Crossref, Google Scholar
21. Greses J, Hilton PA, Barlow CY. Plume attenuation under high power Nd:yttritium aluminum garnet laser welding. J Laser Appl 2004;16:9–15. Crossref, Google Scholar
22. Shcheglov PY, Uspenskiy SA, Gumenyuk AV, et al. Plume attenuation of laser radiation during high power fiber laser welding. Laser Phys Lett 2011;8:475–480. Crossref, Google Scholar
23. Yang P, Liou KN. Effective refractive index for determining ray propagation in an absorbing dielectric particle. J Quant Spectrosc Radiat Transf 2009;110:300–306. Crossref, Google Scholar
24. Barber PW. Absorption and scattering of light by small particles. J Colloid Interface Sci 1984;98:290–291. Google Scholar
25. Hu ZR, Chen X, Yang G, et al. Metal transfer in wire feeding-based electron beam 3D printing: Modes, dynamics, and transition criterion. Int J Heat Mass Transf 2018;126:877–887. Crossref, Google Scholar
26. David SA, Babu SS, Vitek JM. Welding: Solidification and microstructure. JOM 2013;55:14–20. Crossref, Google Scholar
27. Zhong ML, Liu W. Laser surface cladding: The state of the art and challenges. Proc Inst Mech Eng Part C J Mech Eng Sci 2010;224:1041–1060. Crossref, Google Scholar
28. Kobryn PA, Semiatin S. Microstructure and texture evolution during solidification processing of Ti-6Al-4V. J Mater Process Technol 2003;135:330–339. Crossref, Google Scholar
29. Debroy T, David S. Physical processes in fusion welding. Rev Mod Phys 1995;67:85–112. Crossref, Google Scholar
30. Lee YS, Nordin M, Babu SS, et al. Effect of fluid convection on dendrite arm spacing in laser deposition. Metall Trans B 2014;45:1520–1528. Crossref, Google Scholar
31. Rappaz M, David SA, Vitek JM, et al. Development of microstructures in Fe15Ni15Cr single crystal electron beam welds. Metall Trans A 1989;20:1125–1138. Crossref, Google Scholar
레이저 분말 베드 퓨전(L-PBF) 적층 제조(AM)는 우수한 기계적 특성으로 그물 모양에 가까운 복잡한 부품을 생산할 수 있습니다. 그러나 빌드 실패 및 다공성과 같은 결함으로 이어지는 원치 않는 잔류 응력 및 왜곡이 L-PBF의 광범위한 적용을 방해하고 있습니다.
L-PBF의 잠재력을 최대한 실현하기 위해 잔류 변형, 용융 풀 및 다공성 형성을 예측하는 다중 규모 모델링 방법론이 개발되었습니다. L-PBF의 잔류 변형 및 응력을 부품 규모에서 예측하기 위해 고유 변형 방법을 기반으로 하는 다중 규모 프로세스 모델링 프레임워크가 제안됩니다.
고유한 변형 벡터는 마이크로 스케일에서 충실도가 높은 상세한 다층 프로세스 시뮬레이션에서 추출됩니다. 균일하지만 이방성인 변형은 잔류 왜곡 및 응력을 예측하기 위해 준 정적 평형 유한 요소 분석(FEA)에서 레이어별로 L-PBF 부품에 적용됩니다.
부품 규모에서의 잔류 변형 및 응력 예측 외에도 분말 규모의 다중물리 모델링을 수행하여 공정 매개변수, 예열 온도 및 스패터링 입자에 의해 유도된 용융 풀 변동 및 결함 형성을 연구합니다. 이러한 요인과 관련된 용융 풀 역학 및 다공성 형성 메커니즘은 시뮬레이션 및 실험을 통해 밝혀졌습니다.
제안된 부품 규모 잔류 응력 및 왜곡 모델을 기반으로 경로 계획 방법은 큰 잔류 변형 및 건물 파손을 방지하기 위해 주어진 형상에 대한 레이저 스캐닝 경로를 조정하기 위해 개발되었습니다.
연속 및 아일랜드 스캐닝 전략을 위한 기울기 기반 경로 계획이 공식화되고 공식화된 컴플라이언스 및 스트레스 최소화 문제에 대한 전체 감도 분석이 수행됩니다. 이 제안된 경로 계획 방법의 타당성과 효율성은 AconityONE L-PBF 시스템을 사용하여 실험적으로 입증되었습니다.
또한 기계 학습을 활용한 데이터 기반 프레임워크를 개발하여 L-PBF에 대한 부품 규모의 열 이력을 예측합니다. 본 연구에서는 실시간 열 이력 예측을 위해 CNN(Convolutional Neural Network)과 RNN(Recurrent Neural Network)을 포함하는 순차적 기계 학습 모델을 제안합니다.
유한 요소 해석과 비교하여 100배의 예측 속도 향상이 달성되어 실제 제작 프로세스보다 빠른 예측이 가능하고 실시간 온도 프로파일을 사용할 수 있습니다.
Laser powder bed fusion (L-PBF) additive manufacturing (AM) is capable of producing complex parts near net shape with good mechanical properties. However, undesired residual stress and distortion that lead to build failure and defects such as porosity are preventing broader applications of L-PBF. To realize the full potential of L-PBF, a multiscale modeling methodology is developed to predict residual deformation, melt pool, and porosity formation. To predict the residual deformation and stress in L-PBF at part-scale, a multiscale process modeling framework based on inherent strain method is proposed.
Inherent strain vectors are extracted from detailed multi-layer process simulation with high fidelity at micro-scale. Uniform but anisotropic strains are then applied to L-PBF part in a layer-by-layer fashion in a quasi-static equilibrium finite element analysis (FEA) to predict residual distortion and stress. Besides residual distortion and stress prediction at part scale, multiphysics modeling at powder scale is performed to study the melt pool variation and defect formation induced by process parameters, preheating temperature and spattering particles. Melt pool dynamics and porosity formation mechanisms associated with these factors are revealed through simulation and experiments.
Based on the proposed part-scale residual stress and distortion model, path planning method is developed to tailor the laser scanning path for a given geometry to prevent large residual deformation and building failures. Gradient based path planning for continuous and island scanning strategy is formulated and full sensitivity analysis for the formulated compliance- and stress-minimization problem is performed.
The feasibility and effectiveness of this proposed path planning method is demonstrated experimentally using the AconityONE L-PBF system. In addition, a data-driven framework utilizing machine learning is developed to predict the thermal history at part-scale for L-PBF.
In this work, a sequential machine learning model including convolutional neural network (CNN) and recurrent neural network (RNN), long shortterm memory unit, is proposed for real-time thermal history prediction. A 100x prediction speed improvement is achieved compared to the finite element analysis which makes the prediction faster than real fabrication process and real-time temperature profile available.
Figure 1.1: Schematic Overview of Metal Laser Powder Bed Fusion Process [2]Figure 1.2: Commercial Powder Bed Fusion SystemsFigure 1.3: Commercial Metal Components Fabricated by Powder Bed Fusion Additive Manufacturing: (a) GE Fuel Nozzle; (b) Stryker Hip Biomedical Implant.Figure 2.1: Proposed Multiscale Process Simulation FrameworkFigure 2.2: (a) Experimental Setup for In-situ Thermocouple Measurement in the EOS M290 Build Chamber; (b) Themocouple Locations on the Bottom Side of the Substrate.Figure 2.3: (a) Finite Element Model for Single Layer Thermal Analysis; (b) Deposition LayerFigure 2.4: Core-skin layer: (a) Surface Morphology; (b) Scanning Strategy; (c) Transient Temperature Distribution and Temperature History at (d) Point 1; (e) Point 2 and (f) Point 3Figure 2.5: (a) Scanning Orientation of Each Layer; (b) Finite Element Model for Micro-scale Representative VolumeFigure 2.6: Bottom Layer (a) Thermal History; (b) Plastic Strain and (c) Elastic Strain Evolution HistoryFigure 2.7: Bottom Layer Inherent Strain under Default Process Parameters along Horizontal Scanning PathFigure 2.8: Snapshots of the Element Activation ProcessFigure 2.9: Double Cantilever Beam Structure Built by the EOS M290 DMLM Process (a) Before and (b)
After Cutting off; (c) Faro Laser ScanArm V3 for Distortion MeasurementFigure 2.10: Square Canonical Structure Built by the EOS M290 DMLM ProcessFigure 2.11: Finite Element Mesh for the Square Canonical and Snapshots of Element Activation ProcessFigure 2.12: Simulated Distortion Field for the Double Cantilever Beam before Cutting off the Supports: (a) Inherent Strain Method; (b) Simufact Additive 3.1Figure 3.10: Snapshots of Temperature Profile for Single Track in Keyhole Regime (P = 250W and V = 0.5m/s) at the Preheating Temperature of 100 °Cs) at the Preheating Temperature of 500 °CFigure 3.15: Melt Pool Cross Section Comparison Between Simulation and Experiment for Single Track
Bibliography
[1] I. Astm, ASTM52900-15 Standard Terminology for Additive Manufacturing—General Principles—Terminology, ASTM International, West Conshohocken, PA 3(4) (2015) 5. [2] W.E. King, A.T. Anderson, R.M. Ferencz, N.E. Hodge, C. Kamath, S.A. Khairallah, A.M. Rubenchik, Laser powder bed fusion additive manufacturing of metals; physics, computational, and materials challenges, Applied Physics Reviews 2(4) (2015) 041304. [3] W. Yan, Y. Lu, K. Jones, Z. Yang, J. Fox, P. Witherell, G. Wagner, W.K. Liu, Data-driven characterization of thermal models for powder-bed-fusion additive manufacturing, Additive Manufacturing (2020) 101503. [4] K. Dai, L. Shaw, Thermal and stress modeling of multi-material laser processing, Acta Materialia 49(20) (2001) 4171-4181. [5] K. Dai, L. Shaw, Distortion minimization of laser-processed components through control of laser scanning patterns, Rapid Prototyping Journal 8(5) (2002) 270-276. [6] S.S. Bo Cheng, Kevin Chou, Stress and deformation evaluations of scanning strategy effect in selective laser melting, Additive Manufacturing (2017). [7] C. Fu, Y. Guo, Three-dimensional temperature gradient mechanism in selective laser melting of Ti-6Al-4V, Journal of Manufacturing Science and Engineering 136(6) (2014) 061004. [8] P. Prabhakar, W.J. Sames, R. Dehoff, S.S. Babu, Computational modeling of residual stress formation during the electron beam melting process for Inconel 718, Additive Manufacturing 7 (2015) 83-91. [9] A. Hussein, L. Hao, C. Yan, R. Everson, Finite element simulation of the temperature and stress fields in single layers built without-support in selective laser melting, Materials & Design (1980-2015) 52 (2013) 638-647. [10] P.Z. Qingcheng Yang, Lin Cheng, Zheng Min, Minking Chyu, Albert C. To, articleFinite element modeling and validation of thermomechanicalbehavior of Ti-6Al-4V in directed energy deposition additivemanufacturing, Additive Manufacturing (2016). [11] E.R. Denlinger, J. Irwin, P. Michaleris, Thermomechanical Modeling of Additive Manufacturing Large Parts, Journal of Manufacturing Science and Engineering 136(6) (2014) 061007. [12] E.R. Denlinger, M. Gouge, J. Irwin, P. Michaleris, Thermomechanical model development and in situ experimental validation of the Laser Powder-Bed Fusion process, Additive Manufacturing 16 (2017) 73-80. [13] V.J. Erik R Denlinger, G.V. Srinivasan, Tahany EI-Wardany, Pan Michaleris, Thermal modeling of Inconel 718 processed with powder bed fusionand experimental validation using in situ measurements, Additive Manufacturing 11 (2016) 7-15. [14] N. Patil, D. Pal, H.K. Rafi, K. Zeng, A. Moreland, A. Hicks, D. Beeler, B. Stucker, A Generalized Feed Forward Dynamic Adaptive Mesh Refinement and Derefinement Finite Element Framework for Metal Laser Sintering—Part I: Formulation and Algorithm Development, Journal of Manufacturing Science and Engineering 137(4) (2015) 041001. [15] D. Pal, N. Patil, K.H. Kutty, K. Zeng, A. Moreland, A. Hicks, D. Beeler, B. Stucker, A Generalized Feed-Forward Dynamic Adaptive Mesh Refinement and Derefinement FiniteElement Framework for Metal Laser Sintering—Part II: Nonlinear Thermal Simulations and Validations, Journal of Manufacturing Science and Engineering 138(6) (2016) 061003. [16] N. Keller, V. Ploshikhin, New method for fast predictions of residual stress and distortion of AM parts, Solid Freeform Fabrication Symposium, Austin, Texas, 2014, pp. 1229-1237. [17] S.A. Khairallah, A.T. Anderson, A. Rubenchik, W.E. King, Laser powder-bed fusion additive manufacturing: Physics of complex melt flow and formation mechanisms of pores, spatter, and denudation zones, Acta Materialia 108 (2016) 36-45. [18] M.J. Matthews, G. Guss, S.A. Khairallah, A.M. Rubenchik, P.J. Depond, W.E. King, Denudation of metal powder layers in laser powder bed fusion processes, Acta Materialia 114 (2016) 33-42. [19] A.A. Martin, N.P. Calta, S.A. Khairallah, J. Wang, P.J. Depond, A.Y. Fong, V. Thampy, G.M. Guss, A.M. Kiss, K.H. Stone, Dynamics of pore formation during laser powder bed fusion additive manufacturing, Nature communications 10(1) (2019) 1987. [20] R. Shi, S.A. Khairallah, T.T. Roehling, T.W. Heo, J.T. McKeown, M.J. Matthews, Microstructural control in metal laser powder bed fusion additive manufacturing using laser beam shaping strategy, Acta Materialia (2019). [21] S.A. Khairallah, A.A. Martin, J.R. Lee, G. Guss, N.P. Calta, J.A. Hammons, M.H. Nielsen, K. Chaput, E. Schwalbach, M.N. Shah, Controlling interdependent meso-nanosecond dynamics and defect generation in metal 3D printing, Science 368(6491) (2020) 660-665. [22] W. Yan, W. Ge, Y. Qian, S. Lin, B. Zhou, W.K. Liu, F. Lin, G.J. Wagner, Multi-physics modeling of single/multiple-track defect mechanisms in electron beam selective melting, Acta Materialia 134 (2017) 324-333. [23] S. Shrestha, Y. Kevin Chou, A Numerical Study on the Keyhole Formation During Laser Powder Bed Fusion Process, Journal of Manufacturing Science and Engineering 141(10) (2019). [24] S. Shrestha, B. Cheng, K. Chou, An Investigation into Melt Pool Effective Thermal Conductivity for Thermal Modeling of Powder-Bed Electron Beam Additive Manufacturing. [25] D. Rosenthal, Mathematical theory of heat distribution during welding and cutting, Welding journal 20 (1941) 220-234. [26] P. Promoppatum, S.-C. Yao, P.C. Pistorius, A.D. Rollett, A comprehensive comparison of the analytical and numerical prediction of the thermal history and solidification microstructure of Inconel 718 products made by laser powder-bed fusion, Engineering 3(5) (2017) 685-694. [27] M. Tang, P.C. Pistorius, J.L. Beuth, Prediction of lack-of-fusion porosity for powder bed fusion, Additive Manufacturing 14 (2017) 39-48. [28] T. Moran, P. Li, D. Warner, N. Phan, Utility of superposition-based finite element approach for part-scale thermal simulation in additive manufacturing, Additive Manufacturing 21 (2018) 215-219. [29] Y. Yang, M. Knol, F. van Keulen, C. Ayas, A semi-analytical thermal modelling approach for selective laser melting, Additive Manufacturing 21 (2018) 284-297. [30] B. Cheng, S. Shrestha, K. Chou, Stress and deformation evaluations of scanning strategy effect in selective laser melting, Additive Manufacturing 12 (2016) 240-251. [31] L.H. Ahmed Hussein, Chunze Yan, Richard Everson, Finite element simulation of the temperature and stress fields in single layers built without-support in selective laser melting, Materials and Design 52 (2013) 638-647. [32] H. Peng, D.B. Go, R. Billo, S. Gong, M.R. Shankar, B.A. Gatrell, J. Budzinski, P. Ostiguy, R. Attardo, C. Tomonto, Part-scale model for fast prediction of thermal distortion in DMLS additive manufacturing; Part 2: a quasi-static thermo-mechanical model, Austin, Texas (2016). [33] M.F. Zaeh, G. Branner, Investigations on residual stresses and deformations in selective laser melting, Production Engineering 4(1) (2010) 35-45. [34] C. Li, C. Fu, Y. Guo, F. Fang, A multiscale modeling approach for fast prediction of part distortion in selective laser melting, Journal of Materials Processing Technology 229 (2016) 703- 712. [35] C. Li, Z. Liu, X. Fang, Y. Guo, On the Simulation Scalability of Predicting Residual Stress and Distortion in Selective Laser Melting, Journal of Manufacturing Science and Engineering 140(4) (2018) 041013. [36] S. Afazov, W.A. Denmark, B.L. Toralles, A. Holloway, A. Yaghi, Distortion Prediction and Compensation in Selective Laser Melting, Additive Manufacturing 17 (2017) 15-22. [37] Y. Lee, W. Zhang, Modeling of heat transfer, fluid flow and solidification microstructure of nickel-base superalloy fabricated by laser powder bed fusion, Additive Manufacturing 12 (2016) 178-188. [38] L. Scime, J. Beuth, A multi-scale convolutional neural network for autonomous anomaly detection and classification in a laser powder bed fusion additive manufacturing process, Additive Manufacturing 24 (2018) 273-286. [39] L. Scime, J. Beuth, Using machine learning to identify in-situ melt pool signatures indicative of flaw formation in a laser powder bed fusion additive manufacturing process, Additive Manufacturing 25 (2019) 151-165. [40] X. Xie, J. Bennett, S. Saha, Y. Lu, J. Cao, W.K. Liu, Z. Gan, Mechanistic data-driven prediction of as-built mechanical properties in metal additive manufacturing, npj Computational Materials 7(1) (2021) 1-12. [41] C. Wang, X. Tan, S. Tor, C. Lim, Machine learning in additive manufacturing: State-of-theart and perspectives, Additive Manufacturing (2020) 101538. [42] J. Li, R. Jin, Z.Y. Hang, Integration of physically-based and data-driven approaches for thermal field prediction in additive manufacturing, Materials & Design 139 (2018) 473-485. [43] M. Mozaffar, A. Paul, R. Al-Bahrani, S. Wolff, A. Choudhary, A. Agrawal, K. Ehmann, J. Cao, Data-driven prediction of the high-dimensional thermal history in directed energy deposition processes via recurrent neural networks, Manufacturing letters 18 (2018) 35-39. [44] A. Paul, M. Mozaffar, Z. Yang, W.-k. Liao, A. Choudhary, J. Cao, A. Agrawal, A real-time iterative machine learning approach for temperature profile prediction in additive manufacturing processes, 2019 IEEE International Conference on Data Science and Advanced Analytics (DSAA), IEEE, 2019, pp. 541-550. [45] S. Clijsters, T. Craeghs, J.-P. Kruth, A priori process parameter adjustment for SLM process optimization, Innovative developments on virtual and physical prototyping, Taylor & Francis Group., 2012, pp. 553-560. [46] R. Mertens, S. Clijsters, K. Kempen, J.-P. Kruth, Optimization of scan strategies in selective laser melting of aluminum parts with downfacing areas, Journal of Manufacturing Science and Engineering 136(6) (2014) 061012. [47] J.-P. Kruth, J. Deckers, E. Yasa, R. Wauthlé, Assessing and comparing influencing factors of residual stresses in selective laser melting using a novel analysis method, Proceedings of the institution of mechanical engineers, Part B: Journal of Engineering Manufacture 226(6) (2012) 980-991. [48] Y. Lu, S. Wu, Y. Gan, T. Huang, C. Yang, L. Junjie, J. Lin, Study on the microstructure, mechanical property and residual stress of SLM Inconel-718 alloy manufactured by differing island scanning strategy, Optics & Laser Technology 75 (2015) 197-206. [49] E. Foroozmehr, R. Kovacevic, Effect of path planning on the laser powder deposition process: thermal and structural evaluation, The International Journal of Advanced Manufacturing Technology 51(5-8) (2010) 659-669. [50] L.H. Ahmed Hussein, Chunze Yan, Richard Everson, Finite element simulation of the temperature and stress fields in single layers built without-support in selective laser melting, Materials and Design (2013). [51] J.-P. Kruth, M. Badrossamay, E. Yasa, J. Deckers, L. Thijs, J. Van Humbeeck, Part and material properties in selective laser melting of metals, Proceedings of the 16th international symposium on electromachining, 2010, pp. 1-12. [52] L. Thijs, K. Kempen, J.-P. Kruth, J. Van Humbeeck, Fine-structured aluminium products with controllable texture by selective laser melting of pre-alloyed AlSi10Mg powder, Acta Materialia 61(5) (2013) 1809-1819. [53] D. Ding, Z.S. Pan, D. Cuiuri, H. Li, A tool-path generation strategy for wire and arc additive manufacturing, The international journal of advanced manufacturing technology 73(1-4) (2014) 173-183. [54] B.E. Carroll, T.A. Palmer, A.M. Beese, Anisotropic tensile behavior of Ti–6Al–4V components fabricated with directed energy deposition additive manufacturing, Acta Materialia 87 (2015) 309-320. [55] D. Ding, Z. Pan, D. Cuiuri, H. Li, A practical path planning methodology for wire and arc additive manufacturing of thin-walled structures, Robotics and Computer-Integrated Manufacturing 34 (2015) 8-19. [56] D. Ding, Z. Pan, D. Cuiuri, H. Li, S. van Duin, N. Larkin, Bead modelling and implementation of adaptive MAT path in wire and arc additive manufacturing, Robotics and Computer-Integrated Manufacturing 39 (2016) 32-42. [57] R. Ponche, O. Kerbrat, P. Mognol, J.-Y. Hascoet, A novel methodology of design for Additive Manufacturing applied to Additive Laser Manufacturing process, Robotics and ComputerIntegrated Manufacturing 30(4) (2014) 389-398. [58] D.E. Smith, R. Hoglund, Continuous fiber angle topology optimization for polymer fused fillament fabrication, Annu. Int. Solid Free. Fabr. Symp. Austin, TX, 2016. [59] J. Liu, J. Liu, H. Yu, H. Yu, Concurrent deposition path planning and structural topology optimization for additive manufacturing, Rapid Prototyping Journal 23(5) (2017) 930-942. [60] Q. Xia, T. Shi, Optimization of composite structures with continuous spatial variation of fiber angle through Shepard interpolation, Composite Structures 182 (2017) 273-282. [61] C. Kiyono, E. Silva, J. Reddy, A novel fiber optimization method based on normal distribution function with continuously varying fiber path, Composite Structures 160 (2017) 503-515. [62] C.J. Brampton, K.C. Wu, H.A. Kim, New optimization method for steered fiber composites using the level set method, Structural and Multidisciplinary Optimization 52(3) (2015) 493-505. [63] J. Liu, A.C. To, Deposition path planning-integrated structural topology optimization for 3D additive manufacturing subject to self-support constraint, Computer-Aided Design 91 (2017) 27- 45. [64] H. Shen, J. Fu, Z. Chen, Y. Fan, Generation of offset surface for tool path in NC machining through level set methods, The International Journal of Advanced Manufacturing Technology 46(9-12) (2010) 1043-1047. [65] C. Zhuang, Z. Xiong, H. Ding, High speed machining tool path generation for pockets using level sets, International Journal of Production Research 48(19) (2010) 5749-5766. [66] K.C. Mills, Recommended values of thermophysical properties for selected commercial alloys, Woodhead Publishing2002. [67] S.S. Sih, J.W. Barlow, The prediction of the emissivity and thermal conductivity of powder beds, Particulate Science and Technology 22(4) (2004) 427-440. [68] L. Dong, A. Makradi, S. Ahzi, Y. Remond, Three-dimensional transient finite element analysis of the selective laser sintering process, Journal of materials processing technology 209(2) (2009) 700-706. [69] J.J. Beaman, J.W. Barlow, D.L. Bourell, R.H. Crawford, H.L. Marcus, K.P. McAlea, Solid freeform fabrication: a new direction in manufacturing, Kluwer Academic Publishers, Norwell, MA 2061 (1997) 25-49. [70] G. Bugeda Miguel Cervera, G. Lombera, Numerical prediction of temperature and density distributions in selective laser sintering processes, Rapid Prototyping Journal 5(1) (1999) 21-26. [71] T. Mukherjee, W. Zhang, T. DebRoy, An improved prediction of residual stresses and distortion in additive manufacturing, Computational Materials Science 126 (2017) 360-372. [72] A.J. Dunbar, E.R. Denlinger, M.F. Gouge, P. Michaleris, Experimental validation of finite element modeling for laser powderbed fusion deformation, Additive Manufacturing 12 (2016) 108-120. [73] J. Goldak, A. Chakravarti, M. Bibby, A new finite element model for welding heat sources, Metallurgical and Materials Transactions B 15(2) (1984) 299-305. [74] J. Liu, Q. Chen, Y. Zhao, W. Xiong, A. To, Quantitative Texture Prediction of Epitaxial Columnar Grains in Alloy 718 Processed by Additive Manufacturing, Proceedings of the 9th International Symposium on Superalloy 718 & Derivatives: Energy, Aerospace, and Industrial Applications, Springer, 2018, pp. 749-755. [75] J. Irwin, P. Michaleris, A line heat input model for additive manufacturing, Journal of Manufacturing Science and Engineering 138(11) (2016) 111004. [76] M. Gouge, J. Heigel, P. Michaleris, T. Palmer, Modeling forced convection in the thermal simulation of laser cladding processes, International Journal of Advanced Manufacturing Technology 79 (2015). [77] J. Heigel, P. Michaleris, E. Reutzel, Thermo-mechanical model development and validation of directed energy deposition additive manufacturing of Ti–6Al–4V, Additive manufacturing 5 (2015) 9-19. [78] E.R. Denlinger, J.C. Heigel, P. Michaleris, Residual stress and distortion modeling of electron beam direct manufacturing Ti-6Al-4V, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture 229(10) (2015) 1803-1813. [79] X. Liang, Q. Chen, L. Cheng, Q. Yang, A. To, A modified inherent strain method for fast prediction of residual deformation in additive manufacturing of metal parts, 2017 Solid Freeform Fabrication Symposium Proceedings, Austin, Texas, 2017. [80] X. Liang, L. Cheng, Q. Chen, Q. Yang, A. To, A Modified Method for Estimating Inherent Strains from Detailed Process Simulation for Fast Residual Distortion Prediction of Single-Walled Structures Fabricated by Directed Energy Deposition, Additive Manufacturing 23 (2018) 471-486. [81] L. Sochalski-Kolbus, E.A. Payzant, P.A. Cornwell, T.R. Watkins, S.S. Babu, R.R. Dehoff, M. Lorenz, O. Ovchinnikova, C. Duty, Comparison of residual stresses in Inconel 718 simple parts made by electron beam melting and direct laser metal sintering, Metallurgical and Materials Transactions A 46(3) (2015) 1419-1432. [82] P. Mercelis, J.-P. Kruth, Residual stresses in selective laser sintering and selective laser melting, Rapid Prototyping Journal 12(5) (2006) 254-265. [83] N. Hodge, R. Ferencz, J. Solberg, Implementation of a thermomechanical model for the simulation of selective laser melting, Computational Mechanics 54(1) (2014) 33-51. [84] A.S. Wu, D.W. Brown, M. Kumar, G.F. Gallegos, W.E. King, An experimental investigation into additive manufacturing-induced residual stresses in 316L stainless steel, Metallurgical and Materials Transactions A 45(13) (2014) 6260-6270. [85] C. Li, J. liu, Y. Guo, Efficient predictive model of part distortion and residual stress in selective laser melting, Solid Freeform Fabrication 2016, 2017. [86] Y. Zhao, Y. Koizumi, K. Aoyagi, D. Wei, K. Yamanaka, A. Chiba, Molten pool behavior and effect of fluid flow on solidification conditions in selective electron beam melting (SEBM) of a biomedical Co-Cr-Mo alloy, Additive Manufacturing 26 (2019) 202-214. [87] J.-H. Cho, S.-J. Na, Implementation of real-time multiple reflection and Fresnel absorption of laser beam in keyhole, Journal of Physics D: Applied Physics 39(24) (2006) 5372. [88] Q. Guo, C. Zhao, M. Qu, L. Xiong, L.I. Escano, S.M.H. Hojjatzadeh, N.D. Parab, K. Fezzaa, W. Everhart, T. Sun, In-situ characterization and quantification of melt pool variation under constant input energy density in laser powder-bed fusion additive manufacturing process, Additive Manufacturing (2019). [89] E. Assuncao, S. Williams, D. Yapp, Interaction time and beam diameter effects on the conduction mode limit, Optics and Lasers in Engineering 50(6) (2012) 823-828. [90] R. Cunningham, C. Zhao, N. Parab, C. Kantzos, J. Pauza, K. Fezzaa, T. Sun, A.D. Rollett, Keyhole threshold and morphology in laser melting revealed by ultrahigh-speed x-ray imaging, Science 363(6429) (2019) 849-852. [91] W. Tan, N.S. Bailey, Y.C. Shin, Investigation of keyhole plume and molten pool based on a three-dimensional dynamic model with sharp interface formulation, Journal of Physics D: Applied Physics 46(5) (2013) 055501. [92] W. Tan, Y.C. Shin, Analysis of multi-phase interaction and its effects on keyhole dynamics with a multi-physics numerical model, Journal of Physics D: Applied Physics 47(34) (2014) 345501. [93] R. Fabbro, K. Chouf, Keyhole modeling during laser welding, Journal of applied Physics 87(9) (2000) 4075-4083. [94] Q. Guo, C. Zhao, M. Qu, L. Xiong, S.M.H. Hojjatzadeh, L.I. Escano, N.D. Parab, K. Fezzaa, T. Sun, L. Chen, In-situ full-field mapping of melt flow dynamics in laser metal additive manufacturing, Additive Manufacturing 31 (2020) 100939. [95] Y. Ueda, K. Fukuda, K. Nakacho, S. Endo, A new measuring method of residual stresses with the aid of finite element method and reliability of estimated values, Journal of the Society of Naval Architects of Japan 1975(138) (1975) 499-507. [96] M.R. Hill, D.V. Nelson, The inherent strain method for residual stress determination and its application to a long welded joint, ASME-PUBLICATIONS-PVP 318 (1995) 343-352. [97] H. Murakawa, Y. Luo, Y. Ueda, Prediction of welding deformation and residual stress by elastic FEM based on inherent strain, Journal of the society of Naval Architects of Japan 1996(180) (1996) 739-751. [98] M. Yuan, Y. Ueda, Prediction of residual stresses in welded T-and I-joints using inherent strains, Journal of Engineering Materials and Technology, Transactions of the ASME 118(2) (1996) 229-234. [99] L. Zhang, P. Michaleris, P. Marugabandhu, Evaluation of applied plastic strain methods for welding distortion prediction, Journal of Manufacturing Science and Engineering 129(6) (2007) 1000-1010. [100] M. Bugatti, Q. Semeraro, Limitations of the Inherent Strain Method in Simulating Powder Bed Fusion Processes, Additive Manufacturing 23 (2018) 329-346. [101] L. Cheng, X. Liang, J. Bai, Q. Chen, J. Lemon, A. To, On Utilizing Topology Optimization to Design Support Structure to Prevent Residual Stress Induced Build Failure in Laser Powder Bed Metal Additive Manufacturing, Additive Manufacturing (2019). [102] Q. Chen, X. Liang, D. Hayduke, J. Liu, L. Cheng, J. Oskin, R. Whitmore, A.C. To, An inherent strain based multiscale modeling framework for simulating part-scale residual deformation for direct metal laser sintering, Additive Manufacturing 28 (2019) 406-418. [103] S. Osher, J.A. Sethian, Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations, Journal of computational physics 79(1) (1988) 12-49. [104] M.Y. Wang, X. Wang, D. Guo, A level set method for structural topology optimization, Computer methods in applied mechanics and engineering 192(1) (2003) 227-246. [105] G. Allaire, F. Jouve, A.-M. Toader, Structural optimization using sensitivity analysis and a level-set method, Journal of computational physics 194(1) (2004) 363-393. [106] Y. Wang, Z. Luo, Z. Kang, N. Zhang, A multi-material level set-based topology and shape optimization method, Computer Methods in Applied Mechanics and Engineering 283 (2015) 1570-1586. [107] P. Dunning, C. Brampton, H. Kim, Simultaneous optimisation of structural topology and material grading using level set method, Materials Science and Technology 31(8) (2015) 884-894. [108] P. Liu, Y. Luo, Z. Kang, Multi-material topology optimization considering interface behavior via XFEM and level set method, Computer methods in applied mechanics and engineering 308 (2016) 113-133. [109] J. Liu, Q. Chen, Y. Zheng, R. Ahmad, J. Tang, Y. Ma, Level set-based heterogeneous object modeling and optimization, Computer-Aided Design (2019). [110] J. Liu, Q. Chen, X. Liang, A.C. To, Manufacturing cost constrained topology optimization for additive manufacturing, Frontiers of Mechanical Engineering 14(2) (2019) 213-221. [111] Z. Kang, Y. Wang, Integrated topology optimization with embedded movable holes based on combined description by material density and level sets, Computer methods in applied mechanics and engineering 255 (2013) 1-13. [112] P.D. Dunning, H. Alicia Kim, A new hole insertion method for level set based structural topology optimization, International Journal for Numerical Methods in Engineering 93(1) (2013) 118-134. [113] J.A. Sethian, A fast marching level set method for monotonically advancing fronts, Proceedings of the National Academy of Sciences 93(4) (1996) 1591-1595. [114] J.A. Sethian, Level set methods and fast marching methods: evolving interfaces in computational geometry, fluid mechanics, computer vision, and materials science, Cambridge university press1999. [115] C. Le, J. Norato, T. Bruns, C. Ha, D. Tortorelli, Stress-based topology optimization for continua, Structural and Multidisciplinary Optimization 41(4) (2010) 605-620. [116] A. Takezawa, G.H. Yoon, S.H. Jeong, M. Kobashi, M. Kitamura, Structural topology optimization with strength and heat conduction constraints, Computer Methods in Applied Mechanics and Engineering 276 (2014) 341-361. [117] S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural computation 9(8) (1997) 1735-1780. [118] A. Krizhevsky, I. Sutskever, G.E. Hinton, Imagenet classification with deep convolutional neural networks, Advances in neural information processing systems 25 (2012) 1097-1105. [119] K. Simonyan, A. Zisserman, Very deep convolutional networks for large-scale image recognition, arXiv preprint arXiv:1409.1556 (2014). [120] K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770-778. [121] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. Bernstein, Imagenet large scale visual recognition challenge, International journal of computer vision 115(3) (2015) 211-252. [122] S. Ren, K. He, R. Girshick, J. Sun, Faster r-cnn: Towards real-time object detection with region proposal networks, Advances in neural information processing systems 28 (2015) 91-99. [123] E.J. Schwalbach, S.P. Donegan, M.G. Chapman, K.J. Chaput, M.A. Groeber, A discrete source model of powder bed fusion additive manufacturing thermal history, Additive Manufacturing 25 (2019) 485-498. [124] D.G. Duffy, Green’s functions with applications, Chapman and Hall/CRC2015. [125] J. Martínez-Frutos, D. Herrero-Pérez, Efficient matrix-free GPU implementation of fixed grid finite element analysis, Finite Elements in Analysis and Design 104 (2015) 61-71. [126] F. Dugast, P. Apostolou, A. Fernandez, W. Dong, Q. Chen, S. Strayer, R. Wicker, A.C. To, Part-scale thermal process modeling for laser powder bed fusion with matrix-free method and GPU computing, Additive Manufacturing 37 (2021) 101732. [127] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A.N. Gomez, Ł. Kaiser, I. Polosukhin, Attention is all you need, Advances in neural information processing systems, 2017, pp. 5998-6008. [128] J. Devlin, M.-W. Chang, K. Lee, K. Toutanova, Bert: Pre-training of deep bidirectional transformers for language understanding, arXiv preprint arXiv:1810.04805 (2018).
Optimization of Solar CCHP Systems with Collector Enhanced by Porous Media and Nanofluid
Navid Tonekaboni,1Mahdi Feizbahr,2 Nima Tonekaboni,1Guang-Jun Jiang,3,4 and Hong-Xia Chen3,4
Abstract
태양열 집열기의 낮은 효율은 CCHP(Solar Combined Cooling, Heating, and Power) 사이클의 문제점 중 하나로 언급될 수 있습니다. 태양계를 개선하기 위해 나노유체와 다공성 매체가 태양열 집열기에 사용됩니다.
다공성 매질과 나노입자를 사용하는 장점 중 하나는 동일한 조건에서 더 많은 에너지를 흡수할 수 있다는 것입니다. 이 연구에서는 평균 일사량이 1b인 따뜻하고 건조한 지역의 600 m2 건물의 전기, 냉방 및 난방을 생성하기 위해 다공성 매질과 나노유체를 사용하여 태양열 냉난방 복합 발전(SCCHP) 시스템을 최적화했습니다.
본 논문에서는 침전물이 형성되지 않는 lb = 820 w/m2(이란) 정도까지 다공성 물질에서 나노유체의 최적량을 계산하였다. 이 연구에서 태양열 집열기는 구리 다공성 매체(95% 다공성)와 CuO 및 Al2O3 나노 유체로 향상되었습니다.
나노유체의 0.1%-0.6%가 작동 유체로 물에 추가되었습니다. 나노유체의 0.5%가 태양열 집열기 및 SCCHP 시스템에서 가장 높은 에너지 및 엑서지 효율 향상으로 이어지는 것으로 밝혀졌습니다.
본 연구에서 포물선형 집열기(PTC)의 최대 에너지 및 엑서지 효율은 각각 74.19% 및 32.6%입니다. 그림 1은 태양 CCHP의 주기를 정확하게 설명하기 위한 그래픽 초록으로 언급될 수 있습니다.
The low efficiency of solar collectors can be mentioned as one of the problems in solar combined cooling, heating, and power (CCHP) cycles. For improving solar systems, nanofluid and porous media are used in solar collectors. One of the advantages of using porous media and nanoparticles is to absorb more energy under the same conditions. In this research, a solar combined cooling, heating, and power (SCCHP) system has been optimized by porous media and nanofluid for generating electricity, cooling, and heating of a 600 m2 building in a warm and dry region with average solar radiation of Ib = 820 w/m2 in Iran. In this paper, the optimal amount of nanofluid in porous materials has been calculated to the extent that no sediment is formed. In this study, solar collectors were enhanced with copper porous media (95% porosity) and CuO and Al2O3 nanofluids. 0.1%–0.6% of the nanofluids were added to water as working fluids; it is found that 0.5% of the nanofluids lead to the highest energy and exergy efficiency enhancement in solar collectors and SCCHP systems. Maximum energy and exergy efficiency of parabolic thermal collector (PTC) riches in this study are 74.19% and 32.6%, respectively. Figure 1 can be mentioned as a graphical abstract for accurately describing the cycle of solar CCHP.
1. Introduction
Due to the increase in energy consumption, the use of clean energy is one of the important goals of human societies. In the last four decades, the use of cogeneration cycles has increased significantly due to high efficiency. Among clean energy, the use of solar energy has become more popular due to its greater availability [1]. Low efficiency of energy production, transmission, and distribution system makes a new system to generate simultaneously electricity, heating, and cooling as an essential solution to be widely used. The low efficiency of the electricity generation, transmission, and distribution system makes the CCHP system a basic solution to eliminate waste of energy. CCHP system consists of a prime mover (PM), a power generator, a heat recovery system (produce extra heating/cooling/power), and thermal energy storage (TES) [2]. Solar combined cooling, heating, and power (SCCHP) has been started three decades ago. SCCHP is a system that receives its propulsive force from solar energy; in this cycle, solar collectors play the role of propulsive for generating power in this system [3].
Increasing the rate of energy consumption in the whole world because of the low efficiency of energy production, transmission, and distribution system causes a new cogeneration system to generate electricity, heating, and cooling energy as an essential solution to be widely used. Building energy utilization fundamentally includes power required for lighting, home electrical appliances, warming and cooling of building inside, and boiling water. Domestic usage contributes to an average of 35% of the world’s total energy consumption [4].
Due to the availability of solar energy in all areas, solar collectors can be used to obtain the propulsive power required for the CCHP cycle. Solar energy is the main source of energy in renewable applications. For selecting a suitable area to use solar collectors, annual sunshine hours, the number of sunny days, minus temperature and frosty days, and the windy status of the region are essentially considered [5]. Iran, with an average of more than 300 sunny days, is one of the suitable countries to use solar energy. Due to the fact that most of the solar radiation is in the southern regions of Iran, also the concentration of cities is low in these areas, and transmission lines are far apart, one of the best options is to use CCHP cycles based on solar collectors [6]. One of the major problems of solar collectors is their low efficiency [7]. Low efficiency increases the area of collectors, which increases the initial cost of solar systems and of course increases the initial payback period. To increase the efficiency of solar collectors and improve their performance, porous materials and nanofluids are used to increase their workability.
There are two ways to increase the efficiency of solar collectors and mechanical and fluid improvement. In the first method, using porous materials or helical filaments inside the collector pipes causes turbulence of the flow and increases heat transfer. In the second method, using nanofluids or salt and other materials increases the heat transfer of water. The use of porous materials has grown up immensely over the past twenty years. Porous materials, especially copper porous foam, are widely used in solar collectors. Due to the high contact surface area, porous media are appropriate candidates for solar collectors [8]. A number of researchers investigated Solar System performance in accordance with energy and exergy analyses. Zhai et al. [9] reviewed the performance of a small solar-powered system in which the energy efficiency was 44.7% and the electrical efficiency was 16.9%.
Abbasi et al. [10] proposed an innovative multiobjective optimization to optimize the design of a cogeneration system. Results showed the CCHP system based on an internal diesel combustion engine was the applicable alternative at all regions with different climates. The diesel engine can supply the electrical requirement of 31.0% and heating demand of 3.8% for building.
Jiang et al. [11] combined the experiment and simulation together to analyze the performance of a cogeneration system. Moreover, some research focused on CCHP systems using solar energy. It integrated sustainable and renewable technologies in the CCHP, like PV, Stirling engine, and parabolic trough collector (PTC) [2, 12–15].
Wang et al. [16] optimized a cogeneration solar cooling system with a Rankine cycle and ejector to reach the maximum total system efficiency of 55.9%. Jing et al. analyzed a big-scale building with the SCCHP system and auxiliary heaters to produced electrical, cooling, and heating power. The maximum energy efficiency reported in their work is 46.6% [17]. Various optimization methods have been used to improve the cogeneration system, minimum system size, and performance, such as genetic algorithm [18, 19].
Hirasawa et al. [20] investigated the effect of using porous media to reduce thermal waste in solar systems. They used the high-porosity metal foam on top of the flat plate solar collector and observed that thermal waste decreased by 7% due to natural heat transfer. Many researchers study the efficiency improvement of the solar collector by changing the collector’s shapes or working fluids. However, the most effective method is the use of nanofluids in the solar collector as working fluid [21]. In the experimental study done by Jouybari et al. [22], the efficiency enhancement up to 8.1% was achieved by adding nanofluid in a flat plate collector. In this research, by adding porous materials to the solar collector, collector efficiency increased up to 92% in a low flow regime. Subramani et al. [23] analyzed the thermal performance of the parabolic solar collector with Al2O3 nanofluid. They conducted their experiments with Reynolds number range 2401 to 7202 and mass flow rate 0.0083 to 0.05 kg/s. The maximum efficiency improvement in this experiment was 56% at 0.05 kg/s mass flow rate.
Shojaeizadeh et al. [24] investigated the analysis of the second law of thermodynamic on the flat plate solar collector using Al2O3/water nanofluid. Their research showed that energy efficiency rose up to 1.9% and the exergy efficiency increased by a maximum of 0.72% compared to pure water. Tiwari et al. [25] researched on the thermal performance of solar flat plate collectors for working fluid water with different nanofluids. The result showed that using 1.5% (optimum) particle volume fraction of Al2O3 nanofluid as an absorbing medium causes the thermal efficiency to enhance up to 31.64%.
The effect of porous media and nanofluids on solar collectors has already been investigated in the literature but the SCCHP system with a collector embedded by both porous media and nanofluid for enhancing the ratio of nanoparticle in nanofluid for preventing sedimentation was not discussed. In this research, the amount of energy and exergy of the solar CCHP cycles with parabolic solar collectors in both base and improved modes with a porous material (copper foam with 95% porosity) and nanofluid with different ratios of nanoparticles was calculated. In the first step, it is planned to design a CCHP system based on the required load, and, in the next step, it will analyze the energy and exergy of the system in a basic and optimize mode. In the optimize mode, enhanced solar collectors with porous material and nanofluid in different ratios (0.1%–0.7%) were used to optimize the ratio of nanofluids to prevent sedimentation.
2. Cycle Description
CCHP is one of the methods to enhance energy efficiency and reduce energy loss and costs. The SCCHP system used a solar collector as a prime mover of the cogeneration system and assisted the boiler to generate vapor for the turbine. Hot water flows from the expander to the absorption chiller in summer or to the radiator or fan coil in winter. Finally, before the hot water wants to flow back to the storage tank, it flows inside a heat exchanger for generating domestic hot water [26].
For designing of solar cogeneration system and its analysis, it is necessary to calculate the electrical, heating (heating load is the load required for the production of warm water and space heating), and cooling load required for the case study considered in a residential building with an area of 600 m2 in the warm region of Iran (Zahedan). In Table 1, the average of the required loads is shown for the different months of a year (average of electrical, heating, and cooling load calculated with CARRIER software).Table 1The average amount of electric charges, heating load, and cooling load used in the different months of the year in the city of Zahedan for a residential building with 600 m2.
According to Table 1, the maximum magnitude of heating, cooling, and electrical loads is used to calculate the cogeneration system. The maximum electric load is 96 kW, the maximum amount of heating load is 62 kW, and the maximum cooling load is 118 kW. Since the calculated loads are average, all loads increased up to 10% for the confidence coefficient. With the obtained values, the solar collector area and other cogeneration system components are calculated. The cogeneration cycle is capable of producing 105 kW electric power, 140 kW cooling capacity, and 100 kW heating power.
2.1. System Analysis Equations
An analysis is done by considering the following assumptions:(1)The system operates under steady-state conditions(2)The system is designed for the warm region of Iran (Zahedan) with average solar radiation Ib = 820 w/m2(3)The pressure drops in heat exchangers, separators, storage tanks, and pipes are ignored(4)The pressure drop is negligible in all processes and no expectable chemical reactions occurred in the processes(5)Potential, kinetic, and chemical exergy are not considered due to their insignificance(6)Pumps have been discontinued due to insignificance throughout the process(7)All components are assumed adiabatic
Schematic shape of the cogeneration cycle is shown in Figure 1 and all data are given in Table 2.
Figure 1Schematic shape of the cogeneration cycle.Table 2Temperature and humidity of different points of system.
Based on the first law of thermodynamic, energy analysis is based on the following steps.
First of all, the estimated solar radiation energy on collector has been calculated:where α is the heat transfer enhancement coefficient based on porous materials added to the collector’s pipes. The coefficient α is increased by the porosity percentage, the type of porous material (in this case, copper with a porosity percentage of 95), and the flow of fluid to the collector equation.
Collector efficiency is going to be calculated by the following equation [9]:
Total energy received by the collector is given by [9]
In the last step based on thermodynamic second law, exergy efficiency has been calculated from the following equation and the above-mentioned calculated loads [9]:
3. Porous Media
The porous medium that filled the test section is copper foam with a porosity of 95%. The foams are determined in Figure 2 and also detailed thermophysical parameters and dimensions are shown in Table 3.
Figure 2Copper foam with a porosity of 95%.Table 3Thermophysical parameters and dimensions of copper foam.
In solar collectors, copper porous materials are suitable for use at low temperatures and have an easier and faster manufacturing process than ceramic porous materials. Due to the high coefficient conductivity of copper, the use of copper metallic foam to increase heat transfer is certainly more efficient in solar collectors.
Porous media and nanofluid in solar collector’s pipes were simulated in FLOW-3D software using the finite-difference method [27]. Nanoparticles Al2O3 and CUO are mostly used in solar collector enhancement. In this research, different concentrations of nanofluid are added to the parabolic solar collectors with porous materials (copper foam with porosity of 95%) to achieve maximum heat transfer in the porous materials before sedimentation. After analyzing PTC pipes with the nanofluid flow in FLOW-3D software, for energy and exergy efficiency analysis, Carrier software results were used as EES software input. Simulation PTC with porous media inside collector pipe and nanofluids sedimentation is shown in Figure 3.
Figure 3Simulation PTC pipes enhanced with copper foam and nanoparticles in FLOW-3D software.
3.1. Nano Fluid
In this research, copper and silver nanofluids (Al2O3, CuO) have been added with percentages of 0.1%–0.7% as the working fluids. The nanoparticle properties are given in Table 4. Also, system constant parameters are presented in Table 4, which are available as default input in the EES software.Table 4Properties of the nanoparticles [9].
System constant parameters for input in the software are shown in Table 5.Table 5System constant parameters.
The thermal properties of the nanofluid can be obtained from equations (18)–(21). The basic fluid properties are indicated by the index (bf) and the properties of the nanoparticle silver with the index (np).
The density of the mixture is shown in the following equation [28]:where ρ is density and ϕ is the nanoparticles volume fraction.
The specific heat capacity is calculated from the following equation [29]:
The thermal conductivity of the nanofluid is calculated from the following equation [29]:
The parameter β is the ratio of the nanolayer thickness to the original particle radius and, usually, this parameter is taken equal to 0.1 for the calculated thermal conductivity of the nanofluids.
The mixture viscosity is calculated as follows [30]:
In all equations, instead of water properties, working fluids with nanofluid are used. All of the above equations and parameters are entered in the EES software for calculating the energy and exergy of solar collectors and the SCCHP cycle. All calculation repeats for both nanofluids with different concentrations of nanofluid in the solar collector’s pipe.
4. Results and Discussion
In the present study, relations were written according to Wang et al. [16] and the system analysis was performed to ensure the correctness of the code. The energy and exergy charts are plotted based on the main values of the paper and are shown in Figures 4 and 5. The error rate in this simulation is 1.07%.
Figure 4Verification charts of energy analysis results.
Figure 5Verification charts of exergy analysis results.
We may also investigate the application of machine learning paradigms [31–41] and various hybrid, advanced optimization approaches that are enhanced in terms of exploration and intensification [42–55], and intelligent model studies [56–61] as well, for example, methods such as particle swarm optimizer (PSO) [60, 62], differential search (DS) [63], ant colony optimizer (ACO) [61, 64, 65], Harris hawks optimizer (HHO) [66], grey wolf optimizer (GWO) [53, 67], differential evolution (DE) [68, 69], and other fusion and boosted systems [41, 46, 48, 50, 54, 55, 70, 71].
At the first step, the collector is modified with porous copper foam material. 14 cases have been considered for the analysis of the SCCHP system (Table 6). It should be noted that the adding of porous media causes an additional pressure drop inside the collector [9, 22–26, 30, 72]. All fourteen cases use copper foam with a porosity of 95 percent. To simulate the effect of porous materials and nanofluids, the first solar PTC pipes have been simulated in the FLOW-3D software and then porous media (copper foam with porosity of 95%) and fluid flow with nanoparticles (AL2O3 and CUO) are generated in the software. After analyzing PTC pipes in FLOW-3D software, for analyzing energy and exergy efficiency, software outputs were used as EES software input for optimization ratio of sedimentation and calculating energy and exergy analyses.Table 6Collectors with different percentages of nanofluids and porous media.
In this research, an enhanced solar collector with both porous media and Nanofluid is investigated. In the present study, 0.1–0.5% CuO and Al2O3 concentration were added to the collector fully filled by porous media to achieve maximum energy and exergy efficiencies of solar CCHP systems. All steps of the investigation are shown in Table 6.
Energy and exergy analyses of parabolic solar collectors and SCCHP systems are shown in Figures 6 and 7.
Figure 6Energy and exergy efficiencies of the PTC with porous media and nanofluid.
Figure 7Energy and exergy efficiency of the SCCHP.
Results show that the highest energy and exergy efficiencies are 74.19% and 32.6%, respectively, that is achieved in Step 12 (parabolic collectors with filled porous media and 0.5% Al2O3). In the second step, the maximum energy efficiency of SCCHP systems with fourteen steps of simulation are shown in Figure 7.
In the second step, where 0.1, −0.6% of the nanofluids were added, it is found that 0.5% leads to the highest energy and exergy efficiency enhancement in solar collectors and SCCHP systems. Using concentrations more than 0.5% leads to sediment in the solar collector’s pipe and a decrease of porosity in the pipe [73]. According to Figure 7, maximum energy and exergy efficiencies of SCCHP are achieved in Step 12. In this step energy efficiency is 54.49% and exergy efficiency is 18.29%. In steps 13 and 14, with increasing concentration of CUO and Al2O3 nanofluid solution in porous materials, decreasing of energy and exergy efficiency of PTC and SCCHP system at the same time happened. This decrease in efficiency is due to the formation of sediment in the porous material. Calculations and simulations have shown that porous materials more than 0.5% nanofluids inside the collector pipe cause sediment and disturb the porosity of porous materials and pressure drop and reduce the coefficient of performance of the cogeneration system. Most experience showed that CUO and AL2O3 nanofluids with less than 0.6% percent solution are used in the investigation on the solar collectors at low temperatures and discharges [74]. One of the important points of this research is that the best ratio of nanofluids in the solar collector with a low temperature is 0.5% (AL2O3 and CUO); with this replacement, the cost of solar collectors and SCCHP cycle is reduced.
5. Conclusion and Future Directions
In the present study, ways for increasing the efficiency of solar collectors in order to enhance the efficiency of the SCCHP cycle are examined. The research is aimed at adding both porous materials and nanofluids for estimating the best ratio of nanofluid for enhanced solar collector and protecting sedimentation in porous media. By adding porous materials (copper foam with porosity of 95%) and 0.5% nanofluids together, high efficiency in solar parabolic collectors can be achieved. The novelty in this research is the addition of both nanofluids and porous materials and calculating the best ratio for preventing sedimentation and pressure drop in solar collector’s pipe. In this study, it was observed that, by adding 0.5% of AL2O3 nanofluid in working fluids, the energy efficiency of PTC rises to 74.19% and exergy efficiency is grown up to 32.6%. In SCCHP cycle, energy efficiency is 54.49% and exergy efficiency is 18.29%.
In this research, parabolic solar collectors fully filled by porous media (copper foam with a porosity of 95) are investigated. In the next step, parabolic solar collectors in the SCCHP cycle were simultaneously filled by porous media and different percentages of Al2O3 and CuO nanofluid. At this step, values of 0.1% to 0.6% of each nanofluid were added to the working fluid, and the efficiency of the energy and exergy of the collectors and the SCCHP cycle were determined. In this case, nanofluid and the porous media were used together in the solar collector and maximum efficiency achieved. 0.5% of both nanofluids were used to achieve the biggest efficiency enhancement.
In the present study, as expected, the highest efficiency is for the parabolic solar collector fully filled by porous material (copper foam with a porosity of 95%) and 0.5% Al2O3. Results of the present study are as follows:(1)The average enhancement of collectors’ efficiency using porous media and nanofluids is 28%.(2)Solutions with 0.1 to 0.5% of nanofluids (CuO and Al2O3) are used to prevent collectors from sediment occurrence in porous media.(3)Collector of solar cogeneration cycles that is enhanced by both porous media and nanofluid has higher efficiency, and the stability of output temperature is more as well.(4)By using 0.6% of the nanofluids in the enhanced parabolic solar collectors with copper porous materials, sedimentation occurs and makes a high-pressure drop in the solar collector’s pipe which causes decrease in energy efficiency.(5)Average enhancement of SCCHP cycle efficiency is enhanced by both porous media and nanofluid 13%.
Nomenclature
:
Solar radiation
a:
Heat transfer augmentation coefficient
A:
Solar collector area
Bf:
Basic fluid
:
Specific heat capacity of the nanofluid
F:
Constant of air dilution
:
Thermal conductivity of the nanofluid
:
Thermal conductivity of the basic fluid
:
Viscosity of the nanofluid
:
Viscosity of the basic fluid
:
Collector efficiency
:
Collector energy receives
:
Auxiliary boiler heat
:
Expander energy
:
Gas energy
:
Screw expander work
:
Cooling load, in kilowatts
:
Heating load, in kilowatts
:
Solar radiation energy on collector, in Joule
:
Sanitary hot water load
Np:
Nanoparticle
:
Energy efficiency
:
Heat exchanger efficiency
:
Sun exergy
:
Collector exergy
:
Natural gas exergy
:
Expander exergy
:
Cooling exergy
:
Heating exergy
:
Exergy efficiency
:
Steam mass flow rate
:
Hot water mass flow rate
:
Specific heat capacity of water
:
Power output form by the screw expander
Tam:
Average ambient temperature
:
Density of the mixture.
Greek symbols
ρ:
Density
ϕ:
Nanoparticles volume fraction
β:
Ratio of the nanolayer thickness.
Abbreviations
CCHP:
Combined cooling, heating, and power
EES:
Engineering equation solver.
Data Availability
For this study, data were generated by CARRIER software for the average electrical, heating, and cooling load of a residential building with 600 m2 in the city of Zahedan, Iran.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work was partially supported by the National Natural Science Foundation of China under Contract no. 71761030 and Natural Science Foundation of Inner Mongolia under Contract no. 2019LH07003.
References
A. Fudholi and K. Sopian, “Review on solar collector for agricultural produce,” International Journal of Power Electronics and Drive Systems (IJPEDS), vol. 9, no. 1, p. 414, 2018.View at: Publisher Site | Google Scholar
G. Yang and X. Zhai, “Optimization and performance analysis of solar hybrid CCHP systems under different operation strategies,” Applied Thermal Engineering, vol. 133, pp. 327–340, 2018.View at: Publisher Site | Google Scholar
J. Wang, Z. Han, and Z. Guan, “Hybrid solar-assisted combined cooling, heating, and power systems: a review,” Renewable and Sustainable Energy Reviews, vol. 133, p. 110256, 2020.View at: Publisher Site | Google Scholar
Y. Tian and C. Y. Zhao, “A review of solar collectors and thermal energy storage in solar thermal applications,” Applied Energy, vol. 104, pp. 538–553, 2013.View at: Publisher Site | Google Scholar
J. M. Hassan, Q. J. Abdul-Ghafour, and M. F. Mohammed, “CFD simulation of enhancement techniques in flat plate solar water collectors,” Al-Nahrain Journal for Engineering Sciences, vol. 20, no. 3, pp. 751–761, 2017.View at: Google Scholar
M. Jahangiri, O. Nematollahi, A. Haghani, H. A. Raiesi, and A. Alidadi Shamsabadi, “An optimization of energy cost of clean hybrid solar-wind power plants in Iran,” International Journal of Green Energy, vol. 16, no. 15, pp. 1422–1435, 2019.View at: Publisher Site | Google Scholar
I. H. Yılmaz and A. Mwesigye, “Modeling, simulation and performance analysis of parabolic trough solar collectors: a comprehensive review,” Applied Energy, vol. 225, pp. 135–174, 2018.View at: Google Scholar
F. Wang, J. Tan, and Z. Wang, “Heat transfer analysis of porous media receiver with different transport and thermophysical models using mixture as feeding gas,” Energy Conversion and Management, vol. 83, pp. 159–166, 2014.View at: Publisher Site | Google Scholar
H. Zhai, Y. J. Dai, J. Y. Wu, and R. Z. Wang, “Energy and exergy analyses on a novel hybrid solar heating, cooling and power generation system for remote areas,” Applied Energy, vol. 86, no. 9, pp. 1395–1404, 2009.View at: Publisher Site | Google Scholar
M. H. Abbasi, H. Sayyaadi, and M. Tahmasbzadebaie, “A methodology to obtain the foremost type and optimal size of the prime mover of a CCHP system for a large-scale residential application,” Applied Thermal Engineering, vol. 135, pp. 389–405, 2018.View at: Google Scholar
R. Jiang, F. G. F. Qin, X. Yang, S. Huang, and B. Chen, “Performance analysis of a liquid absorption dehumidifier driven by jacket-cooling water of a diesel engine in a CCHP system,” Energy and Buildings, vol. 163, pp. 70–78, 2018.View at: Publisher Site | Google Scholar
F. A. Boyaghchi and M. Chavoshi, “Monthly assessments of exergetic, economic and environmental criteria and optimization of a solar micro-CCHP based on DORC,” Solar Energy, vol. 166, pp. 351–370, 2018.View at: Publisher Site | Google Scholar
F. A. Boyaghchi and M. Chavoshi, “Multi-criteria optimization of a micro solar-geothermal CCHP system applying water/CuO nanofluid based on exergy, exergoeconomic and exergoenvironmental concepts,” Applied Thermal Engineering, vol. 112, pp. 660–675, 2017.View at: Publisher Site | Google Scholar
B. Su, W. Han, Y. Chen, Z. Wang, W. Qu, and H. Jin, “Performance optimization of a solar assisted CCHP based on biogas reforming,” Energy Conversion and Management, vol. 171, pp. 604–617, 2018.View at: Publisher Site | Google Scholar
F. A. Al-Sulaiman, F. Hamdullahpur, and I. Dincer, “Performance assessment of a novel system using parabolic trough solar collectors for combined cooling, heating, and power production,” Renewable Energy, vol. 48, pp. 161–172, 2012.View at: Publisher Site | Google Scholar
J. Wang, Y. Dai, L. Gao, and S. Ma, “A new combined cooling, heating and power system driven by solar energy,” Renewable Energy, vol. 34, no. 12, pp. 2780–2788, 2009.View at: Publisher Site | Google Scholar
Y.-Y. Jing, H. Bai, J.-J. Wang, and L. Liu, “Life cycle assessment of a solar combined cooling heating and power system in different operation strategies,” Applied Energy, vol. 92, pp. 843–853, 2012.View at: Publisher Site | Google Scholar
J.-J. Wang, Y.-Y. Jing, and C.-F. Zhang, “Optimization of capacity and operation for CCHP system by genetic algorithm,” Applied Energy, vol. 87, no. 4, pp. 1325–1335, 2010.View at: Publisher Site | Google Scholar
L. Ali, “LDA–GA–SVM: improved hepatocellular carcinoma prediction through dimensionality reduction and genetically optimized support vector machine,” Neural Computing and Applications, vol. 87, pp. 1–10, 2020.View at: Google Scholar
S. Hirasawa, R. Tsubota, T. Kawanami, and K. Shirai, “Reduction of heat loss from solar thermal collector by diminishing natural convection with high-porosity porous medium,” Solar Energy, vol. 97, pp. 305–313, 2013.View at: Publisher Site | Google Scholar
E. Bellos, C. Tzivanidis, and Z. Said, “A systematic parametric thermal analysis of nanofluid-based parabolic trough solar collectors,” Sustainable Energy Technologies and Assessments, vol. 39, p. 100714, 2020.View at: Publisher Site | Google Scholar
H. J. Jouybari, S. Saedodin, A. Zamzamian, M. E. Nimvari, and S. Wongwises, “Effects of porous material and nanoparticles on the thermal performance of a flat plate solar collector: an experimental study,” Renewable Energy, vol. 114, pp. 1407–1418, 2017.View at: Publisher Site | Google Scholar
J. Subramani, P. K. Nagarajan, S. Wongwises, S. A. El-Agouz, and R. Sathyamurthy, “Experimental study on the thermal performance and heat transfer characteristics of solar parabolic trough collector using Al2O3 nanofluids,” Environmental Progress & Sustainable Energy, vol. 37, no. 3, pp. 1149–1159, 2018.View at: Publisher Site | Google Scholar
E. Shojaeizadeh, F. Veysi, and A. Kamandi, “Exergy efficiency investigation and optimization of an Al2O3-water nanofluid based Flat-plate solar collector,” Energy and Buildings, vol. 101, pp. 12–23, 2015.View at: Publisher Site | Google Scholar
A. K. Tiwari, P. Ghosh, and J. Sarkar, “Solar water heating using nanofluids–a comprehensive overview and environmental impact analysis,” International Journal of Emerging Technology and Advanced Engineering, vol. 3, no. 3, pp. 221–224, 2013.View at: Google Scholar
D. R. Rajendran, E. Ganapathy Sundaram, P. Jawahar, V. Sivakumar, O. Mahian, and E. Bellos, “Review on influencing parameters in the performance of concentrated solar power collector based on materials, heat transfer fluids and design,” Journal of Thermal Analysis and Calorimetry, vol. 140, no. 1, pp. 33–51, 2020.View at: Publisher Site | Google Scholar
M. Feizbahr, C. Kok Keong, F. Rostami, and M. Shahrokhi, “Wave energy dissipation using perforated and non perforated piles,” International Journal of Engineering, vol. 31, no. 2, pp. 212–219, 2018.View at: Google Scholar
K. Khanafer and K. Vafai, “A critical synthesis of thermophysical characteristics of nanofluids,” International Journal of Heat and Mass Transfer, vol. 54, no. 19-20, pp. 4410–4428, 2011.View at: Publisher Site | Google Scholar
K. Farhana, K. Kadirgama, M. M. Rahman et al., “Improvement in the performance of solar collectors with nanofluids – a state-of-the-art review,” Nano-Structures & Nano-Objects, vol. 18, p. 100276, 2019.View at: Publisher Site | Google Scholar
M. Turkyilmazoglu, “Condensation of laminar film over curved vertical walls using single and two-phase nanofluid models,” European Journal of Mechanics-B/Fluids, vol. 65, pp. 184–191, 2017.View at: Publisher Site | Google Scholar
X. Zhang, J. Wang, T. Wang, R. Jiang, J. Xu, and L. Zhao, “Robust feature learning for adversarial defense via hierarchical feature alignment,” Information Sciences, vol. 2020, 2020.View at: Google Scholar
X. Zhang, T. Wang, W. Luo, and P. Huang, “Multi-level fusion and attention-guided CNN for image dehazing,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 1, 2020.View at: Publisher Site | Google Scholar
X. Zhang, M. Fan, D. Wang, P. Zhou, and D. Tao, “Top-k feature selection framework using robust 0-1 integer programming,” IEEE Transactions on Neural Networks and Learning Systems, vol. 1, pp. 1–15, 2020.View at: Publisher Site | Google Scholar
X. Zhang, D. Wang, Z. Zhou, and Y. Ma, “Robust low-rank tensor recovery with rectification and alignment,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 1, pp. 238–255, 2019.View at: Google Scholar
X. Zhang, R. Jiang, T. Wang, and J. Wang, “Recursive neural network for video deblurring,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 1, 2020.View at: Publisher Site | Google Scholar
X. Zhang, T. Wang, J. Wang, G. Tang, and L. Zhao, “Pyramid channel-based feature attention network for image dehazing,” Computer Vision and Image Understanding, vol. 1, 2020.View at: Google Scholar
M. Mirmozaffari, “Machine learning algorithms based on an optimization model,” 2020.View at: Google Scholar
M. Mirmozaffari, M. Yazdani, A. Boskabadi, H. Ahady Dolatsara, K. Kabirifar, and N. Amiri Golilarz, “A novel machine learning approach combined with optimization models for eco-efficiency evaluation,” Applied Sciences, vol. 10, no. 15, p. 5210, 2020.View at: Publisher Site | Google Scholar
M. Vosoogha and A. Addeh, “An intelligent power prediction method for wind energy generation based on optimized fuzzy system,” Computational Research Progress in Applied Science & Engineering (CRPASE), vol. 5, pp. 34–43, 2019.View at: Google Scholar
A. Javadi, N. Mikaeilvand, and H. Hosseinzdeh, “Presenting a new method to solve partial differential equations using a group search optimizer method (GSO),” Computational Research Progress in Applied Science and Engineering, vol. 4, no. 1, pp. 22–26, 2018.View at: Google Scholar
F. J. Golrokh, Gohar Azeem, and A. Hasan, “Eco-efficiency evaluation in cement industries: DEA malmquist productivity index using optimization models,” ENG Transactions, vol. 1, pp. 1–8, 2020.View at: Google Scholar
H. Yu, “Dynamic Gaussian bare-bones fruit fly optimizers with abandonment mechanism: method and analysis,” Engineering with Computers, vol. 1, pp. 1–29, 2020.View at: Google Scholar
C. Yu, “SGOA: annealing-behaved grasshopper optimizer for global tasks,” Engineering with Computers, vol. 1, pp. 1–28, 2021.View at: Google Scholar
W. Shan, Z. Qiao, A. A. Heidari, H. Chen, H. Turabieh, and Y. Teng, “Double adaptive weights for stabilization of moth flame optimizer: balance analysis, engineering cases, and medical diagnosis,” Knowledge-Based Systems, vol. 1, p. 106728, 2020.View at: Google Scholar
J. Tu, H. Chen, J. Liu et al., “Evolutionary biogeography-based whale optimization methods with communication structure: towards measuring the balance,” Knowledge-Based Systems, vol. 212, p. 106642, 2021.View at: Publisher Site | Google Scholar
Y. Zhang, “Towards augmented kernel extreme learning models for bankruptcy prediction: algorithmic behavior and comprehensive analysis,” Neurocomputing, vol. 1, 2020.View at: Google Scholar
Y. Zhang, R. Liu, X. Wang, H. Chen, and C. Li, “Boosted binary Harris hawks optimizer and feature selection,” Engineering with Computers, vol. 1, pp. 1–30, 2020.View at: Google Scholar
H.-L. Chen, G. Wang, C. Ma, Z.-N. Cai, W.-B. Liu, and S.-J. Wang, “An efficient hybrid kernel extreme learning machine approach for early diagnosis of Parkinson’s disease,” Neurocomputing, vol. 184, pp. 131–144, 2016.View at: Publisher Site | Google Scholar
L. Hu, G. Hong, J. Ma, X. Wang, and H. Chen, “An efficient machine learning approach for diagnosis of paraquat-poisoned patients,” Computers in Biology and Medicine, vol. 59, pp. 116–124, 2015.View at: Publisher Site | Google Scholar
L. Shen, H. Chen, Z. Yu et al., “Evolving support vector machines using fruit fly optimization for medical data classification,” Knowledge-Based Systems, vol. 96, pp. 61–75, 2016.View at: Publisher Site | Google Scholar
J. Xia, H. Chen, Q. Li et al., “Ultrasound-based differentiation of malignant and benign thyroid Nodules: an extreme learning machine approach,” Computer Methods and Programs in Biomedicine, vol. 147, pp. 37–49, 2017.View at: Publisher Site | Google Scholar
C. Li, L. Hou, B. Y. Sharma et al., “Developing a new intelligent system for the diagnosis of tuberculous pleural effusion,” Computer Methods and Programs in Biomedicine, vol. 153, pp. 211–225, 2018.View at: Publisher Site | Google Scholar
X. Zhao, X. Zhang, Z. Cai et al., “Chaos enhanced grey wolf optimization wrapped ELM for diagnosis of paraquat-poisoned patients,” Computational Biology and Chemistry, vol. 78, pp. 481–490, 2019.View at: Publisher Site | Google Scholar
M. Wang and H. Chen, “Chaotic multi-swarm whale optimizer boosted support vector machine for medical diagnosis,” Applied Soft Computing Journal, vol. 88, 2020.View at: Publisher Site | Google Scholar
X. Xu and H.-L. Chen, “Adaptive computational chemotaxis based on field in bacterial foraging optimization,” Soft Computing, vol. 18, no. 4, pp. 797–807, 2014.View at: Publisher Site | Google Scholar
R. U. Khan, X. Zhang, R. Kumar, A. Sharif, N. A. Golilarz, and M. Alazab, “An adaptive multi-layer botnet detection technique using machine learning classifiers,” Applied Sciences, vol. 9, no. 11, p. 2375, 2019.View at: Publisher Site | Google Scholar
A. Addeh, A. Khormali, and N. A. Golilarz, “Control chart pattern recognition using RBF neural network with new training algorithm and practical features,” ISA Transactions, vol. 79, pp. 202–216, 2018.View at: Publisher Site | Google Scholar
N. Amiri Golilarz, H. Gao, R. Kumar, L. Ali, Y. Fu, and C. Li, “Adaptive wavelet based MRI brain image de-noising,” Frontiers in Neuroscience, vol. 14, p. 728, 2020.View at: Publisher Site | Google Scholar
N. A. Golilarz, H. Gao, and H. Demirel, “Satellite image de-noising with Harris hawks meta heuristic optimization algorithm and improved adaptive generalized Gaussian distribution threshold function,” IEEE Access, vol. 7, pp. 57459–57468, 2019.View at: Publisher Site | Google Scholar
M. Eisazadeh and J. Rezapour, “Multi-objective optimization of the composite sheets using PSO algorithm,” 2017.View at: Google Scholar
I. Bargegol, M. Nikookar, R. V. Nezafat, E. J. Lashkami, and A. M. Roshandeh, “Timing optimization of signalized intersections using shockwave theory by genetic algorithm,” Computational Research Progress in Applied Science & Engineering, vol. 1, pp. 160–167, 2015.View at: Google Scholar
B. Bai, Z. Guo, C. Zhou, W. Zhang, and J. Zhang, “Application of adaptive reliability importance sampling-based extended domain PSO on single mode failure in reliability engineering,” Information Sciences, vol. 546, pp. 42–59, 2021.View at: Publisher Site | Google Scholar
J. Liu, C. Wu, G. Wu, and X. Wang, “A novel differential search algorithm and applications for structure design,” Applied Mathematics and Computation, vol. 268, pp. 246–269, 2015.View at: Publisher Site | Google Scholar
X. Zhao, D. Li, B. Yang, C. Ma, Y. Zhu, and H. Chen, “Feature selection based on improved ant colony optimization for online detection of foreign fiber in cotton,” Applied Soft Computing, vol. 24, pp. 585–596, 2014.View at: Publisher Site | Google Scholar
D. Zhao, “Chaotic random spare ant colony optimization for multi-threshold image segmentation of 2D Kapur entropy,” Knowledge-Based Systems, vol. 24, p. 106510, 2020.View at: Google Scholar
H. Chen, A. A. Heidari, H. Chen, M. Wang, Z. Pan, and A. H. Gandomi, “Multi-population differential evolution-assisted Harris hawks optimization: framework and case studies,” Future Generation Computer Systems, vol. 111, pp. 175–198, 2020.View at: Publisher Site | Google Scholar
J. Hu, H. Chen, A. A. Heidari et al., “Orthogonal learning covariance matrix for defects of grey wolf optimizer: insights, balance, diversity, and feature selection,” Knowledge-Based Systems, vol. 213, p. 106684, 2021.View at: Publisher Site | Google Scholar
G. Sun, B. Yang, Z. Yang, and G. Xu, “An adaptive differential evolution with combined strategy for global numerical optimization,” Soft Computing, vol. 24, pp. 1–20, 2019.View at: Google Scholar
G. Sun, C. Li, and L. Deng, “An adaptive regeneration framework based on search space adjustment for differential evolution,” Neural Computing and Applications, vol. 24, pp. 1–17, 2021.View at: Google Scholar
A. Addeh and M. Iri, “Brain tumor type classification using deep features of MRI images and optimized RBFNN,” ENG Transactions, vol. 2, pp. 1–7, 2021.View at: Google Scholar
F. J. Golrokh and A. Hasan, “A comparison of machine learning clustering algorithms based on the DEA optimization approach for pharmaceutical companies in developing countries,” Soft Computing, vol. 1, pp. 1–8, 2020.View at: Google Scholar
H. Tyagi, P. Phelan, and R. Prasher, “Predicted efficiency of a low-temperature nanofluid-based direct absorption solar collector,” Journal of Solar Energy Engineering, vol. 131, no. 4, 2009.View at: Publisher Site | Google Scholar
S. Rashidi, M. Bovand, and J. A. Esfahani, “Heat transfer enhancement and pressure drop penalty in porous solar heat exchangers: a sensitivity analysis,” Energy Conversion and Management, vol. 103, pp. 726–738, 2015.View at: Publisher Site | Google Scholar
N. Akram, R. Sadri, S. N. Kazi et al., “A comprehensive review on nanofluid operated solar flat plate collectors,” Journal of Thermal Analysis and Calorimetry, vol. 139, no. 2, pp. 1309–1343, 2020.View at: Publisher Site | Google Scholar
The low efficiency of Collectors that absorb energy can be mentioned as one of the drawbacks in solar cogeneration cycles. In the present study, solar systems have been improved by adding porous media and Nanofluid to collectors. One advantage of using porous media and nanomaterials is to absorb more energy while the surface area is reduced. In this study, first, solar collectors are enhanced using 90% porosity copper in solar combined cooling, heating and power systems (SCCHP). Second, different percentages of CuO and Al2O3 nano-fluids are added to a flat plate and parabolic collectors to enhance thermal properties. Simulations are performed in different modes (simple parabolic collectors, simple flat plate collectors, improved flat plate collectors, parabolic collectors with porous media, and flat plate and parabolic collectors with different density of CuO and Al2O3 nanofluids). A case study is investigated for warm and dry regions with mean solar radiation Ib = 820 w / m2 in Iran. The maximum energy and exergy efficiencies are 60.12% and 18.84%, respectively, that is related to enhanced parabolic solar collectors with porous media and nanofluids. Adding porous media and nano-fluids increases an average 14.4% collector energy efficiency and 8.08% collector exergy efficiency.
[1] Center TU. Annual report on China building energy efficiency. China Construction Industry Press (In Chinese). 2016.
[2] Tonekaboni N, Salarian H, Fatahian E, Fatahian H. Energy and exergy economic analysis of cogeneration cycle of homemade CCHP with PVT collector. Canadian Journal of Basic and Applied Sciences 2015;3:224-233.
[3] Hassan JM, Abdul-Ghafour QJ, Mohammed MF. CFD simulation of enhancement techniques in flat plate solar water collectors. Al-Nahrain Journal for Engineering Sciences 2017;20:751-761.
[4] Sopian K, Daud WR, Othman MY, Yatim B. Thermal performance of the double-pass solar collector with and without porous media. Renewable Energy 1999;18:557-564. https://doi.org/10.1016/S0960-1481(99)00007-5
[5] Feizbahr M, Kok Keong C, Rostami F, Shahrokhi M. Wave energy dissipation using perforated and non perforated piles. International Journal of Engineering 2018;31:212-219. https://doi.org/10.5829/ije.2018.31.02b.04
[6] Tian Y, Zhao CY. A review of solar collectors and thermal energy storage in solar thermal applications. Applied Energy 2013;104:538-553. https://doi.org/10.1016/j.apenergy.2012.11.051
[7] Wang F, Tan J, Wang Z. Heat transfer analysis of porous media receiver with different transport and thermophysical models using mixture as feeding gas. Energy Conversion and Management 2014;83:159-166. https://doi.org/10.1016/j.enconman.2014.03.068
[8] Korti AI. Numerical 3-D heat flow simulations on double-pass solar collector with and without porous media. Journal of Thermal Engineering 2015;1:10-23. https://doi.org/10.18186/jte.86295
[9] Sharma N, Diaz G. Performance model of a novel evacuated-tube solar collector based on minichannels. Solar Energy 2011;85:881-890. https://doi.org/10.1016/j.solener.2011.02.001
[10] Tyagi VV, Kaushik SC, Tyagi SK. Advancement in solar photovoltaic/thermal (PV/T) hybrid collector technology. Renewable and Sustainable Energy Reviews 2012;16:1383-1398. https://doi.org/10.1016/j.rser.2011.12.013
[11] Zhai H, Dai YJ, Wu JY, Wang RZ. Energy and exergy analyses on a novel hybrid solar heating, cooling and power generation system for remote areas. Applied Energy 2009;86:1395-1404. https://doi.org/10.1016/j.apenergy.2008.11.020
[12] Wang J, Dai Y, Gao L, Ma S. A new combined cooling, heating and power system driven by solar energy. Renewable Energy 2009;34:2780-2788. https://doi.org/10.1016/j.renene.2009.06.010
[13] Jing YY, Bai H, Wang JJ, Liu L. Life cycle assessment of a solar combined cooling heating and power system in different operation strategies. Applied Energy 2012;92:843-853. https://doi.org/10.1016/j.apenergy.2011.08.046
[14] Temir G, Bilge D. Thermoeconomic analysis of a trigeneration system. applied thermal engineering. Applied Thermal Engineering 2004;24:2689-2699. https://doi.org/10.1016/j.applthermaleng.2004.03.014
[15] Wang JJ, Jing YY, Zhang CF. Optimization of capacity and operation for CCHP system by genetic algorithm. Applied Energy 2010;87:1325-1335. https://doi.org/10.1016/j.apenergy.2009.08.005
[16] Kleinstreuer C, Chiang H. Analysis of a porous-medium solar collector. Heat Transfer Engineering 1990;11:45-55. https://doi.org/10.1080/01457639008939728
[17] Mbaye M, Bilgen E. Natural convection and conduction in porous wall, solar collector systems without vents. Jornal of Solar Energy Engineering 1992;114:40-46. https://doi.org/10.1115/1.2929980
[18] Hirasawa S, Tsubota R, Kawanami T, Shirai K. Reduction of heat loss from solar thermal collector by diminishing natural convection with high-porosity porous medium. Solar Energy 2013;97:305-313. https://doi.org/10.1016/j.solener.2013.08.035
[19] Jouybari HJ, Saedodin S, Zamzamian A, Nimvari ME, Wongwises S. Effects of porous material and nanoparticles on the thermal performance of a flat plate solar collector: an experimental study. Renewable Energy 2017;114:1407-1418. https://doi.org/10.1016/j.renene.2017.07.008
[20] Subramani J, Nagarajan PK, Wongwises S, El‐Agouz SA, Sathyamurthy R. Experimental study on the thermal performance and heat transfer characteristics of solar parabolic trough collector using Al2O3 nanofluids. Environmental Progress & Sustainable Energy 2018;37:1149-1159. https://doi.org/10.1002/ep.12767
[21] Yousefi T, Veysi F, Shojaeizadeh E, Zinadini S. An experimental investigation on the effect of Al2O3–H2O nanofluid on the efficiency of flat-plate solar collectors. Renewable Energy 2012;39:293-298. https://doi.org/10.1016/j.renene.2011.08.056
[22] Tyagi H, Phelan P, Prasher R. Predicted efficiency of a low-temperature nanofluid-based direct absorption solar collector. Journal of Solar Energy Engineering 2009;131:041004. https://doi.org/10.1115/1.3197562
[23] Shojaeizadeh E, Veysi F, Kamandi A. Exergy efficiency investigation and optimization of an Al2O3–water nanofluid based Flat-plate solar collector. Energy and Buildings 2015;101:12-23. https://doi.org/10.1016/j.enbuild.2015.04.048
[24] Tiwari AK, Ghosh P, Sarkar J. Solar water heating using nanofluids–a comprehensive overview and environmental impact analysis. International Journal of Emerging Technology and Advanced Engineering 2013;3:221-224. [25] Akram N, Sadri R, Kazi SN, Zubir MN, Ridha M, Ahmed W, et al. A comprehensive review on nanofluid operated solar flat plate collectors. Journal of Thermal Analysis and Calorimetry 2020;139:1309-1343. https://doi.org/10.1007/s10973-019-08514-z
[26] Lemington N. Study of solar driven adsorption cooling potential in Indonesia. Journal of Thermal Engineering 2017;3:1044-1051. https://doi.org/10.18186/thermal.290257
[27] Tong Y, Lee H, Kang W, Cho H. Energy and exergy comparison of a flat-plate solar collector using water, Al2O3 nanofluid, and CuO nanofluid. Applied Thermal Engineering 2019;159:113959. https://doi.org/10.1016/j.applthermaleng.2019.113959
[28] Khanafer K, Vafai K. A critical synthesis of thermophysical characteristics of nanofluids. International Journal of Heat And Mass Transfer 2011;54:4410-4428. https://doi.org/10.1016/j.ijheatmasstransfer.2011.04.048
[29] Farhana K, Kadirgama K, Rahman MM, Ramasamy D, Noor MM, Najafi G, et al. Improvement in the performance of solar collectors with nanofluids—A state-of-the-art review. Nano-Structures & Nano-Objects 2019;18:100276. https://doi.org/10.1016/j.nanoso.2019.100276
[30] Turkyilmazoglu M. Condensation of laminar film over curved vertical walls using single and two-phase nanofluid models. European Journal of Mechanics-B/Fluids 2017;65:184-91. https://doi.org/10.1016/j.euromechflu.2017.04.007
[31] Chen CC, Huang PC. Numerical study of heat transfer enhancement for a novel flat-plate solar water collector using metal-foam blocks. International Journal of Heat And Mass Transfer 2012;55:6734-6756. https://doi.org/10.1016/j.ijheatmasstransfer.2012.06.082
[32] Huang PC, Chen CC, Hwang HY. Thermal enhancement in a flat-plate solar water collector by flow pulsation and metal-foam blocks. International Journal of Heat and Mass Transfer 2013;61:696-720. https://doi.org/10.1016/j.ijheatmasstransfer.2013.02.037
[33] Hajipour M, Dehkordi AM. Mixed-convection flow of Al2O3–H O nanofluid in a channel partially filled with porous metal foam: experimental and numerical study. Experimental Thermal and Fluid Science 2014;53:49-56. https://doi.org/10.1016/j.expthermflusci.2013.11.002
[34] Rashidi S, Bovand M, Esfahani JA. Heat transfer enhancement and pressure drop penalty in porous solar heat exchangers: a sensitivity analysis. Energy Conversion and Management 2015;103:726-738. https://doi.org/10.1016/j.enconman.2015.07.019
[35] Manikandan GK, Iniyan S, Goic R. Enhancing the optical and thermal efficiency of a parabolic trough collector–A review. Applied Energy 2019;235:1524-1540. https://doi.org/10.1016/j.apenergy.2018.11.048
실제 작동 조건에서 석탄 연소 회전 시멘트 가마의 클링커 형성은 방사선에 대한 Monte Carlo 방법, 가마 벽의 에너지 방정식에 대한 유한 체적 코드 및 클링커에 대한 화학 반응을 포함한 에너지 보존 방정식 및 종에 대한 새로운 코드. 기상의 온도 장, 벽으로의 복사 열유속, 가마 및 클링커 온도에 대한 예측 간의 반복적인 절차는 내부 벽 온도의 분포를 명시적으로 예측하는 데 사용됩니다. 여기에는 열 흐름 계산이 포함됩니다. 수갑. 가스와 가마 벽 사이의 주요 열 전달 모드는 복사에 의한 것이며 내화물을 통해 환경으로 손실되는 열은 입력 열의 약 10%이고 추가로 40%는 장입 가열 및 클링커 형성. 예측은 실제 규모의 시멘트 가마에서 경험과 제한된 측정을 기반으로 한 경향과 일치합니다.
키워드
산업용 CFD, 로타리 가마, 클링커 형성, 복사 열전달, Industrial CFD, Rotary kilns, Clinker formation, Radiative heat transfer
1 . 소개
시멘트 산업은 에너지의 주요 소비자이며, 미국에서 산업 사용자의 총 화석 연료 소비량의 약 1.4%를 차지하며 [1] 일반적인 비에너지 사용량은 제조된 클링커 1kg당 약 3.2MJ [2] 입니다. CaCO 3 → CaO + CO 2 반응이 일어나기 때문입니다., 클링커 형성의 첫 번째 단계는 높은 흡열성입니다. 시멘트 가마에서 에너지를 절약하기 위한 현재의 경향은 일반적으로 길이가 약 100m이고 직경이 약 5m인 회전 실린더인 가마를 떠나는 배기 가스로부터 에너지를 보다 효율적으로 회수하는 것과 저열량 연료의 사용에 중점을 둡니다. 값. 2-5초 정도의 화염 체류 시간을 허용하고 2200K의 높은 온도에 도달하는 회전 가마의 특성은 또한 시멘트 가마를 유기 폐기물 및 용제에 대한 상업용 소각로에 대한 경쟁력 있는 대안으로 만듭니다 [3]. 클링커의 형성이 이러한 2차 액체 연료의 사용으로 인한 화염의 변화로부터 어떤 식으로든 영향을 받지 않도록 하고, 대기 중으로 방출되는 오염 물질의 양에 대한 현재 및 미래 제한을 준수할 수 있도록, 화염 구조의 세부 사항과 화염에서 고체 충전물로의 열 전달을 더 잘 이해할 필요가 있습니다.
최근 시멘트 가마 4 , 5 , 6 , 7 에서 유동장 및 석탄 연소의 이론적 모델링복사 열 전달을 포함한 전산 유체 역학(CFD) 코드를 사용하여 달성되었습니다. 이러한 결과는 시멘트 가마에 대한 최초의 결과였으며 화염 길이, 산소 소비 등과 관련하여 실험적으로 관찰된 경향을 재현했기 때문에 그러한 코드가 수용 가능한 정확도로 대규모 산업용 용광로에 사용될 수 있음을 보여주었습니다. 킬른과 클링커는 포함하지 않았고, 벽온도의 경계조건은 가스온도와 용액영역의 열유속에 영향을 미치므로 계산에 필요한 경계조건은 예측하지 않고 실험적 측정에 기초하였다. 기상에 대한 CFD 솔루션은 앞으로의 주요 단계이지만 회전 가마를 포괄적으로 모델링하는 데만으로는 충분하지 않습니다.
내화물의 열 전달과 전하에 대한 세부 사항은 다양한 저자 8 , 9 , 10 , 11에 의해 조사되었습니다 . 충전물(보통 잘 혼합된 것으로 가정)은 노출된 표면에 직접 복사되는 열 외에도 전도에 의해 가마 벽에서 가열됩니다. 가장 완전한 이론적 노력에서, 가마 벽 (내화물)에 대한 3 차원 열전도 방정식을 해결하고, 두 개 또는 세 개의 인접하는 영역으로 한정 한 좌표 축 방향에서 어느 방사선 방사선 열전달 영역 모델과 결합 [ 10] 또는 자세히 해결 [11]. 그러나 클링커 형성 중에 일어나는 화학 반응은 고려되지 않았고 기체 상이 균일한 온도로 고정되어 필요한 수준의 정확도로 처리되지 않았습니다.
최종적으로 연소에 의해 방출되는 에너지(일부)를 받는 고체 전하가 화학 반응을 거쳐 최종 제품인 클링커를 형성합니다. 이것들은 [12]에 설명된 주요 특징에 대한 단순화된 모델과 함께 시멘트 화학 문헌에서 광범위한 조사의 주제였습니다 . 그 작업에서, 고체 온도 및 조성의 축 방향 전개를 설명하는 odes가 공식화되고 해결되었지만, 전하에 대한 열유속 및 따라서 클링커 형성 속도를 결정하는 가스 및 벽 온도는 1차원으로 근사되었습니다. 자세한 화염 계산이 없는 모델.
화염, 벽 및 장입물에 대한 위의 이론적 모델 중 어느 것도 회전식 가마 작동을 위한 진정한 예측 도구로 충분하지 않다는 것이 분명합니다. 국부 가스 온도(CFD 계산 결과 중 하나)는 벽 온도에 크게 의존합니다. 클링커 형성은 에너지를 흡수하므로 지역 가스 및 벽 온도에 따라 달라지며 둘 다 화염에 의존합니다. 벽은 화염에서 클링커로의 순 열 전달에서 “중개자” 역할을 하며, 내화재 두께에 따라 환경으로 피할 수 없는 열 손실이 발생합니다. 이러한 상호 의존성은 가마의 거동에 중요하며 개별 프로세스를 개별적으로 계산하는 데 중점을 두었기 때문에 문헌에서 발견된 수학적 모델로는 다루기 어렵습니다.
본 논문에서 우리는 위에 설명된 유형의 세 가지 개별 모델을 결합하여 수행되는 회전식 시멘트 가마에서 발생하는 대부분의 공정에 대한 포괄적인 모듈식 모델을 제시합니다. 우리 작업은 4 , 5 , 6 , 7 에서와 같이 석탄 연소를 위한 다차원 CFD 코드로 기체 상태를 처리합니다 . 10 , 11 에서와 같이 가마 벽의 3차원 열전도 방정식을 풉니다 . 9 , 12 와 유사한 모델로 잘 혼합된 전하 온도 및 조성을 해결합니다.. 3개의 모듈(화염, 벽, 전하)은 내화물에 입사하는 열유속의 축 분포에 대해 수렴이 달성될 때까지 반복적으로 계산됩니다. 충전 온도 및 구성. 따라서 이전 작업에 비해 현재의 주요 이점은 완전성에 있습니다. 이는 가스-킬른-클링커 시스템의 다양한 부분에서 에너지 흐름의 정량화를 통해 킬른 작동에 대한 더 나은 이해를 가능하게 하고 여기에서 사용된 방법을 건조 및 소각과 같은 다른 회전 킬른 응용 분야에 적용할 수 있게 합니다.
이 문서의 특정 목적은 회전식 시멘트 가마에 대한 포괄적인 모델을 제시하고 화염에서 클링커로의 에너지 플럭스와 가마에서 열 손실을 정량화하는 것입니다. 이 문서의 나머지 부분은 다음과 같이 구성됩니다. 2장 에서는 다양한 모델과 해법을 제시하고 3장 에서는 그 결과를 제시하고 논의한다 . 여기에는 본격적인 회전식 시멘트 가마의 제한된 측정값과의 비교가 포함됩니다. 이 논문은 가장 중요한 결론의 요약으로 끝납니다.
2 . 모델 공식화
2.1 . 개요
Fig. 1 은 시멘트 로터리 킬른의 단면을 보여준다. 가마의 회전은 전하의 움직임을 유도하여 후자를 대략적으로 잘 혼합되도록 합니다 [10] , 여기에서 채택할 가정입니다. 우리는 이 코팅을 클링커와 유사한 물리적 특성의 고체 재료로 모델링하여 가마 내화물에 부착된 클링커의 존재를 허용할 것입니다. 우리는 이 층의 두께가 가마를 따라 균일하다고 가정합니다. 이것은 아마도 지나치게 단순화한 것일 수 있지만 관련 데이터를 사용할 수 없습니다. 모델 설명을 진행하기 전에 그림 2 에 개략적으로 표시된 회전식 가마의 다양한 에너지 흐름을 이해하는 것이 중요합니다 .
석탄 연소에 의해 방출되는 에너지(단위 시간당)( Q 석탄 )는 배기 가스(Δ H 가스 )와 함께 가마 밖으로 흘러 가마 벽에 직접 복사( Q rad ) 및 대류( Q conv )됩니다. 공급 및 배기 덕트( Q rad,1 + Q rad,2 ) 에 대한 축 방향의 복사에 의해 작은 부분이 손실됩니다 . 전하 가마 시스템은 복사( Q rad ) 및 대류( Q conv )에 의해 가스로부터 에너지(Δ H cl )를 흡수 하고 주변으로 열을 잃습니다( Q손실 ). 전체 에너지 균형에서 개별 항의 계산, 즉(1a)큐석탄=ΔH가스-Q라드-Q전환-Q일, 1-Q일, 2,(1b)큐라드+Q전환=ΔH클+Q손실여기에서 다음 섹션에 설명된 대로 가스, 가마 및 클링커에 대한 이산화 에너지를 국부적으로 해결함으로써 수행됩니다.
2.2 . CFD 코드
가스 운동량, 종 농도 및 에너지의 Favre 평균 방정식은 표준 k – ε 모델을 사용하여 방사 모듈(RAD-3D)과 함께 상업적으로 이용 가능한 축대칭 CFD 코드(FLOW-3D)에 의해 해결됩니다. [13] . 기하학이 실제로 3차원이고 벽 온도의 각도 분포가 존재하지만 합리적인 시간과 현재 워크스테이션에서 완전한 3으로 솔루션을 얻을 수 있도록 기체상을 축대칭으로 취급합니다. -D를 요구하는 해상도로 계산하려면 슈퍼컴퓨터에 의존해야 합니다. FLOW-3D에서 사용되는 다양한 하위 모델의 일부 기능과 벽 경계 조건에 대한 특수 처리는 다음과 같습니다.
2.2.1 . 석탄 연소
Rossin-Rammler 크기 분포(45μm 평균 직경, 1.3 지수 [6] )를 따르는 석탄 입자 는 CPU 시간을 줄이기 위해 솔루션 영역(즉, 확률적 구성 요소 없이)에서 결정론적으로 추적되었지만 분산을 과소 평가하는 단점이 있습니다 . 14] . 입자는 2-반응 모델에 따라 휘발되도록 허용되었고 휘발성 연소는 무한히 빠른 것으로 간주되었습니다. 석탄 연소에 대한 설명의 세부 사항은 FLOW-3D에서 석탄 휘발 및 열분해의 “표준” 상수 집합이 합리적인 결과를 제공하고 Ref. [5] .
2.2.2 . 복사와 대류
가스의 복사 강도는 RAD-3D 모듈을 사용하여 80,000개의 입자로 Monte-Carlo 방법으로 계산되었습니다. 가마는 반경 방향으로 7개, 축 방향으로 19개(크기가 0.1 × 1.0 m와 0.2 × 5.0 m 사이)로 불균일한 구역으로 나뉘었으며 각 구역 에서 방사선 강도가 균일하다고 가정했습니다. 방사선 모듈의 출력은 내부적으로 FLOW-3D에 대한 유체 계산에 인터페이스되고 외부적으로 벽 및 클링커에 대한 코드에 인터페이스되었습니다( 섹션 2.3 섹션 2.4 참조). 방사선 패키지의 이산화된 구역은 CFD 그리드의 셀보다 훨씬 커야 하므로 구역에 온도 평균이 형성될 수 있는 많은 셀이 포함될 수 있다는 점을 이해하는 것이 중요합니다. 상대적으로 조잡한 복사 구역의 분해능과 Monte-Carlo 방법의 통계적 특성은 구역의 복사 열유속이 더 미세한 구역화 및 더 많은 입자로 몇 번의 실행에 의해 결정된 바와 같이 최대 약 10%까지 부정확할 수 있음을 의미합니다. 또한 경계면에 입사하는 열유속은 영역 크기보다 미세한 분해능으로 결정할 수 없으므로 복사 열유속은 벽에 인접한 19개 영역 각각의 중심에서만 계산됩니다. 0.15m -1 의 흡수 계수는 Ref.[11] . 엄밀히 말하면, 흡수 계수는 국부적 가스 조성과 온도의 함수이므로 균일하지 않아야 합니다. 그러나 가스 조성은 가마의 일부만 차지하는 화염 내에서만 변 하므로( 3절 참조 ) 균일한 흡수 계수를 가정하는 것이 합리적입니다. 또한, 현재 버전의 소프트웨어는 FLOW-3D의 반복 프로세스 동안 이 요소의 자동 재조정을 허용하지 않습니다. 여기서 로컬 가스 특성이 계산되므로 일정하고 균일한 흡수 계수가 필요합니다.
최종적으로, 벽에서 대류 열전달이 플로우 3D 패키지에서 표준 출력 표준 “벽 기능”제형에 혼입 난류 경계층에 대한 식에 기초하고,의 속도 경계 조건과 유사한 K – ε 모델. FLOW-3D 및 RAD-3D에서 입력으로 사용하고 출력으로 계산된 다양한 양은 그림 3에 개략적으로 표시 됩니다.
2.2.3 . 그리드
반경 방향 47개, 축 방향 155개 노드를 갖는 불균일한 격자를 사용하였으며 격자 독립성 연구를 수행한 결과 충분하다고 판단하였다. 유사한 크기의 그리드도 Refs에서 적절한 것으로 밝혀졌습니다. 4 , 5 , 6 , 7 . 매우 높은 축 방향 및 소용돌이 속도로 인해 석탄 버너 유정에 가까운 지역을 해결하기 위해 특별한 주의를 기울였습니다. HP 715/100MHz 워크스테이션에서 이 그리드의 일반적인 CPU 시간은 10시간이었습니다.
2.2.4 . 경계 조건
벽 온도에 대한 경계 조건은 기체상 및 복사 솔버 모두에 필요하다는 것을 인식하는 것이 중요합니다. 아래에서는 4 , 5 , 6 , 7 을 규정하기 보다는 축대칭 그리드에 대한 이 온도 분포를 예측하는 대략적인 방법을 설명합니다 .
내벽 온도 T w ( R in , x , ϕ ) 의 각도 분포 가 알려져 있다고 가정합니다 . 그런 다음 전체 3차원 문제를 “동등한” 축대칭 문제로 줄이기 위해 가상의 내벽 온도 T RAD ( x )는(2)2πε에티4라드(x) = ε클∫0ㄷ티4클(엑스)디ϕ + ε에∫ㄷ2π티4에(아르 자형~에, x, ϕ)디ϕ”효과적인” 경계 조건으로 사용할 수 있습니다. T RAD ( x )는 방위각으로 평균화된 “복사 가중” 온도입니다. 필요한 경계 조건으로 이 온도를 사용하는 것은 복사가 열 전달을 지배한다는 기대에 의해 동기가 부여됩니다(후반부 확인, 섹션 3.4 ). 따라서 전체 3차원 문제와 이 “유효한” 축대칭 문제에서 가스에서 가마로의 전체 에너지 흐름은 거의 동일할 것으로 예상됩니다. 식 의 사용 (2) 축대칭 코드로 기체상 및 복사장을 계산할 수 있으므로 엔지니어링 워크스테이션을 사용하여 문제를 다루기 쉽습니다.
고려되는 가마의 규모와 온도에서 가스는 광학적으로 두꺼운 것으로 간주될 수 있습니다. 솔루션(나중에 제시됨)은 평균 경로 길이(즉, “광자”의 모든 에너지가 흡수되기 전의 평균 길이)가 약 3.2m임을 보여주며, 이는 가마 내경 4.1m보다 작습니다. 이것은 내벽에 입사하는 복사 플럭스가 국부적 벽과 가스 온도에 강하게 의존하고 더 먼 축 또는 방위각 위치에서 벽의 온도에 약하게만 의존함을 의미합니다. 이것은 기체상에 사용된 축대칭 근사에 대한 신뢰를 줍니다. 그것은 또한 Refs의 “구역 방법”을 의미합니다. 8 , 9 , 10표면에 입사하는 방사선이 1-2 구역 길이보다 더 먼 축 위치와 무관한 것으로 간주되는 경우에는 충분했을 것입니다.
2.3 . 가마 온도
내부 소성로 표면 온도 T w ( R in , x , ϕ )는 Eq. 에서 필요합니다 . (2) 및 가마 벽 에너지 방정식의 솔루션 결과의 일부입니다. 각속도 ω로 회전하는 좌표계 에서 후자는 [10] 이 됩니다 .(3)ω∂(ϱ에씨피티에)∂ϕ=1아르 자형∂∂아르 자형에게에아르 자형∂티에∂아르 자형+1아르 자형2∂∂ϕ에게에∂티에∂ϕ+∂∂엑스에게에∂티에∂엑스경계 조건에 따라(3a)r=R~에,Θ<ϕ⩽2π:에게∂티에∂아르 자형=q라드(x)+q전환(엑스),(3b)r=R~에, 0 <ϕ⩽Θ:에게∂티에∂아르 자형=qw–cl(x, ϕ) = hw–cl티클(x)-T에(아르 자형~에, x, ϕ),(3c)r=R밖, 0 <ϕ⩽2π:.케이∂티에∂아르 자형=h쉿티쉿-T∞+ ε쉿티4쉿-T4∞.
전도도, 밀도 및 비열용량에 대한 값은 실제 가마에 사용되는 내화물 재료에 대한 제조업체 정보에서 가져옵니다 [15] . 외부 쉘 온도 T sh = T w ( R out , x , ϕ )는 x 및 ϕ 에 따라 달라질 수 있습니다 .
위 방정식에 대한 몇 가지 의견이 있습니다. 에서는 식. (3a) 에서 열유속의 방위각 의존성이 제거되었습니다. 이전에 언급했듯이 흐름은 광학적으로 두꺼운 것으로 간주됩니다. 즉, 화염이 너무 방사되고 너무 넓기 때문에 벽면 요소가 화염을 가로질러 반대쪽 벽을 “보지” 않습니다. 따라서 q rad ( x , ϕ ) 의 계산은 다른 각도 위치로부터의 복사를 포함할 필요 없이 T 가스 ( r , x ) 및 로컬 T w ( R in , x , ϕ )를 기반으로 할 수 있습니다. 여기부터 qrad ( x )는 Eq. 의 방위각 평균 온도를 기반으로 하는 축대칭 RAD-3D 솔루션에서 가져옵니다 . (2) , 결과적인 q rad ( x )는 어떤 의미에서 방위각으로 평균된 열유속입니다. 식 따라서 (3a) 는 우리가 이 열유속을 모든 ϕ 에 등분포한다는 것을 의미합니다 . Eq 에서 q rad 의 각도 변화를 무시한다는 점에 유의하십시오 . (3a) 는 Refs. [10] 또는 [11] 이 우선되어야 합니다.
소성로와 장입물 사이의 열전달 계수 h w-cl 은 소성로의 에너지 흐름과 온도를 정확하게 예측하는 데 중요하지만 잘 알려져 있지 않습니다. 500 W / m의 전형적인 값 이 K는 여기에 제시된 결과 사용되고있다 [8] . 계산된 T w ( r , x , ϕ ) 및 T RAD ( x) 이 계수의 선택에 따라 달라지지만 예측은 질적으로 변하지 않습니다. 껍질에서 대기로의 열 전달은 복사와 별도로 강제 및 자연 대류를 통해 발생합니다. 자연 대류에 대한 열전달 계수는 Ref. [11] , 현재 조건에서 약 5 W/m 2 K의 일반적인 값 을 사용합니다. 그러나 쉘에 불어오는 외부 팬은 과열을 피하기 위해 산업에서 종종 사용되며 이러한 효과는 총 h sh =30 W/m 2 K 를 사용하여 여기에서 모델링 되었습니다. 방사율에는 다음 값이 사용되었습니다. ε w = ε cl = 0.9 및 ε sh = 0.8.
식 (3) 은 가마의 방사형 기울기가 훨씬 더 가파르기 때문에 방위각 및 축 전도를 무시한 후 명시적 유한 체적 방법으로 해결되었습니다. 방사형으로 50개 노드와 축 방향으로 19개 노드가 있는 균일하지 않은 그리드가 사용되었으며 회전으로 인한 화염에 주기적으로 노출되는 표면으로 인해 발생하는 빠른 온도 변화를 따르기 위해 내부 표면에서 적절한 방사형 분해능이 사용되었습니다. 동일한 이유로 사용 된 작은 단계(Δ ϕ = π /100)는 가마의 큰 열 관성과 함께 가마 벽 온도가 수렴되도록 하기 위해 2시간 정도의 CPU 시간이 필요했습니다.
2.4 . 수갑
가마에 대한 모델의 마지막 부분은 클링커 온도 및 조성 보존 방정식에 관한 것으로, 축 방향 기울기만 고려하고 전도는 무시합니다.(4)씨피V클디(ϱ클티클)디엑스=−엘wclㄷㅏ클∫0ㄷ큐w–cl(x, ϕ)디ϕ +엘gclㅏ클큐라드(x)+q전환(엑스)−∑나Nsp아르 자형나시간0, 나는에프+씨피티,(5)V클디(ϱ클와이나)디엑스=r나,(6)V클디ϱ클디엑스=−r무엇2,여기서 A cl 은 속도 V cl 로 흐르는 전하가 덮는 단면적 이며 둘 다 일정하다고 가정하고 L gcl =2 R in sin( Θ /2) 전하로 덮인 섹터의 현( 그림 1 ) , L WCL = Θ R 에서는 , N SP 화학 종의 수와 r에 난을 (kg / m의 형성 속도 순 3 종의) I를 . 전하의 밀도는 Eq를 감소시킵니다 . (6) CO 2 에 대한 질량 손실로 인한하소하는 동안 초기 값은 총 질량 유량이 ϱ cl V cl A cl 과 같도록 선택되었습니다 . 참고 ρ (CL)이 있다 하지 전하 느슨하게 포장 된 입자로 이루어지는 것으로 생각 될 수있는 바와 같이, 충전 재료 밀도하지만 벌크 밀도. 우리는 또한 전하의 실제 입상 흐름 패턴을 조사하는 것보다 적은 것은 모델의 신뢰성에 크게 추가되지 않는 임시 설명 [10] 이라고 믿기 때문에 전하의 전도를 무시 합니다. 전하는 CaCO 3 , CaO, SiO 2 , Al 2 O 3 , Fe 로 구성된 것으로 가정합니다.2 O 3 , C2S, C3S, C3A 및 C4AF로, 마지막 4종은 클링커화 중에 형성된 복합 염에 대해 시멘트 화학자가 사용하는 특수 표기법으로 표시됩니다. 다음과 같은 화학 반응을 가정합니다 [12] .
(나)
CaCO3→높은+무엇2
k = 108특급(−175728/RT)
(Ⅱ)
높은+2SiO2→C2S
k = 107특급(−240000/RT)
(Ⅲ)
높은+C2S→C3S
k = 109특급(−420000/RT)
(IV)
3높은+로2그만큼3→C3A
k = 108특급(−310000/RT)
(V)
4높은+로2그만큼3+철2그만큼3→Q4AF
k = 108특급(−330000/RT)
상기 시행 착오에 의해 선택되는 아 레니 우스 식에 사용되는 사전 지수 인자 및 활성화 온도는 카코에 대한 활성화 에너지를 제외하고, 가마의 출구에서의 전하의 예상 조성물을 얻었다 (3) 에서 촬영 한 분해 참조 [16] . 우리는 이러한 반응이 임시 모델임을 강조합니다. 실제로 고체상의 화학반응은 다양한 종의 결정들 사이의 계면에서 일어나며 확산이 제한적 이지만 [17] , 클링커 화학에 대한 상세한 처리는 본 연구의 범위를 벗어난다.
클링커 형성의 마지막 단계로 간주되는 반응 (III)은 고온에서 액상이 존재할 때만 발생합니다. 클링커의 용융은 액체 분획 Y fus 에 대해서도 해결함으로써 모델링되었습니다 .(7)엘소란V클디(ϱ클와이소란)디엑스=RHS의식(4)만약 T의 CL이 융해 온도와 같거나보다 커진다 T의 FUS 와 T의 FUS 의 = 1560 K. 상한 Y의 FUS = 0.3 수행 하였다 [17] 상기 식을. (7) 무시되었다.
상미분 방정식, , , , , Gear 방식과 통합되었습니다. 가마 온도에 대한 유한 체적 코드( 2.3절 )와 클링커에 대한 코드는 반복적으로 해결되었으며( 그림 4 ), 이는 벽 클링커 열유속 q w–cl ( x , ϕ ).
2.5 . 최종 커플링
전체 문제(가스, 가마, 장입)는 반복 방식으로 해결되었습니다. T RAD 의 균일한 분포에서 시작 하여 기체상은 q rad ( x ) 및 q conv ( x ) 의 축 분포를 제공하도록 해결되었습니다 . 이것들은 다음에서 사용되었습니다., , , , , 그 솔루션의 새로운 추정 결과 T RAD ( X 통해) 식. (2) . 그런 다음 FLOW3D-RAD3D 실행이 6차 다항식 피팅의 계수 형태로 프로그램에 도입된 새로운 경계 조건으로 반복되었습니다. 의 연속 추정치 사이에 0.5 미만의 밑에 이완 인자 T RAD ( X)는 벽 온도에 대한 복사 열유속의 민감도가 크기 때문에 필요한 것으로 밝혀졌습니다. 일반적으로 HP 715 워크스테이션에서 10일 정도의 총 CPU 시간에 해당하는 내벽 온도(연속 반복이 40K 이상 변하지 않을 때 정의됨)의 수렴을 달성하기 위해 이러한 단계 사이에 약 10번의 반복이 필요했습니다. . 그림 5 는 균일한 값(1600K)에서 시작하여 최종 프로파일까지 T RAD ( x ) 의 수렴 이력을 보여줍니다 .
2.6 . 가마 조건
사용된 일부 매개변수에 대한 작동 조건 및 값은 표 1 표 2 표 3에 나와 있습니다. 이 값은 시멘트 회전 가마의 전형입니다.
표 1 . 공기 및 석탄 입자 입구 조건
축
수송
소용돌이
중고등 학년
석탄
m (kg/s)
2.253
1.759
2.910
45.930
4.0
유 (m/s)
77.1
36.5
76.1
12.73
36.5
V (m/s)
−20.7
0
63.9
0
0
W (m/s)
0
0
112.8
0
0
티 (케이)
318
383
318
1273
383
표 2 . 클링커 조성(질량 분율)
밀가루
가마 입구
가마 출구
m (kg/s)
50.374
39.815
32.775
티 (케이)
−
1100
1785
CACO 3
0.7947
0.40218
0
높은
0
0.33801
0.0229
그런가 2
0.1434
0.18143
0
알 2 O 3
0.0349
0.0442
0
철 2 O 3
0.0270
0.03416
0
C2S
0
0
0.1808
C3S
0
0
0.5981
C3A
0
0
0.0731
Q4AF
0
0
0.1242
소성 인자
0
0.6
1.0
소성 계수 카코의 비율을 3 의 CaO로 변환 된 FARINE있다.
표 3 . 재료 속성 및 기타 매개변수
ω (래드/초)
0.5
V의 CL (m / s)
0.035
T ∞ (K)
300
h sh (W/m 2 K)
30
h w–cl (W/m 2 K)
500
ε w , ε cl
0.9
ε 쉬
0.8
C의 P (클링커) (킬로 / kg K)
1.5
ϱ cl (kg/m 3 )
1200
L fus (kJ/kg)
418.4
c p (벽) (kJ/kg K)
1.5
ϱ w (kg/m 3 )
1600–3000
k는 w (W / m K)
0.6–3.0
석탄 열 방출(kJ/kg)
25475
3 . 결과 및 토론
이 섹션에서는 먼저 화염 구조에 대한 정보와 함께 예측된 공기역학적 패턴의 세부사항을 제시합니다. 소성로 내화물의 온도 분포와 클링커 조성의 변화를 설명합니다. 이 섹션은 가마의 전체 에너지 균형과 가능한 모델 개선에 대한 논의로 끝납니다.
3.1 . 화염 구조
그림 6 은 명확성을 위해 방사상 좌표가 과장된 온도의 등고선 플롯을 보여줍니다. 석탄은 주입 지점에서 약 1m 지점에서 약간 축에서 벗어나 점화되며 최대 화염 온도(약 2400K)는 경험에 따라 약 40m 하류에서 도달합니다 [15] . 완전한 입자 소진에 대한 가장 긴 시간은 버너에서 45m에 해당하는 약 1.4초였습니다. 방사형 온도 프로파일( 그림 7 ) 은 온도의 상당한 불균일성이 있음을 보여주지만 출구 프로파일이 본질적으로 평평해짐에 따라 하류에서 감소합니다. 또한 벽에 인접한 가스가 더 차가운 열 경계층이 존재한다는 것이 분명합니다.석탄 노즐에서 최대 30m까지 벽보다 이것은 이 영역에서 대류에 의한 열 전달이 음(즉, 기체 쪽으로)임을 의미하며, 3.4절 에서 더 자세히 논의된 지점 입니다.
버너 출구 바로 하류에 길이가 약 1 버너 직경인 재순환 구역이 있는데( 그림 8 ), 여기에서 화염이 더 하류에서 발화하기 때문에 소용돌이 안정화 화염 [7] 에서와 같이 화염 안정화에 기여하지 않습니다 . 그러나 액체 연료를 사용할 때는 중요할 수 있으므로 버너에 가까운 그리드의 세부 사항을 강조해야 합니다. 버너에서 처음 몇 미터는 매우 높은 전단력과 높은 난류 에너지 생산을 포함하며 이것이 그리드 미세 조정을 강조하는 또 다른 이유입니다. 휘발성 물질 연소 영역( x =10m, r =1m) 에서 k 및 ε 의 일반적인 예측 값 은 24.3 및 142m 2 /s입니다.3 , 각각. 대규모 난류 시간은 171ms이고 Kolmogorov 시간 규모는 1.1ms입니다. 휘발성 물질의 연소는 0.1ms(일반적인 탄화수소 연료) 정도의 시간 규모에서 발생하며, 이는 가마의 소규모 난류 시간보다 10배 더 짧습니다. 따라서 이 흐름에서 연소에 대한 유한 속도 동역학을 포함할 필요는 없으며 “혼합 연소” 근사가 합리적입니다.
3.2 . 가마 온도 분포
중심선에서 계산된 가스 온도, 온도 T RAD ( x ) 및 클링커 온도는 그림 9 에서 비교됩니다 . 최고 가스 온도는 25~40m 사이에 위치하며 내화 내부 표면 온도도 최고점입니다. 클링커는 놀랍게도 가마에서 나오기 전 마지막 몇 미터 동안 벽보다 뜨겁 습니다. 복사에 의해 내화물에 입사하는 열유속은 대류에 의한 것보다 1-2 배 더 높으며( 그림 10 ) 가마의 처음 10m에 대한 총 열 전달 은 가스를 향 합니다. 이 관찰의 중요성은 나중에 논의됩니다.
대류로 인한 에너지 플럭스는 화염에서 가마까지의 전체 에너지 플럭스의 매우 작은 부분인 것으로 밝혀졌습니다( 그림 10 ). 여기서 예측된 대류의 작은 기여는 Ref. [11] . 그 작업에서 대류 열 전달 계산에 사용된 가스 온도는 가마 단면의 평균이었고 따라서 축 근처에 있는 화염의 기여로 인해 벽 부근의 온도보다 훨씬 높았습니다. . 여기에서 우리는 온도와 가스 속도 및 난류 운동 에너지의 국부적 값을 기반으로 하는 보다 정확한 열전달 계수를 사용했기 때문에 보다 정확한 결과를 기대합니다.
예측된 벽 온도는 모든 방향에서 불균일합니다. Fig. 11 은 가마가 회전함에 따라 화염에 노출되었을 때 벽이 가스에 의해 연속적으로 가열되고 클링커에 열을 공급하여 냉각되는 것을 보여준다. 이것은 약 100K의 일반적인 각도 온도 변화를 갖는 대부분의 가마 길이에 해당됩니다. 대조적으로 버너에 가까우면 벽 은 (0 < ϕ < π /2) 동안 클링커에서 열을 얻고 다음으로 열을 잃 습니다. 노출될 때의 가스( π /2 < ϕ < 2 π ). 벽과 클링커 온도가 같으면서 방위각 변화가 없는 경우가 발생할 수 있습니다( 그림 11 , x = 17.5m). 이 온도 변화가 작은 것으로 간주될 수 있지만 벽에서 클링커까지의 열유속을 계산하는 위치에 있으려면 전체 3차원 내벽 온도 분포를 계산해야 합니다(0 < ϕ 범위에서 발생 < π /2).
그림 12 는 ϕ에 독립적인 외부(쉘) 온도와 함께 고체의 큰 비열로 인해 각도 방향의 변화 영역이 벽으로 약 1cm만 확장됨을 보여줍니다( 그림 12b) .. 벽 온도 방사 분포는 가스 온도, 입사 방사선 및 내화 재료의 특성이 변하기 때문에 축 방향 거리에 따라 달라집니다. 정확한 예측을 위해서는 내화물에 부착된 클링커 코팅의 두께에 대한 정확한 지식이 필요합니다. 여기에서 우리는 이 코팅을 클링커와 유사한 물성을 가진 균일한 두께의 재료로 취급했습니다. 그러나 이 코팅층의 실제 물리적 특성과 두께 분포에 관한 실험 데이터를 사용하여 예측의 신뢰성이 향상될 것입니다.
마지막으로, 그림 13 은 외부 쉘 온도가 화염 영역에서 최고조에 달하고 대략적으로 실험 경향을 따른다는 것을 보여줍니다 [15] . 외부 가마 외피는 다양한 강철 두께, 방사율(외피 착색으로 인한) 및 열 전달 계수(송풍기 간격으로 인한)를 갖고 가마는 가변 내화 두께(에 의한 침식으로 인해)를 갖기 때문에 정확한 비교는 의미가 없습니다. 클링커), 여기에 사용된 가정과 반대입니다. 전체 규모 가마는 또한 차등 코팅 및 내화 침식으로 인한 최대 ±100K의 쉘 온도 각도 변동을 보여줍니다 [15] . 따라서 우리는 그림 13 의 일치 가 실제 가마의 복잡성을 고려할 때 예상할 수 있는 만큼 우수 하다고 믿습니다 .
이 섹션에 제시된 예측은 가마 내부의 열 전달 경로에 대한 다음 그림을 뒷받침합니다. 대부분의 가마 길이에서 장입물은 화염으로부터의 복사와 벽으로부터의 열 전도에 의해 가열되고 있습니다. 장입물이 내화물보다 더 차갑기 때문입니다. 가마가 회전함에 따라 내화물은 화염에 노출될 때 열을 얻고 이를 클링커에 공급합니다( 그림 11 ). 벽의 이 “재생” 작용은 Refs. 9 , 10 및 현재 결과에서 재현되었습니다. 그러나 버너 근처에서 반대 에너지 흐름이 발생합니다( 그림 11 , 작은 x). 여기의 가스는 아직 충분히 뜨겁지 않아 내화물이나 장입물에 에너지를 공급하지 않습니다. 이 영역에서 벽은 다가오는 전하에 의해 열을 얻으므로 고체가 없을 때보다 더 뜨겁게 유지됩니다. 벽과 전하가 대류와 복사에 의해 가스에 열을 공급합니다. 우리는 이것을 “음의 재생” 작용으로 식별할 수 있으며 가마의 더 높은 온도 영역( x > 15m) 에서 클링커에 의해 흡수된 에너지에 의해 유지됩니다 . 전반적으로 클링커는 x > 15 m 에서 열을 흡수 하고 0 < x < 15 m 에서 일부를 가스로 되돌려 줍니다.
이 상호 작용은 간단하지 않으며 쉽게 예상할 수 없습니다. 이는 예를 들어 고체를 액체 연료로 대체하여 화염을 수정하면 열유속 분포를 변경하여 최종 클링커 온도에 중대한 영향을 미칠 수 있음을 의미합니다. 현재의 포괄적인 모델이 제공하는 세부 사항은 가마에서 이러한 변화를 평가하는 데 도움이 될 것입니다.
3.3 . 클링커 온도 및 조성
클링커 온도( 그림 9 )는 가장 높은 화염 온도에 도달하는 축 방향 위치에서 거의 최고조에 달하며 클링커는 약 1780K에서 킬른에 존재하며 이는 시멘트 킬른에서 실험 측정값에 가까운 값입니다 [15] . 초기 및 최종 클링커 조성은 표 2 에 나와 있으며 실제 가마에서 작동 값에 가깝습니다 [15] . 다양한 클링커 성분의 축방향 분포( 그림 14 )는 완전한 하소를 위해 고체 유입구에서 약 25m, C2S, C3A 및 C4AF 생성을 위해 추가로 10m가 소요됨을 보여줍니다. 첫 번째 액체상은 x 에서 발견됩니다.=50m이고 액화는 경험과 일치하는 예측인 매우 직후에 완료됩니다 [17] . 클링커화 반응(R-III)은 모델에서 액체가 나타날 때 시작되는 것으로 가정되었으며, 그림 14 에서 클링커화에는 나머지 길이의 거의 전체가 완료되어야 한다는 것이 분명 합니다. 예측은 전체적으로 시멘트 가마 운영의 경험과 일치하며 여기에 사용된 화학적 및 물리적 매개변수가 현실적인 값을 가지고 있음을 의미합니다.
3.4 . 글로벌 에너지 균형
전지구적 에너지 균형은 기체상(FLOW-3D 및 RAD-3D에 의한)과 소성로 장입 시스템에 대한 솔루션에서 쉽게 계산할 수 있으며 표 4 에 나와 있습니다. CFD 코드는 방사 모듈과 함께 에너지를 약 2%까지 절약합니다. 작은 것으로 간주되는 이 오류는 주로 RAD-3D의 영역 이산화와 Monte-Carlo 계산의 유한한 입자 수로 인해 발생하는 오류에 기인하며 CPU 시간을 희생하여 개선할 수 있습니다. 소성로-클링커 계산의 정확도는 더 나쁩니다. 소성로-클링커 시스템에 입력되는 에너지의 약 10% 오류( Q rad + Q conv )입니다. 이는 수렴된 솔루션이 식 (3) , 그리고 보다 정확한 암시적 솔버에 의해 개선될 수 있습니다.
시멘트 회전식 가마의 에너지 사용에 관한 몇 가지 흥미로운 결론은 표 4 의 결과를 통해 얻을 수 있습니다 . 연소에 의해 방출되는 에너지의 약 40%는 전하 가열 및 클링커 형성에 필요하고 약 10%는 내화물을 통해 대기로 손실됩니다. 나머지의 대부분은 본질적으로 배기 가스와 함께 소성로 밖으로 흐릅니다. 이 중 일부는 소성로 외부의 예비 하소기 및 사이클론에서 회수됩니다. 내부 가마 벽과 장입 온도를 자세히 다루는 여기에 제시된 포괄적인 모델에 의존하지 않고는 국지적 가스 온도를 정확하게 예측하고 이에 따라 향후 연구에서 오염 물질 형성을 예측하는 것이 불가능하다는 것이 분명합니다.
3.5 . 논의
여기에 제시된 회전식 시멘트 가마 작동에 대한 포괄적인 모델의 결과는 합리적이며 실험적으로 관찰된 경향을 재현합니다. 이전 모델링 작업에 비해 이 작업의 주요 이점은 가마에서 발생하는 대부분의 물리적 프로세스를 포함한다는 점입니다. 특히, 가스 온도와 클링커로의 열유속 및 이에 따른 클링커 형성을 결정하는 데 가장 중요한 양인 내벽 온도는 실험 데이터를 사용하여 규정된 것이 아니라 예측되었습니다. 이 특정 기능은 현재 모델을 진정한 예측형으로 만듭니다.
우리는 전체 3차원 문제를 공기역학에 대한 “동등한” 축대칭 문제로 줄이는 방법을 포함했습니다( 식 (2) ). 이를 통해 현재 워크스테이션에서 솔루션을 얻을 수 있습니다. 모델의 모듈식 특성, 즉 공기역학, 복사, 가마 및 장입에 대한 별도의 코드는 해당 모듈만 수정하면 다른 회전 가마 응용 프로그램(예: 소각 및 건조)에도 사용할 수 있음을 의미합니다. 예를 들어, 고형 폐기물의 소각은 현재 코드로 모델링할 수 있지만 적절한 화학, .
실험 데이터와의 상세한 비교는 이용 가능한 측정이 거의 없고 현지 시멘트 회사에서 제공한 경험적 데이터로 제한되어 매우 어렵습니다 [15] . 비교는 앞서 지적한 바와 같이 출구 클링커 조성과 온도가 산업적 경험( 표 2 ) 이내 이고, 배기 가스 조성은 공장 굴뚝에서 측정된 값에 가깝고(“가짜 공기” 희석을 허용한 후), 가마 외피 온도는 측정 범위 내에 있습니다( 그림 13 ). 이 동의는 모델이 프로세스의 정확한 표현임을 시사합니다.
더 높은 정확도의 예측을 달성하려면 모델의 다양한 부분에서 개선이 필요합니다. 내화물의 정확한 두께(즉, 내화물과 부착된 클링커)를 설정해야 합니다. 이는 가마 벽을 통해 주변으로 열 손실이 발생하여 외부 쉘 온도에 영향을 미치기 때문입니다. 새 내화물이 있는 가마에서 쉘 온도 측정과 자세한 비교가 이루어져야 합니다(불균일한 코팅 두께가 방지되도록). 벽 재료의 물리적 특성(열용량, 밀도, 전도도)의 적절한 값을 사용해야 합니다. 가장 큰 불확실성은 클링커 코팅의 가정된 특성에 관한 것입니다. 내벽 표면의 방사율과 가스의 흡수 계수를 더 자세히 조사해야 합니다. 가마에 입사하는 복사 열유속에 영향을 미치므로 벽 온도에 영향을 줄 수 있습니다. 클링커의 온도는 사용된 비열 용량에 따라 달라지므로 정확한 평가에 각별한 주의가 필요합니다. 화염의 국지적 온도와 종 구성에 대한 지식은 CFD 코드를 검증하는 데 매우 유용할 것이지만 그러한 적대적인 환경에서 측정은 분명히 달성하기 매우 어렵습니다. 마지막으로 클링커 화학 및 전하 이동은 개선할 수 있는 영역입니다. 그러한 적대적인 환경에서의 측정은 분명히 달성하기 매우 어렵습니다. 마지막으로 클링커 화학 및 전하 이동은 개선할 수 있는 영역입니다. 그러한 적대적인 환경에서의 측정은 분명히 달성하기 매우 어렵습니다. 마지막으로 클링커 화학 및 전하 이동은 개선할 수 있는 영역입니다.
이러한 모든 잠재적 개선과 모델과 관련된 불확실성에도 불구하고 가마의 모든 에너지 경로가 적절한 세부 사항으로 모델링되었기 때문에 전체 동작은 최소한 질적으로 정확합니다. 클링커 출구 구성, 쉘 온도 및 배기 가스 구성과 같은 중요한 양은 허용 가능한 정확도로 예측됩니다. 이 모델은 버너, 연료 유형, 품질 및 수량, 예비 하소 수준( 표 2 ) 또는 고형물 유량 등의 변경과 같은 많은 상황에서 산업계에 매우 유용할 것으로 예상됩니다 . 소성로 운영자는 최종 클링커 구성이 여전히 허용 가능하고 현재의 포괄적인 모델이 이 방향에 도움이 될 수 있는지 확인해야 합니다.
4 . 결론
실제 작동 조건에서 석탄 연소 회전 시멘트 가마의 클링커 형성은 석탄 화염과 가마 사이의 열 교환, 가마와 역류 고체 사이의 열 교환, 고형물을 최종 제품(클링커)으로 변환합니다. 방사선에 대한 Monte-Carlo 방법을 포함하는 축대칭 CFD 코드(상용 패키지 FLOW-3D)가 기상에 사용되었습니다. 가마 벽의 온도는 유한 체적 열전도 코드로 계산되었으며 클링커에 대한 종 및 에너지 보존 방정식도 공식화 및 해결되었습니다. 기체 온도 필드에 대한 예측 사이의 반복적인 절차, 벽에 대한 복사 열 유속, 가마 및 클링커 온도는 실험에서 이러한 정보를 사용한 이전 모델링 노력과 달리 내벽 온도 분포를 명시적으로 계산하는 데 사용되었습니다. 접선 좌표에 대한 통합은 CFD 코드에 필요한 경계 조건으로 사용되는 “유효” 내벽 온도의 축 분포를 초래했습니다. 이 절차를 통해 클링커로의 열 흐름 계산이 가능하고 축대칭 CFD 코드로 3차원 문제를 대략적으로 처리할 수 있습니다. CFD 코드에 필요한 경계 조건으로 사용됩니다. 이 절차를 통해 클링커로의 열 흐름 계산이 가능하고 축대칭 CFD 코드로 3차원 문제를 대략적으로 처리할 수 있습니다. CFD 코드에 필요한 경계 조건으로 사용됩니다. 이 절차를 통해 클링커로의 열 흐름 계산이 가능하고 축대칭 CFD 코드로 3차원 문제를 대략적으로 처리할 수 있습니다.
결과는 복사가 가스와 가마 벽 사이의 대부분의 열 전달을 설명하는 반면 내화물을 통한 환경으로의 열 손실은 입력 열의 약 10%를 설명한다는 것을 보여줍니다. 화학 반응과 충전물의 가열은 연소 에너지의 약 40%를 흡수합니다. 따라서 이러한 사항을 반드시 고려해야 합니다. 예측은 실제 규모의 시멘트 가마에서 얻은 경험과 측정값을 기반으로 한 경향과 일치합니다.
감사의 말
이 작업은 과학 및 기술을 위한 그리스 사무국 프로젝트 EPET-II/649의 자금 지원을 받았습니다. Mr.P에게 진심으로 감사드립니다. 시멘트 가마에 관한 지침 및 데이터는 그리스 TITAN SA의 Panagiotopoulos에게 문의하십시오.
References 1 S.R. Turns, An Introduction to Combustion, Concepts and Applications, McGraw-Hill, New York, 1996 Google Scholar 2 V. Johansen, T.V. Kouznetsova, Clinker formation and new processes, Presented at the Ninth International Congress on the Chemistry of Cement, India, 1992; also RAMBOLL Bulletin No. 42, 1993 Google Scholar 3 Basel Convention, UNEP Document No. 93-7758, 1993 Google Scholar 4 N.C Markatos Mathematical modelling of single and two-phase flow problems in the process industries Revue de l’Institut Français du Pétrole, 48 (1993), p. 631 View PDFCrossRefView Record in ScopusGoogle Scholar 5 T. Avgeropoulos, J.P. Glekas, C. Papadopoulos, Numerical simulation of the combustion aerodynamics inside a rotary cement kiln, in: Pilavachi (Ed.), Energy Efficiency in Process Technology, Elsevier, London, 1993, p. 767 Google Scholar 6 F.C. Lockwood, B. Shen, T. Lowes, Numerical study of petroleum coke fired cement kiln flames, Presented at the Third International Conference on Combustion Technologies for a Clean Environment, Lisbon, 1995 Google Scholar 7 F.C. Lockwood, B. Shen, Performance predictions of pulverised-coal flames of power station furnace and cement kiln types, Twenty-Fifth Symposium International on Combustion, The Combustion Institute, 1994 p. 503 Google Scholar 8 P.V Barr, J.K Brimacombe, A.P Watkinson A heat-transfer model for the rotary kiln: Part II, development of the cross-section model Metallurgical Transactions B, 20B (1989), p. 403 View Record in ScopusGoogle Scholar 9 V Frisch, R Jeschar Possibilities for optimizing the burning process in rotary cement kilns Zement-Kalk-Gips, 36 (1983), p. 549 View Record in ScopusGoogle Scholar 10 A.A Boateng, P.V Barr A thermal model for the rotary kiln including heat transfer within the bed Int. J. Heat Mass Transfer, 39 (1996), p. 2131 ArticleDownload PDFView Record in ScopusGoogle Scholar 11 M.G. Carvahlo, T. Farias, A. Martius, A three-dimensional modelling of the radiative heat transfer in a cement kiln, in: Carvahlo et al. (Eds.), Combustion Technologies for a Clean Environment, Gordon and Breach, London, 1995, p. 146 Google Scholar 12 H.A Spang A dynamic model of a cement kiln Automatica, 8 (1972), p. 309 ArticleDownload PDFView Record in ScopusGoogle Scholar 13 CFDS, FLOW-3D Users Manual, AEA Harwell, UK Google Scholar 14 E Mastorakos, J.J McGuirk, A.M.K.P Taylor The origin of turbulence acquired by heavy particles in a round, turbulent jet Part. Part. Syst. Charact., 7 (1990), p. 203 View PDFCrossRefView Record in ScopusGoogle Scholar 15 P. Panagiotopoulos, TITAN S.A. Cement Company, Personal communication, 1996 Google Scholar 16 M.S Murthy, B.R Harish, K.S Rajanandam, K.Y Ajoy Pavan Kumar Investigation on the kinetics of thermal decomposition of calcium carbonate Chem. Eng. Sci., 49 (1996), p. 2198 Google Scholar 17 V. Johansen, Cement production and chemistry, Presented at the Symposium on Cement Manufacturing and Chemistry, Anaheim, November 1989; also RAMBOLL Bulletin No. 41, 1993 Google Scholar 1 Also at Department of Mechanical Engineering, University of Patras, Greece.
2 Also at Department of Chemical Engineering, University of Patras, Greece.
곡면에서 GMAW 기반 적층 가공의 용접 성형 특성은 중력의 영향을 크게 받습니다. 성형면의 경사각이 크면 혹 비드(hump bead)와 같은 심각한 결함이 발생합니다.
본 논문에서는 양생면에서 용접 비드 형성의 형성 특성과 제어 방법을 연구하기 위해 용접 용융 풀 유동 역학의 전산 모델을 수립하고 제안된 모델을 검증하기 위해 증착 실험을 수행하였습니다.
결과는 용접 비드 경사각(α)이 증가함에 따라 역류의 속도가 증가하고 상향 용접의 경우 α > 60°일 때 불규칙한 험프 결함이 나타나는 것으로 나타났습니다.
상부 과잉 액체의 하향 압착력과 하부 상향 유동의 반동력과 표면장력 사이의 상호작용은 용접 혹 형성의 주요 요인이었다. 하향 용접의 경우 양호한 형태를 얻을 수 있었으며, 용접 비드 경사각이 증가함에 따라 용접 높이는 감소하고 용접 폭은 증가하였습니다.
하향 및 상향 용접을 위한 곡면의 용융 거동 및 성형 특성을 기반으로 험프 결함을 제어하기 위해 위브 용접을 통한 증착 방법을 제안하였습니다.
성형 궤적의 변화로 인해 용접 방향의 중력 성분이 크게 감소하여 용융 풀 흐름의 안정성이 향상되었으며 복잡한 표면에서 안정적이고 일관된 용접 비드를 얻는 데 유리했습니다.
하향 용접과 상향 용접 사이의 단일 비드의 치수 편차는 7% 이내였으며 하향 및 상향 혼합 혼합 비드 중첩 증착에서 비드의 변동 편차는 0.45로 GMAW 기반 적층 제조 공정에서 허용될 수 있었습니다.
이러한 발견은 GMAW를 기반으로 하는 곡선 적층 적층 제조의 용접 비드 형성 제어에 기여했습니다.
The weld forming characteristics of GMAW-based additive manufacturing on curved surface are dramatically influenced by gravity. Large inclined angle of the forming surface would lead to severe defects such as hump bead. In this paper, a computational model of welding molten pool flow dynamics was established to research the forming characteristic and control method of weld bead forming on cured surface, and deposition experiments were conducted to verify the proposed model. Results indicated that the velocity of backward flows increased with the increase of weld bead tilt angle (α) and irregular hump defects appeared when α > 60° for upward welding. The interaction between the downward squeezing force of the excess liquid at the top and the recoil force of the upward flow at the bottom and the surface tension were primary factors for welding hump formation. For downward welding, a good morphology shape could be obtained, and the weld height decreased and the weld width increased with the increase of weld bead tilt angle. Based on the molten behaviors and forming characteristics on curved surface for downward and upward welding, the method of deposition with weave welding was proposed to control hump defects. Gravity component in the welding direction was significantly reduced due to the change of forming trajectory, which improved the stability of the molten pool flow and was beneficial to obtain stable and consistent weld bead on complex surface. The dimensional deviations of the single bead between downward and upward welding were within 7% and the fluctuation deviation of the bead in multi-bead overlapping deposition with mixing downward and upward welding was 0.45, which could be acceptable in GMAW-based additive manufacturing process. These findings contributed to the weld bead forming control of curve layered additive manufacturing based on GMAW.
7.Xie FB, Chen LF, Li ZY, Tang K (2020) Path smoothing and feed rate planning for robotic curved layer additive manufacturing. Robot Comput Integr Manuf 65. https://doi.org/10.1016/j.rcim.2020.101967
8.Ding YY, Dwivedi R, Kovacevic R (2017) Process planning for 8-axis robotized laser-based direct metal deposition system: a case on building revolved part. Robot Comput Integr Manuf 44:67–76. https://doi.org/10.1016/j.rcim.2016.08.008ArticleGoogle Scholar
12.Yuan L, Pan ZX, Ding DH, He FY, Duin SV, Li HJ, Li WH (2020) Investigation of humping phenomenon for the multi-directional robotic wire and arc additive manufacturing. Robot Comput Integr Manuf 63. https://doi.org/10.1016/j.rcim.2019.101916
13.Nguyen MC, Medale M, Asserin O, Gounand S, Gilles P (2017) Sensitivity to welding positions and parameters in GTA welding with a 3D multiphysics numerical model. Numer Heat Transf Part A Appl 71:233–249. https://doi.org/10.1080/10407782.2016.1264747ArticleGoogle Scholar
17.Philip Y, Xu ZY, Wang Y, Wang R, Ye X (2019) Investigation of humping defect formation in a lap joint at a high-speed hybrid laser-GMA welding. Results Phys 13. https://doi.org/10.1016/j.rinp.2019.102341
18.Hu ZQ, Qin XP, Shao T, Liu HM (2018) Understanding and overcoming of abnormity at start and end of the weld bead in additive manufacturing with GMAW. Int J Adv Manuf Technol 95:2357–2368. https://doi.org/10.1007/s00170-017-1392-9ArticleGoogle Scholar
19.Tang SY, Wang GL, Huang C, Li RS, Zhou SY, Zhang HO (2020) Investigation, modeling and optimization of abnormal areas of weld beads in wire and arc additive manufacturing. Rapid Prototyp J 26:1183–1195. https://doi.org/10.1108/RPJ-08-2019-0229ArticleGoogle Scholar
20.Bai X, Colegrove P, Ding J, Zhou XM, Diao CL, Bridgeman P, Honnige JR, Zhang HO, Williams S (2018) Numerical analysis of heat transfer and fluid flow in multilayer deposition of PAW-based wire and arc additive manufacturing. Int J Heat Mass Transf 124:504–516. https://doi.org/10.1016/j.ijheatmasstransfer.2018.03.085ArticleGoogle Scholar
21.Siewert E, Schein J, Forster G (2013) Determination of enthalpy, temperature, surface tension and geometry of the material transfer in PGMAW for the system argon-iron. J Phys D Appl Phys 46. https://doi.org/10.1088/0022-3727/46/22/224008
23.Fachinotti VD, Cardona A (2008) Semi-analytical solution of the thermal field induced by a moving double-ellipsoidal welding heat source in a semi-infinite body. Mec Comput XXVII:1519–1530
24.Nguyen NT, Mai YW, Simpson S, Ohta A (2004) Analytical approximate solution for double ellipsoidal heat source in finite thick plate. Weld J 83:82–93Google Scholar
25.Goldak J, Chakravarti A, Bibby M (1985) A double ellipsoid finite element model for welding heat sources. IIW Doc. No. 212-603-85
29.Zhan XH, Zhang D, Liu XB, Chen J, Wei YH, Liu RP (2017) Comparison between weave bead welding and multi-layer multi-pass welding for thick plate Invar steel. Int J Adv Manuf Technol 88:2211–2225. https://doi.org/10.1007/s00170-016-8926-4ArticleGoogle Scholar
30.Xu GX, Li L, Wang JY, Zhu J, Li PF (2018) Study of weld formation in swing arc narrow gap vertical GMA welding by numerical modeling and experiment. Int J Adv Manuf Technol 96:1905–1917. https://doi.org/10.1007/s00170-018-1729-zArticleGoogle Scholar
Keyhole Formation by Laser Drilling in Laser Powder Bed Fusion of Ti6Al4V Biomedical Alloy: Mesoscopic Computational Fluid Dynamics Simulation versus Mathematical Modelling Using Empirical Validation
Asif Ur Rehman 1,2,3,* ,† , Muhammad Arif Mahmood 4,* ,† , Fatih Pitir 1 , Metin Uymaz Salamci 2,3 , Andrei C. Popescu 4 and Ion N. Mihailescu 4
Abstract
LPBF(Laser Powder Bed fusion) 공정에서 작동 조건은 열 분포를 기반으로 레이저 유도 키홀 영역을 결정하는 데 필수적입니다. 얕은 구멍과 깊은 구멍으로 분류되는 이러한 영역은 LPBF 프로세스에서 확률과 결함 형성 강도를 제어합니다.
LPBF 프로세스의 핵심 구멍을 연구하고 제어하기 위해 수학적 및 CFD(전산 유체 역학) 모델이 제공됩니다. CFD의 경우 이산 요소 모델링 기법을 사용한 유체 체적 방법이 사용되었으며, 분말 베드 보이드 및 표면에 의한 레이저 빔 흡수를 포함하여 수학적 모델이 개발되었습니다.
동적 용융 풀 거동을 자세히 살펴봅니다. 실험적, CFD 시뮬레이션 및 분석적 컴퓨팅 결과 간에 정량적 비교가 수행되어 좋은 일치를 얻습니다.
LPBF에서 레이저 조사 영역 주변의 온도는 높은 내열성과 분말 입자 사이의 공기로 인해 분말층 주변에 비해 급격히 상승하여 레이저 횡방향 열파의 이동이 느려집니다. LPBF에서 키홀은 에너지 밀도에 의해 제어되는 얕고 깊은 키홀 모드로 분류될 수 있습니다. 에너지 밀도를 높이면 얕은 키홀 구멍 모드가 깊은 키홀 구멍 모드로 바뀝니다.
깊은 키홀 구멍의 에너지 밀도는 다중 반사와 키홀 구멍 내의 2차 반사 빔의 집중으로 인해 더 높아져 재료가 빠르게 기화됩니다.
깊은 키홀 구멍 모드에서는 온도 분포가 높기 때문에 액체 재료가 기화 온도에 가까우므로 얕은 키홀 구멍보다 구멍이 형성될 확률이 훨씬 높습니다. 온도가 급격히 상승하면 재료 밀도가 급격히 떨어지므로 비열과 융해 잠열로 인해 유체 부피가 증가합니다.
그 대가로 표면 장력을 낮추고 용융 풀 균일성에 영향을 미칩니다.
In the laser powder bed fusion (LPBF) process, the operating conditions are essential in determining laser-induced keyhole regimes based on the thermal distribution. These regimes, classified into shallow and deep keyholes, control the probability and defects formation intensity in the LPBF process. To study and control the keyhole in the LPBF process, mathematical and computational fluid dynamics (CFD) models are presented. For CFD, the volume of fluid method with the discrete element modeling technique was used, while a mathematical model was developed by including the laser beam absorption by the powder bed voids and surface. The dynamic melt pool behavior is explored in detail. Quantitative comparisons are made among experimental, CFD simulation and analytical computing results leading to a good correspondence. In LPBF, the temperature around the laser irradiation zone rises rapidly compared to the surroundings in the powder layer due to the high thermal resistance and the air between the powder particles, resulting in a slow travel of laser transverse heat waves. In LPBF, the keyhole can be classified into shallow and deep keyhole mode, controlled by the energy density. Increasing the energy density, the shallow keyhole mode transforms into the deep keyhole mode. The energy density in a deep keyhole is higher due to the multiple reflections and concentrations of secondary reflected beams within the keyhole, causing the material to vaporize quickly. Due to an elevated temperature distribution in deep keyhole mode, the probability of pores forming is much higher than in a shallow keyhole as the liquid material is close to the vaporization temperature. When the temperature increases rapidly, the material density drops quickly, thus, raising the fluid volume due to the specific heat and fusion latent heat. In return, this lowers the surface tension and affects the melt pool uniformity.
Keywords: laser powder bed fusion; computational fluid dynamics; analytical modelling; shallow and deep keyhole modes; experimental correlation
Figure 1. Powder bed schematic with voids.Figure 2. (a) Scanning electron microscopy images of Ti6Al4V powder particles and (b) simulated powder bed using
discrete element modellingFigure 3. Temperature field contour formation at various time intervals (a) 0.695 ms, (b) 0.795 ms,
(c) 0.995 ms and (d) 1.3 ms.Figure 4. Detailed view of shallow depth melt mode with temperature field at 0.695 msFigure 5. Melt flow stream traces formation at various time intervals (a) 0.695 ms, (b) 0.795 ms,
(c) 0.995 ms and (d) 1.3 msFigure 6. Density evolution of the melt pool at various time intervals (a) 0.695 ms, (b) 0.795 ms,
(c) 0.995 ms and (d) 1.3 ms.Figure 7. Un-melted and melted regions at different time intervals (a) 0.695 ms, (b) 0.795 ms,
(c) 0.995 ms and (d) 1.3 msFigure 8. Transformation from shallow depth melt flow to deep keyhole formation when laser power increased from
(a) 170 W to (b) 200 WFigure 9. Stream traces and laser beam multiple reflections in deep keyhole melt flow modeFigure 10. A comparison between analytical and CFD simulation results for peak thermal distribution value in the deep
keyhole formationFigure 11. A comparison among experiments [49], CFD and analytical simulations for deep keyhole
top width and bottom width
References
Kok, Y.; Tan, X.P.; Wang, P.; Nai, M.L.S.; Loh, N.H.; Liu, E.; Tor, S.B. Anisotropy and heterogeneity of microstructure and mechanical properties in metal additive manufacturing: A critical review. Mater. Des. 2018, 139, 565–586. [CrossRef]
Ansari, P.; Salamci, M.U. On the selective laser melting based additive manufacturing of AlSi10Mg: The process parameter investigation through multiphysics simulation and experimental validation. J. Alloys Compd. 2022, 890, 161873. [CrossRef]
Guo, N.; Leu, M.C. Additive manufacturing: Technology, applications and research needs. Front. Mech. Eng. 2013, 8, 215–243. [CrossRef]
Mohsin Raza, M.; Lo, Y.L. Experimental investigation into microstructure, mechanical properties, and cracking mechanism of IN713LC processed by laser powder bed fusion. Mater. Sci. Eng. A 2021, 819, 141527. [CrossRef]
Dezfoli, A.R.A.; Lo, Y.L.; Raza, M.M. Prediction of Epitaxial Grain Growth in Single-Track Laser Melting of IN718 Using Integrated Finite Element and Cellular Automaton Approach. Materials 2021, 14, 5202. [CrossRef]
Tiwari, S.K.; Pande, S.; Agrawal, S.; Bobade, S.M. Selection of selective laser sintering materials for different applications. Rapid Prototyp. J. 2015, 21, 630–648. [CrossRef]
Liu, F.H. Synthesis of bioceramic scaffolds for bone tissue engineering by rapid prototyping technique. J. Sol-Gel Sci. Technol. 2012, 64, 704–710. [CrossRef]
Ur Rehman, A.; Sglavo, V.M. 3D printing of geopolymer-based concrete for building applications. Rapid Prototyp. J. 2020, 26, 1783–1788. [CrossRef]
Ur Rehman, A.; Sglavo, V.M. 3D printing of Portland cement-containing bodies. Rapid Prototyp. J. 2021. ahead of print. [CrossRef]
Popovich, A.; Sufiiarov, V. Metal Powder Additive Manufacturing. In New Trends in 3D Printing; InTech: Rijeka, Croatia, 2016.
Jia, T.; Zhang, Y.; Chen, J.K.; He, Y.L. Dynamic simulation of granular packing of fine cohesive particles with different size distributions. Powder Technol. 2012, 218, 76–85. [CrossRef]
Ansari, P.; Ur Rehman, A.; Pitir, F.; Veziroglu, S.; Mishra, Y.K.; Aktas, O.C.; Salamci, M.U. Selective Laser Melting of 316L Austenitic Stainless Steel: Detailed Process Understanding Using Multiphysics Simulation and Experimentation. Metals 2021, 11, 1076. [CrossRef]
Ur Rehman, A.; Tingting, L.; Liao, W. 4D Printing; Printing Ceramics from Metals with Selective Oxidation. Patent No. W0/2019/052128, 21 March 2019.
Ullah, A.; Wu, H.; Ur Rehman, A.; Zhu, Y.; Liu, T.; Zhang, K. Influence of laser parameters and Ti content on the surface morphology of L-PBF fabricated Titania. Rapid Prototyp. J. 2021, 27, 71–80. [CrossRef]
Ur Rehman, A. Additive Manufacturing of Ceramic Materials and Combinations with New Laser Strategies. Master’s Thesis, Nanjing University of Science and Technology, Nanjing, China, 2017.
Wong, K.V.; Hernandez, A. A Review of Additive Manufacturing. ISRN Mech. Eng. 2012, 2012, 1–10. [CrossRef]
Körner, C. Additive manufacturing of metallic components by selective electron beam melting—A review. Int. Mater. Rev. 2016, 61, 361–377. [CrossRef]
Fayazfar, H.; Salarian, M.; Rogalsky, A.; Sarker, D.; Russo, P.; Paserin, V.; Toyserkani, E. A critical review of powder-based additive manufacturing of ferrous alloys: Process parameters, microstructure and mechanical properties. Mater. Des. 2018, 144, 98–128. [CrossRef]
Everton, S.K.; Hirsch, M.; Stavroulakis, P.I.; Leach, R.K.; Clare, A.T. Review of in-situ process monitoring and in-situ metrology for metal additive manufacturing. Mater. Des. 2016, 95, 431–445. [CrossRef]
Sing, S.L.; An, J.; Yeong, W.Y.; Wiria, F.E. Laser and electron-beam powder-bed additive manufacturing of metallic implants: A review on processes, materials and designs. J. Orthop. Res. 2016, 34, 369–385. [CrossRef] [PubMed]
Olakanmi, E.O.; Cochrane, R.F.; Dalgarno, K.W. A review on selective laser sintering/melting (SLS/SLM) of aluminium alloy powders: Processing, microstructure, and properties. Prog. Mater. Sci. 2015, 74, 401–477. [CrossRef]
Mahmood, M.A.; Popescu, A.C.; Hapenciuc, C.L.; Ristoscu, C.; Visan, A.I.; Oane, M.; Mihailescu, I.N. Estimation of clad geometry and corresponding residual stress distribution in laser melting deposition: Analytical modeling and experimental correlations. Int. J. Adv. Manuf. Technol. 2020, 111, 77–91. [CrossRef]
Mahmood, M.A.; Popescu, A.C.; Oane, M.; Ristoscu, C.; Chioibasu, D.; Mihai, S.; Mihailescu, I.N. Three-jet powder flow and laser–powder interaction in laser melting deposition: Modelling versus experimental correlations. Metals 2020, 10, 1113. [CrossRef]
King, W.; Anderson, A.T.; Ferencz, R.M.; Hodge, N.E.; Kamath, C.; Khairallah, S.A. Overview of modelling and simulation of metal powder bed fusion process at Lawrence Livermore National Laboratory. Mater. Sci. Technol. 2015, 31, 957–968. [CrossRef]
Gong, H.; Rafi, K.; Gu, H.; Starr, T.; Stucker, B. Analysis of defect generation in Ti-6Al-4V parts made using powder bed fusion additive manufacturing processes. Addit. Manuf. 2014, 1, 87–98. [CrossRef]
Frazier, W.E. Metal additive manufacturing: A review. J. Mater. Eng. Perform. 2014, 23, 1917–1928. [CrossRef]
Panwisawas, C.; Qiu, C.L.; Sovani, Y.; Brooks, J.W.; Attallah, M.M.; Basoalto, H.C. On the role of thermal fluid dynamics into the evolution of porosity during selective laser melting. Scr. Mater. 2015, 105, 14–17. [CrossRef]
Qian, Y.; Yan, W.; Lin, F. Parametric study and surface morphology analysis of electron beam selective melting. Rapid Prototyp. J. 2018, 24, 1586–1598. [CrossRef]
Panwisawas, C.; Perumal, B.; Ward, R.M.; Turner, N.; Turner, R.P.; Brooks, J.W.; Basoalto, H.C. Keyhole formation and thermal fluid flow-induced porosity during laser fusion welding in titanium alloys: Experimental and modelling. Acta Mater. 2017, 126, 251–263. [CrossRef]
Panwisawas, C.; Sovani, Y.; Turner, R.P.; Brooks, J.W.; Basoalto, H.C.; Choquet, I. Modelling of thermal fluid dynamics for fusion welding. J. Mater. Process. Technol. 2018, 252, 176–182. [CrossRef]
Martin, A.A.; Calta, N.P.; Hammons, J.A.; Khairallah, S.A.; Nielsen, M.H.; Shuttlesworth, R.M.; Sinclair, N.; Matthews, M.J.; Jeffries, J.R.; Willey, T.M.; et al. Ultrafast dynamics of laser-metal interactions in additive manufacturing alloys captured by in situ X-ray imaging. Mater. Today Adv. 2019, 1, 100002. [CrossRef]
Cunningham, R.; Zhao, C.; Parab, N.; Kantzos, C.; Pauza, J.; Fezzaa, K.; Sun, T.; Rollett, A.D. Keyhole threshold and morphology in laser melting revealed by ultrahigh-speed x-ray imaging. Science 2019, 363, 849–852. [CrossRef] [PubMed]
Tang, C.; Tan, J.L.; Wong, C.H. A numerical investigation on the physical mechanisms of single track defects in selective laser melting. Int. J. Heat Mass Transf. 2018, 126, 957–968. [CrossRef]
Mirkoohi, E.; Ning, J.; Bocchini, P.; Fergani, O.; Chiang, K.-N.; Liang, S. Thermal Modeling of Temperature Distribution in Metal Additive Manufacturing Considering Effects of Build Layers, Latent Heat, and Temperature-Sensitivity of Material Properties. J. Manuf. Mater. Process. 2018, 2, 63. [CrossRef]
Oane, M.; Sporea, D. Temperature profiles modeling in IR optical components during high power laser irradiation. Infrared Phys. Technol. 2001, 42, 31–40. [CrossRef]
Cleary, P.W.; Sawley, M.L. DEM modelling of industrial granular flows: 3D case studies and the effect of particle shape on hopper discharge. Appl. Math. Model. 2002, 26, 89–111. [CrossRef]
Parteli, E.J.R.; Pöschel, T. Particle-based simulation of powder application in additive manufacturing. Powder Technol. 2016, 288, 96–102. [CrossRef]
Cao, L. Numerical simulation of the impact of laying powder on selective laser melting single-pass formation. Int. J. Heat Mass Transf. 2019, 141, 1036–1048. [CrossRef]
Tian, Y.; Yang, L.; Zhao, D.; Huang, Y.; Pan, J. Numerical analysis of powder bed generation and single track forming for selective laser melting of SS316L stainless steel. J. Manuf. Process. 2020, 58, 964–974. [CrossRef]
Lee, Y.S.; Zhang, W. Modeling of heat transfer, fluid flow and solidification microstructure of nickel-base superalloy fabricated by laser powder bed fusion. Addit. Manuf. 2016, 12, 178–188. [CrossRef]
Tang, M.; Pistorius, P.C.; Beuth, J.L. Prediction of lack-of-fusion porosity for powder bed fusion. Addit. Manuf. 2017, 14, 39–48. [CrossRef]
Promoppatum, P.; Yao, S.C.; Pistorius, P.C.; Rollett, A.D. A Comprehensive Comparison of the Analytical and Numerical Prediction of the Thermal History and Solidification Microstructure of Inconel 718 Products Made by Laser Powder-Bed Fusion. Engineering 2017, 3, 685–694. [CrossRef]
Rosenthal, D. Mathematical Theory of Heat Distribution During Welding and Cutting. Weld. J. 1941, 20, 220–234.
Chen, Q.; Zhao, Y.Y.; Strayer, S.; Zhao, Y.Y.; Aoyagi, K.; Koizumi, Y.; Chiba, A.; Xiong, W.; To, A.C. Elucidating the Effect of Preheating Temperature on Melt Pool Morphology Variation in Inconel 718 Laser Powder Bed Fusion via Simulation and Experiment. Available online: https://www.sciencedirect.com/science/article/pii/S2214860420310149#bb8 (accessed on 30 April 2021).
Ur Rehman, A.; Pitir, F.; Salamci, M.U. Laser Powder Bed Fusion (LPBF) of In718 and the Impact of Pre-Heating at 500 and 1000 ◦C: Operando Study. Materials 2021, 14, 6683. [CrossRef] [PubMed]
Ur Rehman, A.; Pitir, F.; Salamci, M.U. Full-Field Mapping and Flow Quantification of Melt Pool Dynamics in Laser Powder Bed Fusion of SS316L. Materials 2021, 14, 6264. [CrossRef] [PubMed]
Gong, H.; Gu, H.; Zeng, K.; Dilip, J.J.S.; Pal, D.; Stucker, B.; Christiansen, D.; Beuth, J.; Lewandowski, J.J. Melt Pool Characterization for Selective Laser Melting of Ti-6Al-4V Pre-alloyed Powder. In Proceedings of the International Solid Freeform Fabrication Symposium, Austin, TX, USA, 10–12 August 2014; 2014; pp. 256–267.
Song, B.; Dong, S.; Liao, H.; Coddet, C. Process parameter selection for selective laser melting of Ti6Al4V based on temperature distribution simulation and experimental sintering. Int. J. Adv. Manuf. Technol. 2012, 61, 967–974. [CrossRef]
선택적 레이저 용융 동안 CP-Ti의 열 거동을 연구하기 위해 무작위 패킹 분말 베드 모델을 기반으로 하는 메조스코프 시뮬레이션이 설정되었습니다. 레이저와 분말의 상호 작용에 따른 용융 풀의 특성과 레이저 출력이 용융 풀의 열 거동, 유체 역학 및 표면 형태 변화에 미치는 영향을 연구했습니다.
결과는 레이저 출력이 증가함에 따라 최대 온도, 온도 변화율, 용융 풀의 수명 및 크기가 크게 향상되었음을 보여줍니다. 또한 본 연구에서는 이중궤도 하의 용융지의 특성과 열거동을 주로 연구하였다.
두 번째 트랙의 용융 풀의 최대 온도, 수명 및 길이와 너비는 첫 번째 트랙보다 더 높고 레이저 출력이 증가함에 따라 용융 풀에서 두 번째 트랙의 길이 너비 비율이 증가함을 알 수 있습니다. 더 커집니다.
A mesoscopic simulation based on random packing powder bed model was established to study the heat behavior of CP-Ti during selective laser melting. The characteristics of the molten pool under the interaction of laser and powder, and the influence of laser power on the thermal behavior, hydrodynamics and surface morphology evolution of the molten pool were studied. The results show that with the increase of laser power, the maximum temperature, temperature change rate, lifetime of molten pool and size are greatly improved. In addition, the characteristics and heat behavior of the molten pool under the double track are mainly studied in this study. It is found that the maximum temperature, lifetime, and the length and width of the molten pool of the second track are higher than those in the first, and with the increase of laser power, the length width ratio of the second track in molten pool becomes larger.
Multiphysics Modeling of Thermal Behavior of Commercial Pure Titanium Powder During Selective Laser Melting
Keywords
Additive manufacturing
Selective laser melting
Numerical simulation
Thermal behavior
References
1.D.K. Pattanayak, A. Fukuda, T. Matsushita, M. Takemoto, S. Fujibayashi, K. Sasaki, N. Nishida, T. Nakamura, T. Kokubo, Acta Biomater. 7, 1398 (2011)CASArticleGoogle Scholar
2.A.K. Patnaik, N. Poondla, C.C. Menzemer, T.S. Srivatsan, Mater. Sci. Eng. A 590, 390 (2014)CASArticleGoogle Scholar
3.D.D. Gu, Y.C. Hagedorn, W. Meiners, G.B. Meng, R.J.S. Batista, K. Wissenbach, R. Poprawe, Acta Mater. 60, 3849 (2012)CASArticleGoogle Scholar
4.H. Attar, M. Calin, L.C. Zhang, S. Scudino, J. Eckert, Mater. Sci. Eng. A 593, 170 (2014)CASArticleGoogle Scholar
5.J. Shen, B. Chen, J. Umeda, K. Kondoh, Mater. Sci. Eng. A 716, 1 (2018)CASArticleGoogle Scholar
6.E. Santos, K. Osakada, M. Shiomi, M. Morita, F. Abe, Fabrication of titanium dental implants by selective laser melting. in Proceedings of the 5th International Symposium on Laser Precision Microfabrication, Nara, 11–14 May 2004
26.Y.S. Lee, W. Zhang, Modeling of heat transfer, Addit. Manuf. 12, 178 (2016)CASGoogle Scholar
27.I. Kovaleva, O. Kovalev, I. Smurov, Phys. Procedia 56, 400 (2014)ArticleGoogle Scholar
28.Y.S. Lee, W. Zhang, Mesoscopic simulation of heat transfer and fluid flow in laser powder bed additive manufacturing. in Proceedings of 26th Solid Freeform Fabrication Symposium, Austin, 10-12 August 2015
29.W. Yan, W. Ge, Y. Qian, S. Lin, B. Zhou, W.K. Liu, Acta Mater. 134, 324 (2017)CASArticleGoogle Scholar
30.I. Yadroitsev, A. Gusarov, I. Yadroitsava, I. Smurov, J. Mater. Process. Tech. 210, 1624 (2010)CASArticleGoogle Scholar
43.H. Sakaguchi, E. Ozaki, T. Igarashi, Int. J. Mod. Phys. B 7, 1949 (1993)
44.Flow3D: Version 11 0.1.2: User Manual, Flow Science, Santa Fe, NM, USA, (2014)
45.S. Kolossov, E. Boillat, R. Glardon, P. Fischer, M. Locher, Int. J. Mach. Tool. Manu. 44, 117 (2004)ArticleGoogle Scholar
46.V.R. Voller, A.D. Brent, C. Prakash, Int. J. Heat Mass Tran. 32, 1719 (1989)CASArticleGoogle Scholar
47.Y.-C. Wu, C.-H. San, C.-H. Chang, H.-J. Lin, R. Marwan, S. Baba, W.-S. Hwang, J. Mater. Process. Tech. 254, 72 (2018)ArticleGoogle Scholar
48.B. Cheng, X. Li, C. Tuile, A. Ilin, H. Willeck, U. Hartel, Multi-physics modeling of single-track scanning in selective laser melting: powder compaction effect. in Proceedings of 29th Annual International Solid Freeform Fabrication Symposium-An Additive Manufacturing Conference, Austin, 13–15 August 2018
49.B. Liu, G. Fang, L. Lei, W. Liu, Appl. Math. Model. 79, 506 (2020)ArticleGoogle Scholar
50.S. Lee, J. Kim, J. Choe, S.-W. Kim, J.-K. Hong, Y.S. Choi, Met. Mater. Int. 27, 78 (2021)
WU Jingxia1 , ZHANG Chunjin2,3 (1. Xi’an Water Conservancy Survey Design Institute, Xi’an 710054, Shaanxi, China; 2. Key Laboratory of Yellow River Sediment Research, M. W. R. , Yellow River Institute of Hydraulic Research, Zhengzhou 450003, Henan, China; 3. State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, Jiangsu, China)
수치 시뮬레이션을 통해 오지 여수로 터널의 수리적 특성 연구의 타당성을 탐색하기 위해 황하 Xiaolangdi 수질 관리 프로젝트의 2번 오지 여수로 터널을 연구 대상으로 취한 다음 오지의 수리 특성 설계 및 점검 홍수 수준 조건에서 여수로 터널은 RNG k-ε 난류 모델을 사용하여 배출 용량, 터널 크라운 잔류 공간, 단면 유속, 압전 수두, 유동 캐비테이션 수, 제트 흐름 범위 및 1 ∶ 40의 일반 수리 모델과 결합된 세굴 구덩이 깊이, 시뮬레이션 값과 실험 값 모두 비교됩니다.
연구결과 모의실험값이 실험값과 일치하여 오지 여수로터널의 수리적 특성을 수치모사를 통해 탐색할 수 있음을 확인하였다. 여수로터널 내부의 흐름은 안정적이고 터널 크라운 잔류 공간은 개방 흐름과 완전 흐름의 교대 흐름 패턴이 없는 25% 이상입니다.
체크 홍수 수위에서 시뮬레이션 값과 유량 계수의 실험 값은 모두 설계에서보다 높으므로 배출 용량은 홍수 제어 관련 설계 요구 사항을 충족할 수 있습니다. 오지 단면과 플립 단면의 유동 캐비테이션 수는 캐비테이션 손상이 발생할 가능성이 작기 때문에 캐비테이션 침식을 줄이기 위한 적절한 적절한 조치가 채택될 필요가 있습니다.
유압 모델의 고르지 않은 표면에 부압이 발생하면 표면 구조에 관련주의를 기울일 필요가 있습니다. 연구 결과는 여수로 터널의 설계 및 건설에 대한 관련 참고 및 이론적 근거를 제공할 수 있습니다.
Keywords
Xiaolangdi Water Control Project; ogee spillway tunnel; simulative calculation; hydraulic characteristics; turbulent model
Fig. 1 Layout of spillway tunnelFig. 4 Hydraulic modelingFig. 6 Sectional surface profile distributionsFig. 7 Comparison between simulated results and experimental
results for flow velocity of section-cross
参考文献(References)
[1] 谢省宗, 吴一红, 陈文学. 我国高坝泄洪消能新技术的研究和创 新[J]. 水利学报, 2016, 47(3): 324-336. XIE Shengzong, WU Yihong, CHEN Wenxue. New technology and innovation on flood discharge and energy dissipation of high dams in China [J]. Journal of Hydraulic Engineering, 2016, 47( 3): 324- 336. [2] 刘嘉夫, 齐昕. 龙抬头水电站泄洪洞水力特性研究[ J]. 水利水 电技术, 2019, 50(2): 139-143. LIU Jiafu, QI Xin. Study on hydraulic characteristics of ogee spillway tunnel of hydropower station [ J]. Water Resources and Hydropower Engineering, 2019, 50(2): 139-143. [3] 范灵, 张宏伟, 刘之平, 等. 明流泄洪洞布置形式对水力特性影 响的数值研究[J]. 水力发电学报, 2009, 28(3): 126-131. FAN Ling, ZHANG Hongwei, LIU Zhiping, et al. Numerical study on hydraulic characteristic of free surface flow in spillway tunnel with different configuration [ J ]. Journal of Hydroelectric Engineering, 2009, 28(3): 126-131. [4] 张春晋, 李永业, 孙西欢. 明流泄洪洞水力特性的二维数值模拟 与试验研究[J]. 长江科学院院报, 2016, 33(1): 54-60. ZHANG Chunjin, LI Yongye, SUN Xihuan. Two-dimensional numerical simulation and experimental research of hydraulic characteristics in spillway tunnel with free water surface [ J]. Journal of Yangtze River Scientific Research Institute, 2016, 33(1): 54-60. [5] 徐国宾, 章环境, 刘昉, 等. 龙抬头泄洪洞水力特性的数值模拟 [J]. 长江科学院院报, 2015, 32(1): 84-87. XU Guobin, ZHANG Huanjing, LIU Fang, et al. Numerical simulation on hydraulic characteristic of high head ogee spillway tunnel [J]. Journal of Yangtze River Scientific Research Institute, 2015, 32(1): 84-87. [6] 陈瑞华, 杨吉健, 马麟, 等. 小湾水电站泄洪洞洞身数值模拟 [J]. 排灌机械工程学报, 2017, 35(6): 488-494. CHEN Ruihua, YANG Jijian, MA Lin, et al. Numerical simulation of tunnel of Xiaowan Hydropower Station [ J]. Journal of Drainage and Irrigation Machinery Engineering, 2017, 35(6): 488-494. [7] 翟保林, 刘亚坤. 高水头明流泄洪洞三维数值模拟[ J]. 水利与 建筑工程学报, 2017, 15(3): 31-34. ZHAI Baolin, LIU Yakun. 3-D Numerical simulation of high water head spillway tunnel with free surface [ J ]. Journal of Water Resources and Architectural Engineering, 2017, 15(3): 31-34. [8] 姜 攀, 尹进步, 何武全, 等. 有压泄洪洞弯道压力特性数值模拟 与试验研究[J]. 水力发电, 2016, 42(2): 49-53. JIANG Pan, YIN Jinbu, HE Wuquan, et al. Numerical simulation and experimental research on pressure characteristic of curved section of pressure spillway tunnel [J]. Water Power, 2016, 42(2): 49-53. [9] 邓 军, 许唯临, 雷军, 等. 高水头岸边泄洪洞水力特性的数值模 拟[J]. 水利学报, 2005(10): 1209-1212. DENG Jun, XU Weilin, LEI Jun, et al. Numerical simulation of hydraulic characteristics of high head spillway tunnel [J]. Journal of Hydraulic Engineering, 2005(10): 1209-1212. [10] 史晓薇, 王长新, 李琳. 高流速泄洪隧洞水力特性的三维数值模 拟[J]. 新疆农业大学学报, 2015, 38(6): 495-501. SHI Xiaowei, WANG Changxin, LI Lin. Three dimensional numerical simulation of hydraulic characteristics of spillway tunnel with high flow velocity [ J]. Journal of Xinjiang Agricultural University, 2015, 38 (6): 495-501. [11] 叶茂, 伍平, 王波, 等. 泄洪洞掺气水流的数值模拟研究[J]. 水 力发电学报, 2014, 33(4): 105-110. YE Mao, WU Ping, WANG Bo, et al. Numerical simulation of aerated flow in hydraulic tunnel [ J ]. Journal of Hydroelectric Engineering, 2014, 33(4): 105-110. [12] 胡涛, 王均星, 杜少磊. 大流量泄洪洞掺气坎水力特性数值模拟 [J]. 武汉大学学报(工学版), 2014, 47(5): 615-620. HU Tao, WANG Junxing, DU Shaolei. Numerical simulation of hydraulic characteristics of aerators in spillway tunnel with large discharge [J]. Engineering Journal of Wuhan University, 2014, 47 (5): 615-620. [13] 孙鹏飞, 姜哲, 崔维成, 等. 基于 CFD 的全海深载人潜水器直航 阻力性能研究[J]. 中国造船, 2019, 60(2): 77-87. SUN Pengfei, JIANG Zhe, CUI Weicheng, et al. Numerical simulation of a full ocean depth manned submersible based on CFD method [J]. Shipbuilding of China, 2019, 60(2): 77-87. [14] 宛鹏翔, 范俊, 韩省思, 等. 冲击射流流动换热超大涡模拟研究 [J]. 推进技术, 2020, 41(10): 2237-2247. WAN Pengxiang, FAN Jun, HAN Xingsi, et al. Very-large eddy simulation of impinging jet flow and heat transfer [ J]. Journal of Propulsion Technology, 2020, 41(10): 2237-2247. [15] 李国杰, 黄萌, 陈斌. 基于 PISO 算法的非结构化网格 VOF 算法 [J]. 工程热物理学报, 2013, 34(3): 476-479. LI Guojie, HUANG Meng, CHEN Bing. VOF method on unstructured grid using PISO algorithm [ J]. Journal of Engineering Thermophysics, 2013, 34(3): 476-479. [16] 董玮, 何庆南, 梁武科, 等. 双蜗壳离心泵泵腔轴向宽度与流动
DONG Wei, HE Qingnan, LIANG Wuke, et al. Relationship between axial width and flow characteristics of pump chamber in double volute centrifugal pump [ J ]. Journal of Northwestern Polytechnical University, 2020, 38(6): 1322-1329. [17] 陈恺, 张震宇, 王同光, 等. 基于 CFD 的水平轴风力机叶尖小翼 增功研究[J]. 太阳能学报, 2021, 42(1): 272-278. CHEN Kai, ZHANG Zhenyu, WANG Tongguang, et al. CFD-Based power enhancement of winglets for horizontal-axis wind turbines [ J]. Acta Energiae Solaris Sinica, 2021, 42(1): 272-278. [18] 张志君, 金柱男, 辛相锦, 等. 基于 VOF 方法的湿式离合器润滑 油路 CFD 数值模拟[J]. 东北大学学报(自然科学版), 2020, 41 (5): 716-722. ZHANG Zhijun, JIN Zhunan, XIN Xiangjin, et al. VOF method based CFD numerical simulation for wet clutch lubricating oil passage [ J]. Journal of Northeastern University (Natural Science), 2020, 41 (5): 716-722. [19] 罗永钦, 刁明军, 何大明, 等. 高坝明流泄洪洞掺气减蚀三维数 值模拟分析[J]. 水科学进展, 2012, 23(1): 110-116. LUO Yongqin, DIAO Mingjun, HE Daming, et al. Numerical simulation of aeration and cavitation in high dam spillway tunnels [ J]. Advances in Water Science, 2012, 23(1): 110-116. [20] 许文海, 党彦, 李国栋, 等. 双洞式溢洪洞三维流动的数值模拟 [J]. 水力发电学报, 2007(1): 56-60. XU Wenhai, DANG Yan, LI Guodong, et al. Three dimensional numerical simulation of the bi-tunnel spillway flow [ J]. Journal of Hydroelectric Engineering, 2007(1): 56-60. [21] 李爱华, 王腾, 刘沛清. 溪洛渡坝区岩石河床冲刷过程数值模拟 [J]. 水力发电学报, 2012, 31(5): 154-158. LI Aihua, WANG Teng, LIU Peiqing. Numerical simulation of rock bed scour behind the dam of Xiluodu hydropower station [J]. Journal of Hydroelectric Engineering, 2012, 31(5): 154-15
Rasoul Daneshfaraz*, Ehsan Aminvash**, Silvia Di Francesco***, Amir Najibi**, John Abraham****
토목공학의 수치해석법
Abstract
The main purpose of this study is to provide a method to increase energy dissipation on an inclined drop. Therefore, three types of rough elements with cylindrical, triangular and batshaped geometries are used on the inclined slope in the relative critical depth range of 0.128 to 0.36 and the effect of the geometry of these elements is examined using Flow 3D software. The results showed demonstrate that the downstream relative depth obtained from the numerical analysis is in good agreement with the laboratory results. The application of rough elements on the inclined drop increased the downstream relative depth and also the relative energy dissipation. The application of rough elements on the sloping surface of the drop significantly reduced the downstream Froude number, so that the Froude number in all models ranging from 4.7~7.5 to 1.45~3.36 also decreased compared to the plain drop. Bat-shaped elements are structurally smaller in size, so the use of these elements, in addition to dissipating more energy, is also economically viable.
이 연구의 주요 목적은 경사진 낙하에서 에너지 소산을 증가시키는 방법을 제공하는 것입니다. 따라서 0.128 ~ 0.36의 상대 임계 깊이 범위에서 경사면에 원통형, 삼각형 및 박쥐 모양의 형상을 가진 세 가지 유형의 거친 요소가 사용되며 이러한 요소의 형상의 영향은 Flow 3D 소프트웨어를 사용하여 조사됩니다. 결과는 수치 분석에서 얻은 하류 상대 깊이가 실험실 결과와 잘 일치함을 보여줍니다. 경 사진 낙하에 거친 요소를 적용하면 하류 상대 깊이와 상대 에너지 소산이 증가했습니다. 낙차 경사면에 거친 요소를 적용하면 하류의 Froude 수를 크게 감소시켜 4.7~7.5에서 1.45~3.36 범위의 모든 모델에서 Froude 수도 일반 낙차에 비해 감소했습니다. 박쥐 모양의 요소는 구조적으로 크기가 더 작기 때문에 더 많은 에너지를 분산시키는 것 외에도 이러한 요소를 사용하는 것이 경제적으로도 가능합니다.
Keywords: Downstream depth, Energy dissipation, Froude number, Inclined drop, Roughness elements
Introduction
급수 네트워크 시스템, 침식 수로, 수처리 시스템 및 경사가 큰 경우 흐름 에너지를 더 잘 제어하기 위해 경사 방울을 사용할 수 있습니다. 낙하 구조는 지반의 자연 경사를 설계 경사로 변환하여 에너지 소산, 유속 감소 및 수심 증가를 유발합니다. 따라서 흐름의 하류 에너지를 분산 시키기 위해 에너지 분산 구조를 사용할 수 있습니다. 난기류와 혼합된 물과 공기의 형성은 에너지 소비를 증가 시키는 효과적인 방법입니다. 흐름 경로에서 거칠기 요소를 사용하는 것은 에너지 소산을 위한 알려진 방법입니다. 이러한 요소는 흐름 경로에 배치됩니다. 그들은 종종 에너지 소산을 증가시키기 위해 다른 기하학적 구조와 배열을 가지고 있습니다. 이 연구의 목적은 직사각형 경사 방울에 대한 거칠기 요소의 영향을 조사하는 것입니다.
Fig. 1: Model made in Ardabil, IranFig. 2: Geometric and hydraulic parameters of an inclined drop
equipped with roughness elementsFig. 3: Views of the incline with (a) Bat-shaped, (b)
Cylindrical, (c) Triangular roughness elementsFig. 4: Geometric profile of inclined drop and boundary
conditions with the bat-shape roughness elementFig. 5: Variation of the RMSE varying cell sizeFig. 6: Numerical and laboratory comparison of the
downstream relative depthFig. 7: Flow profile on inclined drop in discharge of 5 L/s: (a)
Without roughness elements; (b) Bat-shaped roughness
element; (c) Cylindrical roughness element; (d) Triangular
roughness elementFig. 8: Relative edge depth versus the relative critical depthFlow on the inclined drop with bat-shaped elements: (a)
Non-submerged flowFig. 9: Flow on the inclined drop with bat-shaped elements: (b) Submerged flowFig. 10: Relative downstream depth versus the relative
critical depthFig. 11: Relative downstream depth versus the relative critical
depth
Conclusions
현재 연구에서 FLOW-3D 소프트웨어를 사용하여 한 높이, 한 각도, 밀도 15% 및 지그재그 배열에서 삼각형, 원통형 및 박쥐 모양의 형상을 가진 세 가지 유형의 거칠기 요소를 사용하여 경사 낙하 수리학적 매개변수에 대한 거칠기 요소 형상의 영향 평가되었다. VOF 방법을 사용하여 자유 표면 흐름을 시뮬레이션하고 초기에 3개의 난류 모델 RNG, k-ɛ 및 kω를 검증에 사용하고 이를 검토한 후 RNG 방법을 사용하여 다른 모델을 시뮬레이션했습니다. 1- 수치 결과에서 얻은 부드러운 경사 방울의 하류 상대 깊이는 실험실 데이터와 매우 좋은 상관 관계가 있으며 원통형 요소가 장착 된 경사 방울의 상대 에지 깊이 값이 가장 높았습니다. 2- 하류 상대깊이는 임계상대깊이가 증가함에 따라 상승하는 경향을 나타내어 박쥐형 요소를 구비한 경사낙하와 완만한 경사낙하가 각각 하류상대깊이가 가장 높고 가장 낮았다. 3- 하류 깊이의 증가로 인해 상대적 임계 깊이가 증가함에 따라 상대적 에너지 소산이 감소합니다. 한편, 가장 높은 에너지 소산은 박쥐 모양의 요소가 장착된 경사 낙하와 관련이 있으며 가장 낮은 에너지 소산은 부드러운 낙하와 관련이 있습니다. 삼각형, 원통형 및 박쥐 모양의 거친 요소가 장착된 드롭은 부드러운 드롭보다 각각 65%, 76% 및 85% 더 많은 흐름 에너지를 소산합니다. 4- 낙차의 경사면에 거친 요소를 적용하여 다운 스트림 Froude 수를 크게 줄여 4.7 ~ 7.5에서 1.45 ~ 3.36까지의 모든 모델에서 Froude 수가 부드러운 낙하에 비해 감소했습니다. 또한, 다른 원소보다 부피가 작은 박쥐 모양의 거칠기의 부피로 인해 이러한 유형의 거칠기를 사용하는 것이 경제적입니다.
References
References: [1] Abbaspour, A., Shiravani, P., and Hosseinzadeh dalir, A., “Experimental study of the energy dissipation on the rough ramps”, ISH journal of hydraulic engineering, 2019, p. 1-9. [2] Abraham, J.P., Sparrow, E.M., Gorman, J.M., Zhao, Y., and Minkowycz, W.J., “Application of an Intermittency model for laminar, transitional, and turbulent internal flows”, Journal of Fluids Engineering, vol. 141, 2019, paper no. 071204. [3] Ahmad, Z., Petappa, N.M., and Westrich, B., “Energy dissipation on block ramps with staggered boulders”, Journal of hydraulic engineering, vol. 135(6), 2009, p. 522-526. [4] Babaali, H.R., Shamsai, A., and Vosoughifar, H.R., “Computational modeling of the hydraulic jump in the stilling basin with convergence walls using CFD codes”, Arabian Journal for Science and Engineering, vol. 40(2), 2014, p. 381-395. [5] Castillo, L.G., Carrillo, J.M., and Cacía, J.T., “Numerical simulations and laboratory measurements in hydraulic jumps”, International conference on hydroinformatics. (2014, August) New York city. [6] Daneshfaraz, R., Aminvash, E., Esmaeli, R., Sadeghfam, S., and Abraham, J., “Experimental and numerical investigation for energy dissipation of supercritical flow in sudden contractions”, Journal of groundwater science and engineering, vol. 8(4), 2020a, p. 396-406. [7] Daneshfaraz, R., Aminvash, E., Ghaderi, A., Kuriqi, A., and Abraham, J., “Three-dimensional investigation of hydraulic properties of vertical drop in the presence of step and grid dissipators”, Symmetry, vol. 13 (5), 2021a, p. 895. [8] Daneshfaraz, R., Aminvash, E., Ghaderi, A., Abraham, J., and Bagherzadeh, M., “SVM performance for predicting the effect of horizontal screen diameters on the hydraulic parameters of a vertical drop”, Applied sciences, vol. 11 (9), 2021b, p. 4238. [9] Daneshfaraz, R., Bagherzadeh, M., Esmaeeli, R., Norouzi, R., and Abraham, J. “Study of the performance of support vector machine for predicting vertical drop hydraulic parameters in the presence of dual horizontal screens”, Water supply, vol 21(1), 2021c, p. 217-231. [10] Daneshfaraz, R., and Ghaderi, A., “Numerical investigation of inverse curvature ogee spillways”, Civil engineering journal, vol. 3(11), 2017, p. 1146-1156. [11] Daneshfaraz, R., Majedi Asl, M., and Bagherzadeh, M., “Experimental Investigation of the Energy Dissipation and the Downstream Relative Depth of Pool in the Sloped Gabion Drop and the Sloped simple Drop”, AUT Journal of Civil Engineering, 2020b (In persian). [12] Daneshfaraz, R., Majedi Asl, M., Bazyar, A., Abraham, J., Norouzi, R., “The laboratory study of energy dissipation in inclined drops equipped with a screen”, Journal of Applied Water Engineering and Research, 2020c, p. 1-10. [13] Daneshfaraz, R., Minaei, O., Abraham, J., Dadashi, S., and Ghaderi, A., “3-D Numerical simulation of water flow over a broad-crested weir with openings”, ISH Journal of Hydraulic Engineering, 2019, p.1-9. [14] Daneshfaraz, R., Sadeghfam, S., and Kashani, M., “Numerical simulation of flow over stepped spillways”, Research in civil engineering and environmental engineering, vol. 2(4), 2014, p. 190-198. [15] Ghaderi, A., Abbasi, S., Abraham, J., and Azamathulla, H.M., “Efficiency of trapezoidal labyrinth shaped stepped spillways”, Flow measurement and instrumentation, vol. 72, 2020a. [16] Ghaderi, A., Daneshfaraz, R., Dasineh, M., and Di Francesco, S., “Energy dissipation and hydraulics of flow over trapezoidaltriangular labyrinth weirs”, Water, vol. 12(7), 2020b, p. 1-18. [17] Ghaderi, A., Daneshfaraz, R., Torabi, M., Abraham, and Azamathulla, H.M. “Experimental investigation on effective scouring parameters downstream from stepped spillways”, Water supply, vol. 20(4), 2020c, p. 1-11. [18] Ghare, A.D., Ingle, R.N., Porey, P.D., and Gokhale, S.S. “Block ramp design for efficient energy dissipation”, Journal of energy dissipation, vol. 136(1), 2010, p. 1-5. [19] Gorman, J.M., Sparrow, E.M., Smith, C.J., Ghoash, A., Abraham, J.P., Daneshfaraz, R., Rezezadeh, J., “In-bend pressure drop and post-bend heat transfer for a bend with partial blockage at its inlet”, Numerical Heat Transfer A, vol, 73, 2018, p. 743-767. [20] Jamil, M., and Khan, S.A., “Theorical study of hydraulic jump in circular channel section”, ISH journal of hydraulic engineering, vol. 16(1), 2010, p. 1-10. [21] Katourani, S., and Kashefipour, S.M., “Effect of the geometric characteristics of baffle on hydraulic flow condition in baffled apron drop”, Irrigation sciences and engineering, vol. 37(2), 2012, p. 51-59. [22] Lai, Y.G., and Wu, K.A., “Three-dimensional flow and sediment transport model for free surface open channel flow on unstructured flexible meshes”, Fluids, vol. 4(1), 2019, p. 1-19.
[23] Nayebzadeh, B., Lotfollahi yaghin, M.A., and Daneshfaraz, R., “Numerical investigation of hydraulic characteristics of vertical drops with screens and gradually wall expanding”, Amirkabir journal of civil engineering, 2020 (In Persian). [24] Nurouzi, R., Daneshfaraz, R., and Bazyar, A., “The study of energy dissipation due to the use of vertical screen in the downstream of inclined drop by adaptive Neuro-Fuzzy inference system (ANFIS)”, AUT journal of civil engineering, 2019, (In Persian). [25] Ohtsu, I., and Yasuda, Y., “Hydraulic jump in sloping channel”, Journal of hydraulic engineering, vol. 117(7), 1991, p. 905-921. [26] Olsen, L., Abraham, J.P., Cheng, L.K., Gorman, J.M., and Sparrow, E.M., “Summary of forced-convection fluid flow and heat transfer for square cylinders of different aspect ratios ranging from the cube to a two-dimensional cylinder”, Advances in Heat Transfer, Vol. 51, 2019, p. 351-457. [27] Pagliara, S., Das, R., and Palermo, M., “Energy dissipation on submerged block ramps”, Journal of irrigation and drainage engineering, vol. 134(4), 2008, p.527-532. [28] Pagliara, S., and Palermo, M., “Effect of stilling basin geometry on the dissipative process in the presence of block ramps”, Journal of irrigation and drainage engineering, vol. 138(11), 2012, p. 1027-1031. [29] Simsek, O., Akoz, M.S, and Soydan, N.G., “Numerical validation of open channel flow over a curvilinear broad-creasted weir”, Progress in computational fluid dynamics an international journal, vol. 16(6), 2016, p. 364-378. [30] Sharif, N., and Rostami, A., “Experimental and numerical study of the effect of flow sepration on dissipating energy in compound bucket”, APCBEE procedia, vol. 9, 2014, p. 334-338. [31] Sparrow, E.M., Tong, J.C.K., and Abraham, J.P., “Fluid flow in a system with separate laminar and turbulent zones”, Numerical Heat Transfer A, vol. 53(4), 2008, p. 341-353. [32] Sparrow, E.M., Gorman, J.M., Abraham, J.P., and Minkowycz, W.J., “Validation of turbulence models for numerical simulation of fluid flow and convective heat transfers”, Advances in Heat Transfer, vol. 49, 2017, p. 397-421. [33] Wagner, W.E., “Hydraulic model studies of the check intake structure-potholes East canal”, Bureau of reclamation hydraulic laboratory report hyd, 1956, 411.
냉각 속도 및 온도 구배와 같은 FLOW-3D AM 데이터를 미세 구조 모델에 입력하여 결정 성장 및 수상 돌기 암 간격을 예측할 수 있습니다.
레이저 파우더 베드 융합으로 제작 된 니켈 기반 초합금의 열전달, 유체 흐름 및 응고 미세 구조 모델링
오하이오 주립 대학의 연구원들은 니켈 기반 초합금의 미세 구조 진화를 예측하기 위해 용융 풀과 고체 / 액체 인터페이스의 적절한 위치에서 열 구배 및 냉각 속도 데이터를 추출했습니다.
참조 : YS Lee and W. Zhang, Modeling of heat transfer, fluid flow and solidification microstructure of nickel-base superalloy fabricated by laser powder bed fusion , S2214-8604 (16) 30087-2, doi.org/10.1016/j.addma .2016.05.003 , ADDMA 86.
열 응력 | Thermal Stresses
FLOW-3D AM 시뮬레이션의 결과를 ABAQUS 또는 MSC NASTRAN과 같은 FEA 소프트웨어에 입력하여 추가 열 응력 분석을 실행할 수 있습니다. 여기에서 T- 조인트의 레이저 용접 시뮬레이션 결과를 추가 응력 분석을 위해 ABAQUS로 가져 오는 방법을 볼 수 있습니다. 마찬가지로 LPBF 시뮬레이션에서 응고 된 용융 풀 데이터의 결과를 사용하여 다른 FEA 소프트웨어에서 열 응력 및 왜곡 분석을 연구 할 수 있습니다.
DED (Directed Energy Deposition)는 레이저 또는 전자 빔과 같은 에너지 소스를 사용하여 가열 및 융합되는 와이어 또는 분말을 증착하여 부품을 만드는 적층 제조 공정입니다. FLOW-3D AM 은 분말 또는 와이어 이송 속도 및 크기 특성, 레이저 출력 및 스캔 속도와 같은 공정 매개 변수를 고려하여 DED 공정을 시뮬레이션 할 수 있습니다. 또한, 기판과 분말 재료의 서로 다른 합금에 대해 독립적 인 열 물리적 재료 특성을 정의하여 다중 재료 DED 프로세스를 시뮬레이션 할 수 있습니다.
레이저 물리학의 구현과 열 전달, 응고, 표면 장력, 차폐 가스 효과 및 반동 압력을 포함한 압력 효과를 통해 연구원은 결과 용접 비드의 강도 및 균일성에 대한 공정 매개 변수의 영향을 정확하게 분석 할 수 있습니다. 또한 이러한 시뮬레이션을 여러 레이어로 확장하여 후속 레이어 간의 융합을 분석 할 수 있습니다.
일부 연구자들은 부품을 만들기 위해 더 넓은 범위의 처리 조건을 사용하여 하이브리드 와이어 분말 기반 DED 시스템을 찾고 있습니다. 예를 들어, 이 시뮬레이션은 다양한 분말 및 와이어 이송 속도를 가진 하이브리드 시스템을 살펴봅니다.
와이어 기반 DED | Wire Based DED
와이어 기반 DED는 분말 기반 DED보다 처리량이 높고 낭비가 적지만 재료 구성 및 증착 방향 측면에서 유연성이 떨어집니다. FLOW-3D AM 은 와이어 기반 DED의 처리 결과를 이해하는데 유용하며 최적화 연구를 통해 빌드에 대한 와이어 이송 속도 및 직경과 같은 최상의 처리 매개 변수를 찾을 수 있습니다.
FLOW-3D AM은 레이저 파우더 베드 융합 (L-PBF), 바인더 제트 및 DED (Directed Energy Deposition)와 같은 적층 제조 공정 ( additive manufacturing )을 시뮬레이션하고 분석하는 CFD 소프트웨어입니다. FLOW-3D AM 의 다중 물리 기능은 공정 매개 변수의 분석 및 최적화를 위해 분말 확산 및 압축, 용융 풀 역학, L-PBF 및 DED에 대한 다공성 형성, 바인더 분사 공정을 위한 수지 침투 및 확산에 대해 매우 정확한 시뮬레이션을 제공합니다.
3D 프린팅이라고도하는 적층 제조(additive manufacturing)는 일반적으로 층별 접근 방식을 사용하여, 분말 또는 와이어로 부품을 제조하는 방법입니다. 금속 기반 적층 제조 공정에 대한 관심은 지난 몇 년 동안 시작되었습니다. 오늘날 사용되는 3 대 금속 적층 제조 공정은 PBF (Powder Bed Fusion), DED (Directed Energy Deposition) 및 바인더 제트 ( Binder jetting ) 공정입니다. FLOW-3D AM 은 이러한 각 프로세스에 대한 고유 한 시뮬레이션 통찰력을 제공합니다.
파우더 베드 융합 및 직접 에너지 증착 공정에서 레이저 또는 전자 빔을 열원으로 사용할 수 있습니다. 두 경우 모두 PBF용 분말 형태와 DED 공정용 분말 또는 와이어 형태의 금속을 완전히 녹여 융합하여 층별로 부품을 형성합니다. 그러나 바인더 젯팅(Binder jetting)에서는 결합제 역할을 하는 수지가 금속 분말에 선택적으로 증착되어 층별로 부품을 형성합니다. 이러한 부품은 더 나은 치밀화를 달성하기 위해 소결됩니다.
FLOW-3D AM 의 자유 표면 추적 알고리즘과 다중 물리 모델은 이러한 각 프로세스를 높은 정확도로 시뮬레이션 할 수 있습니다. 레이저 파우더 베드 융합 (L-PBF) 공정 모델링 단계는 여기에서 자세히 설명합니다. DED 및 바인더 분사 공정에 대한 몇 가지 개념 증명 시뮬레이션도 표시됩니다.
레이저 파우더 베드 퓨전 (L-PBF)
LPBF 공정에는 유체 흐름, 열 전달, 표면 장력, 상 변화 및 응고와 같은 복잡한 다중 물리학 현상이 포함되어 공정 및 궁극적으로 빌드 품질에 상당한 영향을 미칩니다. FLOW-3D AM 의 물리적 모델은 질량, 운동량 및 에너지 보존 방정식을 동시에 해결하는 동시에 입자 크기 분포 및 패킹 비율을 고려하여 중규모에서 용융 풀 현상을 시뮬레이션합니다.
FLOW-3D DEM 및 FLOW-3D WELD 는 전체 파우더 베드 융합 공정을 시뮬레이션하는 데 사용됩니다. L-PBF 공정의 다양한 단계는 분말 베드 놓기, 분말 용융 및 응고,이어서 이전에 응고 된 층에 신선한 분말을 놓는 것, 그리고 다시 한번 새 층을 이전 층에 녹이고 융합시키는 것입니다. FLOW-3D AM 은 이러한 각 단계를 시뮬레이션하는 데 사용할 수 있습니다.
파우더 베드 부설 공정
FLOW-3D DEM을 통해 분말 크기 분포, 재료 특성, 응집 효과는 물론 롤러 또는 블레이드 움직임 및 상호 작용과 같은 기하학적 효과와 관련된 분말 확산 및 압축을 이해할 수 있습니다. 이러한 시뮬레이션은 공정 매개 변수가 후속 인쇄 공정에서 용융 풀 역학에 직접적인 영향을 미치는 패킹 밀도와 같은 분말 베드 특성에 어떻게 영향을 미치는지에 대한 정확한 이해를 제공합니다.
다양한 파우더 베드 압축을 달성하는 한 가지 방법은 베드를 놓는 동안 다양한 입자 크기 분포를 선택하는 것입니다. 아래에서 볼 수 있듯이 세 가지 크기의 입자 크기 분포가 있으며, 이는 가장 높은 압축을 제공하는 Case 2와 함께 다양한 분말 베드 압축을 초래합니다.
세 가지 다른 입자 크기 분포를 사용하여 파우더 베드 배치세 가지 다른 입자 크기 분포를 사용한 분말 베드 압축
입자-입자 상호 작용, 유체-입자 결합 및 입자 이동 물체 상호 작용은 FLOW-3D DEM을 사용하여 자세히 분석 할 수도 있습니다 . 또한 입자간 힘을 지정하여 분말 살포 응용 분야를 보다 정확하게 연구 할 수도 있습니다.
이 FLOW-3D AM 시뮬레이션은 이산 요소 방법 (DEM)을 사용하여 역 회전하는 원통형 롤러로 인한 분말 확산을 연구합니다. 비디오 시작 부분에서 빌드 플랫폼이 위로 이동하는 동안 분말 저장소가 아래로 이동합니다. 그 직후, 롤러는 분말 입자 (초기 위치에 따라 색상이 지정됨)를 다음 층이 녹고 구축 될 준비를 위해 구축 플랫폼으로 펼칩니다. 이러한 시뮬레이션은 저장소에서 빌드 플랫폼으로 전송되는 분말 입자의 선호 크기에 대한 추가 통찰력을 제공 할 수 있습니다.
Melting | 파우더 베드 용해
DEM 시뮬레이션에서 파우더 베드가 생성되면 STL 파일로 추출됩니다. 다음 단계는 CFD를 사용하여 레이저 용융 공정을 시뮬레이션하는 것입니다. 여기서는 레이저 빔과 파우더 베드의 상호 작용을 모델링 합니다. 이 프로세스를 정확하게 포착하기 위해 물리학에는 점성 흐름, 용융 풀 내의 레이저 반사 (광선 추적을 통해), 열 전달, 응고, 상 변화 및 기화, 반동 압력, 차폐 가스 압력 및 표면 장력이 포함됩니다. 이 모든 물리학은 이 복잡한 프로세스를 정확하게 시뮬레이션하기 위해 TruVOF 방법을 기반으로 개발되었습니다.
레이저 출력 200W, 스캔 속도 3.0m / s, 스폿 반경 100μm에서 파우더 베드의 용융 풀 분석.
용융 풀이 응고되면 FLOW-3D AM 압력 및 온도 데이터를 Abaqus 또는 MSC Nastran과 같은 FEA 도구로 가져와 응력 윤곽 및 변위 프로파일을 분석 할 수도 있습니다.
Multilayer | 다층 적층 제조
용융 풀 트랙이 응고되면 DEM을 사용하여 이전에 응고된 층에 새로운 분말 층의 확산을 시뮬레이션 할 수 있습니다. 유사하게, 레이저 용융은 새로운 분말 층에서 수행되어 후속 층 간의 융합 조건을 분석 할 수 있습니다.
해석 진행 절차는 첫 번째 용융층이 응고되면 입자의 두 번째 층이 응고 층에 증착됩니다. 새로운 분말 입자 층에 레이저 공정 매개 변수를 지정하여 용융 풀 시뮬레이션을 다시 수행합니다. 이 프로세스를 여러 번 반복하여 연속적으로 응고된 층 간의 융합, 빌드 내 온도 구배를 평가하는 동시에 다공성 또는 기타 결함의 형성을 모니터링 할 수 있습니다.
LPBF의 키홀 링 | Keyholing in LPBF
키홀링 중 다공성은 어떻게 형성됩니까? 이것은 TU Denmark의 연구원들이 FLOW-3D AM을 사용하여 답변한 질문이었습니다. 레이저 빔의 적용으로 기판이 녹으면 기화 및 상 변화로 인한 반동 압력이 용융 풀을 압박합니다. 반동 압력으로 인한 하향 흐름과 레이저 반사로 인한 추가 레이저 에너지 흡수가 공존하면 폭주 효과가 발생하여 용융 풀이 Keyholing으로 전환됩니다. 결국, 키홀 벽을 따라 온도가 변하기 때문에 표면 장력으로 인해 벽이 뭉쳐져서 진행되는 응고 전선에 의해 갇힐 수 있는 공극이 생겨 다공성이 발생합니다. FLOW-3D AM 레이저 파우더 베드 융합 공정 모듈은 키홀링 및 다공성 형성을 시뮬레이션 하는데 필요한 모든 물리 모델을 보유하고 있습니다.
바인더 분사 (Binder jetting)
Binder jetting 시뮬레이션은 모세관 힘의 영향을받는 파우더 베드에서 바인더의 확산 및 침투에 대한 통찰력을 제공합니다. 공정 매개 변수와 재료 특성은 증착 및 확산 공정에 직접적인 영향을 미칩니다.
Scan Strategy | 스캔 전략
스캔 전략은 온도 구배 및 냉각 속도에 영향을 미치기 때문에 미세 구조에 직접적인 영향을 미칩니다. 연구원들은 FLOW-3D AM 을 사용하여 결함 형성과 응고된 금속의 미세 구조에 영향을 줄 수 있는 트랙 사이에서 발생하는 재 용융을 이해하기 위한 최적의 스캔 전략을 탐색하고 있습니다. FLOW-3D AM 은 하나 또는 여러 레이저에 대해 시간에 따른 방향 속도를 구현할 때 완전한 유연성을 제공합니다.
Beam Shaping | 빔 형성
레이저 출력 및 스캔 전략 외에도 레이저 빔 모양과 열유속 분포는 LPBF 공정에서 용융 풀 역학에 큰 영향을 미칩니다. AM 기계 제조업체는 공정 안정성 및 처리량에 대해 다중 코어 및 임의 모양의 레이저 빔 사용을 모색하고 있습니다. FLOW-3D AM을 사용하면 멀티 코어 및 임의 모양의 빔 프로파일을 구현할 수 있으므로 생산량을 늘리고 부품 품질을 개선하기 위한 최상의 구성에 대한 통찰력을 제공 할 수 있습니다.
이 시뮬레이션에서 스테인리스 강 및 알루미늄 분말은 FLOW-3D AM 이 용융 풀 역학을 정확하게 포착하기 위해 추적하는 독립적으로 정의 된 온도 의존 재료 특성을 가지고 있습니다. 시뮬레이션은 용융 풀에서 재료 혼합을 이해하는 데 도움이됩니다.
다중 재료 용접 사례 연구
이종 금속의 레이저 키홀 용접에서 금속 혼합 조사
GM과 University of Utah의 연구원들은 FLOW-3D WELD 를 사용 하여 레이저 키홀 용접을 통한 이종 금속의 혼합을 이해했습니다. 그들은 반동 압력 및 Marangoni 대류와 관련하여 구리와 알루미늄의 혼합 농도에 대한 레이저 출력 및 스캔 속도의 영향을 조사했습니다. 그들은 시뮬레이션을 실험 결과와 비교했으며 샘플 내의 절단 단면에서 재료 농도 사이에 좋은 일치를 발견했습니다.
이종 금속의 레이저 키홀 용접에서 금속 혼합 조사참조 : Wenkang Huang, Hongliang Wang, Teresa Rinker, Wenda Tan, 이종 금속의 레이저 키홀 용접에서 금속 혼합 조사 , Materials & Design, Volume 195, (2020). https://doi.org/10.1016/j.matdes.2020.109056
방향성 에너지 증착
FLOW-3D AM 의 내장 입자 모델 을 사용하여 직접 에너지 증착 프로세스를 시뮬레이션 할 수 있습니다. 분말 주입 속도와 고체 기질에 입사되는 열유속을 지정함으로써 고체 입자는 용융 풀에 질량, 운동량 및 에너지를 추가 할 수 있습니다. 다음 비디오에서 고체 금속 입자가 용융 풀에 주입되고 기판에서 용융 풀의 후속 응고가 관찰됩니다.
Electromagnetic metal casting (EMC) is a casting technique that uses electromagnetic energy to heat metal powders. It is a faster, cleaner, and less time-consuming operation. Solid metals create issues in electromagnetics since they reflect the electromagnetic radiation rather than consume it—electromagnetic energy processing results in sounded pieces with higher-ranking material properties and a more excellent microstructure solution. For the physical production of the electromagnetic casting process, knowledge of electromagnetic material interaction is critical. Even where the heated material is an excellent electromagnetic absorber, the total heating quality is sometimes insufficient. Numerical modelling works on finding the proper coupled effects between properties to bring out the most effective operation. The main parameters influencing the quality of output of the EMC process are: power dissipated per unit volume into the material, penetration depth of electromagnetics, complex magnetic permeability and complex dielectric permittivity. The contact mechanism and interference pattern also, in turn, determines the quality of the process. Only a few parameters, such as the environment’s temperature, the interference pattern, and the rate of metal solidification, can be controlled by AI models. Neural networks are used to achieve exact outcomes by stimulating the neurons in the human brain. Additive manufacturing (AM) is used to design mold and cores for metal casting. The models outperformed the traditional DFA optimization approach, which is susceptible to local minima. The system works only offline, so real-time analysis and corrections are not yet possible.
Korea Abstract
전자기 금속 주조 (EMC)는 전자기 에너지를 사용하여 금속 분말을 가열하는 주조 기술입니다. 더 빠르고 깨끗하며 시간이 덜 소요되는 작업입니다.
고체 금속은 전자기 복사를 소비하는 대신 반사하기 때문에 전자기학에서 문제를 일으킵니다. 전자기 에너지 처리는 더 높은 등급의 재료 특성과 더 우수한 미세 구조 솔루션을 가진 사운드 조각을 만듭니다.
전자기 주조 공정의 물리적 생산을 위해서는 전자기 물질 상호 작용에 대한 지식이 중요합니다. 가열된 물질이 우수한 전자기 흡수재인 경우에도 전체 가열 품질이 때때로 불충분합니다. 수치 모델링은 가장 효과적인 작업을 이끌어 내기 위해 속성 간의 적절한 결합 효과를 찾는데 사용됩니다.
EMC 공정의 출력 품질에 영향을 미치는 주요 매개 변수는 단위 부피당 재료로 분산되는 전력, 전자기의 침투 깊이, 복합 자기 투과성 및 복합 유전율입니다. 접촉 메커니즘과 간섭 패턴 또한 공정의 품질을 결정합니다. 환경 온도, 간섭 패턴 및 금속 응고 속도와 같은 몇 가지 매개 변수 만 AI 모델로 제어 할 수 있습니다.
신경망은 인간 뇌의 뉴런을 자극하여 정확한 결과를 얻기 위해 사용됩니다. 적층 제조 (AM)는 금속 주조용 몰드 및 코어를 설계하는 데 사용됩니다. 모델은 로컬 최소값에 영향을 받기 쉬운 기존 DFA 최적화 접근 방식을 능가했습니다. 이 시스템은 오프라인에서만 작동하므로 실시간 분석 및 수정은 아직 불가능합니다.
electromagnetic metal casting computation designs Fig1electromagnetic metal casting computation designs Fig2electromagnetic metal casting computation designs Fig3electromagnetic metal casting computation designs Fig4electromagnetic metal casting computation designs Fig5electromagnetic metal casting computation designs Fig6electromagnetic metal casting computation designs Fig7electromagnetic metal casting computation designs Fig8electromagnetic metal casting computation designs Fig9
References
1.J. Sun, W. Wang, Q. Yue, Review on electromagnetic-matter interaction fundamentals and efficient electromagnetic-associated heating strategies. Materials 9(4), 231 (2016). https://doi.org/10.3390/ma9040231ADSArticleGoogle Scholar
2.E. Ghasali, A. Fazili, M. Alizadeh, K. Shirvanimoghaddam, T. Ebadzadeh, Evaluation of microstructure and mechanical properties of Al-TiC metal matrix composite prepared by conventional, electromagnetic and spark plasma sintering methods. Materials 10(11), 1255 (2017). https://doi.org/10.3390/ma10111255ADSArticleGoogle Scholar
10.R. Singh, S. Singh, V. Mahajan, Investigations for dimensional accuracy of investment casting process after cycle time reduction by advancements in shell moulding. Procedia Mater. Sci. 6, 859–865 (2014). https://doi.org/10.1016/j.mspro.2014.07.103ArticleGoogle Scholar
19.V. Gangwar, S. Kumar, V. Singh, H. Singh, Effect of process parameters on hardness of AA-6063 in-situ electromagnetic casting by using taguchi method, in IOP Conference Series: Materials Science and Engineering, vol. 804(1) (IOP Publishing, 2020), p. 012019
20.X. Ye, S. Guo, L. Yang, J. Gao, J. Peng, T. Hu, L. Wang, M. Hou, Q. Luo, New utilization approach of electromagnetic thermal energy: preparation of metallic matrix diamond tool bit by electromagnetic hot-press sintering. J. Alloy. Compd. (2018). https://doi.org/10.1016/j.jallcom.2018.03.183ArticleGoogle Scholar
22.K.L. Glass, D.M. Ashby, U.S. Patent No. 9050656. (U.S. Patent and Trademark Office, Washington, DC, 2015)
23.S. Verma, P. Gupta, S. Srivastava, S. Kumar, A. Anand, An overview: casting/melting of non ferrous metallic materials using domestic electromagnetic oven. J. Mater. Sci. Mech. Eng. 4(4), (2017). p-ISSN: 2393-9095; e-ISSN: 2393-9109
24.S.S. Panda, V. Singh, A. Upadhyaya, D. Agrawal, Sintering response of austenitic (316L) and ferritic (434L) stainless steel consolidated in conventional and electromagnetic furnaces. Scripta Mater. 54(12), 2179–2183 (2006). https://doi.org/10.1016/j.scriptamat.2006.02.034ArticleGoogle Scholar
25.Y. Zhang, S. Yang, S. Wang, X. Liu, L. Li, Microwave/freeze casting assisted fabrication of carbon frameworks derived from embedded upholder in tremella for superior performance supercapacitors. Energy Storage Mater. (2018). https://doi.org/10.1016/j.ensm.2018.08.006ArticleGoogle Scholar
27.M.H. Awida, N. Shah, B. Warren, E. Ripley, A.E. Fathy, Modeling of an industrial Electromagnetic furnace for metal casting applications. 2008 IEEE MTT-S Int. Electromagn. Symp. Digest. (2008). https://doi.org/10.1109/mwsym.2008.4633143ArticleGoogle Scholar
29.E.B. Ripley, J.A. Oberhaus, WWWeb search power page-melting and heat treating metals using electromagnetic heating-the potential of electromagnetic metal processing techniques for a wide variety of metals and alloys is. Ind. Heat. 72(5), 65–70 (2005)Google Scholar
30.J. Campbell, Complete Casting Handbook: Metal Casting Processes, Metallurgy, Techniques and Design (Butterworth-Heinemann, 2015)Google Scholar
31.B. Ravi, Metal Casting: Computer-Aided Design and Analysis, 1st edn. (PHI Learning Ltd, 2005)Google Scholar
34.J. Ha, P. Cleary, V. Alguine, T. Nguyen, Simulation of die filling in gravity die casting using SPH and MAGMAsoft, in Proceedings of 2nd International Conference on CFD in Minerals & Process Industries (1999) pp. 423–428
35.M. Sirviö, M. Woś, Casting directly from a computer model by using advanced simulation software FLOW-3D Cast Ž. Arch. Foundry Eng. 9(1), 79–82 (2009)Google Scholar
38.EKK, Inc. Metal Casting Simulation Software and Consulting Services, CAPCAST Brochure
39.P. Muenprasertdee, Solidification modeling of iron castings using SOLIDCast (2007)
40.CasCAE, CT-CasTest Inc. Oy, Kerava
41.E. Dominguez-Tortajada, J. Monzo-Cabrera, A. Diaz-Morcillo, Uniform electric field distribution in electromagnetic heating applicators by means of genetic algorithms optimization of dielectric multilayer structures. IEEE Trans. Electromagn. Theory Tech. 55(1), 85–91 (2007). https://doi.org/10.1109/tmtt.2006.886913ADSArticleGoogle Scholar
42.B. Warren, M.H. Awida, A.E. Fathy, Electromagnetic heating of metals. IET Electromagn. Antennas Propag. 6(2), 196–205 (2012)ArticleGoogle Scholar
43.S. Ashouri, M. Nili-Ahmadabadi, M. Moradi, M. Iranpour, Semi-solid microstructure evolution during reheating of aluminum A356 alloy deformed severely by ECAP. J. Alloy. Compd. 466(1–2), 67–72 (2008). https://doi.org/10.1016/j.jallcom.2007.11.010ArticleGoogle Scholar
45.R.R. Mishra, A.K. Sharma, A review of research trends in electromagnetic processing of metal-based materials and opportunities in electromagnetic metal casting. Crit. Rev. Solid State Mater. Sci. 41(3), 217–255 (2016). https://doi.org/10.1080/10408436.2016.1142421ADSArticleGoogle Scholar
46.D.K. Ghodgaonkar, V.V. Varadan, V.K. Varadan, Free-space measurement of complex permittivity and complex permeability of magnetic materials at Electromagnetic frequencies. IEEE Trans. Instrum. Meas. 39(2), 387–394 (1990). https://doi.org/10.1109/19.52520ArticleGoogle Scholar
47.J. Baker-Jarvis, E.J. Vanzura, W.A. Kissick, Improved technique for determining complex permittivity with the transmission/reflection method. Microw. Theory Tech. IEEE Trans. 38, 1096–1103 (1990)ADSArticleGoogle Scholar
48.M. Bologna, A. Petri, B. Tellini, C. Zappacosta, Effective magnetic permeability measurementin composite resonator structures. Instrum. Meas. IEEE Trans. 59, 1200–1206 (2010)ArticleGoogle Scholar
49.B. Ravi, G.L. Datta, Metal casting–back to future, in 52nd Indian Foundry Congress, (2004)
51.H. Sekiguchi, Y. Mori, Steam plasma reforming using Electromagnetic discharge. Thin Solid Films 435, 44–48 (2003)ADSArticleGoogle Scholar
52.J. Sun, W. Wang, C. Zhao, Y. Zhang, C. Ma, Q. Yue, Study on the coupled effect of wave absorption and metal discharge generation under electromagnetic irradiation. Ind. Eng. Chem. Res. 53, 2042–2051 (2014)ArticleGoogle Scholar
56.B. Ravi, Metal Casting: Computer-Aided Design and Analysis. (PHI Learning Pvt. Ltd., 2005)
57.J.H. Lee, S.D. Noh, H.-J. Kim, Y.-S. Kang, Implementation of cyber-physical production systems for quality prediction and operation control in metal casting. Sensors 18, 1428 (2018). https://doi.org/10.3390/s18051428ADSArticleGoogle Scholar
58.B. Aksoy, M. Koru, Estimation of casting mold interfacial heat transfer coefficient in pressure die casting process by artificial intelligence methods. Arab. J. Sci. Eng. 45, 8969–8980 (2020). https://doi.org/10.1007/s13369-020-04648-7ArticleGoogle Scholar
61.W. Chen, B. Gutmann, C.O. Kappe, Characterization of electromagnetic-induced electric discharge phenomena in metal-solvent mixtures. ChemistryOpen 1, 39–48 (2012)ArticleGoogle Scholar
62.J. Walker, A. Prokop, C. Lynagh, B. Vuksanovich, B. Conner, K. Rogers, J. Thiel, E. MacDonald, Real-time process monitoring of core shifts during metal casting with wireless sensing and 3D sand printing. Addit. Manuf. (2019). https://doi.org/10.1016/j.addma.2019.02.018ArticleGoogle Scholar
63.G.C. Manjunath Patel, A.K. Shettigar, M.B. Parappagoudar, A systematic approach to model and optimize wear behaviour of castings produced by squeeze casting process. J. Manuf. Process. 32, 199–212 (2018). https://doi.org/10.1016/j.jmapro.2018.02.004ArticleGoogle Scholar
64.G.C. Manjunath Patel, P. Krishna, M.B. Parappagoudar, An intelligent system for squeeze casting process—soft computing based approach. Int. J. Adv. Manuf. Technol. 86, 3051–3065 (2016). https://doi.org/10.1007/s00170-016-8416-8ArticleGoogle Scholar
65.M. Ferguson, R. Ak, Y.T. Lee, K.H. Law, Automatic localization of casting defects with convolutional neural networks, in 2017 IEEE International Conference on Big Data (Big Data) (Boston, MA, USA, 2017), pp. 1726–1735. https://doi.org/10.1109/BigData.2017.8258115.
67.G.C. ManjunathPatel, A.K. Shettigar, P. Krishna, M.B. Parappagoudar, Back propagation genetic and recurrent neural network applications in modelling and analysis of squeeze casting process. Appl. Soft Comput. 59, 418–437 (2017). https://doi.org/10.1016/j.asoc.2017.06.018ArticleGoogle Scholar
69.E. Mares, J. Sokolowski, Artificial intelligence-based control system for the analysis of metal casting properties. J. Achiev. Mater. Manuf. Eng. 40, 149–154 (2010)Google Scholar
70.K.S. Senthil, S. Muthukumaran, C. Chandrasekhar Reddy, Suitability of friction welding of tube to tube plate using an external tool process for different tube diameters—a study. Exp. Tech. 37(6), 8–14 (2013)ArticleGoogle Scholar
72.K.S. Senthil, D.A. Daniel, An investigation of boiler grade tube and tube plate without block by using friction welding process. Mater. Today Proc. 5(2), 8567–8576 (2018)ArticleGoogle Scholar
73.E. Hetmaniok, D. Słota, A. Zielonka, Restoration of the cooling conditions in a three-dimensional continuous casting process using artificial intelligence algorithms. Appl. Math. Modell. 39(16), 4797–4807 (2015). https://doi.org/10.1016/j.apm.2015.03.056ArticleMATHGoogle Scholar
74.C.V. Kumar, S. Muthukumaran, A. Pradeep, S.S. Kumaran, Optimizational study of friction welding of steel tube to aluminum tube plate using an external tool process. Int. J. Mech. Mater. Eng. 6(2), 300–306 (2011)Google Scholar
75.T. Adithiyaa, D. Chandramohan, T. Sathish, Optimal prediction of process parameters by GWO-KNN in stirring-squeeze casting of AA2219 reinforced metal matrix composites. Mater. Today Proc. 150, 1598 (2020). https://doi.org/10.1016/j.matpr.2019.10.051ArticleGoogle Scholar
76.B.P. Pehrson, A.F. Moore (2014). U.S. Patent No. 8708031 (U.S. Patent and Trademark Office, Washington, DC, 2014)
77.Liu, J., & Rynerson, M. L. (2008). U.S. Patent No. 7,461,684. Washington, DC: U.S. Patent and Trademark Office.
79.R.R. Mishra, A.K. Sharma, Effect of solidification environment on microstructure and indentation hardness of Al–Zn–Mg alloy casts developed using electromagnetic heating. Int. J. Metal Cast. 10, 1–13 (2017). https://doi.org/10.1007/s40962-017-0176-1ArticleGoogle Scholar
84.M.P. Reddy, R.A. Shakoor, G. Parande, V. Manakari, F. Ubaid, A.M.A. Mohamed, M. Gupta, Enhanced performance of nano-sized SiC reinforced Al metal matrix nanocomposites synthesized through electromagnetic sintering and hot extrusion techniques. Prog. Nat. Sci. Mater. Int. 27(5), 606–614 (2017). https://doi.org/10.1016/j.pnsc.2017.08.015ArticleGoogle Scholar
86.V. Bist, A.K. Sharma, P. Kumar, Development and microstructural characterisations of the lead casting using electromagnetic technology. Manager’s J. Mech. Eng. 4(4), 6 (2014). https://doi.org/10.26634/jme.4.4.2840ArticleGoogle Scholar
87.A. Sharma, A. Chouhan, L. Pavithran, U. Chadha, S.K. Selvaraj, Implementation of LSS framework in automotive component manufacturing: a review, current scenario and future directions. Mater Today: Proc. (2021). https://doi.org/10.1016/J.MATPR.2021.02.374ArticleGoogle Scholar
Surface roughness of laser powder bed fusion (L-PBF) printed overhang regions is a major contributor to deteriorated shape accuracy/surface quality. This study investigates the mechanisms behind the evolution of surface roughness (Ra) in overhang regions. The evolution of surface morphology is the result of a combination of border track contour, powder adhesion, warp deformation, and dross formation, which is strongly related to the overhang angle (θ). When 0° ≤ θ ≤ 15°, the overhang angle does not affect Ra significantly since only a small area of the melt pool boundaries contacts the powder bed resulting in slight powder adhesion. When 15° < θ ≤ 50°, powder adhesion is enhanced by the melt pool sinking and the increased contact area between the melt pool boundary and powder bed. When θ > 50°, large waviness of the overhang contour, adhesion of powder clusters, severe warp deformation and dross formation increase Ra sharply.
레이저 파우더 베드 퓨전 (L-PBF) 프린팅 오버행 영역의 표면 거칠기는 형상 정확도 / 표면 품질 저하의 주요 원인입니다. 이 연구 는 오버행 영역에서 표면 거칠기 (Ra ) 의 진화 뒤에 있는 메커니즘을 조사합니다 . 표면 형태의 진화는 오버행 각도 ( θ ) 와 밀접한 관련이있는 경계 트랙 윤곽, 분말 접착, 뒤틀림 변형 및 드로스 형성의 조합의 결과입니다 . 0° ≤ θ ≤ 15° 인 경우 , 용융풀 경계의 작은 영역 만 분말 베드와 접촉하여 약간의 분말 접착이 발생하기 때문에 오버행 각도가 R a에 큰 영향을 주지 않습니다 . 15° < θ 일 때 ≤ 50°, 용융 풀 싱킹 및 용융 풀 경계와 분말 베드 사이의 증가된 접촉 면적으로 분말 접착력이 향상됩니다. θ > 50° 일 때 오버행 윤곽의 큰 파형, 분말 클러스터의 접착, 심한 휨 변형 및 드 로스 형성이 Ra 급격히 증가 합니다.
KEYWORDS: Laser powder bed fusion (L-PBF), melt pool dynamics, overhang region, shape deviation, surface roughness
1. Introduction
레이저 분말 베드 융합 (L-PBF)은 첨단 적층 제조 (AM) 기술로, 집중된 레이저 빔을 사용하여 금속 분말을 선택적으로 융합하여 슬라이스 된 3D 컴퓨터 지원에 따라 층별로 3 차원 (3D) 금속 부품을 구축합니다. 설계 (CAD) 모델 (Chatham, Long 및 Williams 2019 ; Tan, Zhu 및 Zhou 2020 ). 재료가 인쇄 층 아래에 존재하는지 여부에 따라 인쇄 영역은 각각 솔리드 영역 또는 돌출 영역으로 분류 될 수 있습니다. 따라서 오버행 영역은 고체 기판이 아니라 분말 베드 바로 위에 건설되는 특수 구조입니다 (Patterson, Messimer 및 Farrington 2017). 오버행 영역은지지 구조를 포함하거나 포함하지 않고 구축 할 수 있으며, 지지대가있는 돌출 영역의 L-PBF는 지지체가 더 낮은 밀도로 구축된다는 점을 제외 하고 (Wang and Chou 2018 ) 고체 기판의 공정과 유사합니다 (따라서 기계적 강도가 낮기 때문에 L-PBF 공정 후 기계적으로 쉽게 제거 할 수 있습니다. 따라서지지 구조로 인쇄 된 오버행 영역은 L-PBF 공정 후 지지물 제거, 연삭 및 연마와 같은 추가 후 처리 단계가 필요합니다.
수평 내부 채널의 제작과 같은 일부 특정 경우에는 공정 후 지지대를 제거하기가 어려우므로 채널 상단 절반의 돌출부 영역을 지지대없이 건설해야합니다 (Hopkinson and Dickens 2000 ). 수평 내부 채널에 사용할 수없는지지 구조 외에도 내부 표면, 특히 등각 냉각 채널 (Feng, Kamat 및 Pei 2021 ) 에서 발생하는 복잡한 3D 채널 네트워크의 경우 표면 마감 프로세스를 구현하는 것도 어렵습니다 . 결과적으로 오버행 영역은 (i) 잔류 응력에 의한 변형, (ii) 계단 효과 (Kuo et al. 2020 ; Li et al. 2020 )로 인해 설계된 모양에서 벗어날 수 있습니다 .) 및 (iii) 원하지 않는 분말 소결로 인한 향상된 표면 거칠기; 여기서, 앞의 두 요소는 일반적으로 mm 길이 스케일에서 ‘매크로’편차로 분류되고 후자는 일반적으로 µm 길이 스케일에서 ‘마이크로’편차로 인식됩니다.
열 응력에 의한 변형은 오버행 영역에서 발생하는 중요한 문제입니다 (Patterson, Messimer 및 Farrington 2017 ). 국부적 인 용융 / 냉각은 용융 풀 내부 및 주변에서 큰 온도 구배를 유도하여 응고 된 층에 집중적 인 열 응력을 유발합니다. 열 응력에 의한 뒤틀림은 고체 영역을 현저하게 변형하지 않습니다. 이러한 영역은 아래의 여러 레이어에 의해 제한되기 때문입니다. 반면에 오버행 영역은 구속되지 않고 공정 중 응력 완화로 인해 상당한 변형이 발생합니다 (Kamat 및 Pei 2019 ). 더욱이 용융 깊이는 레이어 두께보다 큽니다 (이전 레이어도 재용 해되어 빌드 된 레이어간에 충분한 결합을 보장하기 때문입니다 [Yadroitsev et al. 2013 ; Kamath et al.2014 ]),응고 된 두께가 설계된 두께보다 크기 때문에형태 편차 (예 : 드 로스 [Charles et al. 2020 ; Feng et al. 2020 ])가 발생합니다. 마이크로 스케일에서 인쇄 된 표면 (R a 및 S a ∼ 10 μm)은 기계적으로 가공 된 표면보다 거칠다 (Duval-Chaneac et al. 2018 ; Wen et al. 2018 ). 이 문제는고형화 된 용융 풀의 가장자리에 부착 된 용융되지 않은 분말의 결과로 표면 거칠기 (R a )가 일반적으로 약 20 μm인 오버행 영역에서 특히 심각합니다 (Mazur et al. 2016 ; Pakkanen et al. 2016 ).
오버행 각도 ( θ , 빌드 방향과 관련하여 측정)는 오버행 영역의 뒤틀림 편향과 표면 거칠기에 영향을 미치는 중요한 매개 변수입니다 (Kamat and Pei 2019 ; Mingear et al. 2019 ). θ ∼ 45 ° 의 오버행 각도 는 일반적으로지지 구조없이 오버행 영역을 인쇄 할 수있는 임계 값으로 합의됩니다 (Pakkanen et al. 2016 ; Kadirgama et al. 2018 ). θ 일 때이 임계 값보다 크면 오버행 영역을 허용 가능한 표면 품질로 인쇄 할 수 없습니다. 오버행 각도 외에도 레이저 매개 변수 (레이저 에너지 밀도와 관련된)는 용융 풀의 모양 / 크기 및 용융 풀 역학에 영향을줌으로써 오버행 영역의 표면 거칠기에 영향을줍니다 (Wang et al. 2013 ; Mingear et al . 2019 ).
용융 풀 역학은 고체 (Shrestha 및 Chou 2018 ) 및 오버행 (Le et al. 2020 ) 영역 모두에서 수행되는 L-PBF 공정을 포함한 레이저 재료 가공의 일반적인 물리적 현상입니다 . 용융 풀 모양, 크기 및 냉각 속도는 잔류 응력으로 인한 변형과 표면 거칠기에 모두 영향을 미치므로 처리 매개 변수와 표면 형태 / 품질 사이의 다리 역할을하며 용융 풀을 이해하기 위해 수치 시뮬레이션을 사용하여 추가 조사를 수행 할 수 있습니다. 거동과 표면 거칠기에 미치는 영향. 현재까지 고체 영역의 L-PBF 동안 용융 풀 동작을 시뮬레이션하기 위해 여러 연구가 수행되었습니다. 유한 요소 방법 (FEM)과 같은 시뮬레이션 기술 (Roberts et al. 2009 ; Du et al.2019 ), 유한 차분 법 (FDM) (Wu et al. 2018 ), 전산 유체 역학 (CFD) (Lee and Zhang 2016 ), 임의의 Lagrangian-Eulerian 방법 (ALE) (Khairallah and Anderson 2014 )을 사용하여 증발 반동 압력 (Hu et al. 2018 ) 및 Marangoni 대류 (Zhang et al. 2018 ) 현상을포함하는 열 전달 (온도 장) 및 물질 전달 (용융 흐름) 프로세스. 또한 이산 요소법 (DEM)을 사용하여 무작위 분산 분말 베드를 생성했습니다 (Lee and Zhang 2016 ; Wu et al. 2018 ). 이 모델은 분말 규모의 L-PBF 공정을 시뮬레이션했습니다 (Khairallah et al. 2016) 메조 스케일 (Khairallah 및 Anderson 2014 ), 단일 트랙 (Leitz et al. 2017 )에서 다중 트랙 (Foroozmehr et al. 2016 ) 및 다중 레이어 (Huang, Khamesee 및 Toyserkani 2019 )로.
그러나 결과적인 표면 거칠기를 결정하는 오버행 영역의 용융 풀 역학은 문헌에서 거의 관심을받지 못했습니다. 솔리드 영역의 L-PBF에 대한 기존 시뮬레이션 모델이 어느 정도 참조가 될 수 있지만 오버행 영역과 솔리드 영역 간의 용융 풀 역학에는 상당한 차이가 있습니다. 오버행 영역에서 용융 금속은 분말 입자 사이의 틈새로 아래로 흘러 용융 풀이 다공성 분말 베드가 제공하는 약한 지지체 아래로 가라 앉습니다. 이것은 중력과 표면 장력의 영향이 용융 풀의 결과적인 모양 / 크기를 결정하는 데 중요하며, 결과적으로 오버행 영역의 마이크로 스케일 형태의 진화에 중요합니다. 또한 분말 입자 사이의 공극, 열 조건 (예 : 에너지 흡수,2019 ; Karimi et al. 2020 ; 노래와 영 2020 ). 표면 거칠기는 (마이크로) 형상 편차를 증가시킬뿐만 아니라 주기적 하중 동안 미세 균열의 시작 지점 역할을함으로써 기계적 강도를 저하시킵니다 (Günther et al. 2018 ). 오버행 영역의 높은 표면 거칠기는 (마이크로) 정확도 / 품질에 대한 엄격한 요구 사항이있는 부품 제조에서 L-PBF의 적용을 제한합니다.
본 연구는 실험 및 시뮬레이션 연구를 사용하여 오버행 영역 (지지물없이 제작)의 미세 형상 편차 형성 메커니즘과 표면 거칠기의 기원을 체계적이고 포괄적으로 조사합니다. 결합 된 DEM-CFD 시뮬레이션 모델은 경계 트랙 윤곽, 분말 접착 및 뒤틀림 변형의 효과를 고려하여 오버행 영역의 용융 풀 역학과 표면 형태의 형성 메커니즘을 나타 내기 위해 개발되었습니다. 표면 거칠기 R의 시뮬레이션 및 단일 요인 L-PBF 인쇄 실험을 사용하여 오버행 각도의 함수로 연구됩니다. 용융 풀의 침몰과 관련된 오버행 영역에서 분말 접착의 세 가지 메커니즘이 식별되고 자세히 설명됩니다. 마지막으로, 인쇄 된 오버행 영역에서 높은 표면 거칠기 문제를 완화 할 수 있는 잠재적 솔루션에 대해 간략하게 설명합니다.
The shape and size of the L-PBF printed samples are illustrated in Figure 1Figure 2. Borders in the overhang region depending on the overhang angle θFigure 3. (a) Profile of the volumetric heat source, (b) the model geometry of single-track printing on a solid substrate (unit: µm), and (c) the comparison of melt pool dimensions obtained from the experiment (right half) and simulation (left half) for a calibrated optical penetration depth of 110 µm (laser power 200 W and scan speed 800 mm/s, solidified layer thickness 30 µm, powder size 10–45 µm).Figure 4. The model geometry of an overhang being L-PBF processed: (a) 3D view and (b) right view.Figure 5. The cross-sectional contour of border tracks in a 45° overhang region.Figure 6. Evolution of melt pool in the overhang region (θ = 45°, P = 100 W, v = 1000 mm/s, the streamlines are shown by arrows).Figure 7. The overhang contour is contributed by (a) only outer borders when θ ≤ 60° (b) both inner borders and outer borders when θ > 60°.Figure 8. Schematic of powder adhesion on a 45° overhang region.Figure 9. The L-PBF printed samples with various overhang angle (a) θ = 0° (cube), (b) θ = 30°, (c) θ = 45°, (d) θ = 55° and (e) θ = 60°.Figure 10. Two mechanisms of powder adhesion related to the overhang angle: (a) simulation-predicted, θ = 45°; (b) simulation-predicted, θ = 60°; (c, e) optical micrographs, θ = 45°; (d, f) optical micrographs, θ = 60°. (e) and (f) are partial enlargement of (c) and (d), respectively.Figure 11. Simulation-predicted surface morphology in the overhang region at different overhang angle: (a) θ = 15°, (b) θ = 30°, (c) θ = 45°, (d) θ = 60° and (e) θ = 80° (Blue solid lines: simulation-predicted contour; red dashed lines: the planar profile of designed overhang region specified by the overhang angles).Figure 12. Effect of overhang angle on surface roughness Ra in overhang regionsFigure 13. Surface morphology of L-PBF printed overhang regions with different overhang angle: (a) θ = 15°, (b) θ = 30°, (c) θ = 45° and (d) θ = 60° (overhang border parameters: P = 100 W, v = 1000 mm/s).Figure 14. Effect of (a) laser power (scan speed = 1000 mm/s) and (b) scan speed (lase power = 100 W) on surface roughness Ra in overhang regions (θ = 45°, laser power and scan speed referred to overhang border parameters, and the other process parameters are listed in Table 2).
References
Cai, Chao, Chrupcala Radoslaw, Jinliang Zhang, Qian Yan, Shifeng Wen, Bo Song, and Yusheng Shi. 2019. “In-Situ Preparation and Formation of TiB/Ti-6Al-4V Nanocomposite via Laser Additive Manufacturing: Microstructure Evolution and Tribological Behavior.” Powder Technology 342: 73–84. doi:10.1016/j.powtec.2018.09.088. [Crossref], [Web of Science ®], [Google Scholar]
Cai, Chao, Wei Shian Tey, Jiayao Chen, Wei Zhu, Xingjian Liu, Tong Liu, Lihua Zhao, and Kun Zhou. 2021. “Comparative Study on 3D Printing of Polyamide 12 by Selective Laser Sintering and Multi Jet Fusion.” Journal of Materials Processing Technology 288 (August 2020): 116882. doi:10.1016/j.jmatprotec.2020.116882. [Crossref], [Web of Science ®], [Google Scholar]
Cai, Chao, Xu Wu, Wan Liu, Wei Zhu, Hui Chen, Jasper Dong Qiu Chua, Chen Nan Sun, Jie Liu, Qingsong Wei, and Yusheng Shi. 2020. “Selective Laser Melting of Near-α Titanium Alloy Ti-6Al-2Zr-1Mo-1V: Parameter Optimization, Heat Treatment and Mechanical Performance.” Journal of Materials Science and Technology 57: 51–64. doi:10.1016/j.jmst.2020.05.004. [Crossref], [Web of Science ®], [Google Scholar]
Charles, Amal, Ahmed Elkaseer, Lore Thijs, and Steffen G. Scholz. 2020. “Dimensional Errors Due to Overhanging Features in Laser Powder Bed Fusion Parts Made of Ti-6Al-4V.” Applied Sciences 10 (7): 2416. doi:10.3390/app10072416. [Crossref], [Google Scholar]
Chatham, Camden A., Timothy E. Long, and Christopher B. Williams. 2019. “A Review of the Process Physics and Material Screening Methods for Polymer Powder Bed Fusion Additive Manufacturing.” Progress in Polymer Science 93: 68–95. doi:10.1016/j.progpolymsci.2019.03.003. [Crossref], [Web of Science ®], [Google Scholar]
Du, Yang, Xinyu You, Fengbin Qiao, Lijie Guo, and Zhengwu Liu. 2019. “A Model for Predicting the Temperature Field during Selective Laser Melting.” Results in Physics 12 (November 2018): 52–60. doi:10.1016/j.rinp.2018.11.031. [Crossref], [Web of Science ®], [Google Scholar]
Duval-Chaneac, M. S., S. Han, C. Claudin, F. Salvatore, J. Bajolet, and J. Rech. 2018. “Experimental Study on Finishing of Internal Laser Melting (SLM) Surface with Abrasive Flow Machining (AFM).” Precision Engineering 54 (July 2017): 1–6. doi:10.1016/j.precisioneng.2018.03.006. [Crossref], [Web of Science ®], [Google Scholar]
Feng, Shaochuan, Shijie Chen, Amar M. Kamat, Ru Zhang, Mingji Huang, and Liangcai Hu. 2020. “Investigation on Shape Deviation of Horizontal Interior Circular Channels Fabricated by Laser Powder Bed Fusion.” Additive Manufacturing 36 (December): 101585. doi:10.1016/j.addma.2020.101585. [Crossref], [Web of Science ®], [Google Scholar]
Feng, Shaochuan, Chuanzhen Huang, Jun Wang, Hongtao Zhu, Peng Yao, and Zhanqiang Liu. 2017. “An Analytical Model for the Prediction of Temperature Distribution and Evolution in Hybrid Laser-Waterjet Micro-Machining.” Precision Engineering 47: 33–45. doi:10.1016/j.precisioneng.2016.07.002. [Crossref], [Web of Science ®], [Google Scholar]
Feng, Shaochuan, Amar M. Kamat, and Yutao Pei. 2021. “Design and Fabrication of Conformal Cooling Channels in Molds: Review and Progress Updates.” International Journal of Heat and Mass Transfer. doi:10.1016/j.ijheatmasstransfer.2021.121082. [Crossref], [PubMed], [Web of Science ®], [Google Scholar]
Flow-3D V11.2 Documentation. 2016. Flow Science, Inc. [Crossref], [Google Scholar]
Foroozmehr, Ali, Mohsen Badrossamay, Ehsan Foroozmehr, and Sa’id Golabi. 2016. “Finite Element Simulation of Selective Laser Melting Process Considering Optical Penetration Depth of Laser in Powder Bed.” Materials and Design 89: 255–263. doi:10.1016/j.matdes.2015.10.002. [Crossref], [Web of Science ®], [Google Scholar]
“Geometrical Product Specifications (GPS) — Surface Texture: Profile Method — Rules and Procedures for the Assessment of Surface Texture (ISO 4288).” 1996. International Organization for Standardization. https://www.iso.org/standard/2096.html. [Google Scholar]
Günther, Johannes, Stefan Leuders, Peter Koppa, Thomas Tröster, Sebastian Henkel, Horst Biermann, and Thomas Niendorf. 2018. “On the Effect of Internal Channels and Surface Roughness on the High-Cycle Fatigue Performance of Ti-6Al-4V Processed by SLM.” Materials & Design 143: 1–11. doi:10.1016/j.matdes.2018.01.042. [Crossref], [Web of Science ®], [Google Scholar]
Hopkinson, Neil, and Phill Dickens. 2000. “Conformal Cooling and Heating Channels Using Laser Sintered Tools.” In Solid Freeform Fabrication Conference, 490–497. Texas. doi:10.26153/tsw/3075. [Crossref], [Google Scholar]
Hu, Zhiheng, Haihong Zhu, Changchun Zhang, Hu Zhang, Ting Qi, and Xiaoyan Zeng. 2018. “Contact Angle Evolution during Selective Laser Melting.” Materials and Design 139: 304–313. doi:10.1016/j.matdes.2017.11.002. [Crossref], [Web of Science ®], [Google Scholar]
Hu, Cheng, Kejia Zhuang, Jian Weng, and Donglin Pu. 2019. “Three-Dimensional Analytical Modeling of Cutting Temperature for Round Insert Considering Semi-Infinite Boundary and Non-Uniform Heat Partition.” International Journal of Mechanical Sciences 155 (October 2018): 536–553. doi:10.1016/j.ijmecsci.2019.03.019. [Crossref], [Web of Science ®], [Google Scholar]
Huang, Yuze, Mir Behrad Khamesee, and Ehsan Toyserkani. 2019. “A New Physics-Based Model for Laser Directed Energy Deposition (Powder-Fed Additive Manufacturing): From Single-Track to Multi-Track and Multi-Layer.” Optics & Laser Technology 109 (August 2018): 584–599. doi:10.1016/j.optlastec.2018.08.015. [Crossref], [Web of Science ®], [Google Scholar]
Kadirgama, K., W. S. W. Harun, F. Tarlochan, M. Samykano, D. Ramasamy, Mohd Zaidi Azir, and H. Mehboob. 2018. “Statistical and Optimize of Lattice Structures with Selective Laser Melting (SLM) of Ti6AL4V Material.” International Journal of Advanced Manufacturing Technology 97 (1–4): 495–510. doi:10.1007/s00170-018-1913-1. [Crossref], [Web of Science ®], [Google Scholar]
Kamat, Amar M, and Yutao Pei. 2019. “An Analytical Method to Predict and Compensate for Residual Stress-Induced Deformation in Overhanging Regions of Internal Channels Fabricated Using Powder Bed Fusion.” Additive Manufacturing 29 (March): 100796. doi:10.1016/j.addma.2019.100796. [Crossref], [Web of Science ®], [Google Scholar]
Kamath, Chandrika, Bassem El-Dasher, Gilbert F. Gallegos, Wayne E. King, and Aaron Sisto. 2014. “Density of Additively-Manufactured, 316L SS Parts Using Laser Powder-Bed Fusion at Powers up to 400 W.” International Journal of Advanced Manufacturing Technology 74 (1–4): 65–78. doi:10.1007/s00170-014-5954-9. [Crossref], [Web of Science ®], [Google Scholar]
Karimi, J., C. Suryanarayana, I. Okulov, and K. G. Prashanth. 2020. “Selective Laser Melting of Ti6Al4V: Effect of Laser Re-Melting.” Materials Science and Engineering A (July): 140558. doi:10.1016/j.msea.2020.140558. [Crossref], [Web of Science ®], [Google Scholar]
Khairallah, Saad A., and Andy Anderson. 2014. “Mesoscopic Simulation Model of Selective Laser Melting of Stainless Steel Powder.” Journal of Materials Processing Technology 214 (11): 2627–2636. doi:10.1016/j.jmatprotec.2014.06.001. [Crossref], [Web of Science ®], [Google Scholar]
Khairallah, Saad A., Andrew T. Anderson, Alexander Rubenchik, and Wayne E. King. 2016. “Laser Powder-Bed Fusion Additive Manufacturing: Physics of Complex Melt Flow and Formation Mechanisms of Pores, Spatter, and Denudation Zones.” Edited by Adedeji B. Badiru, Vhance V. Valencia, and David Liu. Acta Materialia 108 (April): 36–45. doi:10.1016/j.actamat.2016.02.014. [Crossref], [Web of Science ®], [Google Scholar]
Kuo, C. N., C. K. Chua, P. C. Peng, Y. W. Chen, S. L. Sing, S. Huang, and Y. L. Su. 2020. “Microstructure Evolution and Mechanical Property Response via 3D Printing Parameter Development of Al–Sc Alloy.” Virtual and Physical Prototyping 15 (1): 120–129. doi:10.1080/17452759.2019.1698967. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
Le, K. Q., C. H. Wong, K. H. G. Chua, C. Tang, and H. Du. 2020. “Discontinuity of Overhanging Melt Track in Selective Laser Melting Process.” International Journal of Heat and Mass Transfer 162 (December): 120284. doi:10.1016/j.ijheatmasstransfer.2020.120284. [Crossref], [Web of Science ®], [Google Scholar]
Lee, Y. S., and W. Zhang. 2016. “Modeling of Heat Transfer, Fluid Flow and Solidification Microstructure of Nickel-Base Superalloy Fabricated by Laser Powder Bed Fusion.” Additive Manufacturing 12: 178–188. doi:10.1016/j.addma.2016.05.003. [Crossref], [Web of Science ®], [Google Scholar]
Leitz, K. H., P. Singer, A. Plankensteiner, B. Tabernig, H. Kestler, and L. S. Sigl. 2017. “Multi-Physical Simulation of Selective Laser Melting.” Metal Powder Report 72 (5): 331–338. doi:10.1016/j.mprp.2016.04.004. [Crossref], [Google Scholar]
Li, Jian, Jing Hu, Yi Zhu, Xiaowen Yu, Mengfei Yu, and Huayong Yang. 2020. “Surface Roughness Control of Root Analogue Dental Implants Fabricated Using Selective Laser Melting.” Additive Manufacturing 34 (September 2019): 101283. doi:10.1016/j.addma.2020.101283. [Crossref], [Web of Science ®], [Google Scholar]
Li, Yingli, Kun Zhou, Pengfei Tan, Shu Beng Tor, Chee Kai Chua, and Kah Fai Leong. 2018. “Modeling Temperature and Residual Stress Fields in Selective Laser Melting.” International Journal of Mechanical Sciences 136 (February): 24–35. doi:10.1016/j.ijmecsci.2017.12.001. [Crossref], [Web of Science ®], [Google Scholar]
Mazur, MacIej, Martin Leary, Matthew McMillan, Joe Elambasseril, and Milan Brandt. 2016. “SLM Additive Manufacture of H13 Tool Steel with Conformal Cooling and Structural Lattices.” Rapid Prototyping Journal 22 (3): 504–518. doi:10.1108/RPJ-06-2014-0075. [Crossref], [Web of Science ®], [Google Scholar]
Mingear, Jacob, Bing Zhang, Darren Hartl, and Alaa Elwany. 2019. “Effect of Process Parameters and Electropolishing on the Surface Roughness of Interior Channels in Additively Manufactured Nickel-Titanium Shape Memory Alloy Actuators.” Additive Manufacturing 27 (October 2018): 565–575. doi:10.1016/j.addma.2019.03.027. [Crossref], [Web of Science ®], [Google Scholar]
Pakkanen, Jukka, Flaviana Calignano, Francesco Trevisan, Massimo Lorusso, Elisa Paola Ambrosio, Diego Manfredi, and Paolo Fino. 2016. “Study of Internal Channel Surface Roughnesses Manufactured by Selective Laser Melting in Aluminum and Titanium Alloys.” Metallurgical and Materials Transactions A 47 (8): 3837–3844. doi:10.1007/s11661-016-3478-7. [Crossref], [Web of Science ®], [Google Scholar]
Patterson, Albert E., Sherri L. Messimer, and Phillip A. Farrington. 2017. “Overhanging Features and the SLM/DMLS Residual Stresses Problem: Review and Future Research Need.” Technologies 5 (4): 15. doi:10.3390/technologies5020015. [Crossref], [Web of Science ®], [Google Scholar]
Roberts, I. A., C. J. Wang, R. Esterlein, M. Stanford, and D. J. Mynors. 2009. “A Three-Dimensional Finite Element Analysis of the Temperature Field during Laser Melting of Metal Powders in Additive Layer Manufacturing.” International Journal of Machine Tools and Manufacture 49 (12–13): 916–923. doi:10.1016/j.ijmachtools.2009.07.004. [Crossref], [Web of Science ®], [Google Scholar]
Shrestha, Subin, and Kevin Chou. 2018. “Computational Analysis of Thermo-Fluid Dynamics with Metallic Powder in SLM.” In CFD Modeling and Simulation in Materials Processing 2018, edited by Laurentiu Nastac, Koulis Pericleous, Adrian S. Sabau, Lifeng Zhang, and Brian G. Thomas, 85–95. Cham, Switzerland: Springer Nature. doi:10.1007/978-3-319-72059-3_9. [Crossref], [Google Scholar]
Sing, S. L., and W. Y. Yeong. 2020. “Laser Powder Bed Fusion for Metal Additive Manufacturing: Perspectives on Recent Developments.” Virtual and Physical Prototyping 15 (3): 359–370. doi:10.1080/17452759.2020.1779999. [Taylor & Francis Online], [Web of Science ®], [Google Scholar]
Šmilauer, Václav, Emanuele Catalano, Bruno Chareyre, Sergei Dorofeenko, Jérôme Duriez, Nolan Dyck, Jan Eliáš, et al. 2015. Yade Documentation. 2nd ed. The Yade Project. doi:10.5281/zenodo.34073. [Crossref], [Google Scholar]
Tan, Pengfei, Fei Shen, Biao Li, and Kun Zhou. 2019. “A Thermo-Metallurgical-Mechanical Model for Selective Laser Melting of Ti6Al4V.” Materials & Design 168 (April): 107642. doi:10.1016/j.matdes.2019.107642. [Crossref], [Web of Science ®], [Google Scholar]
Tan, Lisa Jiaying, Wei Zhu, and Kun Zhou. 2020. “Recent Progress on Polymer Materials for Additive Manufacturing.” Advanced Functional Materials 30 (43): 1–54. doi:10.1002/adfm.202003062. [Crossref], [Web of Science ®], [Google Scholar]
Wang, Xiaoqing, and Kevin Chou. 2018. “Effect of Support Structures on Ti-6Al-4V Overhang Parts Fabricated by Powder Bed Fusion Electron Beam Additive Manufacturing.” Journal of Materials Processing Technology 257 (February): 65–78. doi:10.1016/j.jmatprotec.2018.02.038. [Crossref], [Web of Science ®], [Google Scholar]
Wang, Di, Yongqiang Yang, Ziheng Yi, and Xubin Su. 2013. “Research on the Fabricating Quality Optimization of the Overhanging Surface in SLM Process.” International Journal of Advanced Manufacturing Technology 65 (9–12): 1471–1484. doi:10.1007/s00170-012-4271-4. [Crossref], [Web of Science ®], [Google Scholar]
Wen, Peng, Maximilian Voshage, Lucas Jauer, Yanzhe Chen, Yu Qin, Reinhart Poprawe, and Johannes Henrich Schleifenbaum. 2018. “Laser Additive Manufacturing of Zn Metal Parts for Biodegradable Applications: Processing, Formation Quality and Mechanical Properties.” Materials and Design 155: 36–45. doi:10.1016/j.matdes.2018.05.057. [Crossref], [Web of Science ®], [Google Scholar]
Wu, Yu-che, Cheng-hung San, Chih-hsiang Chang, Huey-jiuan Lin, Raed Marwan, Shuhei Baba, and Weng-Sing Hwang. 2018. “Numerical Modeling of Melt-Pool Behavior in Selective Laser Melting with Random Powder Distribution and Experimental Validation.” Journal of Materials Processing Technology 254 (November 2017): 72–78. doi:10.1016/j.jmatprotec.2017.11.032. [Crossref], [Web of Science ®], [Google Scholar]
Yadroitsev, I., P. Krakhmalev, I. Yadroitsava, S. Johansson, and I. Smurov. 2013. “Energy Input Effect on Morphology and Microstructure of Selective Laser Melting Single Track from Metallic Powder.” Journal of Materials Processing Technology 213 (4): 606–613. doi:10.1016/j.jmatprotec.2012.11.014. [Crossref], [Web of Science ®], [Google Scholar]
Yu, Wenhui, Swee Leong Sing, Chee Kai Chua, and Xuelei Tian. 2019. “Influence of Re-Melting on Surface Roughness and Porosity of AlSi10Mg Parts Fabricated by Selective Laser Melting.” Journal of Alloys and Compounds 792: 574–581. doi:10.1016/j.jallcom.2019.04.017. [Crossref], [Web of Science ®], [Google Scholar]
Zhang, Dongyun, Pudan Zhang, Zhen Liu, Zhe Feng, Chengjie Wang, and Yanwu Guo. 2018. “Thermofluid Field of Molten Pool and Its Effects during Selective Laser Melting (SLM) of Inconel 718 Alloy.” Additive Manufacturing 21 (100): 567–578. doi:10.1016/j.addma.2018.03.031. [Crossref], [Web of Science ®], [Google Scholar]
Won-Ik Cho, Peer Woizeschke Bremer Institut für angewandte Strahltechnik GmbH, Klagenfurter Straße 5, Bremen 28359, Germany
Received 30 July 2020, Revised 3 October 2020, Accepted 18 October 2020, Available online 1 November 2020.
Abstract
Molten pool flow and heat transfer in a laser welding process using beam oscillation and filler wire feeding were calculated using computational fluid dynamics (CFD). There are various indirect methods used to analyze the molten pool dynamics in fusion welding. In this work, based on the simulation results, the surface fluctuation was directly measured to enable a more intuitive analysis, and then the signal was analyzed using the Fourier transform and wavelet transform in terms of the beam oscillation frequency and buttonhole formation. The 1st frequency (2 x beam oscillation frequency, the so-called chopping frequency), 2nd frequency (4 x beam oscillation frequency), and beam oscillation frequency components were the main components found. The 1st and 2nd frequency components were caused by the effect of the chopping process and lumped line energy. The beam oscillation frequency component was related to rapid, unstable molten pool behavior. The wavelet transform effectively analyzed the rapid behaviors based on the change of the frequency components over time.
Korea Abstract
빔 진동 및 필러 와이어 공급을 사용하는 레이저 용접 공정에서 용융 풀 흐름 및 열 전달은 CFD (전산 유체 역학)를 사용하여 계산되었습니다. 용융 용접에서 용융 풀 역학을 분석하는 데 사용되는 다양한 간접 방법이 있습니다.
본 연구에서는 시뮬레이션 결과를 바탕으로 보다 직관적 인 분석이 가능하도록 표면 변동을 직접 측정 한 후 빔 발진 주파수 및 버튼 홀 형성 측면에서 푸리에 변환 및 웨이블릿 변환을 사용하여 신호를 분석했습니다.
1 차 주파수 (2 x 빔 발진 주파수, 이른바 초핑 주파수), 2 차 주파수 (4 x 빔 발진 주파수) 및 빔 발진 주파수 성분이 발견 된 주요 구성 요소였습니다. 1 차 및 2 차 주파수 성분은 쵸핑 공정과 집중 라인 에너지의 영향으로 인해 발생했습니다.
빔 진동 주파수 성분은 빠르고 불안정한 용융 풀 동작과 관련이 있습니다. 웨이블릿 변환은 시간 경과에 따른 주파수 구성 요소의 변화를 기반으로 빠른 동작을 효과적으로 분석했습니다.
1 . 소개
융합 용접에서 용융 풀 역학은 용접 결함과 시각적 이음새 품질에 직접적인 영향을 미칩니다. 이러한 역학을 연구하기 위해 고속 카메라를 사용하는 직접 방법과 광학 또는 음향 신호를 사용하는 간접 방법과 같은 다양한 측정 방법을 사용하여 여러 실험 방법을 고려했습니다. 시간 도메인의 원래 신호는 특별히 주파수 도메인에서 변환 된 신호로 변환되어 용융 풀 동작에 영향을 미치는 주파수 성분을 분석합니다. Kotecki et al. (1972)는 고속 카메라를 사용하여 가스 텅스텐 아크 용접에서 용융 풀을 관찰했습니다. [1]. 그들은 120Hz 리플 DC 출력을 가진 용접 전원을 사용할 때 용융 풀 진동 주파수가 120Hz임을 보여주었습니다. 전원을 끈 후 진동 주파수는 용융 풀의 고유 주파수를 나타내는 용융 풀 크기와 관련이 있습니다. 진동은 응고 중에 용접 표면 스케일링을 생성했습니다. Zacksenhouse and Hardt (1983)는 레이저 섀도 잉 동작 측정 기술을 사용하여 가스 텅스텐 아크 용접에서 완전히 관통 된 용융 풀의 동작을 측정했습니다 [2] . 그들은 2.5mm 두께의 강판에서 6mm 풀 반경 (고정 용접)에 대해 용융 풀의 고유 주파수가 18.9Hz라는 것을 발견했습니다. Semak et al. (1995) 고속 카메라를 사용하여 레이저 스폿 용접에서 용융 풀 및 키홀 역학 조사 [3]. 그들은 깊이가 약 3mm이고 반경이 약 3mm 인 용융 풀에서 200Hz의 낮은 체적 진동 주파수를 관찰했습니다. 0.45mm Aendenroomer와 den Ouden (1998)은 강철의 펄스 가스 텅스텐 아크 용접에서 용융 풀 진동을보고했습니다 [4] . 그들은 침투 깊이에 따라 진동 모드 변화를 보였고 주파수는 50Hz에서 150Hz 사이에서 변화했습니다. 주파수는 완전히 침투 된 용융 풀에서 더 낮았습니다. Hermans와 den Ouden (1999)은 단락 가스 금속 아크 용접에서 용융 풀 진동을 분석했습니다. [5]. 그들은 용융 풀의 단락 주파수와 고유 주파수가 같을 때 부분적으로 침투 된 용융 풀의 경우 공정 안정성이 향상되었음을 보여주었습니다. Yudodibroto et al. (2004)는 가스 텅스텐 아크 용접에서 용융 풀 진동에 대한 필러 와이어의 영향을 조사했습니다 [6] . 그들은 금속 전달이 특히 부분적으로 침투 된 용융 풀에서 진동 거동을 방해한다는 것을 보여주었습니다. Geiger et al. (2009) 레이저 키홀 용접에서 발광 분석 [7]. 신호의 주파수 분석을 사용하여 용융 풀 (1.5kHz 미만)과 키홀 (약 3kHz)에 해당하는 진동 주파수 범위를 찾았습니다. Kägeler와 Schmidt (2010)는 레이저 용접에서 용융 풀 크기의 변화를 관찰하기 위해 고속 카메라를 사용했습니다 [8] . 그들은 용융 풀에서 지배적 인 저주파 진동 성분 (100Hz 미만)을 발견했습니다. Shi et al. (2015) 고속 카메라를 사용하여 펄스 가스 텅스텐 아크 용접에서 용융 풀 진동 주파수 분석 [9]. 그들은 용접 침투 깊이가 작을수록 용융 풀의 진동 빈도가 더 높다는 것을 보여주었습니다. 추출 된 진동 주파수는 완전 용입 용접의 경우 85Hz 미만 이었지만 부분 용입 용접의 경우 110Hz에서 125Hz 사이였습니다. Volpp와 Vollertsen (2016)은 레이저 키홀 역학을 분석하기 위해 광학 신호를 사용했습니다 [10] . 그들은 공간 레이저 강도 분포로 인해 0.8에서 154 kHz 사이의 고주파 범위에서 피크를 발견했습니다. 위에서 언급 한 실험적 접근법은 공정 조건, 측정 방법 및 측정 된 위치에 따라 수십 Hz에서 수십 kHz까지 광범위한 용융 풀 역학에 대한 결과를 보여 주었다는 점에 유의해야합니다.
융합 용접에서 용융 풀 역학을 연구하기 위해 분석 접근 방식도 사용되었습니다. Zacksenhouse와 Hardt (1983)는 2.5mm 두께의 강판에서 대칭형 완전 관통 용융 풀의 고유 진동수를 계산했습니다 [2] . 매스 스프링 해석 모델을 사용하여 용융 풀 반경 6mm (고정 용접)에 대해 20.4Hz (실험에서 18.9Hz)의 고유 진동수와 3mm 풀 반경 (연속 용접)에 대해 40Hz의 고유 진동수를 예측했습니다. ). Postacioglu et al. (1989)는 원통형 용융 풀과 키홀을 가정하여 레이저 용접의 용융 풀에서 키홀 진동의 고유 진동수를 계산했습니다 .. 특정 열쇠 구멍 모양의 경우 약 900Hz의 기본 주파수가 계산되었습니다. Postacioglu et al. (1991)은 또한 레이저 용접에서 용접 속도를 고려하기 위해 타원형 용융 풀의 고유 진동수를 계산했습니다 [12] . 그들은 타원형 용융 풀의 모양이 고유 진동수에 영향을 미친다는 것을 보여주었습니다. 고유 진동수는 축의 길이 비율이 낮았으며, 즉 타원의 반장 축과 반 단축의 비율이 낮았습니다. Kroos et al. (1993)은 축 대칭 용융 풀과 키홀을 가정하여 레이저 키홀 용접의 동적 거동에 대한 이론적 모델을 개발했습니다 .. 키홀 폐쇄 시간은 0.1ms였으며 안정성 분석은 약 500Hz의 주파수에서 공진과 같은 진동을 예측했습니다. Maruo와 Hirata (1993)는 완전 관통 아크 용접에서 용융 풀을 모델링했습니다 [14] . 그들은 녹은 웅덩이가 정적 타원 모양을 가지고 있다고 가정했습니다. 그들은 고유 진동수와 진동 모드 사이의 관계를 조사하고 용융 풀 크기가 감소함에 따라 고유 진동수가 증가한다는 것을 보여주었습니다. Klein et al. (1994)는 원통형 키홀 모양을 사용하여 완전 침투 레이저 용접에서 키홀 진동을 연구했습니다 [15] . 그들은 점성 감쇠로 인해 키홀 진동이 낮은 kHz 범위로 제한된다는 것을 보여주었습니다. Klein et al. (1996)은 또한 레이저 출력의 작은 변동이 강한 키홀 진동으로 이어질 수 있음을 보여주었습니다[16] . 그들은 키홀 진동의 주요 공진 주파수 범위가 500 ~ 3500Hz라는 것을 발견했습니다. Andersen et al. (1997)은 고정 가스 텅스텐 아크 용접 [17] 에서 고정 된 원통형 모양을 가정하여 용융 풀의 고유 진동수를 예측 했으며 완전 용입 용접에서 용융 풀 폭이 증가함에 따라 감소하는 것으로 나타났습니다. 3.175mm 두께의 강판의 경우 주파수는 20Hz ~ 100Hz 범위였습니다. 위에 표시된 분석 방법은 일반적으로 단순한 용융 풀 모양을 가정하고 고유 진동수를 계산했습니다. 이것은 단순한 용융 풀 모양으로 고정 용접 공정을 분석하는 데 충분하지만 대부분의 용접 사례를 설명하는 과도 용접 공정에서 용융 풀 역학 분석에는 적합하지 않습니다.
반면에 수치 접근 방식은 고온 및 강한 빛과 같은 실험적 제한없이 자세한 정보를 제공하기 때문에 용융 풀 역학을 분석하는 이점이 있습니다. 전산 유체 역학 (CFD)의 수치 시뮬레이션 기술이 발전함에 따라 용융 풀 역학 분석에 대한 많은 연구가 수행되었습니다. 실제 용융 표면 변화는 VOF (체적 부피) 방법을 사용하여 계산할 수 있습니다. Cho et al. (2010) CO 2 레이저-아크 하이브리드 용접 공정을 위한 수학적 모델 개발 [18], 구형 방울이 생성 된 금속 와이어의 용융 과정이 와이어 공급 속도와 일치한다고 가정합니다. 그들은 필러 와이어가 희석되는 용융 풀 동작을 보여주었습니다. Cho et al. (2012)는 높은 빔 품질과 높은 금속 흡수율로 인해 업계에서 널리 사용되는 디스크 레이저 키홀 용접으로 수학적 모델을 확장했습니다 [19] . 그들은 열쇠 구멍에서 레이저 광선 번들의 다중 반사를 고려하고 용융 풀에서 keyholing과 같은 빠른 표면 변화를 자세히보고했습니다. 최근 CFD 시뮬레이션은 험핑 (Otto et al., 2016 [20] ) 및 기공 (Lin et al., 2017 [21] )과 같은보다 구체적인 현상을 분석하는데도 사용되었습니다 .) 레이저 용접에서. 그러나 용융 풀 역학과 관련된 연구는 거의 수행되지 않았습니다. Ko et al. (2000)은 수치 시뮬레이션을 사용하여 가스 텅스텐 아크 용접 풀의 동적 거동을 조사했습니다 [22] . 그들은 완전히 침투 된 용융 풀이 부분적으로 침투 된 풀보다 낮은 주파수에서 진동한다는 것을 보여주었습니다. 진동은 수십 분의 1 초 내에 무시할 수있는 크기로 감쇠되었습니다. Geiger et al. (2009)는 또한 수치 시뮬레이션을 사용하여 레이저 용접에서 용융 풀 거동을 보여주었습니다 [7]. 그들은 계산 된 증발 속도를 주파수 분석에 사용하여 공정에서 나오는 빛의 실험 결과와 비교했습니다. 판금 레이저 용접에서 중요한 공간 빔 진동 및 추가 필러 재료가있는 공정에 대한 용융 풀 역학에 대한 연구도 불충분합니다. Hu et al. (2018)은 금속 전달 메커니즘을 밝히기 위해 전자빔 3D 프린팅에서 와이어 공급 모델링을 수행했습니다. 그들은 주로 열 입력에 의해 결정되는 액체 브리지 전이, 액적 전이 및 중간 전이의 세 가지 유형의 금속 전달 모드를 보여주었습니다 .. Meng et al. (2020)은 레이저 빔 용접에서 용융 풀에 필러 와이어에 의해 추가 된 추가 요소의 전자기 교반 효과를 모델링했습니다. 용가재의 연속적인 액체 브릿지 이동이 가정되었고, 그 결과 전자기 교반의 영향이 키홀 깊이에 미미한 반면 필러 와이어 혼합을 향상 시켰습니다 [24] . Cho et al. (2017) 용접 방향에 수직 인 1 차원 빔 진동과 용접 라인을 따라 공급되는 필러 와이어를 사용하여 레이저 용접을위한 시뮬레이션 모델 개발 [25]. 그들은 시뮬레이션을 사용하여 특정 용접 현상, 즉 용융 풀의 단추 구멍 형성을 보여주었습니다. Cho et al. (2018)은 다중 반사 수와 전력 흡수량의 푸리에 변환을 사용하여 주파수 영역에서 소위 쵸핑 주파수 (2 x 빔 발진 주파수) 성분을 발견했습니다 [26] . 그러나 그들은 용융 풀 역학을 분석하기 위해 간접 신호를 사용했습니다. 따라서보다 직관적 인 분석을 위해서는 표면의 변동을 직접 측정해야합니다.
이 연구는 이전 연구에서 개발 된 레이저 용접 모델을 사용하여 3 차원 과도 CFD 시뮬레이션을 수행하여 빔 진동 및 필러 와이어 공급을 포함한 레이저 용접 공정에서 용융 풀 역학을 조사합니다. 용융 된 풀 표면의 시간적 변화는 시뮬레이션 결과에서 추출되었습니다. 추출 된 데이터는 주파수 영역뿐만 아니라 시간-주파수 영역에서도 분석되었습니다. 신호 처리를 통해 도출 된 결과는 특징적인 용융 풀 역학을 나타내며 빔 진동 주파수 및 단추 구멍 형성 측면에서 레이저 용접의 역학을 줄일 수있는 잠재력을 제공합니다.
2 . 방법론
그림 1도 1은 용접 방향에 수직 인 1 차원 빔 진동과 용접 라인을 따라 공급되는 필러 와이어를 사용하는 레이저 용접 프로세스의 개략적 설명을 보여줍니다. 1mm 두께의 알루미늄 합금 (AlSi1MgMn) 시트는 시트 표면에 초점을 맞춘 멀티 kW 파이버 레이저 (YLR-8000S, IPG Photonics, USA)를 사용하여 용접되었습니다. 시트는 에어 갭이있는 맞대기 이음으로 정렬되었습니다. 1 차원 스캐너 (ILV DC-Scanner, Ingenieurbüro für Lasertechnik + Verschleiss-Schutz (ILV), 독일)를 사용하여 레이저 빔의 1 차원 정현파 진동을 실현했습니다. 이 스캔 시스템에서 최대 진동 폭은 250Hz의 진동 주파수에서 1.4mm입니다. 오정렬에 대한 공차를 개선하기 위해 동일한 최대 너비 값이 사용되었습니다. 와이어 공급 시스템은 1을 공급했습니다. 2mm 직경의 알루미늄 합금 (AlSi5) 필러 와이어를 일정한 공급 속도로 에어 갭을 채 웁니다. 1mm 에어 갭의 경우 와이어 이송 속도는 용접 속도의 1.5 배 값으로 설정되었으며 참조 실험 조건은 문헌에서 얻었습니다 (Schultz, 2015 참조).[27] ).
CFD 시뮬레이션은 레이저 용접에서 열 전달 및 용융 풀 동작을 계산하기 위해 수행되었습니다. 그림 2 는 CFD 시뮬레이션을위한 계산 영역을 보여줍니다. 실온에서 1.2mm 직경의 필러 와이어가 공급되고 레이저 빔이 진동했습니다. 1mm 두께의 공작물이 용접 속도로 왼쪽에서 오른쪽으로 이동했습니다. 0.1mm의 최소 메쉬 크기가 도메인에서 생성되었습니다. 침투 깊이가 더 깊은 이전 연구의 메쉬 테스트 결과는 0.2mm 이하의 메쉬 크기로 시뮬레이션 정확도가 확보 된 것으로 나타 났으므로 [28] 본 연구에서 사용 된 메쉬 크기가 적절할 수 있습니다. 도메인을 구성하는 세포의 수는 약 120 만 개였습니다. 1 번 테이블사용 된 레이저 용접 매개 변수를 보여줍니다. 용융 풀 역학 측면에서 다양한 진동 주파수와 에어 갭 크기가 고려되었으며 12 개의 용접 사례가 표 2 에 나와 있습니다. 표 3 은 시뮬레이션에 사용 된 알루미늄 합금과 순수 알루미늄 (Cho et al., 2018 [26] )의 표면 장력 계수를 제외하고 온도와 무관 한 열-물리적 재료 특성을 보여줍니다 . 여기서 표면 장력 계수는 액체 온도에서 온도와 표면 장력 계수 사이의 선형 관계를 가진 유일한 온도 의존적 특성이었습니다.
시뮬레이션을 위해 단상 뉴턴 유체와 비압축성 층류가 가정되었습니다. 질량, 운동량 및 에너지 보존의 지배 방정식을 해결하여 계산 영역에서 속도, 압력 및 온도 분포를 얻었습니다. VOF 방법은 자유 표면 경계를 찾는 데 사용되었습니다. 스칼라 보존 방정식을 추가로 도입하여 용융 풀에서 충전재의 부피 분율을 계산했습니다. 시뮬레이션에 사용 된 레이저 용접의 수학적 모델은 다음과 같습니다. 레이저 빔은 가우스와 같은 전력 밀도 분포를 기반으로 697 개의 광선 에너지 번들로 나뉩니다. 광선 추적 방법을 사용하여 다중 반사를 고려했습니다. 재료에 대한 레이저 빔의 반사 (또는 흡수) 에너지는 프레 넬 반사 모델을 사용하여 계산되었습니다. 온도에 따른 흡수율의 변화를 고려 하였다. 혼합물의 흡수율은베이스 및 충전제 물질 분획의 가중 평균을 사용하여 계산되었습니다. 반동 압력과 부력도 고려되었습니다. 경계 조건으로 에너지와 압력의 균형은 VOF 방법으로 계산 된 자유 표면에서 고려되었습니다. 레이저 용접 모델과 지배 방정식은 FLOW-3D v.11.2 (2017), Flow Science, Inc.에서 유한 차분 방법과 유한 체적 방법을 사용하여 이산화되고 해결되었습니다. 경계 조건으로 에너지와 압력의 균형은 VOF 방법으로 계산 된 자유 표면에서 고려되었습니다. 레이저 용접 모델과 지배 방정식은 FLOW-3D v.11.2 (2017), Flow Science, Inc.에서 유한 차분 방법과 유한 체적 방법을 사용하여 이산화되고 해결되었습니다. 경계 조건으로 에너지와 압력의 균형은 VOF 방법으로 계산 된 자유 표면에서 고려되었습니다. 레이저 용접 모델과 지배 방정식은 FLOW-3D v.11.2 (2017), Flow Science, Inc.에서 유한 차분 방법과 유한 체적 방법을 사용하여 이산화되고 해결되었습니다.[29] . 계산에는 48GB RAM이 장착 된 Intel® Xeon® 프로세서 E5649로 구성된 워크 스테이션이 사용되었습니다. 계산 시스템을 사용하여 0.2 초 레이저 용접을 시뮬레이션하는 데 약 18 시간이 걸렸습니다. 지배 방정식 (Cho and Woizeschke, 2020 [30] ) 및 레이저 용접 모델 (Cho et al., 2018 [26] )에 대한 자세한 설명은 부록 A 에서 확인할 수 있습니다 .
그림 3 은 용융 풀 변동의 직접 측정에 대한 개략적 설명을 보여줍니다. 용융 풀의 역학을 분석하기 위해 시뮬레이션 중에 용융 풀 표면의 시간적 변동 운동을 측정했습니다. 상단 및 하단 표면 모두에서 10kHz의 샘플링 주파수로 변동을 측정 한 반면, 측정 위치는 X 축의 레이저 빔 위치에서 2mm 떨어진 용접 중심선에있었습니다. 그림 4시간 신호를 분석하는 데 사용되는 푸리에 변환 및 웨이블릿 변환의 개략적 설명을 보여줍니다. 측정 된 시간 신호는 고속 푸리에 변환 (FFT) 방법을 사용하여 주파수 영역으로 변환되었습니다. 결과는 측정 기간 동안 평균화 된 주파수 성분의 크기를 보여줍니다. 웨이블릿 변환 방법은 시간-주파수 영역에서 국부적 인 특성을 찾는 데 사용되었습니다. 결과는 주파수 구성 요소의 크기뿐만 아니라 시간 변화도 보여줍니다.
3 . 결과
이 연구 에서는 표 2에 표시된 12 가지 용접 사례 를 시뮬레이션했습니다. 그림 5 는 3 차원 시뮬레이션 결과를 평면도 와 바닥면으로 보여줍니다. 결과는 용융 된 풀의 거동에 따라 분류 할 수 있습니다 : 단추 구멍 형성 없음 (녹색), 안정 또는 불안정 단추 구멍 있음 (파란색), 불안정한 단추 구멍으로 인한 구멍 결함 (빨간색). 일반적인 열쇠 구멍보다 훨씬 큰 직경을 가진 단추 구멍은 레이저 용접의 특정 진동 조건에서 나타날 수 있습니다 (Vollertsen, 2016 [31]). 진동 주파수가 증가함에 따라 용접 이음 부 코스 및 스케일링 측면에서 시각적 이음새 품질이 향상되었습니다. 고주파에서 스케일링은 무시할 수있을 정도 였고 코스는 균질했습니다. 언더컷 결함의 발생도 감소했습니다. 그러나 관통 결함 부족 (case 7, case 10)이 나타났다. 에어 갭은 단추 구멍 형성에 중요했습니다. 에어 갭 크기가 증가함에 따라 단추 구멍이 더 쉽게 형성되었지만 구멍 결함으로 더 쉽게 남아 있습니다. 안정적인 단추 구멍 형성은 고려 된 공극 조건의 좁은 영역에서만 나타납니다.
그림 6 은 시뮬레이션과 실험에서 융합 영역의 모양을 보여줍니다. 버튼 홀이없는 경우 1, 불안정한 버튼 홀 형성이있는 경우 8, 안정적인 버튼 홀 형성이있는 경우 11의 3 가지 경우에 대해 시뮬레이션 결과와 실험 결과를 비교하여 유사성을 나타냈다. 본 연구에서 고려한 용접 조건의 경우 표면 품질 결과는 Fig. 5 와 같이 큰 차이를 보였으 나 단면 융착 영역 [26] 과 형상은 큰 차이를 보이지 않았다.
무화과. 7 과 8 은 각각 100Hz와 250Hz의 진동 주파수에서 시뮬레이션 결과를 기반으로 분석 된 용융 풀 역학과 시뮬레이션 및 실험 결과를 보여줍니다. 이전 연구에서 볼 수 있듯이 레이저 빔의 진동 주파수는 단추 구멍 형성과 밀접한 관련이 있습니다 (Cho et al., 2018 [26] 참조 ). 그림 7 (a) 및 (b)는 각각 시뮬레이션 및 실험을 기반으로 한 진동 주파수 100Hz에서 대표적인 용융 풀 동작을 보여줍니다. 완전히 관통 된 키홀 및 버튼 홀 형성은 관찰되지 않았으며 응고 후 거친 비드 표면이 남았습니다. 그림 7(c)와 (d)는 각각 윗면과 바닥면의 표면 변동에 대한 시뮬레이션 결과를 기반으로 한 용융 풀 역학 분석을 보여줍니다. 샘플링 데이터는 상단 표면이 공작물의 상단 표면 위치에서 평균적으로 변동하는 반면 하단 표면은 공작물의 하단 표면 위치에서 평균적으로 변동하는 것으로 나타났습니다. 표면 변동의 푸리에 변환 및 웨이블릿 변환 결과는 명확한 1 차 주파수 (2 x 빔 발진 주파수, 이른바 초핑 주파수, Cho et al., 2018 [26] 참조 ) 및 2 차 주파수 (4 x 빔 발진)를 보여줍니다. 주파수) 두 표면의 구성 요소, 그러나 바닥 표면과 첫 번째에 대한 결과주파수 성분이 더 강합니다. 반면 그림 8 (a)와 (b)에서 보는 바와 같이 250Hz의 진동 주파수에서 시뮬레이션과 실험 결과는 안정된 버튼 홀 형성과 응고 후 매끄러운 비드 표면을 나타냈다. 그림 8 의 샘플링 신호의 진폭은 그림 7 의 진폭 보다 작으며 푸리에 변환 및 웨이블릿 변환의 결과에서 중요한 주파수 성분이 발견되지 않았습니다.
Fig. 9 는 진동 주파수 200Hz에서 시뮬레이션 결과를 바탕으로 분석 된 용융 풀 역학과 시뮬레이션 및 실험 결과를 보여준다. 이 주파수에서 Fig. 9 (a)와 (b) 에서 보는 바와 같이 , 시뮬레이션과 실험 모두에서 불안정한 buttonhole 거동이 관찰되었다. 바닥면에서 샘플링 데이터의 푸리에 변환 및 웨이블릿 변환의 결과 빔 발진 주파수 성분이 발견되었습니다.
4 . 토론
시뮬레이션 및 실험 결과는 비드 표면 품질이 향상되고 빔 진동 주파수가 증가함에 따라 버튼 홀이 형성되는 것으로 나타났습니다. 표면의 변동 데이터에 대한 푸리에 변환 및 웨이블릿 변환의 결과에 따라 다음과 같은 주요 주파수 구성 요소가 발견되었습니다. 1 차 및 2 차버튼 홀 형성이없는 주파수, 불안정한 용융 풀 거동이있는 빔 진동 주파수, 안정적인 버튼 홀 형성이있는 중요한 주파수 성분이 없습니다. 이들 중 불안정한 용융 풀 동작과 관련된 빔 진동 주파수 성분은 완전히 관통 된 키홀과 반복적으로 생성 및 붕괴되는 불안정한 버튼 홀의 특성으로 인해 웨이블릿 변환 결과에서 명확한 실선 형태로 나타나지 않았습니다. 분석 결과는 윗면보다 바닥면에서 더 분명했습니다. 이는 필러 와이어 공급 및 키홀 링 공정에서 강한 하향 흐름으로 인해 용융 풀 역학이 바닥 표면 영역에서 더 강했기 때문입니다. 진동 주파수가 증가함에 따라 용융 풀 역학과 상단 표면과 하단 표면 간의 차이가 감소했습니다.
첫 번째 주파수 (2 x 빔 진동 주파수)는이 연구에서 관찰 된 가장 분명한 구성 요소였습니다. Schultz et al. (2018)은 또한 실험을 통해 동일한 성분을 발견했습니다 [32] , 용융 풀 표면 운동에 대한 푸리에 분석을 수행했습니다. 첫 번째 주파수 성분은 빔 발진주기 당 두 개의 주요 이벤트가 있음을 의미합니다. 이것은 레이저 빔이 빔 진동주기 당 두 번 와이어를 절단하거나 절단하는 프로세스와 일치합니다. 용융 된 와이어 팁은 낮은 진동 주파수에서 고르지 않고 날카로운 모서리를 갖는 것으로 나타났습니다 (Cho et al., 2018 [26] ). 이것은 첫 번째 원인이 될 수 있습니다.용융 된 풀에서 지배적이되는 주파수 성분. 진동 주파수가 증가하면 용융 된 와이어 팁이 더 균일 해 지므로 효과가 감소합니다. 용접 방향으로의 정현파 횡 방향 빔 진동을 통한 에너지 집중도 빔 진동주기 당 두 번 발생합니다. 그림 10 은 발진 주파수에 따른 레이저 빔의 라인 에너지 (단위 길이 당 에너지)의 변화를 보여줍니다. 그림 10 b) 의 라인 에너지 는 레이저 출력을 공정 속도로 나누어 계산했습니다. 여기서 처리 속도는(w이자형엘디나는엔지에스피이자형이자형디)2+(디(에스나는엔유에스영형나는디ㅏ엘wㅏV이자형나는엔에프나는지.10ㅏ))디티)2. 낮은 발진 주파수에서 라인 에너지는 발진 폭의 양쪽 끝에 과도하게 집중됩니다. 이러한 집중된 에너지는 과도한 키홀 링 프로세스를 초래하므로 언더컷 결함이 나타날 수있는 높은 흐름 역학이 발생합니다. 진동 주파수가 증가함에 따라 집중 에너지는 더 작은 조각으로 나뉩니다. 따라서 높은 진동 주파수에서 과도한 키홀 링 및 수반되는 언더컷 결함의 발생이 감소되었습니다. 위에서 언급 한 두 가지 현상 (불균일 한 와이어 팁과 집중된 라인 에너지)은 빔 발진주기 당 두 번 발생하며 발진 주파수가 증가하면 그 효과가 감소합니다. 따라서 저주파 에서 2 차 주파수 성분 (4 x 빔 발진 주파수)이 나타나는 것은이 두 현상의 동시 작용입니다.
두 가지 현상 중 첫 번째 주파수 에 대한 주된 효과 는 집중된 라인 에너지입니다. Cho et al. (2018)은 전력 흡수 데이터를 푸리에 변환을 사용하여 분석했을 때 1 차 주파수 성분이 더 우세 해졌고, 2 차 주파수 성분은 발진 주파수가 증가함에 따라 상대적으로 약화 되었음을 보여주었습니다 [26] . 용융 된 와이어 팁은 또한 빈도가 증가함에 따라 더욱 균일 해졌습니다. 결과는 진동 주파수의 증가가 용융 풀에 대한 와이어의 영향을 제거하는 것으로 나타났습니다. 따라서 발진 주파수가 증가함에 따라 라인 에너지 집중의 영향 만 남을 수 있습니다. 그림 10 과 같이, 집중 선 에너지가 작은 조각으로 분할되기 때문에 효과도 감소하지만 최대 값이 변경되지 않았기 때문에 여전히 효과적입니다.
빔 진동 주파수 성분은 불안정한 단추 구멍 및 열쇠 구멍 붕괴를 수반하는 불안정한 용융 풀 동작과 관련이 있습니다. 언더컷 결함이있는 케이스 8 (발진 주파수 200Hz)에서 발진 주파수 성분이 관찰되었습니다. 이것은 특히 완전히 관통 된 열쇠 구멍과 불안정한 단추 구멍에서 불안정한 용융 풀 동작을 보여주었습니다. 경우 10 (진동 주파수 250Hz)의 경우 상대적으로 건강한 비드가 형성 되었으나, 도 11 (a) 와 같이 웨이블릿 변환 결과에서 t1의 시간 간격으로 진동 주파수 성분이 관찰되었다 . 이 시간 간격 t1의 용융 풀 거동은 그림 11에 나와 있습니다.(비). 완전히 관통 된 열쇠 구멍이 즉시 무너지는 것이 분명하게 관찰되었습니다. 이것은 진동 주파수 성분이 불안정한 용융 풀 거동과 밀접한 관련이 있음을 보여줍니다. 발견 된 주파수 성분으로부터 완전히 관통 된 열쇠 구멍과 같은 불안정한 용융 풀 거동을 예측할 수 있습니다. 완전히 관통 된 키홀이 반복적으로 붕괴되기 때문에 빔 진동 주파수 성분은 그림 9 (d) 와 같이 웨이블릿 변환 결과에서 명확한 실선 형태로 보이지 않습니다 .
Cho and Woizeschke (2020)에 따르면 단추 구멍 형성은 자체 지속 가능한 카테 노이드처럼 작용하기 때문에 용융 풀 역학을 감소시킬 수 있습니다 [30] . 그림 12 는 버튼 홀 형성 측면에서 t2의 시간 간격에서 용융 풀 거동의 변화를 보여줍니다. 단추 구멍은 t2의 간헐적 인 부분에만 형성되었습니다. 1st 이후이 시간 동안 웨이블릿 변환의 결과로 주파수 성분이 사라졌고, 버튼 홀 형성은 용융 풀 역학을 줄이는 데 효과적이었습니다. 따라서, 웨이블릿 변환의 결과로 주파수 성분이 지워지는 것을 관찰함으로써 버튼 홀 형성을 예측할 수있다. 이와 관련하여 웨이블릿 변환 기술은 시간에 따른 용융 풀 변화를 나타낼 수 있습니다. 이 기술은 향후 용융 풀 동작을 모니터링하는 데 사용될 수 있습니다.
5 . 결론
CFD 시뮬레이션 결과를 사용하여 빔 진동 및 필러 와이어 공급을 통한 레이저 용접에서 용융 풀 역학을 분석 할 수있었습니다. 용융 풀 표면의 변동 데이터의 푸리에 변환 및 웨이블릿 변환은 여기서 용융 풀 역학을 분석하는 데 사용되었습니다. 결과는 다음과 같은 결론으로 이어집니다.1.
1 차 주파수 (2 x 빔 발진 주파수, 이른바 초핑 주파수), 2 차 주파수 (4 x 빔 발진 주파수) 및 빔 발진 주파수 성분은 푸리에 변환 및 웨이블릿 변환 분석에서 발견 된 주요 성분이었습니다.2.
1 차 주파수와 2 차 주파수 성분 의 출현은 두 가지 사건, 즉 레이저 빔에 의한 필러 와이어의 절단 공정과 집중된 레이저 라인 에너지의 효과의 결과였습니다. 이는 빔 진동주기 당 두 번 발생했습니다. 따라서 두 번째 주파수 성분은 동시 작용으로 인해 발생했습니다. 빔 진동 주파수 성분은 불안정한 용융 풀 동작과 관련이 있습니다. 구성 요소는 열쇠 구멍과 단추 구멍의 붕괴와 함께 나타났습니다.삼.
낮은 발진 주파수에서는 1 차 주파수와 2 차 주파수 성분이 함께 나타 났지만 발진 주파수가 증가함에 따라 그 크기가 함께 감소했습니다. 집중 선 에너지는 주파수가 증가함에 따라 최대 값이 변하지 않는 반면, 잘게 잘린 선단이 평평 해져 그 효과가 사라졌기 때문에 쵸핑 프로세스보다 더 큰 영향을 미쳤습니다.4.
용융 풀 거동의 빠른 시간적 변화는 웨이블릿 변환 방법을 사용하여 분석되었습니다. 따라서이 방법은 열쇠 구멍 및 단추 구멍의 형성 및 붕괴와 같은 일시적인 용융 풀 변화를 해석하는 데 사용할 수 있습니다.
CRediT 저자 기여 성명
조원익 : 개념화, 방법론, 소프트웨어, 검증, 형식 분석, 조사, 데이터 큐 레이션, 글쓰기-원고, 글쓰기-검토 및 편집. Peer Woizeschke : 감독, 프로젝트 관리, 작문-검토 및 편집.
경쟁 관심의 선언
저자는이 논문에보고 된 작업에 영향을 미칠 수있는 경쟁적인 재정적 이해 관계 나 개인적 관계가 없다고 선언합니다.
감사의 말
이 작업은 알루미늄 합금 용접 역량 센터 (Centr-Al)에서 수행되었습니다. Deutsche Forschungsgemeinschaft (DFG, 프로젝트 번호 290705638 , “용접 풀 캐비티를 생성하여 레이저 깊은 용입 용접에서 매끄러운 이음매 표면”) 의 자금은 감사하게도 인정됩니다.
-대량 보존 방정식,(A1)∇·V→=미디엄˙에스ρ어디, V→속도 벡터입니다. ρ밀도이고 미디엄˙에스필러 와이어를 공급하여 질량 소스의 비율입니다. 단위미디엄에스단위 부피당 질량입니다. WFS (와이어 공급 속도) 및 필러 와이어의 직경과 같은 매스 소스 및 필러 와이어 조건,디w계산 영역에서 다음과 같은 관계가 있습니다.(A2)미디엄=∫미디엄에스디V=미디엄0+씨×ρ×W에프에스×π디w24×티어디, 미디엄총 질량, 미디엄0초기 총 질량, V볼륨입니다.씨단위 변환 계수입니다. 티시간입니다.
-운동량 보존 방정식,(A3)∂V→∂티+V→·∇V→=−1ρ∇피+ν∇2V→−케이V→+미디엄˙에스ρ(V에스→−V→)+지어디, 피압력입니다. ν동적 점도입니다. 케이뭉툭한 영역의 다공성 매체 모델에 대한 항력 계수, V에스→질량 소스에 대한 속도 벡터입니다. 지신체 힘으로 인한 신체 가속도입니다.
-에너지 절약 방정식,(A4)∂h∂티+V→·∇h=1ρ∇·(케이∇티)+h˙에스어디, h특정 엔탈피입니다. 케이열전도율, 티온도이고 h˙에스특정 엔탈피 소스로, Eq 의 질량 소스와 연관됩니다 . (A1) . 계산 영역의 총 에너지,이자형다음과 같이 계산됩니다.(A5)이자형=∫미디엄에스h에스디V=∫미디엄에스씨Vw티w디V어디, 씨Vw질량 원의 비열, 티w질량 소스의 온도입니다.
또한, 엔탈피 기반 연속체 모델을 사용하여 고체-액체 상 전이를 고려했습니다.
-VOF 방정식,(A6)∂에프∂티+∇·(V→에프)=에프˙에스어디, 에프유체가 차지하는 부피 분율이며 0과 1 사이의 값을 가지며 에프˙에스질량의 소스와 연결된 유체의 체적 분율의 비율 식. (A1) . 질량 공급원에 해당하는 부피 분율은 다음에 할당됩니다.에프에스.
-스칼라 보존 방정식,(A7)∂Φ∂티+∇·(V→Φ)=Φ˙에스어디, Φ필러 와이어의 스칼라 값입니다. 셀의 유체가 전적으로 필러 와이어로 구성된 경우Φ1이고 유체에 대한 필러 와이어의 부피 분율에 따라 0과 1 사이에서 변경됩니다. Φ˙에스Eq 에서 질량 소스에 연결된 스칼라 소스의 비율입니다 . (A1) . 스칼라 소스는 전적으로 필러 와이어이기 때문에 1에 할당됩니다. 확산 효과는 고려되지 않았습니다.
흡수율을 계산하기 위해 프레 넬 반사 모델을 사용했습니다. ㅏ=1−ρ씨재료의 표면 상에 도시 된 바와 같이 수학 식. (A8) 원 편광 빔의 경우.(A8)ㅏ=1−ρ씨=1−12(ρ에스+ρ피)어디,ρ에스=(엔1씨영형에스θ−피)2+큐2(엔1씨영형에스θ+피)2+큐2,ρ에스=(피−엔1에스나는엔θ티ㅏ엔θ)2+큐2(피+엔1에스나는엔θ티ㅏ엔θ)2+큐2,피2=12{[엔22−케이22−(엔1에스나는엔θ)2]2+2엔22케이22+[엔22−케이22−(엔1에스나는엔θ)2]},큐2=12{[엔22−케이22−(엔1에스나는엔θ)2]2+2엔22케이22−[엔22−케이22−(엔1에스나는엔θ)2]}.어디, 복잡한 인덱스 엔1과 케이1반사 지수와 공기의 흡수 지수이며 엔2과 케이2공작물을위한 것입니다. θ입사각입니다. 도시 된 바와 같이 수학 식. (A9)에서 , 혼합물의 흡수율은 식에서 얻은 모재 및 필러 와이어 분획의 가중 평균이됩니다 . (A7) .(A9)ㅏ미디엄나는엑스티유아르 자형이자형=Φㅏw나는아르 자형이자형+(1−Φ)ㅏ비ㅏ에스이자형어디, ㅏ비ㅏ에스이자형과 ㅏw나는아르 자형이자형각각 비금속과 필러 와이어의 흡수율입니다.
자유 표면 경계에서의 반동 압력 에이 싱은 Eq. (A10) .(A10)피아르 자형(티)≅0.54피에스ㅏ티(티)=0.54피0이자형엑스피(엘V티−티비아르 자형¯티티비)어디, 피에스ㅏ티포화 압력, 피0대기압입니다. 엘V기화의 잠열, 티비끓는 온도이고 아르 자형¯보편적 인 기체 상수입니다.
참고 문헌
D.J. Kotecki, D.L. Cheever, D.G. Howden Mechanism of ripple formation during weld solidification Weld. J., 51 (8) (1972), pp. 386s-391s Google Scholar [2] M. Zacksenhouse, D.E. Hardt Weld pool impedance identification for size measurement and control J. Dyn. Syst. Meas. Control, 105 (3) (1983), pp. 179-184 CrossRefView Record in ScopusGoogle Scholar [3] V.V. Semak, J.A. Hopkins, M.H. McCay, T.D. McCay Melt pool dynamics during laser welding J. Phys. D, 28 (1995), pp. 2443-2450 CrossRefView Record in ScopusGoogle Scholar [4] A.J.R. Aendenroomer, G. den Ouden Weld pool oscillation as a tool for penetration sensing during pulsed GTA welding Weld. J., 77 (5) (1998), pp. 181s-187s Google Scholar [5] M.J.M. Hermans, G. den Ouden Process behavior and stability in short circuit gas metal arc welding Weld. J., 78 (4) (1999), pp. 137-141 View Record in ScopusGoogle Scholar [6] B.Y.B. Yudodibroto, M.J.M. Hermans, Y. Hirata, G. den Ouden Influence of filler wire addition on weld pool oscillation during gas tungsten arc welding Sci. Technol. Weld. Join., 9 (2) (2004), pp. 163-168 View Record in ScopusGoogle Scholar [7] M. Geiger, K.-H. Leitz, H. Koch, A. Otto A 3D transient model of keyhole and melt pool dynamics in laser beam welding applied to the joining of zinc coated sheets Prod. Eng. Res. Dev., 3 (2009), pp. 127-136 CrossRefView Record in ScopusGoogle Scholar [8] C. Kägeler, M. Schmidt Frequency-based analysis of weld pool dynamics and keyhole oscillations at laser beam welding of galvanized steel sheets Phys. Procedia, 5 (2010), pp. 447-453 ArticleDownload PDFView Record in ScopusGoogle Scholar [9] Y. Shi, G. Zhang, X.J. Ma, Y.F. Gu, J.K. Huang, D. Fan Laser-vision-based measurement and analysis of weld pool oscillation frequency in GTAW-P Weld. J., 94 (2015), pp. 176s-187s Google Scholar [10] J. Volpp, F. Vollertsen Keyhole stability during laser welding—part I: modelling and evaluation Prod. Eng.-Res. Dev., 10 (2016), pp. 443-457 CrossRefView Record in ScopusGoogle Scholar [11] N. Postacioglu, P. Kapadia, J. Dowden Capillary waves on the weld pool in penetration welding with a laser J. Phys. D, 22 (1989), pp. 1050-1061 CrossRefView Record in ScopusGoogle Scholar [12] N. Postacioglu, P. Kapadia, J. Dowden Theory of the oscillations of an ellipsoidal weld pool in laser welding J. Phys. D, 24 (1991), pp. 1288-1292 CrossRefView Record in ScopusGoogle Scholar [13] J. Kroos, U. Gratzke, M. Vicanek, G. Simon Dynamic behaviour of the keyhole in laser welding J. Phys. D, 26 (1993), pp. 481-486 View Record in ScopusGoogle Scholar [14] H. Maruo, Y. Hirata Natural frequency and oscillation modes of weld pools. 1st Report: weld pool oscillation in full penetration welding of thin plate Weld. Int., 7 (8) (1993), pp. 614-619 CrossRefView Record in ScopusGoogle Scholar [15] T. Klein, M. Vicanek, J. Kroos, I. Decker, G. Simon Oscillations of the keyhole in penetration laser beam welding J. Phys. D, 27 (1994), pp. 2023-2030 CrossRefView Record in ScopusGoogle Scholar [16] T. Klein, M. Vicanek, G. Simon Forced oscillations of the keyhole in penetration laser beam welding J. Phys. D, 29 (1996), pp. 322-332 View Record in ScopusGoogle Scholar [17] K. Andersen, G.E. Cook, R.J. Barnett, A.M. Strauss Synchronous weld pool oscillation for monitoring and control IEEE Trans. Ind. Appl., 33 (2) (1997), pp. 464-471 View Record in ScopusGoogle Scholar [18] W.-I. Cho, S.-J. Na, M.-H. Cho, J.-S. Lee Numerical study of alloying element distribution in CO2 laser-GMA hybrid welding Comput. Mater. Sci., 49 (2010), pp. 792-800 ArticleDownload PDFView Record in ScopusGoogle Scholar [19] W.-I. Cho, S.-J. Na, C. Thomy, F. Vollertsen Numerical simulation of molten pool dynamics in high power disk laser welding J. Mater. Process. Technol., 212 (2012), pp. 262-275 ArticleDownload PDFView Record in ScopusGoogle Scholar [20] A. Otto, A. Patschger, M. Seiler Numerical and experimental investigations of humping phenomena in laser micro welding Phys. Procedia, 83 (2016), pp. 1415-1423 ArticleDownload PDFView Record in ScopusGoogle Scholar [21] R. Lin, H.-P. Wang, F. Lu, J. Solomon, B.E. Carlson Numerical study of keyhole dynamics and keyhole-induced porosity formation in remote laser welding of Al alloys Int. J. Heat Mass Trans., 108 (2017), pp. 244-256 ArticleDownload PDFView Record in ScopusGoogle Scholar [22] S.H. Ko, C.D. Yoo, D.F. Farson, S.K. Choi Mathematical modeling of the dynamic behavior of gas tungsten arc weld pools Metall. Mater. Trans. B., 31B (2000), pp. 1465-1473 CrossRefView Record in ScopusGoogle Scholar [23] R. Hu, X. Chen, G. Yang, S. Gong, S. Pang Metal transfer in wire feeding-based electron beam 3D printing: modes, dynamics, and transition criterion Int. J. Heat Mass Transf., 126 (2018), pp. 877-887 ArticleDownload PDFView Record in ScopusGoogle Scholar [24] X. Meng, A. Artinov, M. Bachmann, M. Rethmeier Theoretical study of influence of electromagnetic stirring on transport phenomena in wire feed laser beam welding J. Laser Appl., 32 (2020), Article 022026 CrossRefGoogle Scholar [25] W.-I. Cho, V. Schultz, F. Vollertsen Simulation of the buttonhole formation during laser welding with wire feeding and beam oscillation L. Overmeyer, U. Reisgen, A. Ostendorf, M. Schmidt (Eds.), Proceedings of the Lasers in Manufacturing, German Scientific Laser Society, Munich, Germany (2017) Google Scholar [26] W.-I. Cho, V. Schultz, P. Woizeschke Numerical study of the effect of the oscillation frequency in buttonhole welding J. Mater. Process. Technol., 261 (2018), pp. 202-212 ArticleDownload PDFView Record in ScopusGoogle Scholar [27] V. Schultz, T. Seefeld, F. Vollertsen Bridging Large Air Gaps by Laser Welding with Beam Oscillation International Conference on Application of Lasers in Manufacturing, New Delhi, India (2015), pp. 31-32 CrossRefGoogle Scholar [28] W.-I. Cho, S.-J. Na Impact of wavelengths of CO2, disk, and green lasers on fusion zone shape in laser welding of steel J. Weld. Join., 38 (3) (2020), pp. 235-240 CrossRefView Record in ScopusGoogle Scholar [29] FLOW-3D User Manual. 2017. Version 11.2.1.06, Flow Science Inc. Google Scholar [30] W.-I. Cho, P. Woizeschke Analysis of molten pool behavior with buttonhole formation in laser keyhole welding of sheet metal Int. J. Heat Mass Transf., 152 (2020), Article 119528 ArticleDownload PDFView Record in ScopusGoogle Scholar [31] F. Vollertsen Loopless production: definition and examples from joining 69th IIW Annual Assembly and International Conference, Melbourne, Australia (2016) Google Scholar [32] V. Schultz, W.-I. Cho, A. Merkel, P. Woizeschke Deep penetration laser welding with high seam surface quality due to buttonhole welding Proc. of the IIW Annual Assembly, Com. IV, Bali, Indonesia (2018) IIW-Doc. IV-1390-18
Department of mechanical engineering, Technical University of Denmark (DTU), Building 425, 2800 Kgs., Lyngby, Denmark
Received 21 August 2020, Revised 18 November 2020, Accepted 25 November 2020, Available online 15 December 2020.
Abstract
Several different interfacial forces affect the free surface of liquid metals during metal additive manufacturing processes. One of these is thermo-capillarity or the so-called Marangoni effect. In this work, a novel framework is introduced for unraveling the effects of thermo-capillarity on the melt pool morphology/size and its thermo-fluid conditions during the Laser Powder Bed Fusion (L-PBF) process. In this respect, a multi-physics numerical model is developed based on the commercial software package Flow-3D. The model is verified and validated via mesh-independency analysis and by comparison of the predicted melt pool profile with those from lab-scale single-track experiments. Two sets of parametric studies are carried out to find the role of both positive and inverse thermo-capillarity on the melt pool shape and its thermal and fluid dynamics conditions. The thermo-fluid conditions of the melt pool are further investigated using appropriate dimensionless numbers. The results show that for the higher Marangoni number cases, the melt pool temperature drops, and at the same time, the temperature field becomes more uniform. Also, it is shown that at higher Marangoni numbers, temperature gradients decrease, thus reducing the role of conduction in the heat transfer from the melt pool. Furthermore, for the first time, a novel methodology is introduced for the calculation of the melt pool’s average Nusselt number. The average Nusselt numbers calculated for the positive and inverse thermo-capillarity are then used for finding the effective liquid conductivity required for a computationally cheaper pure heat conduction simulation. The results show that the deviation between the average melt pool temperature, using the pure conduction model with effective conductivity, and the one obtained from the advanced fluid dynamics model is less than 2%.
Keywords
Thermo-capillarity, Melt pool, Heat and fluid flow, Numerical model, L-PBF
Korea Abstract
금속 적층 제조 공정 중 액체 금속의 자유 표면에 여러 가지 다른 계면력이 영향을 미칩니다. 이들 중 하나는 열 모세관 또는 소위 Marangoni 효과입니다.
이 작업에서는 L-PBF (Laser Powder Bed Fusion) 공정 중 용융 풀 형태 / 크기 및 열 유동 조건에 대한 열 모세관의 영향을 밝히기 위한 새로운 프레임워크가 도입되었습니다.
이러한 점에서 상용 소프트웨어 패키지 Flow-3D를 기반으로 다중 물리 수치 모델이 개발되었습니다. 모델은 메쉬 독립 분석을 통해 그리고 예측 된 용융 풀 프로필을 실험실 규모의 단일 트랙 실험에서 얻은 프로필과 비교하여 검증 및 검증됩니다.
용융 풀 모양과 열 및 유체 역학 조건에 대한 양 및 역 열 모세관의 역할을 찾기 위해 두 세트의 매개 변수 연구가 수행됩니다. 용융 풀의 열 유동 조건은 적절한 무 차원 숫자를 사용하여 추가로 조사됩니다.
결과는 Marangoni 수가 더 높은 경우 용융 풀 온도가 떨어지고 동시에 온도 필드가 더 균일 해짐을 보여줍니다. 또한 Marangoni 수가 높을수록 온도 구배가 감소하여 용융 풀에서 열 전달에서 전도의 역할이 감소하는 것으로 나타났습니다.
또한 용융 풀의 평균 Nusselt 수를 계산하기위한 새로운 방법론이 처음으로 도입되었습니다. 그런 다음 양수 및 역 열 모세관에 대해 계산 된 평균 Nusselt 수는 계산적으로 더 저렴한 순수 열 전도 시뮬레이션에 필요한 효과적인 액체 전도도를 찾는 데 사용됩니다. 결과는 유효 전도도가 있는 순수 전도 모델을 사용한 평균 용융 풀 온도와 고급 유체 역학 모델에서 얻은 편차가 2 % 미만임을 보여줍니다.
A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig1A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig2A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig3A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig4A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig5A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig6A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig7A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig8A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig9A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig10A fundamental investigation of thermo-capillarity in laser powder bed fusion of metals and alloys Fig11
In this study, the nonlinear effect of contactless bubble–bubble interactions in inertial micropumps is characterized via reduced parameter one-dimensional and three-dimensional computational fluid dynamics (3D CFD) modeling. A one-dimensional pump model is developed to account for contactless bubble-bubble interactions, and the accuracy of the developed one-dimensional model is assessed via the commercial volume of fluid CFD software, FLOW-3D. The FLOW-3D CFD model is validated against experimental bubble dynamics images as well as experimental pump data. Precollapse and postcollapse bubble and flow dynamics for two resistors in a channel have been successfully explained by the modified one-dimensional model. The net pumping effect design space is characterized as a function of resistor placement and firing time delay. The one-dimensional model accurately predicts cumulative flow for simultaneous resistor firing with inner-channel resistor placements (0.2L < x < 0.8L where L is the channel length) as well as delayed resistor firing with inner-channel resistor placements when the time delay is greater than the time required for the vapor bubble to fill the channel cross section. In general, one-dimensional model accuracy suffers at near-reservoir resistor placements and short time delays which we propose is a result of 3D bubble-reservoir interactions and transverse bubble growth interactions, respectively, that are not captured by the one-dimensional model. We find that the one-dimensional model accuracy improves for smaller channel heights. We envision the developed one-dimensional model as a first-order rapid design tool for inertial pump-based microfluidic systems operating in the contactless bubble–bubble interaction nonlinear regime
이 연구에서 관성 마이크로 펌프에서 비접촉 기포-기포 상호 작용의 비선형 효과는 감소 된 매개 변수 1 차원 및 3 차원 전산 유체 역학 (3D CFD) 모델링을 통해 특성화됩니다. 비접촉식 기포-버블 상호 작용을 설명하기 위해 1 차원 펌프 모델이 개발되었으며, 개발 된 1 차원 모델의 정확도는 유체 CFD 소프트웨어 인 FLOW-3D의 상용 볼륨을 통해 평가됩니다.
FLOW-3D CFD 모델은 실험적인 거품 역학 이미지와 실험적인 펌프 데이터에 대해 검증되었습니다. 채널에 있는 두 저항기의 붕괴 전 및 붕괴 후 기포 및 유동 역학은 수정 된 1 차원 모델에 의해 성공적으로 설명되었습니다. 순 펌핑 효과 설계 공간은 저항 배치 및 발사 시간 지연의 기능으로 특징 지어집니다.
1 차원 모델은 내부 채널 저항 배치 (0.2L <x <0.8L, 여기서 L은 채널 길이)로 동시 저항 발생에 대한 누적 흐름과 시간 지연시 내부 채널 저항 배치로 지연된 저항 발생을 정확하게 예측합니다. 증기 방울이 채널 단면을 채우는 데 필요한 시간보다 큽니다.
일반적으로 1 차원 모델 정확도는 저수지 근처의 저항 배치와 1 차원 모델에 의해 포착되지 않는 3D 기포-저수지 상호 작용 및 가로 기포 성장 상호 작용의 결과 인 짧은 시간 지연에서 어려움을 겪습니다. 채널 높이가 작을수록 1 차원 모델 정확도가 향상됩니다. 우리는 개발 된 1 차원 모델을 비접촉 기포-기포 상호 작용 비선형 영역에서 작동하는 관성 펌프 기반 미세 유체 시스템을 위한 1 차 빠른 설계 도구로 생각합니다.
REFERENCES
1.S. Hassan and X. Zhang, “ Design and fabrication of capillary-driven flow device for point-of-care diagnostics,” Biosensors 10, 39 (2020). https://doi.org/10.3390/bios10040039, Google ScholarCrossref 2.Q. Shizhi and H. Bau, “ Magneto-hydrodynamics based microfluidics,” Mech. Res. Commun. 36, 10 (2009). https://doi.org/10.1016/j.mechrescom.2008.06.013, Google ScholarCrossref 3.N. Mishchuk, T. Heldal, T. Volden, J. Auerswald, and H. Knapp, “ Micropump based on electroosmosis of the second kind,” Electrophoresis 30, 3499 (2009). https://doi.org/10.1002/elps.200900271, Google ScholarCrossref 4.J. Snyder, J. Getpreecharsawas, D. Fang, T. Gaborski, C. Striemer, P. Fauchet, D. Borkholder, and J. McGrath, “ High-performance, low-voltage electroosmotic pumps with molecularly thin silicon nanomembranes,” Proc. Nat. Acad. Sci. U. S. A. 110, 18425–18430 (2013). https://doi.org/10.1073/pnas.1308109110, Google ScholarCrossref 5.K. Vinayakumar, G. Nadiger, V. Shetty, S. Dinesh, M. Nayak, and K. Rajanna, “ Packaged peristaltic micropump for controlled drug delivery application,” Rev. Sci. Instrum. 88, 015102 (2017). https://doi.org/10.1063/1.4973513, Google ScholarScitation, ISI 6.D. Duffy, H. Gillis, J. Lin, N. Sheppard, and G. Kellogg, “ Microfabricated centrifugal microfluidic systems: Characterization and multiple enzymatic assays,” Anal. Chem. 71, 4669 (1999). https://doi.org/10.1021/ac990682c, Google ScholarCrossref 7.V. Gnyawali, M. Saremi, M. Kolios, and S. Tsai, “ Stable microfluidic flow focusing using hydrostatics,” Biomicrofluidics 11, 034104 (2017). https://doi.org/10.1063/1.4983147, Google ScholarScitation, ISI 8.J. Lake, K. Heyde, and W. Ruder, “ Low-cost feedback-controlled syringe pressure pumps for microfluidics applications,” PLoS One 12, e0175089 (2017). https://doi.org/10.1371/journal.pone.0175089, Google ScholarCrossref 9.M. I. Mohammed, S. Haswell, and I. Gibson, “ Lab-on-a-chip or chip-in-a-lab: Challenges of commercialization lost in translation,” Procedia Technology 20, 54–59 (2015), proceedings of The 1st International Design Technology Conference, DESTECH2015, Geelong. Google ScholarCrossref 10.E. Torniainen, A. Govyadinov, D. Markel, and P. Kornilovitch, “ Bubble-driven inertial micropump,” Phys. Fluids 24, 122003 (2012). https://doi.org/10.1063/1.4769755, Google ScholarScitation, ISI 11.H. Hoefemann, S. Wadle, N. Bakhtina, V. Kondrashov, N. Wangler, and R. Zengerle, “ Sorting and lysis of single cells by bubblejet technology,” Sens. Actuators, B 168, 442–445 (2012). https://doi.org/10.1016/j.snb.2012.04.005, Google ScholarCrossref 12.B. Hayes, A. Hayes, M. Rolleston, A. Ferreira, and J. Kirsher, “ Pulsatory mixing of laminar flow using bubble-driven micro-pumps,” in Proceedings of the ASME 2018 International Mechanical Engineering Congress and Exposition (2018), Vol. 7. Google ScholarCrossref 13.E. Ory, H. Yuan, A. Prosperetti, S. Popinet, and S. Zaleski, “ Growth and collapse of a vapor bubble in a narrow tube,” Phys. Fluids 12, 1268 (2000). https://doi.org/10.1063/1.870381, Google ScholarScitation, ISI 14.Z. Yin and A. Prosperetti, “‘ Blinking bubble’ micropump with microfabricated heaters,” J. Micromech. Microeng. 15, 1683 (2005). https://doi.org/10.1088/0960-1317/15/9/010, Google ScholarCrossref 15.M. Einat and M. Grajower, “ Microboiling measurements of thermal-inkjet heaters,” J. Microelectromech. Syst. 19, 391 (2010). https://doi.org/10.1109/JMEMS.2010.2040946, Google ScholarCrossref 16.A. Govyadinov, P. Kornilovitch, D. Markel, and E. Torniainen, “ Single-pulse dynamics and flow rates of inertial micropumps,” Microfluid. Nanofluid. 20, 73 (2016). https://doi.org/10.1007/s10404-016-1738-x, Google ScholarCrossref 17.E. Sourtiji and Y. Peles, “ A micro-synthetic jet in a microchannel using bubble growth and collapse,” Appl. Therm. Eng. 160, 114084 (2019). https://doi.org/10.1016/j.applthermaleng.2019.114084, Google ScholarCrossref 18.B. Hayes, A. Govyadinov, and P. Kornilovitch, “ Microfluidic switchboards with integrated inertial pumps,” Microfluid. Nanofluid. 22, 15 (2018). https://doi.org/10.1007/s10404-017-2032-2, Google ScholarCrossref 19.P. Kornilovitch, A. Govyadinov, D. Markel, and E. Torniainen, “ One-dimensional model of inertial pumping,” Phys. Rev. E 87, 023012 (2013). https://doi.org/10.1103/PhysRevE.87.023012, Google ScholarCrossref 20.H. Yuan and A. Prosperetti, “ The pumping effect of growing and collapsing bubbles in a tube,” J. Micromech. Microeng. 9, 402–413 (1999). https://doi.org/10.1088/0960-1317/9/4/318, Google ScholarCrossref 21.J. Zou, B. Li, and C. Ji, “ Interactions between two oscillating bubbles in a rigid tube,” Exp. Therm. Fluid Sci. 61, 105 (2015). https://doi.org/10.1016/j.expthermflusci.2014.10.021, Google ScholarCrossref 22.C. Hirt and B. Nichols, “ Volume of fluid (vof) method for the dynamics of free boundaries,” J. Comput. Phys. 39, 201–225 (1981). https://doi.org/10.1016/0021-9991(81)90145-5, Google ScholarCrossref 23.C. Borgnakke and R. E. Sonntag, Fundamentals of Thermodynamics, 8th ed. ( Wiley, 1999). Google Scholar 24.O. E. Ruiz, “ CFD model of the thermal inkjet droplet ejection process,” in Proceeding of Heat Transfer Summer Conference (2007), Vol. 3. Google ScholarCrossref 25.T. Theofanous, L. Biasi, H. Isbin, and H. Fauske, “ A theoretical study on bubble growth in constant and time-dependent pressure fields,” Chem. Eng. Sci. 24, 885–897 (1969). https://doi.org/10.1016/0009-2509(69)85008-6, Google ScholarCrossref 26.S. Timoshenko and J. Goodier, Theory of Elasticity, 3rd ed. ( McGaw-Hill, Inc., 1970). Google Scholar
Subin Shrestha1 J.B. Speed School of Engineering,University of Louisville,Louisville, KY 40292 e-mail: subin.shrestha@louisville.edu
Y. Kevin Chou J.B. Speed School of Engineering,University of Louisville,Louisville, KY 40292 e-mail: kevin.chou@louisville.edu
LPBF (Laser Powder Bed fusion) 공정 중 용융 풀의 동적 현상은 복잡하고 공정 매개 변수에 민감합니다. 에너지 밀도 입력이 특정 임계 값을 초과하면 키홀이라고 하는 거대한 증기 함몰이 형성 될 수 있습니다.
이 연구는 수치 분석을 통해 LPBF 과정에서 키홀 거동 및 관련 기공 형성을 이해하는 데 중점을 둡니다. 이를 위해 이산 분말 입자가 있는 열 유동 모델이 개발되었습니다.
이산 요소 방법 (DEM)에서 얻은 분말 분포는 계산 영역에 통합되어 FLOW-3D를 사용하는 3D 프로세스 물리학 모델을 개발합니다.
전도 모드 중 용융 풀 형성과 용융의 키홀 모드가 식별되고 설명되었습니다. 높은 에너지 밀도는 증기 기둥의 형성으로 이어지고 결과적으로 레이저 스캔 트랙 아래에 구멍이 생깁니다.
또한 다양한 레이저 출력과 스캔 속도로 인한 Keyhole 모양을 조사합니다. 수치 결과는 동일한 에너지 밀도에서도 레이저 출력이 증가함에 따라 Keyhole크기가 증가 함을 나타냅니다. Keyhole은 더 높은 출력에서 안정되어 레이저 스캔 중 Keyhole 발생을 줄일 수 있습니다.
The dynamic phenomenon of a melt pool during the laser powder bed fusion (LPBF) process is complex and sensitive to process parameters. As the energy density input exceeds a certain threshold, a huge vapor depression may form, known as the keyhole. This study focuses on understanding the keyhole behavior and related pore formation during the LPBF process through numerical analysis. For this purpose, a thermo-fluid model with discrete powder particles is developed. The powder distribution, obtained from a discrete element method (DEM), is incorporated into the computational domain to develop a 3D process physics model using flow-3d. The melt pool formation during the conduction mode and the keyhole mode of melting has been discerned and explained. The high energy density leads to the formation of a vapor column and consequently pores under the laser scan track. Further, the keyhole shape resulted from different laser powers and scan speeds is investigated. The numerical results indicated that the keyhole size increases with the increase in the laser power even with the same energy density. The keyhole becomes stable at a higher power, which may reduce the occurrence of pores during laser scanning.
Keywords: additive manufacturing, keyhole, laser powder bed fusion, porosity
Fig. 1 (a) Powder added to the dispenser platform and
(b) powder particles settled over build plate after the recoating
processFig. 2 3D computational domain used for single-track
simulationFig. 3 Temperature-dependent material properties of Ti-6Al-4VFig. 4 Powder and substrate melting during laser applicationFig. 5 Melt region formed after complete melting and
solidificationFig. 6 Melt pool boundary comparison between the experiment
[25] and the simulationFig. 7 Equilibrium points during the formation of vapor column
[27]Fig. 8 Multiple reflection vectors from the keyhole wallFig. 9 (a) Velocity field, keyhole profile, and breakage of the
keyhole to form bubble and (b) 2D temperature and velocity
field along the longitudinal sectionFig. 10 Fluid flow in the transverse direction during keyhole
meltingFig. 11 Melt pool boundary compared with the experiment [21]
for 195 W laser power and 400 mm/s scan speedFig. 12 Melt region formed after complete melting and
solidificationFig. 13 2D images of the pores formed at the beginning of the
single track and their 3D-rendered morphologyFig. 14 Pore number and volume from a different level of power
with LED = 0.4 J/mm [29]Fig. 15 Keyhole shape at different time steps from different parameters: (a) P = 100 W, v = 250 mm/s,
(b) P = 200 W, v = 500 mm/s, (c) P = 300 W, v = 750 mm/s, and (d) P = 400 W, v = 1000 mm/sFig. 16 Intensity dependence in the relationship between vapor
column and evaporation pressure [27]Fig. 17 Temperature distribution when laser has moved 0.8 mm
with P = 300 W, v = 750 mm/s and P = 400 W, v = 1000 mm/sFig. 18 Melt region with different level of power with LED of
0.4 J/mm
[1] Bauereiß, A., Scharowsky, T., and Körner, C., 2014, “Defect Generation and Propagation Mechanism During Additive Manufacturing by Selective Beam Melting,” J. Mater. Process. Technol., 214(11), pp. 2522–2528. [2] Gong, H., Rafi, K., Gu, H., Starr, T., and Stucker, B., 2014, “Analysis of Defect Generation in Ti–6Al–4V Parts Made Using Powder Bed Fusion Additive Manufacturing Processes,” Add. Manuf., 1(2014), pp. 87–98. [3] Wang, Y., Kamath, C., Voisin, T., and Li, Z., 2018, “A Processing Diagram for High-Density Ti-6Al-4V by Selective Laser Melting,” Rapid Prototyping J., 24 (9), pp. 1469–1478. [4] Khairallah, S. A., and Anderson, A., 2014, “Mesoscopic Simulation Model of Selective Laser Melting of Stainless Steel Powder,” J. Mater. Process. Technol., 214(11), pp. 2627–2636. [5] Yadroitsev, I., Gusarov, A., Yadroitsava, I., and Smurov, I., 2010, “Single Track Formation in Selective Laser Melting of Metal Powders,” J. Mater. Process. Technol., 210(12), pp. 1624–1631. [6] Xia, M., Gu, D., Yu, G., Dai, D., Chen, H., and Shi, Q., 2016, “Influence of Hatch Spacing on Heat and Mass Transfer, Thermodynamics and Laser Processability During Additive Manufacturing of Inconel 718 Alloy,” Int. J. Mach. Tools Manuf., 109(2016), pp. 147–157. [7] Lee, Y., and Zhang, W., 2016, “Modeling of Heat Transfer, Fluid Flow and Solidification Microstructure of Nickel-Base Superalloy Fabricated by Laser Powder bed Fusion,” Add. Manuf., 12(2016), pp. 178–188. [8] Wu, Y.-C., San, C.-H., Chang, C.-H., Lin, H.-J., Marwan, R., Baba, S., and Hwang, W.-S., 2018, “Numerical Modeling of Melt-Pool Behavior in Selective Laser Melting with Random Powder Distribution and Experimental Validation,” J. Mater. Process. Technol., 254(2018), pp. 72–78. [9] Khairallah, S. A., Anderson, A. T., Rubenchik, A., and King, W. E., 2016, “Laser Powder-bed Fusion Additive Manufacturing: Physics of Complex Melt Flow and Formation Mechanisms of Pores, Spatter, and Denudation Zones,” Acta Materialia, 108(2016), pp. 36–45. [10] Tan, J., Tang, C., and Wong, C., 2018, “A Computational Study on Porosity Evolution in Parts Produced by Selective Laser Melting,” Metall. Mater. Trans. A, 49A(8), pp. 3663–3673. [11] Leitz, K.-H., Singer, P., Plankensteiner, A., Tabernig, B., Kestler, H., and Sigl, L. J. M. P. R., 2017, “Multi-Physical Simulation of Selective Laser Melting,” Metal Powder Report, 72(5), pp. 331–338. [12] Zhao, C., Fezzaa, K., Cunningham, R. W., Wen, H., Carlo, F., Chen, L., Rollett, A. D., and Sun, T., 2017, “Real-time Monitoring of Laser Powder Bed Fusion Process Using High-Speed X-ray Imaging and Diffraction,” Sci. Rep., 7(1), p. 3602. [13] Parab, N. D., Zhao, C., Cunningham, R., Escano, L. I., Fezzaa, K., Everhart, W., Rollett, A. D., Chen, L., and Sun, T., 2018, “Ultrafast X-ray Imaging of Laser– Metal Additive Manufacturing Processes,” J. Synchrotron Radiat., 25(5), pp. 1467–1477. [14] Cunningham, R., Zhao, C., Parab, N., Kantzos, C., Pauza, J., Fezzaa, K., Sun, T., and Rollett, A. D., 2019, “Keyhole Threshold and Morphology in Laser Melting Revealed by Ultrahigh-Speed X-Ray Imaging,” Science, 363(6429), pp. 849–852. [15] Shrestha, S., Starr, T., and Chou, K., 2019, “A Study of Keyhole Porosity in Selective Laser Melting: Single Track Scanning With Micro-CT Analysis,” ASME J. Manuf. Sci. Eng., 141(7), pp. 1–23. [16] Ye, J., Rubenchik, A. M., Crumb, M. F., Guss, G., and Matthews, M. J., 2018, “Laser Absorption and Scaling Behavior in Powder Bed Fusion Additive Manufacturing of Metals,” Proceedings of the CLEO: Science and Innovations, Optical Society of America, San Jose, CA, May 13–18, Optical Society of America, p. JW2A.117. [17] Mishra, B., and Rajamani, R. K., 1992, “The Discrete Element Method for the Simulation of Ball Mills,” Appl. Math. Modell., 16(11), pp. 598–604. [18] Yan, W., Qian, Y., Ge, W., Lin, S., Liu, W. K., Lin, F., and Wagner, G. J., 2018, “Meso-Scale Modeling of Multiple-Layer Fabrication Process in Selective Electron Beam Melting: Inter-Layer/Track Voids Formation,” Materials and Design, 141(2018), pp. 210–219. [19] Kloss, C., Goniva, C., Hager, A., Amberger, S., and Pirker, S., 2012, “Models, Algorithms and Validation for Opensource DEM and CFD–DEM,” Prog. Comput. Fluid Dynam. Int. J., 12(2–3), pp. 140–152. [20] Escano, L. I., Parab, N. D., Xiong, L., Guo, Q., Zhao, C., Fezzaa, K., Everhart, W., Sun, T., and Chen, L., 2018, “Revealing Particle-Scale Powder Spreading Dynamics in Powder-Bed-Based Additive Manufacturing Process by High-Speed X-Ray Imaging,” Sci. Rep., 8(1), p. 15079. [21] Gong, H., Gu, H., Zeng, K., Dilip, J., Pal, D., Stucker, B., Christiansen, D., Beuth, J., and Lewandowski, J. J., 2014, “Melt Pool Characterization for Selective Laser Melting of Ti-6Al-4V Pre-Alloyed Powder,” Proceedings of the Solid Freeform Fabrication Symposium, Austin, TX, Aug. 4–6, pp. 256–267. [22] Mills, K. C., 2002, Recommended Values of Thermophysical Properties for Selected Commercial Alloys, Woodhead Publishing, Cambridge, UK. [23] Shrestha, S., and Chou, K., 2017, “A Build Surface Study of Powder-Bed Electron Beam Additive Manufacturing by 3D Thermo-Fluid Simulation and White-Light Interferometry,” Int. J. Mach. Tools Manuf., 121(2017), pp. 37–49. [24] Cho, J.-H., and Na, S.-J., 2006, “Implementation of Real-Time Multiple Reflection and Fresnel Absorption of Laser Beam in Keyhole,” J. Phys. D: Appl. Phys., 39(24), p. 5372. [25] Dilip, J., Zhang, S., Teng, C., Zeng, K., Robinson, C., Pal, D., and Stucker, B., 2017, “Influence of Processing Parameters on the Evolution of Melt Pool, Porosity, and Microstructures in Ti-6Al-4V Alloy Parts Fabricated by Selective Laser Melting,” Prog. Add. Manuf., 2(3), pp. 157–167. [26] Bertoli, U. S., Wolfer, A. J., Matthews, M. J., Delplanque, J.-P. R., and Schoenung, J. M., 2017, “On the Limitations of Volumetric Energy Density as a Design Parameter for Selective Laser Melting,” Mater. Des., 113(2017), pp. 331–340. [27] Kroos, J., Gratzke, U., and Simon, G., 1993, “Towards a Self-Consistent Model of the Keyhole in Penetration Laser Beam Welding,” J. Phys. D: Appl. Phys., 26(3), p. 474. [28] Martin, A., Calta, N., Hammons, J., Khairallah, S., Nielsen, M., Shuttlesworth, R., Sinclair, N., Matthews, M., Jeffries, J., and Willey, T., 2019, “Ultrafast Dynamics of Laser-Metal Interactions in Additive Manufacturing Alloys Captured by In Situ X-Ray Imaging,” Mater. Today Adv., 1(2019), p. 100002. [29] Shrestha, S., Starr, T., and Chou, K., 2018, “Individual and coupled contributions of laser power and scanning speed towards process-induced porosity in selective laser melting,” Proceedings of the Solid Freeform Fabrication Symposium, Austin, TX, Aug. 13–15, pp. 1400–1409. [30] Hann, D., Iammi, J., and Folkes, J., 2011, “A Simple Methodology for Predicting Laser-Weld Properties From Material and Laser Parameters,” J. Phys. D: Appl. Phys., 44(44), p. 445401. [31] Trapp, J., Rubenchik, A. M., Guss, G., and Matthews, M. J., 2017, “In Situ Absorptivity Measurements of Metallic Powders During Laser Powder-bed Fusion Additive Manufacturing,” Appl. Mat. Today, 9(2017), pp. 341–349.
FIGURE 1. – FLOW-3D MODEL OF K-SITE TANK PRESSUR-IZATION.FIGURE 3. – EFFECT OF GRAVITY ON TEMPERATURE CONTOURSFIGURE 6.- EFFECT OF INITIAL PRESSURE ON VELOCITY PROFILEFIGURE 10. – EFFECT OF INITIAL TEMPERATURE ON TEMPERATURE CONTOURSFIGURE 13. – EFFECT OF HEAT TRANSFER ON TEMPERATURE CONTOURSFIGURE 16. – EFFECT OF CONVERGENCE CRITERIA ON TEMPERATURE CONTOURS, 55 PERCENT ULLAGE,
Pi = 17.4 PSI, 6 = 32.2 FT/SECZ, 24 SEC PRESSURIZATIONFIGURE 17. – COMPAR ISON OF CENTERLINE TEMPERATURES USING VARIOUS CONVERGENCE CRITERIA, 55 PERCENT ULLAGE, G = 32,2 FT/SEC2, P;= 17.4 PSI, 24 SEC PRESSURIZATION.FIGURE 19. – EFFECT OF CONVERGENCE CRITERIA ON VELOCITY PROFILE,FIGURE 21. – EFFECT OF MESH SIZE ON TEMPERA-
TURE CONTOURS, 55 PERCENT ULLAGE, Pi = 17.4
PSI, G = 0.0 FT/SEC2, , e = . 02, dt = , 005
SEC, 24 SEC PRESSURIZATION.
이 최종 기술 보고서에는 수상 번호 DE-FC07-01ID14206에 따라 미국 에너지 부에서 부분적으로 자금을 지원 한 “청정 및 에너지 절약 단조 기술을위한 혁신적인 다이 재료 및 윤활 전략”프로젝트에서 수행 된 작업이 포함되어 있습니다. 프로젝트 수행을위한 계약 시간은 2001 년 9 월 30 일부터 2005 년 9 월 29 일까지였습니다. 그러나 DOE / OIT는 2003 년과 2004 년 회계 연도 지난 2 년 동안 자금을 제공 할 수 없었고 프로젝트는 2003-04 회계 연도에 조기 종료되었습니다. 결과적으로 많은 주요 연구 과제가 특정 이정표를 달성하기 위해 수정되거나 완료되지 않고 종료되었습니다. Ohio State University의 산업, 용접 및 시스템 공학 교수 인 Rajiv Shivpuri 박사는이 프로젝트의 프로젝트 책임자이자 수석 조사자였습니다. 이상은 오하이오 주립 대학 연구 재단 (OSURF)에서 관리했습니다. OSURF는 모든 재정 및 행정 문제도 담당했습니다. 재정 보고서는 별도로 제출됩니다. 에너지 부서, 산업 기술 사무소의 프로그램 관리자는 Golden Office의 Mr. Ramesh Jain과 Mr. Dibyajyoti Aichbhowmik이었습니다. 이 프로젝트의 주요 성과는 다음과 같습니다.
• 단조 산업 및 해당 공급 업체와 함께 산업 응용 분야를위한 혁신적인 다이 재료 및 윤활 전략을 탐색하기위한 주요 협력 노력이 수립되었습니다. 여기에는 단조 산업과 협력하는 워크숍과 심포지엄이 포함되었습니다. 단조 산업 전체에 결과를 전파하기 위해 단조 산업 기술 컨퍼런스에서 발표되었습니다.
• 단조 산업 협회와 단조 산업 교육 연구 재단의 후원으로 단조 기술 우수 센터 설립. 이 센터의 일부로 산업, OSU, 오하이오 주 및 DOE 지원과 함께 2 개의 단조 셀이 설치되었습니다. 1300 톤 기계식 프레스 셀과 350 톤 유압 프레스 셀입니다. 이것은 단조 연구에 150 만 달러를 투입 한 것입니다.
• LENS (Laser Enhanced Net Shaping) 기반 니켈 알루미나 이드 코팅 오버레이 (자세한 내용은 부록 A 참조)를 포함하여 혁신적인 다이 코팅이 탐색되었습니다.
• 열간 단조 응용 분야를위한 금형 재료를 최적으로 선택하고 설계하기 위해 혁신적인 실험 설정 및 예측 열 연화 소프트웨어가 개발되었습니다 (부록 B, C 및 D).
• 윤활 전략 및 단일 액적 기반 윤활 모델은 확산 및 열 전달을위한 열간 단조 윤활제의 최적 증착을 위해 개발되었습니다 (부록 E 및 F).
• 윤활유 분해 및 바운스 용 모델이 개발되었습니다. 이 모델은 뜨거운 다이 표면의 흑연 윤활로 인한 공기 및 지하수 오염을 줄이는 데 사용할 수 있습니다.
(부록 G). 이 보고서는 Shivpuri 박사와 Yijun Zhu (연구원)가 작성했습니다. 여기에는 다른 외부 또는 내부 지원과 함께 프로젝트 종료 후 일부 연구 계획 및 프로젝트 기간 동안 완료된 작업에 대한 세부 정보가 포함되어 있습니다.
1.1 프로젝트 목표
이 프로젝트의 목표는 혁신적인 다이 재료 및 윤활 전략을 개발 및 구현하여 다이 수명을 8 배 늘리고, 에너지 투입량을 15 % 줄이며, 부품 당 에너지 비용을 50 % 줄이며, 윤활유에서 나오는 미립자 배출량을 90 % 줄이며, 다이 관련 가동 시간을 90 %까지 늘립니다.
단조 산업, 공급 업체 (철강 및 알루미늄 생산 업체 (IOF), 윤활유, 표면 기술 및 다이 소재 공급 업체) 및 고객 (OEM)에 미치는 최대의 광범위한 에너지 영향을 위해 전략이 선택되었습니다.
여기에는 최적의 윤활제 스프레이 기술, 고급 표면 엔지니어링에 의한 열간 단조의 흑연 제거, 경사 다이 재료 및 다이 엔지니어링, 열간 단조를위한 윤활 및 다이 활성화 등이 포함됩니다.
미국의 단조 산업은 1997 년에 약 120 억 달러였습니다 (DOD 국가 안보). 평가). 제품 총 판매 가치의 약 15 %가 에너지에 할당되며 연간 약 50 조 BTU입니다. 흑연 사용 (열간 단조) 및 냉간 단조 전환 코팅 사용으로 인한 환경 영향은 제품 비용에 20 % 이상 추가 될 것으로 예상됩니다.
Figure 1: Die configuration for a multi-attribute composite die for high die life and self-lubricating surface
BACKGROUND
실온 (저온) 및 고온 (온 및 고온)에서 수행되는 단조는 진화하는 야금, 공구 표면의 마찰 및 금속의 흐름 특성을 포함하는 잘 이해되지 않는 복잡한 현상입니다. 이 프로젝트에서 다루어 진 기술적 장애물은 다음과 같습니다.
• 냉간 및 열간 단조의 윤활 작용에 대한 지식 부족. 윤활유 및 윤활 기술의 선택은 윤활유 및 장비 공급 업체에 맡겨집니다. 이로 인해 윤활유의 과도하고 불량한 사용과 과도한 환경 오염이 발생합니다.
• 고급 단조 응용 분야를위한 새로운 표면 엔지니어링 및 다이 재료 기술의 성숙도가 부족합니다. 실제 생산에서이를 구현하는 데 따른 기술적 및 재정적 위험이 매우 높아 사용을 제한합니다. 이러한 기술의 시장 침투는 거의 존재하지 않습니다.
• 다이와 윤활 시스템의 설계 최적화를위한 계산 도구가 부족합니다.
윤활유 및 다이 소재 기술에서 다음과 같은 전략을 통해 프로젝트 목표를 실현할 계획이었습니다.
• 전략 # 1 : 오염을 제거하고, 윤활제 사용을 줄이며, 다이 냉각 감소로 인한 그물 성형을 가능하게하는 윤활제 스프레이 공정의 최적 설계를위한 시스템 개발. 또한 흑연 기반 윤활유의 필요성을 줄여줍니다.
• 전략 # 2 : 철 및 비철 부품의 온간 단조 (빌릿 가열이 1250F에서 900F로 감소)를위한 다이 수명과 공정을 개선하기 위한 윤활제 및 다이 코팅 가능 요소를 개발합니다. 단조 온도를 낮추면 공차가 개선되고 부품 당 에너지가 크게 절약됩니다.
• 전략 # 3 : 저 마찰 다이 표면 엔지니어링 (DLC (비철) 및 WC / C 코팅)을 사용하여 냉간 단조 빌릿에 인광 코팅을 사용하지 않습니다.
• 전략 # 4 : 열간 단조 금형을위한 고급 표면 클래딩 (렌즈 및 열 스프레이에 의한 단단한 표면) 및 이중 코팅 기술을 개발합니다. 기존의 코팅과 표면 공학 기술은 상당한 이득을 얻지 못했습니다.
• 전략 # 5 : 재료 및 공정 설계를 통해 냉간 및 열간 단조에서 공정 중 다이 고장을 제거하고 예측 다이 유지 보수를위한 소프트웨어를 개발합니다. 이는 스크랩 감소 및 다이 관련 다운 타임에 상당한 영향을 미칩니다.
개발중인 많은 기술은 수치 모델링, 윤활 및 냉각수 기술, 표면 기술, 재료의 신속한 프로토 타이핑, 레이저 기술 등과 같은 교차 절단 R & D 가능 요소를 다루고 있습니다. 이러한 기술은 지원 산업의 로드맵에서도 중요한 기술로 확인되었습니다.
미래의 산업으로. IOF를 위해 250 조 BTU의 에너지 절약과 3500 톤의 오염 물질이 예상됩니다. 프로젝트가 전액 지원을받지 못하고 프로젝트가 2004 년 9 월 30 일에 종료되었으므로 전략 # 1, # 4, # 5 만 추구했습니다. 연구 및 구현에 대한 세부 사항은 부록에 포함되어 있습니다.
Effect of lubricant heat
템퍼링, 마모 및 공구 열화에 대한 단조 윤활유의 효과를 평가하기 위해 다양한 열 전달 계수로 여러 시뮬레이션을 수행했습니다. 컴퓨터 시뮬레이션에 사용 된 열전달 계수의 값은 얻은 값과 일치하며 경우에 따라 Sridhar 등이 오하이오 주립 대학에서 수행 한 테스트에서 추정 한 값입니다. 사용 된 계면 열전달 계수의 값은 12 KW / m2 ° C, 24 KW / m2 ° C 및 33 KW / m2 ° C였으며, 이는 20 부, 30 부 및 100 부 물로 희석 된 수성 흑연 윤활제에 해당합니다 (희석 비율 1:20, 1:30 및 1 : 100). 이러한 각 희석 비율에 대해 3000 및 5000 샷 후 상부 다이의 경도 분포는 그림 C.3, C.4 및 C.5에 나와 있습니다. 희석 비 1:20에 대한 표면 경도 분포는 그림 C.6에 나와 있습니다.
Figure C. 2: stage gear blank forging sequence (Courtesy: Sypris TechnologiesFigure C. 3: Hardness distribution after 3000 and 5000 shots, heat transfer coefficient used = 12 KW/m2°C, press type: mechanical press
F.5.3 Results of the Lubricant Properties
표 F.1은 윤활유의 측정 된 특성을 보여줍니다. DP는 107 및 CA 모세관 작용 방법에서 펜던트 드롭 방법을 나타냅니다. 테스트 된 액체에는 순수한 물이 포함됩니다. 다음과 같은 사실을 관찰 할 수 있습니다. a). 더 높은 표면 장력을 가진 더 높은 희석 비율 회사; 비). 희석 비율이 1 : 1보다 큰 액체의 경우 표면 장력이 물의 장력에 접근합니다. 드롭 펜던트 법으로 추정 한 모든 표면 장력은 동일한 경향을 공유하지만 약 10dynes / cm에 대해 모세관 작용법에 의한 것보다 작다는 것을 알 수 있습니다. 물의 표면 장력이 72.8dynes / cm라는 점을 감안할 때 모세관 작용법에서 얻은 결과가 실제 값에 더 가깝다고 생각합니다.
Figure F. 10: simulation results of lubricant 1:1 with 4mm diameter droplet at impact velocity 10cm/s.Figure F. 12: Experimental results of maxξ v.s. TD. We = 27.Figure G. 1: Dryoff process of a lubricant droplet at film boiling: (a)- (c) fluid dynamic process, (d). quasi-steady dryoff process.
CFD (전산 유체 역학) 전문가가 필요하고 때로는 실행하는데 몇 주가 걸리는 믹싱 시뮬레이션의 시대는 오래 전입니다. 컴퓨팅 및 관련 기술의 엄청난 도약에 힘 입어 Ansys, Comsol 및 Flow Science와 같은 회사는 엔지니어의 데스크톱에 사용하기 쉬운 믹싱 시뮬레이션을 제공하고 있습니다.
“병렬화 및 고성능 컴퓨팅의 발전과 템플릿화는 비전문 화학 엔지니어에게 정확한 CFD 시뮬레이션을 제공했습니다.”라고 펜실베이니아 주 피츠버그에있는 Ansys Inc.의 수석 제품 마케팅 관리자인 Bill Kulp는 말합니다 .
흐름 개선을위한 실용적인 지침이 필요하십니까? 다운로드 화학 처리의 eHandbook을 지금 흐름 도전 싸우는 방법!
예를 들어, 회사는 휴스턴에있는 Nalco Champion과 함께 프로젝트를 시작했습니다. 이 프로젝트는 시뮬레이션 전문가가 아닌 화학 엔지니어에게 Ansys Fluent 및 ACT (분석 제어 기술) 템플릿 기반 시뮬레이션 앱에 대한 액세스 권한을 부여합니다. 새로운 화학 물질을위한 프로세스를 빠르고 효율적으로 확장합니다.
Giving Mixing Its Due
“화학 산업은 CFD와 같은 계산 도구를 사용하여 많은 것을 얻을 수 있지만 혼합 프로세스는 단순하다고 가정하기 때문에 간과되는 경우가 있습니다. 그러나 최신 수치 기법을 사용하여 우수한 성능을 달성하는 흥미로운 방법이 많이 있습니다.”라고 Flow Science Inc. , Santa Fe, NM의 CFD 엔지니어인 Ioannis Karampelas는 말합니다 .
이러한 많은 기술이 회사의 Flow-3D Multiphysics 모델링 소프트웨어 패키지와 전용 포스트 프로세서 시각화 도구 인 FlowSight에 포함되어 있습니다.
“모든 상업용 CFD 패키지는 어떤 형태의 시각화 도구와 번들로 제공되지만 FlowSight는 매우 강력하고 사용하기 쉽고 이해하기 쉽게 설계되었습니다. 예를 들어, 프로세스를 재 설계하려는 엔지니어는 다양한 설계 변경의 효과를 평가하기 위해 매우 직관적인 시각화 도구가 필요합니다.”라고 그는 설명합니다.
이 접근 방식은 실험 측정을 얻기 어려운 공정 (예 : 쉽게 측정 할 수없는 매개 변수 및 독성 물질의 존재로 인해 본질적으로 위험한 공정)을 더 잘 이해하고 최적화하는데 특히 효과적입니다.
동일한 접근 방식은 또한 믹서 관련 장비 공급 업체가 고객 요구에 맞게 제품을보다 정확하게 개발하고 맞춤화하는 데 도움이되었습니다. “이는 불필요한 프로토 타이핑 비용이나 잠재적 인 과도한 엔지니어링을 방지합니다. 두 가지 모두 일부 공급 업체의 문제였습니다.”라고 Karampelas는 말합니다.
CFD 기술 자체는 계속해서 발전하고 있습니다. 예를 들어, 수치 알고리즘의 관점에서 볼 때 구형 입자의 상호 작용이 열 전달을 적절하게 모델링하는 데 중요한 다양한 문제에 대해 이산 요소 모델링을 쉽게 적용 할 수있는 반면, LES 난류 모델은 난류 흐름 패턴을 정확하게 시뮬레이션하는 데 이상적입니다.
컴퓨팅 리소스에 대한 비용과 수요에도 불구하고 Karampelas는 난류 모델의 전체 제품군을 제공 할 수있는 것이 중요하다고 생각합니다. 특히 LES는 이미 대부분의 학계와 일부 산업 (예 : 전력 공학)에서 선택하는 방법이기 때문입니다. .
그럼에도 불구하고 CFD의 사용이 제한적이거나 비실용적 일 수있는 경우는 확실히 있습니다. 여기에는 나노 입자에서 벌크 유체 증발을 모델링하는 것과 같이 관심의 규모가 다른 규모에 따라 달라질 수있는 문제와 중요한 물리적 현상이 아직 알려지지 않았거나 제대로 이해되지 않았거나 아마도 매우 복잡한 문제 (예 : 모델링)가 포함됩니다. 음 펨바 효과”라고 Karampelas는 경고합니다.
반면에 더욱 강력한 하드웨어와 업데이트 된 수치 알고리즘의 출현은 CFD 소프트웨어를 사용하여 과다한 설계 및 최적화 문제를 해결하기위한 최적의 접근 방식이 될 것이라고 그는 믿습니다.
“복잡한 열교환 시스템 및 새로운 혼합 기술과 같이 점점 더 복잡한 공정을 모델링 할 수있는 능력은 가까운 장래에 가능할 수있는 일을 간단히 보여줍니다. 수치적 방법 사용의 주요 이점은 설계자가 상상력에 의해서만 제한되어 소규모 믹서에서 대규모 반응기 및 증류 컬럼에 이르기까지 다양한 화학 플랜트 공정을 최적화 할 수있는 길을 열어 준다는 것입니다. 실험적 또는 경험적 접근 방식은 항상 관련성이 있지만 CFD가 미래의 엔지니어를위한 선택 도구가 될 것이라고 확신합니다.”라고 그는 결론을 내립니다.
드리프트 모델은 밀도가 서로 다른 두 혼합 유체 구성 요소의 상대적 흐름을 설명합니다. 구성 요소는 상이 다를 수도 있고, 상이 같지만(불가침) 유체가 다를 수도 있습니다. 분산된 위상 입자 크기가 클 경우 드리프트 모델의 적용성에 대한 제한이 존재할 수 있습니다. 이러한 제한은 일반적으로 메쉬 셀 크기의 10% 미만으로 분산된 위상 입자 크기를 유지함으로써 방지할 수 있습니다.
배플
얇은 형상 조각을 나타내는데 사용되는 2 차원 개체입니다. 이들은 전처리기에 의해 셀면으로 이동되고 유체의 흐름을 부분적으로 또는 완전히 차단하는 역할을 합니다. 배플은 지정된 열 전달 계수를 가질 수 있으며 배플을 통과하는 양(플럭스 표면)을 측정하는 데 사용할 수 있습니다.
Two-dimensional objects that are used to represent thin pieces of geometry. They are moved by the preprocessor to cell faces and act to partially, or completely block the flow of fluid. Baffles can have heat transfer coefficients specified and can be used to measure quantities that pass through them (a flux surface).
경계 조건
도메인의 범위에서 솔루션을 정의합니다. 경계 위치에서 흐름의 실제 상태를 나타내는 경계 조건을 선택하는 것이 중요합니다.
Defines the solution at the extents of the domain. It is important to choose boundary conditions that represent the true condition of the flow at the boundary location.
CFD
CFD (Computational Fluid Dynamics)는 수치 솔루션을 통해 컴퓨터의 유체 흐름을 시뮬레이션 하는 유체 역학의 한 분야입니다.
Computational Fluid Dynamics (CFD), the branch of fluid mechanics dedicated to simulating the flow of fluid on a computer via numerical solutions.
Complements
Complements를 정의합니다. 예를 들어, 솔리드 구의 complements는 솔리드 재료로 둘러싸인 구형 구멍입니다.
The inverse of a shape defines the complement. For example, the complement of a solid sphere is a spherical hole surrounded by solid material.
Client
클라이언트 컴퓨터는 자신이 FLOW-3D를 실행하고 있지만, FLOW-3D 소프트웨어 라이선스는 다른 컴퓨터 (서버 컴퓨터)에서 획득하는 컴퓨터를 의미합니다.
A client machine is a computer that runs FLOW-3D but acquires the software license from a different machine (the server machine)
Components
Components는 공간의 개체를 정의하며 하위 구성 요소로 구성됩니다. 구성 요소는 열 전도율, 비열 및 표면 거칠기와 같은 재료 특성을 가질 수 있습니다.
Components define objects in space and are comprised of subcomponents. A component can have material properties such as thermal conductivity, specific heat and surface roughness.
Custom result
시뮬레이션 중 또는 완료 후 사용자가 생성한 데이터를 그래픽으로 표시합니다. 생성하려면 사용자가 flsgrf결과 파일을 연 다음 플로팅 매개 변수(예 : 플로팅 할 도메인 부분, 플로팅 할 수량 등)를 선택해야 합니다.
Graphical displays of data generated by the user during the simulation or after it has completed. To generate, the user must open an flsgrf results file and then select the plotting parameter (e.g., portion of domain to plot, quantity to plot, etc.).
Domain
지배 방정식을 풀 영역입니다. 이것은 메쉬의 범위에 의해 정의됩니다.
The region in which the governing equations are to be solved. This is defined by the extents of the mesh.
Diagnostics
전 처리기 및 솔버의 진행 상황과 오류 및 경고에 대한 정보가 포함된 파일 세트입니다.
A suite of files that contain information on the progress of the preprocessor and solver as well as errors and warnings.
EPSI
압력/연속 반복이 어느 지점에서 수렴되는지를 결정하는데 사용된 수렴 기준입니다. 기본 숫자 설정을 사용하면 이 값은 FLOW-3D에 의해 자동으로 계산되며 시간 단계가 증가함에 따라 작아집니다.
The convergence criterion that was used to determine at what point the pressure/continuity iterations have converged. With the default numerical settings, this value is automatically computed by FLOW-3D and becomes smaller as the time step increases.
Existing result
prpplt.* 또는 flsplt.* 파일은 전처리 종료 솔버 실행 종료시 또는 자동으로 생성되는 플롯 파일입니다.
A plot file that is automatically created, either at the end of preprocessing or the end of the solver run- prpplt.* or flsplt.*.
F3D_HOME
FLOW-3D 프로그램 파일이 있는 디렉토리를 정의하는 환경 변수.
Environment variable that defines the directory where the FLOW-3D program files are located.
Floating license
FLOW-3D는 서버 시스템에 라이센스를 액세스하는 각 클라이언트 컴퓨터와 컴퓨터 네트워크에서 실행합니다. 허용하는 라이센스 최대 동시 시뮬레이션 수는 구매한 솔버 토큰 수에 의해 제한됩니다.
A license that allows FLOW-3D to be run on a network of computers with each client machine accessing the license on a server machine. The maximum number of concurrent simulations is limited by the number of solver tokens purchased.
Flsgrf file
솔버가 생성한 결과 파일. 이 파일은 사전에 정의된 시간 간격으로 생성된 정보를 포함하며 그래픽 디스플레이를 생성하는 데 사용됩니다. 사용자 정의 플로팅 중에 포스트 프로세서에서 사용합니다.
Results file produced by the solver. This file contains information produced at predefined time intervals and is used to produce graphical displays. Used by the postprocessor during custom plotting.
Flsplt file
솔버가 자동으로 생성한 플롯 파일입니다. 이 파일에는 시뮬레이션의 히스토리 데이터, 메시 등에 대한 기본 정보와의 $GRAFIC 이름 목록에 사전 정의된 그래픽 요청이 포함되어 prepin.* 파일 안에 있습니다.
Plot file produced automatically by the solver. This file contains basic information on history data, mesh, etc. from the simulation as well as any pre-defined graphics requests in the $GRAFIC namelist in prepin.*.
Fluid #1 surface area
선택한 길이 단위의 자유 표면 영역을 제곱 됩니다. 인터페이스가 예리한 문제에만 해당됩니다.
The free-surface area in the chosen length units squared. This is only relevant for problems with a sharp interface.
Fluid thermal energy
영역에 존재하는 모든 유체에 포함된 총 열 에너지 (에너지 전송이 켜져 있는 시뮬레이션에만 해당).
The total thermal energy contained by all the fluid present in the domain (relevant only for simulations with energy transport turned on).
Free surface
유체와 유체 사이의 인터페이스. FLOW-3D에서 이 인터페이스는 전단이 없는 것으로 가정되며, 이는 빈 공간에 있는 가스가 유체에 무시할 수 있는 트랙션을 발휘함을 의미한다.
The interface between fluid and void. In FLOW-3D , this interface is assumed to be shear-free, meaning that any gas in the void space exerted negligible traction on the fluid.
GUI
” Graphical User Interface”. GUI는 사용자가 FLOW-3D를 제어할 수 있는 그래픽 패널, 대화 상자 및 창을 제공합니다.
“Graphical User Interface”. The GUI presents the graphical panels, dialog boxes and windows that allow the user to control FLOW-3D .
Iteration count
각 시간 단계에서 필요한 압력/연속 반복 횟수입니다. 압력/연속성 반복은 유체 볼륨이 유지되도록 하고 유체 전체에서 올바른 압력을 계산하는 데 필요합니다.
The number of pressure/continuity iterations required at each time step. The pressure/continuity iterations are necessary to ensure that the fluid volume is maintained and to compute the correct pressure throughout the fluid.
License file
사용자가 FLOW-3D 를 실행할 수 있도록 암호화된 정보가 포함된 Flow Science에서 제공하는 전자 파일 입니다.
Electronic file provided by Flow Science that contains encrypted information enabling the user to run FLOW-3D .
License server
플로팅 라이센스 시스템의 작동을 활성화하기 위해 FLEXlm 라이센스 소프트웨어가 설치된 시스템. FLOW-3D는 License Server에 설치할 필요가 없습니다.
Computer on which the FLEXlm licensing software is installed to enable the operation of a floating license system. FLOW-3D does not need to be installed on the license server.
Licensing
FLOW-3D 실행을 제어하는 FLEXlm 소프트웨어.
FLEXlm software that controls the running of FLOW-3D .
Max. residual
압력/연속성 반복의 최종 반복에서 연속성 방정식의 실제 발산. 이 값은 메시지가 나타나지 않는 한 일반적으로 epsi보다 작습니다 .
The actual divergence of the continuity equation on the final iteration of the pressure/continuity iterations. This value is usually smaller than epsi unless the message, pressure iteration did not converge in xxxx iterations appears.
Mean kinetic energy
모든 계산 셀의 운동 에너지의 합을 도메인에 존재하는 총 유체 질량으로 나눈 값입니다. 이 양이 시간이 지남에 따라 변하지 않으면 정상 상태에 도달했음을 나타내는 좋은 지표입니다.
The sum of kinetic energy of all the computational cells, divided by the total mass of fluid present in the domain. When this quantity ceases to change over time, it is a good indicator that steady-state has been reached.
Node-locked license
특정 컴퓨터에 고정된 라이센스. 노드 잠금 라이센스는 네트워크를 통해 액세스 할 수 없으므로 일반적으로 모든 작업을 한 컴퓨터에서 수행해야하는 경우에만 사용됩니다.
A license that is locked to a particular computer. A node-locked license cannot be accessed across a network, and so is typically only used when all work is to be done on one computer.
Non-inertial reference frame
가속화되는 참조 프레임. 비 관성 참조 프레임은 움직이는 컨테이너를 모방하는 데 사용할 수 있습니다.
A frame of reference that is accelerating. A non-inertial reference frame can be used to mimic a moving container.
Pltfsi
1D 및 2D 플롯을 생성하는 FLOW-3D에 포함된 그래픽 디스플레이 프로그램.
Graphics display program included with FLOW-3D that produces 1D and 2D plots.
Postprocessor
FLOW-3D 내의 Postprocessor 프로그램은 FLOW-3D 또는 타사 시각화 프로그램에서 읽을 수 있는 데이터 파일을 생성하거나 타사 소프트웨어 프로그램에서 읽을 텍스트 데이터를 생성하는 솔버 출력 데이터를 처리하는 프로그램입니다.
The program within FLOW-3D that processes the solver output data to produce data files that can be read by FLOW-3D ’s or third-party’s visualization programs, or produce text data to be read by third party software programs.
Prepin file
FLOW-3D 시뮬레이션을 실행하는데 필요한 모든 정보가 포함된 텍스트 파일 입니다. GUI를 사용하거나 텍스트 편집기를 사용하여 수동으로 작성할 수 있습니다.
Text file that contains all of the information necessary to create a FLOW-3D simulation. It can be created using the GUI or manually with a text editor.
Preprocessor
솔버의 실행을 준비하기 위해 입력 파일을 기반으로 메쉬 및 초기 조건을 생성하는 FLOW-3D 내의 프로그램 입니다.
The program within FLOW-3D that generates the mesh and initial conditions based on the input file in preparation for the running of the solver.
Prpgrf file
전처리기에 의해 생성된 결과 파일로 전 처리기의 정보를 포함하며 후 처리기에서 사용자 플롯을 생성하는 데 사용할 수 있습니다. 이 파일은 미리보기 버튼을 선택하거나 시뮬레이션에서 사전 프로세서(runpre 사용)를 실행하는 경우에만 실행됩니다.
Results file produced by the preprocessor. Contains information from the preprocessor and can be used by the postprocessor to create custom plots. This file is produced only when the Preview button is selected or if only the pre-processor is run on the simulation (using runpre).
Prpplt file
전처리기에 의해 자동으로 생성된 파일을 플롯 합니다. 메시, 구성 요소, 초기 조건 및 재료 특성에 대한 정보가 포함되어 있습니다.
Plot file produced automatically by the preprocessor. Contains information on meshing, components, initial conditions and material properties.
Restart simulation
이전 시뮬레이션에서 계속되는 시뮬레이션입니다. 이전 시뮬레이션의 결과는 다시 시작 시뮬레이션을 위한 초기 조건 및 (선택적으로) 경계 조건을 생성하는 데 사용됩니다.
A simulation which continues from a previous simulation. The results from the previous simulation are used to generate the initial conditions and (optionally) boundary conditions for the restart simulation.
Server
라이센스 서버를 호스팅하는 시스템
The machine that hosts the license server.
Stability limit
각 시간 단계에서 사용할 수 있는 최대 시간 단계. 더 큰 시간 단계는 수치적 불안정성과 비물리적 결과로 이어질 것이다.
The maximum time step that can be used during each time step. A larger time step will lead to numerical instabilities and nonphysical results.
STL (Stereolithography) File
.STL 파일 형식은 일련의 삼각형이 있는 솔리드 모델의 표면에 근접한 표준 데이터 전송 형식이다. 삼각형은 가장자리에서 결합해야 하며 일관된 방향을 가리키는 정규식이 있어야 한다.
The .STL file format is a standard data transmission format that approximates the surfaces of a solid model with a series of triangles. The triangles must join at the edges and must have normals that point in a consistent direction.
Solid fraction
응고된 영역의 유체 분율 (응고 모델이 켜져 있는 시뮬레이션에만 해당).
The fraction of fluid in the domain that has become solidified (relevant only for simulations where the solidification model has been turned on).
Solver
입력 파일에 정의된 흐름 문제를 시뮬레이션하는 방정식을 계산하는 FLOW-3D 내의 솔버 프로그램 입니다.
The program within FLOW-3D that solves the system of equations that simulate the flow problem defined in the input file.
STL Viewer
스테레오리소그래피(STL) 파일을 표시하는 특수 유틸리티입니다. STL 파일은 CAD 소프트웨어로 제작되며 3 차원 객체의 표면을 형성하는 많은 삼각형으로 구성됩니다. 의 STL 뷰어 FLOW-3D는 메인 메뉴에서 유틸리티/STL 뷰어를 클릭하여 GUI를 통해 액세스 할 수 있습니다. 그러면 뷰어가 별도의 창에서 열립니다. 메쉬 및 형상 탭에서 STL 파일을 열고 볼 수도 있습니다.
A special utility that displays stereolithography (STL) files. STL files are produced by CAD software and are composed of many triangles that form the surface of a three-dimensional object. The STL Viewer in FLOW-3D is accessible via the GUI by clicking Utilities/STL Viewer in the main menu. This causes the viewer to open in a separate window. STL files can also be opened and viewed in the Meshing and Geometry tab.
Subcomponents
하위 구성 요소는 구성 요소라고하는 더 큰 모양을 형성하기 위해 결합할 수 있는 기하학적 모양입니다. 하위 구성 요소는 재료를 추가하거나 (고체로) 다른 하위 구성 요소에서 재료를 제거하거나 (구멍으로) 또는 모양 외부에 재료를 추가하도록 정의할 수 있습니다.
Subcomponents are geometric shapes that can be combined to form larger shapes, called components. A subcomponent can be defined to add material (as solids), remove material from other subcomponents (as holes), or add material outside of the shape (as a complement).
Time-step size
계산에 사용된 실제 시간 단계. 이 값은 안정성 한계와 같거나 작을 수 있습니다.
The actual time step used in the computation. This value can be equal to or less than the stability limit.
Units
Units are based upon the values set for the physical properties. Items such as mesh block extents and cell lengths automatically conform to the units used for setting these physical properties.
단위는 물리적 특성에 설정된 값을 기반으로 합니다. 메쉬 블록 범위 및 셀 길이와 같은 항목은 이러한 물리적 속성을 설정하는 데 사용되는 단위를 자동으로 따릅니다.
Volume error (%)
주어진 시간에 도메인에 존재하는 총 유체의 백분율로 설명되지 않은 유체 부피의 백분율을 의미합니다. 따라서 단순히 총 부피가 작기 때문에 유체가 시스템 밖으로 배출되는 시뮬레이션에서 큰 비율의 부피 오류가 발생할 수 있습니다.
The percentage of fluid volume not accounted for as a percentage of the total fluid present in the domain at a given time. Therefore, a large percentage volume error can occur for simulations where fluid is draining out of the system simply because the total volume present is small.
Volume of fluid #1
선택한 길이 단위로 입방체에 존재하는 유체 #1의 총 부피입니다. 2 유체 문제의 경우, 유체 #2의 부피는 항상 도메인 부피에서 유체 #1의 부피를 뺀 값입니다.
The total volume of fluid #1 present in the system, in the chosen length units cubed. For two-fluid problems, the volume of fluid #2 is always the domain volume minus the volume of fluid #1.
Wall shear stress
FLOW-3D 옵션은 벽면 및 객체 인터페이스에서 전단 응력 계산을 켜거나 끌 수 있도록 해줍니다. “no-slip” 인터페이스의 효과를 모델링 하려면 벽면 전단 응력을 켜야 합니다.
The FLOW-3D option that allows the user to turn on or off the computation of shear stress at wall and object interfaces. Wall shear stress must be turned on to model the effect of “no-slip” interfaces.
Workspace
작업 공간은 시뮬레이션 프로젝트를 위한 파일 컨테이너입니다. 작업 공간은 사용자가 FLOW-3D 뿐만 아니라 하드 드라이브에서도 작업을 구성하는 데 도움이 됩니다.
A workspace is a file container for simulation projects. Workspaces help the user organize their work, not only within FLOW-3D , but also on their hard drive.
International Thermal Spray Conference – ITSC-2006 Seattle, Washington, U.S.A., May 2006
M. Ivosevic, V. Gupta, R. A. Cairncross, T. E. Twardowski, R. Knight, Drexel University, Philadelphia, Pennsylvania, USA J. A. Baldoni Duke University, North Carolina, USA
Abstract
거친 표면에 대한 입자 충격 및 변형의 3 차원 모델이 HVOF 스프레이 폴리머 입자에 대해 개발되었습니다. 유체 흐름 및 입자 변형은 FLOW-3D® 소프트웨어를 사용하는 유체 부피 (VoF) 방법으로 예측되었습니다. 스플래팅(splatting) 및 최종 스플랫 모양(splat shapes)의 역학에 대한 거칠기의 영향은 몇 가지 프로토타입 거친 표면을 사용하여 탐색 되었습니다 (예: 단계와 그루브)
또한 실제 그릿 블라스팅(grit blasted)된 강철 표면의 광학 간섭 측정에 의해 생성된 보다 사실적인 거친 표면의 수치 표현도 모델에 통합되었습니다. 예측된 스플랫 모양을 그릿 블라스팅 된 강철 기판에 증착된 나일론 11 스플랫의 SEM 이미지와 비교했습니다. 거친 기판은 부드러운 기판의 스플래팅 시뮬레이션에서 거의 관찰되지 않는 손가락 및 기타 비대칭 3 차원 불안정성을 생성했습니다.
Introduction
기판 거칠기가 용사 코팅의 접착력과 접착력을 향상 시킨다는 사실은 잘 알려져 있으며 일반적으로 받아 들여지고 있습니다 [1]. 스프레이하기 전에 기판 표면은 일반적으로 알루미나 또는 SiC와 같은 50 – 300 µm 각 세라믹 입자로 그릿 블라스팅으로 거칠게 처리됩니다.
기판 표면에 증착된 초기 스플랫의 형태는 코팅 / 기판 인터페이스의 무결성과 결과 코팅의 접착 강도에 중요한 역할을합니다. 단단하고 불규칙한 표면에 대한 열 스프레이 액적의 충격 및 변형은 액적 표면의 복잡한 대규모 3 차원 변형이 특징입니다.
충돌하는 물방울의 “스플래싱”이 발생하는 경우, 운지법 또는 위성 입자 생성 및 분리 중 새로운 표면 생성은 일반적으로 축 대칭이 아니므로 사실적인 splat 예측을 위해 3 차원 모델이 필요합니다. 이것은 정확한 3 차원 스플래팅 모델의 개발에 많은 수치적 도전을 야기합니다.
Fauchais et al. [2]는 스플랫 형성 과정과 관련하여 발표 된 논문의 대부분 (~ 98 %)이 매끄러운 표면에 대한 정상적인 액적 충격을 설명한다고보고했습니다. 게시된 작업의 2 % 미만은 매끄러운 표면에 대한 비정상적인 입자 영향과 관련이 있으며 ~ 0.1 %만이 거친 기판과 관련됩니다.
여러 저자 [3, 4]는 2 차원 모델을 사용하여 비평면 표면과 물방울의 상호 작용을 연구했거나 평행 그루브가 있는 표면에 대한 3 차원 충격 [5]을 연구했습니다. 그러나 이 접근법의 주요 단점은 거친 표면에 스플래팅의 비축 대칭 측면을 연구합니다.
최근 Raessi et al. [6] 이전에 개발된 VoF 모델 [7]을 확장하여 평평한 기판에 액적 스플래팅을 프로토 타입 거친 표면과 액적 상호 작용으로 확장했습니다. 표면 거칠기는 규칙적으로 정렬 된 정사각형 블록으로 근사화 되었습니다. Feng et al. [8]은 평평한 표면의 마찰 조건에 의해 표면 거칠기가 근사된 3 차원 Lagrangian 유한 요소 모델을 사용했습니다.
이 접근 방식은 소규모 점성 및 축 대칭 자유 표면 흐름과 관련하여 매우 정확할 수 있지만 fingering 생성 또는 satellites 생성 및 breakups 중 새로운 표면 생성과 관련된 물방울이 튀기는 경계 맞춤 기술에 적합하지 않습니다.
또한, 열 분무에 사용되는 그릿 블라스팅 표면의 평균 표면 거칠기 (Ra)는 일반적으로 50μm의 평균 액적 크기에 비해 ~ 5 ~ 30 % (~ 2 ~ 15μm)입니다. 평평한 표면에 간단한 마찰 흐름.
본 연구의 목표는 임의의 거친 기질에 영향을 미치는 HVOF 분무 중합체 입자의 모델을 개발하는 것이다. 매끄럽지 않은 표면에 대한 입자 분할 모델은 표면의 기하학적 불규칙성이 분할 거동과 최종 분할 형태에 어떻게 영향을 미치는지 더 잘 이해할 수 있게 해줄 것입니다.
HVOF 제트에서 미크론 크기의 공급 원료 입자로의 강제 대류는 높은 대류 열 전달 계수 (h ~ 5000 – 17,000 W / (m2 K))를 특징으로 합니다. 이로 인해 입자 표면 온도가 급격히 증가하지만 폴리머 입자의 높은 내부 열 저항 (높은 Bi 수)은 입자 내부가 동일한 속도로 가열되는 것을 방지합니다. 결과적으로 더 큰 (예 : 90 µm 직경) 나일론 11 입자는 기판에 충격을 주기 전에 코어와 표면 사이에 급격한 온도 구배를 나타냅니다 (그림 1) [9, 10, 11].
Figure 1: Temperature of a 90 µm diameter Nylon 11 particle
with respect to normalized particle radius (r/R) [10].Figure 2: (a) Velocity field within a spreading 90 µm diameter
particle; (Left): velocity magnitude, (Right): velocity vectors,
(b) example Nylon 11 splat deposited via swipe test onto a
room temperature glass slide.
또한 가파른 내부 온도 구배를 가진 HVOF 스프레이 폴리머 입자가 얇은 디스크 중앙에 크고 거의 반구형 인 코어가있는 특징적인 “튀김 달걀”모양으로 퍼졌다고 보고되었습니다 [10]. 이 모양은 저온, 고점도 코어와 고온, 저점도 표면의 유동 특성 간에 큰 방사형 차이가 있음을 나타냅니다.
변형된 입자의 예측 된 모양 (그림 2a)은 유리 슬라이드에 증착된 실험적으로 관찰 된 스플랫과 좋은 질적 일치를 나타 냈습니다 (그림 2b). 액적의 오른쪽에 표시된 속도 장 벡터 (그림 2a)는 저점도 “피부”가 고점도 코어 주위를 흐르면서 특징적인 “튀김 달걀” splat 모양이 형성되었음을 나타냅니다.
이 작업에서 보고된 실험 중에 사용된 HVOF 스프레이 매개 변수는 나일론 11을 증착하는데 사용할 수 있는 일반적인 HVOF 스프레이 매개 변수를 나타냅니다. 그러나 실험 기준 매개 변수를 중심으로 개발된 수치 모델은 개별 스플랫의 흐름 거동을 더 잘 이해하는 데 사용할 수 있습니다. 증착 효율 향상을 위한 공정 최적화를 지원합니다.
Figure 3: Boundary conditions, initial conditions and crosssection of a typical mesh used in Flow-3DFigure 5: Cross section of four steel substrates: (a) polished
with ~1 Pm alumina suspension, (b) grit blasted with #120
grit, (c) grit blasted with #50 grit, (d) grit blasted with #12
grit. Top image shows optical interferometry scan of # 120
grit blasted surface.Figure 6: Nylon-11 splats deposited during a single run over
steel substrates with roughnesses as per Figure 5.Figure 7: Nylon-11 splat on a grit blasted steel substrate, (a)
close up of a peripheral splat finger.Figure 8: Cross-sections of predicted three-dimensional spreading splats for a 90 µm diameter Nylon-11 particle on four different
surface roughnesses (dimensionless time t* = t/(D/v
o
(p))).Figure 9: Predicted three-dimensional spreading splats for a
90 µm diameter Nylon-11 droplet.
Davis, J. R., (Ed.) et al, Handbook of Thermal Spray Technology, ASM International®, 1st Ed., Materials Park, OH, (2004).
Fauchais, P., Fukomoto, M., Vardelle, A. and Vardelle, M., Knowledge Concerning Splat Formation: An Invited Review, Journal of Thermal Spray Technology, 13 (3), pp. 337 – 360, (2004).
Liu, H., Lavernia, E. J. and Rangel, R. H., Modeling of Molten Droplet Impingement on a Non-flat Surface, Acta Metall. Mater, 43(5), pp. 2053 – 2072, (1995).
Sobolev, V. V., Guilemany, J. M. and Martin, A. J., Influence of Surface Roughness on the Flattening of Powder Particles during Thermal Spraying, Journal of Thermal Spray Technology 5(2), pp. 207 – 214, (1996). 5 Patanker, N. A. and Chen, Y., Numerical Simulation of Droplet Shapes on Rough Surfaces, Proc. Int. Conference on Modeling and Simulations of Microsystems – MSM 2002, pp. 116 – 119, (2002) 6 Raessi, M., Mostaghimi, J. and Bussmann, M., “Droplet Impact during the Plasma Spray Coating Process-Effect of Surface Roughness on Splat Shapes,” Proc. 17th Int. Symposium on Plasma Chemistry – ISPC 17, Toronto, Canada, (2005) 7 Pasandideh-Fard, M., Chandra, S. and Mostaghimi, J., A Three-dimensional Model of Droplet Impact and Solidification, Int. J. Heat and Mass Transfer, 45, pp. 2229 – 2242, (2002). 8 Feng, Z. G., Domaszewski, M., Montavon, G. and Coddet, C., Finite Element Analysis of Effect of Substrate Surface Roughness on Liquid Droplet Impact and Flattening Process, J. of Thermal Spray Technology, 11(1), pp. 62-68, (2002). 9 Petrovicova, E., “Structure and Properties of Polymer Nanocomposite Coatings Applied by the HVOF Process,” Ph.D. Dissertation, Drexel University, (1999). 10 Ivosevic, M., Cairncross, R. A., Knight, R., Impact Modeling of Thermally Sprayed Polymer Particles, Proc. ITSC-2005 International Thermal Spray Conference, DVS/IIW/ASM-TSS, Basel, Switzerland, (2005). 11 Bao, Y., Gawne, D. T. and Zhang, T., The Effect of Feedstock Particle Size on the Heat transfer Rates and Properties of Thermally Sprayed Polymer Coatings, Trans. I. M. F., 73(4), pp 119 – 124, (1998). 12 Ivosevic, M., Cairncross, R. A. and Knight, R., “Heating and Impact Modeling of HVOF Sprayed Polymer Particles,” Proc. 2004 International Thermal Spray Conference (ITSC-2004), DVS/IIW/ASM-TSS, Osaka, Japan, (2004). 13 Hirt, C. W. and Nichols, B. D., Volume of Fluid (VoF) Method for the Dynamics of Free Boundaries, Journal of Computational Physics, 39, pp. 201 – 225, (1981).
X.J. Liu a,∗, S.H. Bhavnani b,1, R.A. Overfelt c,2 a United States Steel Corporation, Great Lakes Works, #1 Quality Drive, Ecorse, MI 48229, United States b 213 Ross Hall, Department of Mechanical Engineering, Auburn University, Auburn, AL 36849-5341, United States c 202 Ross Hall, Department of Mechanical Engineering, Materials Engineering Program, Auburn University, Auburn, AL 36849-5341, United States Received 17 April 2006; received in revised form 14 July 2006; accepted 21 August 2006
Keywords: Lost foam casting; Heat transfer coefficient; Gas pressure; VOF-FAVOR
LFC (Loss Foam Casting) 공정에서 부드러운 몰드 충진의 중요성은 오랫동안 인식되어 왔습니다. 충진 공정이 균일할수록 생산되는 주조 제품의 품질이 향상됩니다. 성공적인 컴퓨터 시뮬레이션은 금형 충전 공정에서 복잡한 메커니즘과 다양한 공정 매개 변수의 상호 작용을 더 잘 이해함으로써 새로운 주조 제품 설계의 시도 횟수를 줄이고 리드 타임을 줄이는데 도움이 될 수 있습니다.
이 연구에서는 용융 알루미늄의 유체 흐름과 금속과 발포 폴리스티렌 (EPS) 폼 패턴 사이의 계면 갭에 관련된 열 전달을 시뮬레이션하기 위해 전산 유체 역학 (CFD) 모델이 개발되었습니다.
상업용 코드 FLOW-3D는 VOF (Volume of Fluid) 방법으로 용융 금속의 전면을 추적 할 수 있고 FAVOR (Fractional Area / Volume Ratios) 방법으로 복잡한 부품을 모델링 할 수 있기 때문에 사용되었습니다. 이 코드는 폼 열화 및 코팅 투과성과 관련된 기체 갭 압력을 기반으로 다양한 계면 열 전달 계수 (VHTC)의 효과를 포함하도록 수정되었습니다.
수정은 실험 연구에 대해 검증되었으며 비교는 FLOW-3D의 기본 상수 열 전달 (CHTC) 모델보다 더 나은 일치를 보여주었습니다. 금속 전면 온도는 VHTC 모델에 의해 실험적 불확실성 내에서 예측되었습니다. 몰드 충전 패턴과 1-4 초의 충전 시간 차이는 여러 형상에 대해 CHTC 모델보다 VHTC 모델에 의해 더 정확하게 포착되었습니다. 이 연구는 전통적으로 매우 경험적인 분야에서 중요한 프로세스 및 설계 변수의 효과에 대한 추가 통찰력을 제공했습니다.
지난 20 년 동안 LFC (Loss Foam Casting) 공정은 코어가 필요없는 복잡한 부품을 제조하기 위해 널리 채택되었습니다. 이는 자동차 제조업체가 현재 LFC 기술을 사용하여 광범위한 엔진 블록과 실린더 헤드를 생산하기 때문에 알루미늄 주조 산업에서 특히 그렇습니다.
기본 절차, 적용 및 장점은 [1]에서 찾을 수 있습니다. LFC 프로세스는 주로 숙련 된 실무자의 경험적 지식을 기반으로 개발되었습니다. 발포 폴리스티렌 (EPS) 발포 분해의 수치 모델링은 최근에야 설계 및 공정 변수를 최적화하는 데 유용한 통찰력을 제공 할 수있는 지점에 도달했습니다. LFC 공정에서 원하는 모양의 발포 폴리스티렌 폼 패턴을 적절한 게이팅 시스템이있는 모래 주형에 배치합니다.
폼 패턴은 용융 금속 전면이 패턴으로 진행될 때 붕괴, 용융, 기화 및 열화를 겪습니다. 전진하는 금속 전면과 후퇴하는 폼 패턴 사이의 간격 인 운동 영역은 Warner et al. [2] LFC 프로세스를 모델링합니다. 금형 충진 과정에서 분해 산물은 운동 영역에서 코팅층을 통해 모래로 빠져 나갑니다.
용융 금속과 폼 패턴 사이의 복잡한 반응은 LFC 공정의 시뮬레이션을 극도로 어렵게 만듭니다. SOLA-VOF (SOLution AlgorithmVolume of Fluid) 방법이 Hirt와 Nichols [3]에 의해 처음 공식화 되었기 때문에 빈 금형을 사용한 전통적인 모래 주조 시뮬레이션은 광범위하게 연구되었습니다.
Lost foam 주조 공정은 기존의 모래 주조와 많은 특성을 공유하기 때문에이 새로운 공정을 모델링하는 데 적용된 이론과 기술은 대부분 기존의 모래 주조를 위해 개발 된 시뮬레이션 방법에서 비롯되었습니다. 패턴 분해 속도가 금속성 헤드와 금속 전면 온도의 선형 함수라고 가정함으로써 Wang et al. [4]는 기존의 모래 주조의 기존 컴퓨터 프로그램을 기반으로 복잡한 3D 형상에서 Lost foam 주조 공정을 시뮬레이션했습니다.
Liu et al. [5]는 금속 앞쪽 속도를 예측하기 위한 간단한 1D 수학적 모델과 함께 운동 영역의 배압을 포함했습니다. Mirbagheri et al. [6]은 SOLA-VOF 기술을 기반으로 금속 전면의 자유 표면에 대한 압력 보정 방식을 사용하는 Foam 열화 모델을 개발했습니다.
Kuo et al.에 의해 유사한 배압 방식이 채택되었습니다. [7] 운동량 방정식에서이 힘의 값은 실험 결과에 따라 패턴의 충전 순서를 연구하기 위해 조정되었습니다.
이러한 시뮬레이션의 대부분은 LFC 공정의 충전 속도가 기존의 모래 주조 공정보다 훨씬 느린 것으로 성공적으로 예측합니다. 그러나 Foam 분해의 역할은 대부분 모델의 일부가 아니며 시뮬레이션을 수행하려면 실험 데이터 또는 경험적 함수가 필요합니다.
현재 연구는 일정한 열전달 계수 (CHTC)를 사용하는 상용 코드 FLOW-3D의 기본 LFC 모델을 수정하여 Foam 열화와 관련된 기체 갭 압력에 따라 다양한 열전달 계수 (VHTC)의 영향을 포함합니다. 코팅 투과성. 수정은 여러 공정 변수에 대한 실험 연구에 대해 검증되었습니다.
또한, 손실 된 폼 주조에서 가장 중요한 문제인 결함 형성은 문헌에서 인용 된 수치 작업에서 모델링되지 않았습니다. 접힘, 내부 기공 및 표면 기포와 같은 열분해 결함은 LFC 작업에서 많은 양의 스크랩을 설명합니다. FLOW-3D의 결함 예측 기능은 프로세스를 이해하고 최적화하는데 매우 중요합니다.
Fig. 7. Comparison of mold filling times for a plate pattern with three ingates: (a) measured values by thermometric technique [18]; (b) predicted filling times based
on basic CHTC model with gravity effect; and (c) predicted filing times based on the VHTC model with heat transfer coefficient changing with gas pressure; (d)
mold filling time at the right-and wall of the mold for the plate pattern with three ingates.Fig. 10. Defects formation predicted by (a) basic CHTC model with gravity effect; (b) VHTC model with heat transfer coefficient based on both gas pressure and
coating thickness; and (c) improved model for two ingates. Color represents probability for defects (blue is the lowest and red highest).
[1] S. Shivkumar, L. Wang, D. Apelian, The lost-foam casting of aluminum alloy components, JOM 42 (11) (1990) 38–44. [2] M.H. Warner, B.A. Miller, H.E. Littleton, Pattern pyrolysis defect reduction in lost foam castings, AFS Trans. 106 (1998) 777–785. [3] C.W. Hirt, B.D. Nichols, Volume of Fluid (VOF) method for the dynamics of free boundaries, J. Comp. Phys. 39 (1) (1981) 201–225. [4] C. Wang, A.J. Paul, W.W. Fincher, O.J. Huey, Computational analysis of fluid flow and heat transfer during the EPC process, AFS Trans. 101 (1993) 897–904. [5] Y. Liu, S.I. Bakhtiyarov, R.A. Overfelt, Numerical modeling and experimental verification of mold filling and evolved gas pressure in lost foam casting process, J. Mater. Sci. 37 (14) (2002) 2997–3003. [6] S.M.H. Mirbagheri, H. Esmaeileian, S. Serajzadeh, N. Varahram, P. Davami, Simulation of melt flow in coated mould cavity in the lost foam casting process, J. Mater. Process. Technol. 142 (2003) 493–507. [7] J.-H. Kuo, J.-C. Chen, Y.-N. Pan, W.-S. Hwang, Mold filling analysis in lost foam casting process for aluminum alloys and its experimental validation, Mater. Trans. 44 (10) (2003) 2169–2174. [8] C.W. Hirt, Flow-3D User’s Manual, Flow Science Inc., 2005. [9] E.S. Duff, Fluid flow aspects of solidification modeling: simulation of low pressure die casting, The University of Queensland, Ph.D. Thesis, 1999. [10] X.J. Liu, S.H. Bhavnani, R.A. Overfelt, The effects of foam density and metal velocity on the heat and mass transfer in the lost foam casting process, in: Proceedings of the ASME Summer Heat Transfer Conference, 2003, pp. 317–323. [11] W. Sun, P. Scarber Jr., H. Littleton, Validation and improvement of computer modeling of the lost foam casting process via real time X-ray technology, in: Multiphase Phenomena and CFD Modeling and Simulation in Materials Processes, Minerals, Metals and Materials Society, 2004, pp. 245–251. [12] T.V. Molibog, Modeling of metal/pattern replacement in the lost foam casting process, Materials Engineering, University of Alabama, Birmingham, Ph.D. Thesis, 2002. [13] X.J. Liu, S.H. Bhavnani, R.A. Overfelt, Measurement of kinetic zone temperature and heat transfer coefficient in the lost foam casting process, ASME Int. Mech. Eng. Congr. (2004) 411–418. [14] X. Yao, An experimental analysis of casting formation in the expendable pattern casting (EPC) process, Department of Materials Science and Engineering, Worcester Polytechnic Institute, M.S. Thesis, 1994. [15] M.R. Barkhudarov, C.W. Hirt, Tracking defects, Die Casting Engineer 43 (1) (1999) 44–52. [16] C.W. Hirt, Modeling the Lost Foam Process with Defect PredictionsProgress Report: Lost-Foam Model Extensions, Wicking, Flow Science Inc., 1999. [17] D. Wang, Thermophysical Properties, Solidification Design Center, Auburn University, 2001. [18] S. Shivkumar, B. Gallois, Physico-chemical aspects of the full mold casting of aluminum alloys, part II: metal flow in simple patterns, AFS Trans. 95 (1987) 801–812.
by Vahid Bazargan M.A.Sc., Mechanical Engineering, The University of British Columbia, 2008 B.Sc., Mechanical Engineering, Sharif University of Technology, 2006 B.Sc., Chemical & Petroleum Engineering, Sharif University of Technology, 2006
고착 방울은 평평한 기판에 놓인 액체 방울입니다. 작은 고정 액적이 증발하는 동안 액적의 접촉선은 고정된 접촉 영역이 있는 고정된 단계와 고정된 접촉각이 있는 고정 해제된 단계의 두 가지 단계를 거칩니다. 고정된 접촉 라인이 있는 증발은 액적 내부에서 접촉 라인을 향한 흐름을 생성합니다.
이 흐름은 입자를 운반하고 접촉 선 근처에 침전시킵니다. 이로 인해 일반적으로 관찰되는 “커피 링”현상이 발생합니다. 이 논문은 증발 과정과 고착성 액적의 증발 유도 흐름에 대한 연구를 제공하고 콜로이드 현탁액에서 입자의 침착에 대한 통찰력을 제공합니다. 여기서 우리는 먼저 작은 고착 방울의 증발을 연구하고 증발 과정에서 기판의 열전도도의 중요성에 대해 논의합니다.
현재 증발 모델이 500µm 미만의 액적 크기에 대해 심각한 오류를 생성하는 방법을 보여줍니다. 우리의 모델에는 열 효과가 포함되어 있으며, 특히 증발 잠열의 균형을 맞추기 위해 액적에 열을 제공하는 기판의 열전도도를 포함합니다. 실험 결과를 바탕으로 접촉각의 진화와 관련된 접촉 선의 가상 움직임을 정의하여 고정 및 고정 해제 단계의 전체 증발 시간을 고려합니다.
우리의 모델은 2 % 미만의 오차로 500 µm보다 작은 물방울에 대한 실험 결과와 일치합니다. 또한 유한한 크기의 라인 액적의 증발을 연구하고 증발 중 접촉 라인의 복잡한 동작에 대해 논의합니다. 에너지 공식을 적용하고 접촉 선이 구형 방울의 후퇴 접촉각보다 높은 접촉각을 가진 선 방울의 두 끝에서 후퇴하기 시작 함을 보여줍니다. 그리고 라인 방울 내부의 증발 유도 흐름을 보여줍니다.
마지막으로, 계면 활성제 존재 하에서 접촉 라인의 거동을 논의하고 입자 증착에 대한 Marangoni 흐름 효과에 대해 논의합니다. 열 Marangoni 효과는 접촉 선 근처에 증착 된 입자의 양에 영향을 미치며, 기판 온도가 낮을수록 접촉 선 근처에 증착되는 입자의 양이 많다는 것을 알 수 있습니다.
Figure 1.1: A water droplet with a radius of 1 mm resting on a glass substrate.
The surface of the droplet takes on a spherical cap shape. The contact
angle θ is defined by the balance of the interfacial forces.Figure 2.1: Evaporation modes of sessile droplets on a substrate: (a) evaporation at constant contact angle (de-pinned stage) and (b) evaporation at
constant contact area (pinned stage)Figure 2.2: A sessil droplet with its image can be profiled as the equiconvex
lens formed by two intersecting spheres with radius of a.Figure 2.3: The droplet life time for both evaporation modes derived from
Equation 2.2.Figure 2.4: A probability of escape for vapor molecules at two different sites
of the surface of the droplet for diffusion controlled evaporation. The
random walk path initiated from a vapor molecule is more likely to result in a return to the surface if the starting point is further away from
the edge of the droplet.Figure 2.5: Schematic of the sessile droplet on a substrate. The evaporation
rate at the surface of the droplet is enhanced toward the edge of the
droplet.Figure 2.6: The domain mesh (a) and the solution of the Laplace equation for
diffusion of the water vapor molecule with the concentration of Cv =
1.9×10−8 g/mm3
at the surface of the droplet into the ambient air with
the relative humidity of 55%, i.e. φ = 0.55 (b).Figure 3.1: The portable micro printing setup. A motorized linear stage from
Zaber Technologies Inc. was used to control the place and speed of the
micro nozzle.Figure 4.6: Temperature contours inside the substrate adjacent to the dropletFigure 4.7: The effect of substrate cooling on the evaporation rate, the basic
model shows the same value for all substrates.
[1] R. G. Picknett and R. Bexon, “The evaporation of sessile or pendant drops in still air,” Journal of Colloid and Interface Science, vol. 61, pp. 336–350, Sept. 1977. → pages viii, 8, 9, 18, 42 [2] H. Y. Erbil, “Evaporation of pure liquid sessile and spherical suspended drops: A review,” Advances in Colloid and Interface Science, vol. 170, pp. 67–86, Jan. 2012. → pages 1 [3] R. Sharma, C. Y. Lee, J. H. Choi, K. Chen, and M. S. Strano, “Nanometer positioning, parallel alignment, and placement of single anisotropic nanoparticles using hydrodynamic forces in cylindrical droplets,” Nano Lett., vol. 7, no. 9, pp. 2693–2700, 2007. → pages 1, 54, 71 [4] S. Tokonami, H. Shiigi, and T. Nagaoka, “Review: Micro- and nanosized molecularly imprinted polymers for high-throughput analytical applications,” Analytica Chimica Acta, vol. 641, pp. 7–13, May 2009. →pages 71 [5] A. A. Sagade and R. Sharma, “Copper sulphide (CuxS) as an ammonia gas sensor working at room temperature,” Sensors and Actuators B: Chemical, vol. 133, pp. 135–143, July 2008. → pages [6] W. R. Small, C. D. Walton, J. Loos, and M. in het Panhuis, “Carbon nanotube network formation from evaporating sessile drops,” The Journal of Physical Chemistry B, vol. 110, pp. 13029–13036, July 2006. → pages 71 [7] S. H. Ko, H. Lee, and K. H. Kang, “Hydrodynamic flows in electrowetting,” Langmuir, vol. 24, pp. 1094–1101, Feb. 2008. → pages 42 [8] T. T. Nellimoottil, P. N. Rao, S. S. Ghosh, and A. Chattopadhyay, “Evaporation-induced patterns from droplets containing motile and nonmotile bacteria,” Langmuir, vol. 23, pp. 8655–8658, Aug. 2007. → pages 1 [9] R. Sharma and M. S. Strano, “Centerline placement and alignment of anisotropic nanotubes in high aspect ratio cylindrical droplets of nanometer diameter,” Advanced Materials, vol. 21, no. 1, p. 6065, 2009. → pages 1, 54, 71 [10] V. Dugas, J. Broutin, and E. Souteyrand, “Droplet evaporation study applied to DNA chip manufacturing,” Langmuir, vol. 21, pp. 9130–9136, Sept. → pages 2, 71 [11] Y.-C. Hu, Q. Zhou, Y.-F. Wang, Y.-Y. Song, and L.-S. Cui, “Formation mechanism of micro-flows in aqueous poly(ethylene oxide) droplets on a substrate at different temperatures,” Petroleum Science, vol. 10, pp. 262–268, June 2013. → pages 2, 34, 54 [12] T.-S. Wong, T.-H. Chen, X. Shen, and C.-M. Ho, “Nanochromatography driven by the coffee ring effect,” Analytical Chemistry, vol. 83, pp. 1871–1873, Mar. 2011. → pages 71 [13] J.-H. Kim, S.-B. Park, J. H. Kim, and W.-C. Zin, “Polymer transports inside evaporating water droplets at various substrate temperatures,” The Journal of Physical Chemistry C, vol. 115, pp. 15375–15383, Aug. 2011. → pages 54 [14] S. Choi, S. Stassi, A. P. Pisano, and T. I. Zohdi, “Coffee-ring effect-based three dimensional patterning of Micro/Nanoparticle assembly with a single droplet,” Langmuir, vol. 26, pp. 11690–11698, July 2010. → pages [15] D. Wang, S. Liu, B. J. Trummer, C. Deng, and A. Wang, “Carbohydrate microarrays for the recognition of cross-reactive molecular markers of microbes and host cells,” Nature biotechnology, vol. 20, pp. 275–281, Mar. PMID: 11875429. → pages 2, 54, 71 [16] H. K. Cammenga, “Evaporation mechanisms of liquids,” Current topics in materials science, vol. 5, pp. 335–446, 1980. → pages 3 [17] C. Snow, “Potential problems and capacitance for a conductor bounded by two intersecting spheres,” Journal of Research of the National Bureau of Standards, vol. 43, p. 337, 1949. → pages 9 [18] R. D. Deegan, O. Bakajin, T. F. Dupont, G. Huber, S. R. Nagel, and T. A. Witten, “Contact line deposits in an evaporating drop,” Physical Review E, vol. 62, p. 756, July 2000. → pages 10, 14, 18, 27, 53, 54, 71, 84 [19] H. Hu and R. G. Larson, “Evaporation of a sessile droplet on a substrate,” The Journal of Physical Chemistry B, vol. 106, pp. 1334–1344, Feb. 2002. → pages 12, 18, 29, 43, 44, 48, 49, 53, 61, 71, 84 [20] Y. O. Popov, “Evaporative deposition patterns: Spatial dimensions of the deposit,” Physical Review E, vol. 71, p. 036313, Mar. 2005. → pages 14, 27, 43, 44, 45, 54 [21] H. Gelderblom, A. G. Marin, H. Nair, A. van Houselt, L. Lefferts, J. H. Snoeijer, and D. Lohse, “How water droplets evaporate on a superhydrophobic substrate,” Physical Review E, vol. 83, no. 2, p. 026306,→ pages [22] F. Girard, M. Antoni, S. Faure, and A. Steinchen, “Influence of heating temperature and relative humidity in the evaporation of pinned droplets,” Colloids and Surfaces A: Physicochemical and Engineering Aspects, vol. 323, pp. 36–49, June 2008. → pages 18 [23] Y. Y. Tarasevich, “Simple analytical model of capillary flow in an evaporating sessile drop,” Physical Review E, vol. 71, p. 027301, Feb. 2005. → pages 19, 54, 62, 72 [24] A. J. Petsi and V. N. Burganos, “Potential flow inside an evaporating cylindrical line,” Physical Review E, vol. 72, p. 047301, Oct. 2005. → pages 22, 55, 62, 68, 71 [25] A. J. Petsi and V. N. Burganos, “Evaporation-induced flow in an inviscid liquid line at any contact angle,” Physical Review E, vol. 73, p. 041201, Apr.→ pages 23, 53, 55, 72 [26] H. Masoud and J. D. Felske, “Analytical solution for stokes flow inside an evaporating sessile drop: Spherical and cylindrical cap shapes,” Physics of Fluids, vol. 21, pp. 042102–042102–11, Apr. 2009. → pages 23, 55, 62, 71, 72 [27] H. Hu and R. G. Larson, “Analysis of the effects of marangoni stresses on the microflow in an evaporating sessile droplet,” Langmuir, vol. 21, pp. 3972–3980, Apr. 2005. → pages 24, 28, 53, 54, 56, 62, 68, 71, 72, 74, 84 [28] R. Bhardwaj, X. Fang, and D. Attinger, “Pattern formation during the evaporation of a colloidal nanoliter drop: a numerical and experimental study,” New Journal of Physics, vol. 11, p. 075020, July 2009. → pages 28 [29] A. Petsi, A. Kalarakis, and V. Burganos, “Deposition of brownian particles during evaporation of two-dimensional sessile droplets,” Chemical Engineering Science, vol. 65, pp. 2978–2989, May 2010. → pages 28 [30] J. Park and J. Moon, “Control of colloidal particle deposit patterns within picoliter droplets ejected by ink-jet printing,” Langmuir, vol. 22, pp. 3506–3513, Apr. 2006. → pages 28 [31] H. Hu and R. G. Larson, “Marangoni effect reverses coffee-ring depositions,” The Journal of Physical Chemistry B, vol. 110, pp. 7090–7094, Apr. 2006. → pages 29, 74 [32] K. H. Kang, S. J. Lee, C. M. Lee, and I. S. Kang, “Quantitative visualization of flow inside an evaporating droplet using the ray tracing method,” Measurement Science and Technology, vol. 15, pp. 1104–1112, June 2004. → pages 34 [33] S. T. Beyer and K. Walus, “Controlled orientation and alignment in films of single-walled carbon nanotubes using inkjet printing,” Langmuir, vol. 28, pp. 8753–8759, June 2012. → pages 42, 71 [34] G. McHale, “Surface free energy and microarray deposition technology,” Analyst, vol. 132, pp. 192–195, Feb. 2007. → pages 42 [35] R. Bhardwaj, X. Fang, P. Somasundaran, and D. Attinger, “Self-assembly of colloidal particles from evaporating droplets: Role of DLVO interactions and proposition of a phase diagram,” Langmuir, vol. 26, pp. 7833–7842, June→ pages 42 [36] G. J. Dunn, S. K. Wilson, B. R. Duffy, S. David, and K. Sefiane, “The strong influence of substrate conductivity on droplet evaporation,” Journal of Fluid Mechanics, vol. 623, no. 1, p. 329351, 2009. → pages 44 [37] M. S. Plesset and A. Prosperetti, “Flow of vapour in a liquid enclosure,” Journal of Fluid Mechanics, vol. 78, pp. 433–444, 1976. → pages 44 [38] S. Das, P. R. Waghmare, M. Fan, N. S. K. Gunda, S. S. Roy, and S. K. Mitra, “Dynamics of liquid droplets in an evaporating drop: liquid droplet coffee stain? effect,” RSC Advances, vol. 2, pp. 8390–8401, Aug. 2012. → pages 53 [39] B. J. Fischer, “Particle convection in an evaporating colloidal droplet,” Langmuir, vol. 18, pp. 60–67, Jan. 2002. → pages 54 [40] J. L. Wilbur, A. Kumar, H. A. Biebuyck, E. Kim, and G. M. Whitesides, “Microcontact printing of self-assembled monolayers: applications in microfabrication,” Nanotechnology, vol. 7, p. 452, Dec. 1996. → pages 54 [41] T. Kawase, H. Sirringhaus, R. H. Friend, and T. Shimoda, “Inkjet printed via-hole interconnections and resistors for all-polymer transistor circuits,” Advanced Materials, vol. 13, no. 21, p. 16011605, 2001. → pages 71 [42] B.-J. de Gans, P. C. Duineveld, and U. S. Schubert, “Inkjet printing of polymers: State of the art and future developments,” Advanced Materials, vol. 16, no. 3, p. 203213, 2004. → pages 71 [43] H. Sirringhaus, T. Kawase, R. H. Friend, T. Shimoda, M. Inbasekaran, W. Wu, and E. P. Woo, “High-resolution inkjet printing of all-polymer transistor circuits,” Science, vol. 290, pp. 2123–2126, Dec. 2000. PMID:→ pages [44] D. Soltman and V. Subramanian, “Inkjet-printed line morphologies and temperature control of the coffee ring effect,” Langmuir, vol. 24, pp. 2224–2231, Mar. 2008. → pages 54 [45] R. Tadmor and P. S. Yadav, “As-placed contact angles for sessile drops,” Journal of Colloid and Interface Science, vol. 317, pp. 241–246, Jan. 2008. → pages 56 [46] J. Drelich, “The significance and magnitude of the line tension in three-phase (solid-liquid-fluid) systems,” Colloids and Surfaces A: Physicochemical and Engineering Aspects, vol. 116, pp. 43–54, Sept. 1996. → pages 56 [47] R. Tadmor, “Line energy, line tension and drop size,” Surface Science, vol. 602, pp. L108–L111, July 2008. → pages 69 [48] C.-H. Choi and C.-J. C. Kim, “Droplet evaporation of pure water and protein solution on nanostructured superhydrophobic surfaces of varying heights,” Langmuir, vol. 25, pp. 7561–7567, July 2009. → pages 71 [49] K. F. Baughman, R. M. Maier, T. A. Norris, B. M. Beam, A. Mudalige, J. E. Pemberton, and J. E. Curry, “Evaporative deposition patterns of bacteria from a sessile drop: Effect of changes in surface wettability due to exposure to a laboratory atmosphere,” Langmuir, vol. 26, pp. 7293–7298, May 2010. [50] D. Brutin, B. Sobac, and C. Nicloux, “Influence of substrate nature on the evaporation of a sessile drop of blood,” Journal of Heat Transfer, vol. 134, pp. 061101–061101, May 2012. → pages 71 [51] D. Pech, M. Brunet, P.-L. Taberna, P. Simon, N. Fabre, F. Mesnilgrente, V. Condra, and H. Durou, “Elaboration of a microstructured inkjet-printed carbon electrochemical capacitor,” Journal of Power Sources, vol. 195, pp. 1266–1269, Feb. 2010. → pages 71 [52] J. Bachmann, A. Ellies, and K. Hartge, “Development and application of a new sessile drop contact angle method to assess soil water repellency,” Journal of Hydrology, vol. 231232, pp. 66–75, May 2000. → pages 71 [53] H. Y. Erbil, G. McHale, and M. I. Newton, “Drop evaporation on solid surfaces: constant contact angle mode,” Langmuir, vol. 18, no. 7, pp. 2636–2641, 2002. → pages [54] X. Fang, B. Li, J. C. Sokolov, M. H. Rafailovich, and D. Gewaily, “Hildebrand solubility parameters measurement via sessile drops evaporation,” Applied Physics Letters, vol. 87, pp. 094103–094103–3, Aug.→ pages [55] Y. C. Jung and B. Bhushan, “Wetting behaviour during evaporation and condensation of water microdroplets on superhydrophobic patterned surfaces,” Journal of Microscopy, vol. 229, no. 1, p. 127140, 2008. → pages 71 [56] J. Drelich, J. D. Miller, and R. J. Good, “The effect of drop (bubble) size on advancing and receding contact angles for heterogeneous and rough solid surfaces as observed with sessile-drop and captive-bubble techniques,” Journal of Colloid and Interface Science, vol. 179, pp. 37–50, Apr. 1996. →pages 72, 75 [57] D. Bargeman and F. Van Voorst Vader, “Effect of surfactants on contact angles at nonpolar solids,” Journal of Colloid and Interface Science, vol. 42, pp. 467–472, Mar. 1973. → pages 73 [58] J. Menezes, J. Yan, and M. Sharma, “The mechanism of alteration of macroscopic contact angles by the adsorption of surfactants,” Colloids and Surfaces, vol. 38, no. 2, pp. 365–390, 1989. → pages [59] T. Okubo, “Surface tension of structured colloidal suspensions of polystyrene and silica spheres at the air-water interface,” Journal of Colloid and Interface Science, vol. 171, pp. 55–62, Apr. 1995. → pages 73, 76 [60] R. Pyter, G. Zografi, and P. Mukerjee, “Wetting of solids by surface-active agents: The effects of unequal adsorption to vapor-liquid and solid-liquid interfaces,” Journal of Colloid and Interface Science, vol. 89, pp. 144–153, Sept. 1982. → pages 73 [61] T. Mitsui, S. Nakamura, F. Harusawa, and Y. Machida, “Changes in the interfacial tension with temperature and their effects on the particle size and stability of emulsions,” Kolloid-Zeitschrift und Zeitschrift fr Polymere, vol. 250, pp. 227–230, Mar. 1972. → pages 73 [62] S. Phongikaroon, R. Hoffmaster, K. P. Judd, G. B. Smith, and R. A. Handler, “Effect of temperature on the surface tension of soluble and insoluble surfactants of hydrodynamical importance,” Journal of Chemical & Engineering Data, vol. 50, pp. 1602–1607, Sept. 2005. → pages 73, 80 [63] V. S. Vesselovsky and V. N. Pertzov, “Adhesion of air bubbles to the solid surface,” Zh. Fiz. Khim, vol. 8, pp. 245–259, 1936. → pages 75 [64] Hideo Nakae, Ryuichi Inui, Yosuke Hirata, and Hiroyuki Saito, “Effects of surface roughness on wettability,” Acta Materialia, vol. 46, pp. 2313–2318, Apr. 1998. → pages [65] R. J. Good and M. Koo, “The effect of drop size on contact angle,” Journal of Colloid and Interface Science, vol. 71, pp. 283–292, Sept. 1979. → pages
인접한 물방울 사이의 좋은 야금학적 결합은 droplet 기반 3D 프린팅에서 필수적입니다. 그러나 재용해 메커니즘이 명확하게 마스터되었지만, 콜드 랩은 균일한 알루미늄 액적 증착 제조에서 형성된 부품의 일반적인 내부 결함이며, 이는 응고된 액 적의 표면 형태를 간과하기 때문입니다.
여기에서 처음으로 물방울 사이의 융합에 대한 잔물결과 응고각의 차단 효과가 드러났습니다. 재용해의 자세한 과정을 조사하기 위해 VOF (체적 부피) 방법을 기반으로 3D 수치 모델을 개발했습니다. 실험과 시뮬레이션을 통해 인접한 액적 간의 재 용융 공정은 두 번째 액 적과 기판 사이의 과도 접촉에 따라 두 단계로 나눌 수 있음을 보여줍니다.
첫 번째 단계에서는 재용해 조건이 이론적으로 충족 되더라도 콜드 랩이 형성 될 수 있다는 직관적이지 않은 결과가 관찰됩니다. 이전에 증착된 액적 표면의 잔물결은 새로운 액적과의 직접 접촉을 차단합니다. 두 번째 단계에서는 응고 각도가 90 °보다 클 때 액체 금속이 불완전하게 채워져 바닥 표면에 콜드랩이 형성됩니다. 또한 이러한 콜드 랩은 온도 매개 변수를 개선하여 완전히 피하는 것이 어렵습니다.
이 문제를 해결하기 위해 기판의 열전도 계수를 감소시키는 새로운 전략이 제안 되었습니다. 이 방법은 잔물결을 제거하고 응고 각도를 줄임으로써 물방울 사이의 재용해를 효과적으로 촉진합니다.
Keywords: 3D printing; aluminum droplets; metallurgical bonding; ripples; solidification angle.
Fig. 1. Schematic diagram of (a) experimental setup and (b) process principle of uniform
aluminum droplet deposition manufacturing.Fig. 2. Schematic diagram of the numerical model of two droplets successively depositing on the
substrate.Fig. 3. Comparison of SEM photographs and simulation results of two neighboring aluminum droplets
from (a) top view, (b) side view and (c) bottom view. The scale bar is 100 µm.Fig. 4. Experimental and simulation images of shape evolution during two neighboring droplets
successively impacting at (a) t, (b) t+0.5 ms, (c) t+1 ms, (d) t+2 ms, (e) t+3 ms and (f) t+5 ms.Fig. 5. SEM observation of (a) side view and (b) bottom view of successive deposition of
aluminum droplets; (c) enlarged side view of the section of the printed metal trace in (a); (d)
fracture of two neighboring droplets; (e) cross-section of two droplets successive deposition; (f)
enlarged view of the selected section in (e).Fig. 6. Simulation results of (a) shape evolution and solid fraction distribution in Y- Z middle
cross-section of two successively-deposited droplets; (b) temperature variation with time at three
points (labeled A-C) on the surface of the first droplet during the deposition of the second droplet.
[1] D. Zhang, L. Qi, J. Luo, H. Yi, X. Hou, Direct fabrication of unsupported inclined aluminum pillars based on uniform micro droplets deposition, International Journal of Machine Tools and Manufacture, 116 (2017) 18-24. [2] H. Yi, L. Qi, J. Luo, Y. Jiang, W. Deng, Pinhole formation from liquid metal microdroplets impact on solid surfaces, Applied Physics Letters, 108 (2016) 041601. [3] T. Zhang, X. Wang, T. Li, Q. Guo, J. Yang, Fabrication of flexible copper-based electronics with high-resolution and high-conductivity on paper via inkjet printing, Journal of Materials Chemistry C, 2 (2014) 286-294. [4] T. Zhang, M. Hu, Y. Liu, Q. Guo, X. Wang, W. Zhang, W. Lau, J. Yang, A laser printing based approach for printed electronics, Applied Physics Letters, 108 (2016) 103501. [5] H. Gorter, M. Coenen, M. Slaats, M. Ren, W. Lu, C. Kuijpers, W. Groen, Toward inkjet printing of small molecule organic light emitting diodes, Thin Solid Films, 532 (2013) 11-15. [6] R. Vellacheri, A. Al-Haddad, H. Zhao, W. Wang, C. Wang, Y. Lei, High performance supercapacitor for efficient energy storage under extreme environmental temperatures, Nano Energy, 8 (2014) 231-237. [7] C.W. Visser, R. Pohl, C. Sun, G.W. Römer, B. Hu is in‘t Veld, D. Lohse, Toward 3D printing of pure metals by laser‐induced forward transfer, Advanced materials, 27 (2015) 4087-4092. [8] M. Fang, S. Chandra, C. Park, Heat transfer during deposition of molten aluminum alloy droplets to build vertical columns, Journal of Heat Transfer, 131 (2009) 112101. [9] Q. Xu, V. Gupta, E. Lavernia, Thermal behavior during droplet-based deposition, Acta materialia, 48 (2000) 835-849. [10] W. Liu, G. Wang, E. Matthys, Thermal analysis and measurements for a molten metal drop impacting on a substrate: cooling, solidification and heat transfer coefficient, International Journal of Heat and Mass Transfer, 38 (1995) 1387-1395. [11] R. Rangel, X. Bian, Metal-droplet deposition model including liquid deformation and substrate remelting, International journal of heat and mass transfer, 40 (1997) 2549-2564. [12] B. Kang, Z. Zhao, D. Poulikakos, Solidification of liquid metal droplets impacting sequentially on a solid surface, TRANSACTIONS-AMERICAN SOCIETY OF MECHANICAL ENGINEERS JOURNAL OF HEAT TRANSFER, 116 (1994) 436-436.
오하이오 주립대학의 YS Lee W.Zhang 등에 의한 Mesoscopic Simulation of Heat Transfer and Fluid Flow in Laser Powder Bed Additive Manufactuing는 FLOW-3D를 이용하여 금속 분말층의 레이저에 의한 용융 결합 (L-PBF)을 분석하고 있습니다. 논문에서는 DEM으로 생성한 임의의 분말층을 분석 대상으로 하고, FSJ 에서 개발한 FLOW-3D WELD(레이저 용접 모듈) 모듈을 이용하여 균일한 분말을 바닥에 분사한 후 그 결과를 비교했습니다.
Mesoscopic Simulation of Heat Transfer and Fluid Flow in Laser Powder Bed Additive Manufactuing 해석 모델Mesoscopic Simulation of Heat Transfer and Fluid Flow in Laser Powder Bed Additive Manufactuing 해석 결과
위 사례를 통해 열전도나 용융금속의 거동, 용접속도의 차이에 의한 영향 등, 같은 분말 베드에서도 정상적인 해석이 가능한 것을 알 수 있습니다.
온도분포등의 결과의 차이는 분말층의 차이로 발생될 수 있으며, 향후, FLOW-3D@ DEM(FSJ 개별요소법 모듈)을 이용한 분말층 생성기능도 개발 예정입니다.
열전달은 전도, 대류 및 복사를 통한 열 에너지의 전달입니다. 일반적이지만 매우 중요한 물리적 현상입니다. 재료 특성 및 기타 물리적 현상은 온도 (또는 열에너지)에 매우 민감합니다. FLOW-3D 및 FLOW-3DCAST의 열전달 모델은 전도, 대류 및 복사를 통해 유체 내, 고체 및 공극 내에서 열전달을 처리하는 완전 복합 열전달 방정식을 해석합니다.
또한, 이 모델은 사용자가 다양한 애플리케이션을 모델링 할 수 있도록 유연하고 편리한 옵션을 제공합니다.
명시적 및 암시적인 열전달 옵션을 모두 사용할 수 있습니다. 암시적 방법을 사용하여 명시적 접근과 관련된 시간 단계별 크기의 안정성 제한을 제거 할 수 있습니다. 전도성 또는 열전달의 안정성 제한이 시뮬레이션에서 다른 안정성 제한보다 실질적으로 작을 때, 암시적 방법을 사용하면 계산 효율성이 크게 향상 될 수 있습니다.
각기 다른 매체 사이의 열전달 계수는 흐름 유형에 따라 사용자 정의되거나 자동으로 계산 될 수 있습니다.
1차 및 2차 열에너지의 이류 알고리즘을 모두 사용할 수 있습니다. 1차 옵션은 효율적이고 견고하며 대부분의 열전달 문제에 적합하지만 높은 열 구배가 예상되는 시뮬레이션의 경우 인공적인 열 확산으로 이어질 수 있습니다. 2차 옵션은 가령, 부력 중심의 흐름에서 온도 구배를 해결하는 것이 중요한 상황에 적합합니다.
유체와 고체 사이의 열전달을 모델링하기 위해 여러 가지 옵션을 사용할 수 있습니다 (지정된 열유속에서 전원, 규정 온도까지). 이러한 옵션은 다양한 프로세스 및 응용 프로그램을 모델링 할 수 있는 유연성과 성능을 제공합니다.
다른 물리 모델과 함께 FLOW-3D 및 FLOW-3D CAST의 열전달 모델은 고급 모델링 기능을 위한 견고한 토대가 됩니다. 예를 들어, 액체 / 고체 및 액체 / 증기 상 변화 모델을 사용하여 금속 응고, 물의 건조 및 비등, 분무 냉각을 시뮬레이션 할 수 있습니다. 점성 가열은 고속 점성 흐름에도 포함될 수 있습니다.
FLOW-3D는 자유 표면 유체 흐름 시뮬레이션을 전문으로하는 다중 물리 CFD 소프트웨어입니다. 자유 표면의 동적 진화를 추적하는 소프트웨어의 알고리즘인 VOF (Volume of Fluid) 방법은 Flow Science의 설립자인 Tony Hirt 박사가 개척했습니다.
또한 FLOW-3D에는 금속 주조, 잉크젯 인쇄, 레이저 용접 및 적층 제조 (AM)와 같은 광범위한 응용 분야를 시뮬레이션하기위한 물리 모델이 내장되어 있습니다. 적층 제조 시뮬레이션 소프트웨어, 특히 L-PBF (레이저 파우더 베드 융합 공정)의 현상 유지는 열 왜곡, 잔류 응력 및지지 구조 생성과 같은 부분 규모 모델링에 도움이되는 열 기계 시뮬레이션에 초점을 맞추고 있습니다.
유용하지만 용융 풀 역학 및 볼링 및 다공성과 같은 관련 결함에 대한 정보는 일반적으로 이러한 접근 방식의 영역 밖에 있습니다. 용융 풀 내의 유체 흐름, 열 전달 및 표면 장력이 열 구배 및 냉각 속도에 영향을 미치며 이는 다시 미세 구조 진화에 영향을 미친다는 점을 명심하는 것도 중요합니다.
FLOW-3D와 이산 요소법 (DEM) 및 WELD 모듈을 사용하여 분말 및 용융 풀 규모에서 시뮬레이션 할 수 있습니다. 구현되는 관련 물리학에는 점성 흐름, 열 전달, 응고, 상 변화, 반동 압력, 차폐 가스 압력, 표면 장력, 움직이는 물체 및 분말 / 입자 역학이 포함됩니다. 이러한 접근 방식은 합금에 대한 공정을 성공적으로 개발할 수 있게 하고, AM 기계 제조업체와 AM 기술의 최종 사용자 모두에게 관심있는 미세 구조 진화에 대한 통찰력을 제공하는데 도움이 됩니다.
The realm of operations of FLOW-3D
FLOW-3D는 레이저 분말 베드 융합 (L-PBF), 직접 에너지 증착 (DED) 및 바인더 제트 공정으로 확장되는 기능을 가지고 있습니다. FLOW-3D를 사용하면 분말 확산 및 패킹, 레이저 / 입자 상호 작용, 용융 풀 역학, 표면 형태 및 후속 미세 구조 진화를 정확하게 시뮬레이션 할 수 있습니다. 이러한 기능은 FLOW-3D에 고유하며 계산 효율성이 높은 방식으로 달성됩니다.
예를 들어 1.0mm x 0.4mm x 0.3mm 크기의 계산 영역에서 레이저 빔의 단일 트랙을 시뮬레이션하기 위해 레이저 용융 모델은 단 8 개의 물리적 코어에서 약 2 시간이 걸립니다. FLOW-3D는 모든 관련 물리 구현 간의 격차를 해소하는 동시에 업계 및 연구 표준에서 허용하는 시간 프레임으로 결과를 생성합니다. 분말 패킹, 롤러를 통한 파워 확산, 분말의 레이저 용융, 용융 풀 형성 및 응고를 고려하고 다층 분말 베드 융합 공정을 위해 이러한 단계를 순차적으로 반복하여 FLOW-3D에서 전체 AM 공정을 시뮬레이션 할 수 있습니다.
FLOW-3D의 다층 시뮬레이션은 이전에 응고된 층의 열 이력을 저장한다는 점에서 독특하며, 열 전달을 고려하여 이전에 응고된 층에 확산된 새로운 분말 입자 세트에 대해 시뮬레이션이 수행됩니다. 또한, 응고 된 베드의 열 왜곡 및 잔류 응력은 FLOW-3D를 사용하여 평가할 수 있으며, 보다 복잡한 분석을 수행하기 위해 FLOW-3D의 압력 및 온도 데이터를 Abaqus 및 MSC Nastran과 같은 FEA 소프트웨어로 내보낼 수 있습니다.
Sequence of a multi-layer L-PBF simulation setup in FLOW-3D
Ease of Use
FLOW-3D는 다양한 응용 분야에서 거의 40 년 동안 사용되어 왔습니다. 사용자 피드백을 기반으로 UI 개발자는 소프트웨어를 사용하기 매우 직관적으로 만들었으며 새로운 사용자는 시뮬레이션 설정의 순서를 거의 또는 전혀 어려움없이 이해합니다. 사용자는 FLOW3D에서 구현 된 다양한 모델의 이론에 정통하며 새로운 실험을 설계 할 수 있습니다. 실습 튜토리얼, 비디오 강의, 예제 시뮬레이션 및 기술 노트의 저장소도 사용할 수 있습니다. 사용자가 특정 수준의 경험에 도달하면 고급 수치 교육 및 소프트웨어 사용자 지정 교육을 사용할 수 있습니다.
Available Literature
실험 데이터에 대해 FLOW-3D 모델을 검증하는 몇 가지 독립적으로 발표된 연구가 있습니다. 여기에서 수록된 저널 논문은 레이저 용접 및 적층 제조 공정으로 제한됩니다. 더 많은 참조는 당사 웹 사이트에서 확인할 수 있습니다.
L.J.Zhang, J.X.Zhang, A.Gumenyuk, M.Rethmeier, S.J.Na, Numerical simulation of full penetration laser welding of thick steel plate with high power high brightness laser, Journal of Materials Processing Technology, Volume 214, Issue 8, 2014. A study by researchers from BAM in Germany, KAIST in Korea, and State Key Laboratory of Mechanical Behavior of Materials in China that focuses on keyhole dynamics and full penetration laser welding of steel plates.
Runqi Lin, Hui-ping Wang, Fenggui Lu, Joshua Solomon, Blair E. Carlson, Numerical study of keyhole dynamics and keyhole-induced porosity formation in remote laser welding of Al alloys, International Journal of Heat and Mass Transfer, Volume 108, Part A, 2017. General Motors (GM) and Shangai University collaborated on a study on the influence of welding speed and weld angle of inclination on porosity occurrence in laser keyhole welding.
Koji Tsukimoto, Masashi Kitamura, Shuji Tanigawa, Sachio Shimohata, and Masahiko Mega, Laser Welding Repair for Single Crystal Blades, International Gas Turbine Congress, Tokyo, 2015. Mitsubishi Heavy Industry’s study on laser welding repair using laser cladding for single Ni crystal alloys used in gas turbine blades.
Additive Manufacturing
Yu-Che Wu, Cheng-Hung San, Chih-Hsiang Chang, Huey-Jiuan Lin, Raed Marwan, Shuhei Baba, Weng-Sing Hwang, Numerical modeling of melt-pool behavior in selective laser melting with random powder distribution and experimental validation, Journal of Materials Processing Technology, Volume 254, 2018 This paper discusses powder bed compaction with random packing for different powder-size distributions, and the importance of considering evaporation effects in the melting process to validate the melt pool dimensions.
Lee, Y.S., and W.Zhang, Mesoscopic simulation of heat transfer and fluid flow in laser powder bed additive manufacturing, Proceedings of the Annual International Solid Freeform Fabrication Symposium, Austin, TX, USA. 2015 A study conducted by Ohio State University researchers to understand the influence of process parameters in formation of balling defects.
Y.S. Lee, W.Zhang, Modeling of heat transfer, fluid flow and solidification microstructure of nickel-base superalloy fabricated by laser powder bed fusion, Additive Manufacturing, Volume 12, Part B, 2016 A study conducted by Ohio State University researchers to understand the influence of solidification parameters, calculated from the temperature fields, on solidification morphology and grain size using existing theoretical models in laser powder bed fusion processes.
시뮬레이션을 실행하는 동안 대부분의 사람들은 “내 시뮬레이션이 왜 이렇게 느리게 실행됩니까? 그 이유는 무엇입니까?”라고 묻습니다.
시뮬레이션 런타임에 영향을 미치는 중요한 요소는 시간 단계 크기입니다. 시간 단계 크기가 작으면, 시뮬레이션이 완료 조건에 도달하기 전에 더 많은 계산을 수행해야하므로 런타임이 늘어납니다. 시간 단계의 크기는 종종 명시적 안정성 한계 중 하나에 의해 제어되므로 설정을 변경하지 않고, 늘리는 유일한 방법은 암시적 솔버가 동일하지 않기 때문에 명시적 솔버에서 암시적 솔버로 전환하는 것입니다.
명시적 솔버로서의 시간 단계 제한, 웹 사이트에서 명시적 및 암시적 솔버에 대한 자세한 설명을 읽을 수 있습니다.
솔버 메시지 이해
암시적 솔버를 사용하여 시뮬레이션 런타임을 줄일 수 있는지 확인하는 첫 번째 단계는 솔버 진단 출력을 읽고 이해하는 것입니다. 이는 Solver Messages 파일인 hd3msg. *에 기록되며 Simulation Manager 창 하단에 나타납니다. 처음 몇 개의 열을 살펴 보겠습니다.
솔버 메시지 파일
표 1. 예제 솔버 메시지 파일의 처음 8 개 열.
표 1에는 왼쪽에서 오른쪽으로 다음이 있습니다.
Progress, sim_time: 시뮬레이션의 현재 시간.
Progress, cycle: 시뮬레이션 시작 이후주기 수 (즉, 시간 단계 수)입니다.
Time Step, delt: 이 열은 현재 시간 단계 크기를 나타냅니다.
Time Step, dt_stbl: 이것은 현재 솔루션을 제한하는 명시적 솔버의 안정성 한계입니다. 시간 단계 크기 delt는 항상 이 값보다 작아야합니다.
Time Step, code: 어떤 명시적 솔버 안정성 한계가 시간 단계 크기를 제어하는지 나타내는 2 자리 코드입니다.
Pressure, iter: 해당 시간 단계에 대해 암시적 압력 솔버에서 수행한 반복 횟수입니다. 압력 솔버는 거의 항상 암시적 솔버입니다.
Pressure, res/epsi: 반복 종료시 암시적 압력 솔버에 대한 수렴 기준 (ε)에 대한 잔차의 비율입니다. 이 값은 항상 1보다 작아야합니다.
Heat Transfer, iter: 해당 시간 단계에 대해 암시적 열 전달 솔버에서 수행한 반복 횟수입니다.
Heat Transfer, res/epsi: 반복이 끝날 때 암시적 열 전달 솔버에 대한 수렴 기준 (ε)에 대한 잔여 비율입니다. 이 값은 항상 1보다 작아야합니다.
표 1에 설명 된 진단 출력을 표 2에 설명된 2 자리 진단 코드와 연관 시키면 x 방향 (코드 = cx)의 유체 이류(advection)와 관련된 안정성 한계가 시간 단계의 크기를 제한하고 있음을 나타냅니다.
<참고> 이송 [advection, 移送]해양과학용어사전 기체나 액체가 그에 작용되는 외부로부터의 힘 또는 압력차에 의해 이동하는 현상. (수질오염 등 오염물질 이동과 관련해서는) 물 등 유체(流體) 속에 포함된 오염물질(汚染物質)이 중력(重力)이나 바람 등의 작용(作用)으로 생기는 유체(流體)의 흐름과 함께 이동하는 현상을 말하기도 함.
암시적 열 전달 솔버의 진단은 마지막 두 열 (Heat Transfer, iter 및 Heat Transfer, res / epsi)에서 알 수 있듯이 표 1에도 표시됩니다. 암시적 솔버가 추가되면 반복 횟수와 수렴 기준 (ε)에 대한 잔차 비율을 나타내는 두 개의 유사한 열이 나타납니다. 이 값은 항상 1보다 작아야하며 솔버가 성공적으로 수렴되었음을 나타냅니다. 그렇지 않은 경우 해당 값 옆에 별표 (*)가 표시됩니다.
안정성 코드 표
표 2. 사용 가능한 모든 2 글자 안정성 코드 및 의미.
암시적 방법을 사용하는 경우
때로는 단순히 암시적 솔버로 전환해도 시뮬레이션 런타임이 향상되지 않습니다. 이는 일반적으로 다른 명시적 솔버가 시간 단계 크기가 의미있는 방식으로 증가하지 못하도록 방지할 때 발생합니다. 간단한 지침은 암시적 솔버에 필요한 추가 노력을 보상하고 시뮬레이션 런타임을 줄이기 위해 시간 단계 크기를 5 배 이상 늘려야한다는 것입니다. Solver Summary 파일 인 hd3out. *의 짧은 표시 데이터는 시간 단계 크기의 잠재적 증가를 평가하는 데 사용할 수 있는 몇 가지 정보를 제공합니다. Solver Summary는 Diagnostics 메뉴에서 액세스 할 수 있으며 “Stability limits”를 검색하면 관련 안정성 한계가있는 표를 찾을 수 있습니다. 솔버 요약 안정성 한계 표.
표 3. 솔버 요약 안정성 한계 표. 표면 장력은 시간 단계 값이 가장 작기 때문에 솔루션을 제한합니다.
예제 출력 (표 3)에 제시된 최소 안정성 한계는 표면 장력 솔버와 관련된 것입니다. 암시적 표면 장력 솔버로 변경하면이 안정성 제한이 제거되고 이제 y 방향의 유체 이류에 의해 제한되는 새로운 시간 단계 크기로 이어집니다. 새로운 시간 단계 크기는 이전 시간 단계 크기보다 5 배 이상 큽니다. 이는 암시적 표면 장력 모델을 사용할 때 시뮬레이션이 더 빠르게 실행될 수 있음을 나타냅니다.
그러나 시간 단계 크기에 비례하는 암시적 솔버를 사용하여 솔루션을 찾는 데 필요한 노력과 같이 런타임에 영향을 미치는 다른 요인이 있기 때문에 런타임 감소가 보장되지 않습니다. 따라서 다른 암시적 솔버를 활성화하기 전에 현재 활성화 된 모든 암시적 솔버가 상대적으로 적은 반복으로 수렴되도록 하는 것도 좋은 방법입니다.
마지막으로 시간 단계 크기가 증가함에 따라 솔루션의 정확도가 감소한다는 점에 유의하는 것이 중요합니다. 솔루션 방법을 변경하는 방법
시뮬레이션이 중지되면 그림 1과 같이 Model Setup / Numerics 탭으로 이동하고 원하는 암시적 솔버의 라디오 버튼을 클릭하여 암시적 솔버를 활성화 할 수 있습니다. 이것은 시뮬레이션이 실행될 때 적용됩니다.
명시적 및 암시적 옵션
그림 1. 숫자 탭의 명시 적 / 암시 적 옵션.
시뮬레이션 관리자 탭의 런타임 옵션 대화 상자를 사용하여 시뮬레이션이 실행되는 동안 암시적 솔버를 활성화 할 수도 있습니다. Runtime Options를 클릭하고 Explicit / implicit solvers 탭으로 이동하여 원하는 implicit solver의 라디오 버튼을 클릭 한 다음 Send To Solver를 클릭하십시오. 솔버가 업데이트 되었음을 알리는 대화 상자가 나타납니다.
런타임 옵션 대화 상자
그림 2. 런타임 옵션 대화 상자의 명시 적 / 암시 적 옵션.
결론
시뮬레이션 실행 시간은 완료 조건을 충족하기 위해 수행해야하는 계산 횟수에 따라 결정됩니다. 그러나 계산 수는 시간 단계 크기 및 암시적 솔버에 필요한 계산 노력의 함수이며 후자는 시간 단계 크기의 비선형 함수이기도합니다. 이것은 복잡한 상호 작용이지만, 진단 파일 hd3msg. * 및 hd3out. *에서 솔버가 보고한 정보를 사용하여 런타임을 줄이는 방식으로 숫자 옵션을 신중하게 조정할 수 있습니다.