Machine Learning Algorithms on Density Prediction of Electron Beam Selective Melted Parts

doi:10.3901/JME.2019.15.048

Abstract

Abstract: Electron Beam Selective Melting (EBSM) is a novel additive manufacturing technology, which is developing very fast nowadays. It has many advantages:building up parts with complex morphology; processing under vacuum to get rid of the impurity; its manufactured part with small residual stress and powder recycling. However, it is difficult to establish the relationship between EBSM's processing parameters and parts' properties. Here a series of Inconel 718 cubic specimens are manufactured through adjusting EBSM's scanning speed, beam current, plate temperature and layer thickness. Then the densities of these specimens are measured. Three kinds of machine learning algorithms, including linear regression, support vector regression and neural network, have been utilized to build the relationship between these four processing parameters and density. The results show that:Linear regression has the worst prediction skill as a result of its small model capacity; neural network has a better prediction accuracy, but it is easily overfitting; support vector regression has appropriate model capacity and good physical interpretation, and it behaves best in the density prediction.

Key words: additive manufacturing, density, electron beam selective melting, machine learning, prediction

CLC Number:

TP181

QI Xinbo, LI Changpeng, LI Yang, LIN Feng, LI Yong, CHENG Xuan, CHEN Guofeng. Machine Learning Algorithms on Density Prediction of Electron Beam Selective Melted Parts[J]. Journal of Mechanical Engineering, 2019, 55(15): 48-55.

References

[1] 汤慧萍,王建,逯圣路,等. 电子束选区熔化成形技术研究进展[J]. 中国材料进展,2015(3):49-59. TANG Huiping,WANG Jian,LU Shenglu,et al. Research progress in selective electron beam melting[J]. Materials China,2015(3):49-59
[2] PRABHAKAR P,SAMES W,DEHO R,et al. Computational modeling of residual stress formation during the electron beam melting process for inconel 718[J]. Additive Manufacturing,2015,7:83-91.
[3] MURR L. Metallurgy of additive manufacturing:Examples from electron beam melting[J]. Additive Manufacturing,2015,5:40-53.
[4] MURR LE,GAYTAN SM,RAMIREZ DA,et al. Metal fabrication by additive manufacturing using laser and electron beam melting technologies[J]. Journal of Materials Science & Technology,2012,28(1):1-14.
[5] 刘海涛,赵万华,唐一平. 电子束熔融直接金属成型工艺的研究[J]. 西安交通大学学报,2007,41(11):1307-1310. LIU Haitao,ZHAO Wanhua,TANG Yiping. Process investigation of direct metal fabrication based on electron beam melting[J]. Journal of Xi'an Jiaotong University,2007,41(11):1307-1310.
[6] 颜永年,齐海波,林峰,等. 三维金属零件的电子束选区熔化成形[J]. 机械工程学报,2007,43(6):87-92. YAN Yongnian,QI Haibo,LIN Feng,et al. Produced three-dimensional metal parts by electron beam selective melting[J]. Chinese Journal of Mechanical Engineering,2007,43(6):87-92.
[7] 周斌,张婷,林峰,等. 电子束选区熔化成形ti6al4v和316l不锈钢叶轮体微观组织和力学性能的研究[J]. 稀有金属材料与工程,2018,47(1):175-180. ZHOU Bin,ZHANG Ting,LIN Feng,et al. Study of microstructures and mechanical properties of Ti6Al4V and 316L stainless steel impeller body by electron beam selective melting[J]. Rare Metal Materials and Engineering,2018,47(1):175-180.
[8] SHEN X,YAO J,WANG Y,et al. Density prediction of selective laser sintering parts based on artificial neural network[M]. Berlin,Heidelberg:Springer Berlin Heidelberg,2004.
[9] ZHANG W,MEHTA A,DESAI P S,et al. Machine learning enabled powder spreading process map for metal additive manufacturing (AM)[C]//Solid Freeform Fabrication 2017:Proceedings of the 28th Annual International Solid Freeform Fabrication Symposium-An Additive Manufacturing Conference,August 7-9,2017,Hilton Hotel Austin,Austin,Texas,USA:TMS,2017:1235-1249.
[10] 蔡从中,裴军芳,温玉锋,等. 选择性激光烧结成型件密度的支持矢量回归预测[J]. 物理学报,2009,58(13):8-S14. CAI Congzhong,PEI Junfang,WEN Yufeng,et al. Density prediction of selective laser sintering parts based on support vector regression[J]. Acta Physics Sinica,2009,58(13):8-S14.
[11] FLECK N A,SMITH R A. Effect of density on tensile strength,fracture toughness,and fatigue crack propagation behavior of sintered steel[J]. Powder Metallurgy,1981,24(3):121-125.
[12] CHAWLA N,DENG X,MARRUCCI M,et al. Effect of density on the microstructure and mechanical behavior of powder metallurgy Fe-Mo-Ni steels[J]. Advances in Powder Metallurgy and Particulate Materials,2003,6:7-257.
[13] ASTM. Standard test methods for density of compacted or sintered powder metallurgy (PM) products using archimedes' principle[S]. Philadelphia:ASTM International,2009.
[14] HE H,GARCIA E A. Learning from imbalanced data[J]. IEEE Transactions on Knowledge & Data Engineering,2009,21(9):1263-1284.
[15] ARLOT S,CELISSE A. A survey of cross-validation procedures for model selection[J]. Statistics Surveys,2010,4:40-79.
[16] AKSOY S,HARALICK R M. Feature normalization and likelihood-based similarity measures for image retrieval[J]. Pattern Recognition Letters,2001,22(5):563-582.
[17] PEDREGOSA F,GRAMFORT A,MICHEL V,et al. Scikit-learn:Machine learning in python[J]. J. Mach. Learn Res.,2012,12(10):2825-2830.
[18] PASZKE A,GROSS S,CHINTALA S,et al. Automatic dierentiation in pytorch[C]//31st Conference on Neural Information Processing Systems,Long Beach,CA,USA:NIPS,2017.
[19] HE K,ZHANG X,REN S,et al. Delving deep into rectifers:Surpassing human-level performance on imagenet classification[C]//Proceedings of the IEEE International Conference on Computer Vision. Washington,DC,USA:IEEE Computer Society,2015:1026-1034.
[20] KINGMA D P,BA J. Adam:A method for stochastic optimization[C]//International Conference on Learning Representations (ICLR),May 7-9,2015,San Diego:Ithaca,NY:arXiv.org,2015.
[21] CHONG E K P,ZAK S H. An introduction to optimization[M]. New York:John Wiley & Sons,2013.