en
×

分享给微信好友或者朋友圈

使用微信“扫一扫”功能。
作者简介:

陈儒森,男,1996年出生,硕土研究生。主要研究方向为表面工程与机器学习。E-mail: chenrusen96@qq.com

通讯作者:

张伟,男,1971年出生,博士,教授。主要研究方向为表面工程与再制造工程。E-mail: zhangwei18@fosu.edu.cn

中图分类号:V261;TG665

DOI:10.11933/j.issn.1007-9289.20231115002

参考文献 1
SIDDIQUI A A,DUBEY A K.Recent trends in laser cladding and surface alloying[J].Optics & Laser Technology,2021,134:106619.
参考文献 2
WENG F,CHEN C,YU H.Research status of laser cladding on titanium and its alloys:A review[J].Materials & Design,2014,58:412-425.
参考文献 3
LIU J,YU H,CHEN C,et al.Research and development status of laser cladding on magnesium alloys:A review[J].Optics and Lasers in Engineering,2017,93:195-210.
参考文献 4
高亚丽,路鹏勇,刘宇,等.镁合金表面激光熔覆研究现状[J].中国表面工程,2023,36(3):22-39.GAO Yali,LU Pengyong,LIU Yu,et al.Research status of laser cladding on magnesium alloy[J].China Surface Engineering,2023,36(3):22-39.(in Chinese)
参考文献 5
YANG Q,ZHANG P,LU Q,et al.Application and development of blue and green laser in industrial manufacturing:A review[J].Optics & Laser Technology,2024,170:110202.
参考文献 6
ZHANG H,LIU Y,BAI X,et al.Laser cladding highly corrosion-resistant nano/submicron ultrafine-grained Fe-based composite layers[J].Surface and Coatings Technology,2021,424:127636.
参考文献 7
ARIF Z U,KHALID M Y,UR REHMAN E,et al.A review on laser cladding of high-entropy alloys,their recent trends and potential applications[J].Journal of Manufacturing Processes,2021,68:225-273.
参考文献 8
ZHANG Q,WANG Q,HAN B,et al.Comparative studies on microstructure and properties of CoCrFeMnNi high entropy alloy coatings fabricated by high-speed laser cladding and normal laser cladding[J].Journal of Alloys and Compounds,2023,947:169517.
参考文献 9
ARIF Z U,KHALID M Y,AL RASHID A,et al.Laser deposition of high-entropy alloys:A comprehensive review[J].Optics & Laser Technology,2022,145:107447.
参考文献 10
HALDAR B,SAHA P.Identifying defects and problems in laser cladding and suggestions of some remedies for the same[J].Materials Today:Proceedings,2018,5(5):90-101.
参考文献 11
曹佳俊,常成,邱兆国,等.AISI1045 钢表面激光熔覆 FeCoCrNiAl0.5Ti0.5 涂层的界面特性及摩擦性能[J].中国表面工程,2023,36(2):54-64.CAO Jiajun,CHANG Cheng,QIU Zhaoguo,et al.Interface characteristics and tribological properties of laser cladded FeCoCrNiAl0.5Ti0.5 coating on AISI 1045 steel[J] China Surface Engineering,2023,36(2):54-64.(in Chinese)
参考文献 12
OCELÍK V,EEKMA M,HEMMATI I,et al.Elimination of start/stop defects in laser cladding[J].Surface and Coatings Technology,2012,206(8):2403-2409.
参考文献 13
LI R,FENG A,ZHAO J,et al.Study on process optimization of WC-Ni60A cermet composite coating by laser cladding[J].Materials Today Communications,2023,37:107400.
参考文献 14
LI G,WANG Z,YAO L,et al.Concentration mixing and melt pool solidification behavior during the magnetic field assisted laser cladding of Fe-Cr-based alloy on 45 steel surface[J].Surface and Coatings Technology,2022,445:128732.
参考文献 15
SONG B,YU T,JIANG X,et al.Evolution and convection mechanism of the melt pool formed by V-groove laser cladding[J].Optics & Laser Technology,2021,144:107443.
参考文献 16
何志远,贺文雄,杨海峰,等.铝合金表面激光熔覆研究进展[J].中国表面工程,2021,34(6):33-44.HE Zhiyuan,HE Wenxiong,YANG Haifeng,et al.Research progess in laser cladding on aluminum alloy surface[J].China Surface Engineering,2021,34(6):33-44.(in Chinese)
参考文献 17
郭星星,帅美荣,王建梅,等.基于 NSGA-II 算法的激光熔覆单道成形工艺参数多目标优化[J].中国表面工程,2023,36(3):87-100.GUO Xingxing,SHUAI Meirong,WANG Jianmei,et al.Multi-objective optimization of laser cladding single-pass forming process parameters based on NSGA-II algorithm[J].China Surface Engineering,2023,36(3):87-100.(in Chinese)
参考文献 18
THAWARI N,GULLIPALLI C,CHANDAK A,et al.Influence of laser cladding parameters on distortion,thermal history and melt pool behaviour in multi-layer deposition of stellite 6:in-situ measurement[J].Journal of Alloys and Compounds,2021,860:157894.
参考文献 19
FALLAH V,ALIMARDANI M,CORBIN S F,et al.Temporal development of melt-pool morphology and clad geometry in laser powder deposition[J].Computational Materials Science,2011,50(7):2124-2134.
参考文献 20
WIRTH F,ARPAGAUS S,WEGENER K.Analysis of melt pool dynamics in laser cladding and direct metal deposition by automated high-speed camera image evaluation[J].Additive Manufacturing,2018,21:369-382.
参考文献 21
ZHANG Y M,LIM C W J,TANG C,et al.Numerical investigation on heat transfer of melt pool and clad generation in directed energy deposition of stainless steel[J].International Journal of Thermal Sciences,2021,165:106954.
参考文献 22
LI Y,XU F.Acoustic emission sources localization of laser cladding metallic panels using improved fruit fly optimization algorithm-based independent variational mode decomposition[J].Mechanical Systems and Signal Processing,2022,166:108514.
参考文献 23
郭永明,叶福兴,祁航.超高速激光熔覆技术研究现状及发展趋势[J].中国表面工程,2022,35(6):39-50.GUO Yongming,YE Fuxing,QI Hang,et al.Research status and development of ultra-high speed laser cladding[J].China Surface Engineering,2022,35(6):39-50.(in Chinese)
参考文献 24
JINLONG W,YUXIN M,WENJIE P,et al.Evaluation of the effect of surface roughness parameters on fatigue of TC17 titanium alloy impeller using machine learning algorithm and finite element analysis[J].Engineering Failure Analysis,2023,153:107586.
参考文献 25
HAO W Q,TAN L,YANG X G,et al.A physics-informed machine learning approach for notch fatigue evaluation of alloys used in aerospace[J].International Journal of Fatigue,2023,170:107536.
参考文献 26
ZHAO S,SUN H,PENG F,et al.Feature fusion and distillation embedded sparse Bayesian learning model for in-situ foreknowledge of robotic machining errors[J].Journal of Manufacturing Systems,2023,71:546-564.
参考文献 27
LEE J A,SAGONG M J,JUNG J,et al.Explainable machine learning for understanding and predicting geometry and defect types in Fe-Ni alloys fabricated by laser metal deposition additive manufacturing[J].Journal of Materials Research and Technology,2023,22:413-423.
参考文献 28
SVETLIZKY D,DAS M,ZHENG B,et al.Directed energy deposition(DED)additive manufacturing:physical characteristics,defects,challenges and applications[J].Materials Today,2021,49:271-295.
参考文献 29
ZHOU L,MA G,ZHAO H,et al.Research status and prospect of extreme high-speed laser cladding technology[J].Optics & Laser Technology,2024,168:109800.
参考文献 30
LIN X,ZHU K,FUH J Y H,et al.Metal-based additive manufacturing condition monitoring methods:from measurement to control[J].ISA Transactions,2022,120:147-166.
参考文献 31
CAI Y,XIONG J,CHEN H,et al.A review of in-situ monitoring and process control system in metal-based laser additive manufacturing[J].Journal of Manufacturing Systems,2023,70:309-326.
参考文献 32
FU Y,DOWNEY A R J,YUAN L,et al.Machine learning algorithms for defect detection in metal laser-based additive manufacturing:A review[J].Journal of Manufacturing Processes,2022,75:693-710.
参考文献 33
WANG C,TAN X P,TOR S B,et al.Machine learning in additive manufacturing:state-of-the-art and perspectives[J].Additive Manufacturing,2020,36:101538.
参考文献 34
QIN L,WANG K,LI X,et al.Review of the formation mechanisms and control methods of geometrical defects in laser deposition manufacturing[J].Chinese Journal of Mechanical Engineering:Additive Manufacturing Frontiers,2022,1(4):100052.
参考文献 35
ZHAO S,YUAN K,GUO W,et al.A comparative study of laser metal deposited and forged Ti-6Al-4V alloy:uniaxial mechanical response and vibration fatigue properties[J].International Journal of Fatigue,2020,136:105629.
参考文献 36
WAN H,WANG Q,JIA C,et al.Multi-scale damage mechanics method for fatigue life prediction of additive manufacture structures of Ti-6Al-4V[J].Materials Science and Engineering:A,2016,669:269-278.
参考文献 37
LI J,CHENG X,LI Z,et al.Improving the mechanical properties of Al-5Si-1Cu-Mg aluminum alloy produced by laser additive manufacturing with post-process heat treatments[J].Materials Science and Engineering:A,2018,735:408-417.
参考文献 38
YU X,LIN X,TAN H,et al.Microstructure and fatigue crack growth behavior of Inconel 718 superalloy manufactured by laser directed energy deposition[J].International Journal of Fatigue,2021,143:106005.
参考文献 39
STERLING A J,TORRIES B,SHAMSAEI N,et al.Fatigue behavior and failure mechanisms of direct laser deposited Ti-6Al-4V[J].Materials Science and Engineering:A,2016,655:100-112.
参考文献 40
LEUNG C L A,MARUSSI S,ATWOOD R C,et al.In situ X-ray imaging of defect and molten pool dynamics in laser additive manufacturing[J].Nature Communications,2018,9(1):1355.
参考文献 41
LI S,CHEN B,TAN C,et al.In situ identification of laser directed energy deposition condition based on acoustic emission[J].Optics & Laser Technology,2024,169:110152.
参考文献 42
SONG B,YU T,JIANG X,et al.The relationship between convection mechanism and solidification structure of the iron-based molten pool in metal laser direct deposition[J].International Journal of Mechanical Sciences,2020,165:105207.
参考文献 43
KHAIRALLAH S A,ANDERSON A T,RUBENCHIK A,et al.Laser powder-bed fusion additive manufacturing:physics of complex melt flow and formation mechanisms of pores,spatter,and denudation zones[J].Acta Materialia,2016,108:36-45.
参考文献 44
YANG Z,WANG A,WENG Z,et al.Porosity elimination and heat treatment of diode laser-clad homogeneous coating on cast aluminum-copper alloy[J].Surface and Coatings Technology,2017,321:26-35.
参考文献 45
操龙飞.金属材料的热膨胀特性研究[D].武汉:武汉科技大学,2013.CAO Longfei.Study on thermal expansion properties of steels[D].Wuhan:Wuhan University of Science and Technology,2013.(in Chinese)
参考文献 46
李春彦,张松,康煜平,等.综述激光熔覆材料的若干问题[J].激光杂志,2002,3:5-9.LI Chunyan,ZHANG Song,KANG Yuping,et al.Comment on material system for laser cladding[J].Laser Journal,2002,3:5-9.(in Chinese)
参考文献 47
ZHOU S,ZENG X,HU Q,et al.Analysis of crack behavior for Ni-based WC composite coatings by laser cladding and crack-free realization[J].Applied Surface Science,2008,255(5):1646-1653.
参考文献 48
申发明.超高速激光熔覆AISI431不锈钢涂层组织与耐蚀机理研究[D].哈尔滨:哈尔滨工业大学,2022.SHEN Faming.Research on microstructure and corrosion resistance mechanism of AISI431 stainless steel coating prepared by extra high speed laser cladding[D].Harbin:Harbin Institute of Technology,2022.(in Chinese)
参考文献 49
AUCOTT L,HUANG D,DONG H B,et al.A three-stage mechanistic model for solidification cracking during welding of steel[J].Metallurgical and Materials Transactions A,2018,49(5):1674-1682.
参考文献 50
GAO Z,WANG L,WANG Y,et al.Crack defects and formation mechanism of FeCoCrNi high entropy alloy coating on TC4 titanium alloy prepared by laser cladding[J].Journal of Alloys and Compounds,2022,903:163905.
参考文献 51
JIN K,YANG Z,CHEN P,et al.Dynamic solidification process during laser cladding of IN718:multi-physics model,solute suppressed nucleation and microstructure evolution[J].International Journal of Heat and Mass Transfer,2022,192:122907.
参考文献 52
SCHWERZ C,BIRCHER B A,KÜNG A,et al.In-situ detection of stochastic spatter-driven lack of fusion:application of optical tomography and validation via ex-situ X-ray computed tomography[J].Additive Manufacturing,2023,72:103631.
参考文献 53
XU X,DU J L,LUO K Y,et al.Microstructural features and corrosion behavior of Fe-based coatings prepared by an integrated process of extreme high-speed laser additive manufacturing[J].Surface and Coatings Technology,2021,422:127500.
参考文献 54
FONSECA E B,GABRIEL A H G,ARAÚJO L C,et al.Assessment of laser power and scan speed influence on microstructural features and consolidation of AISI H13 tool steel processed by additive manufacturing[J].Additive Manufacturing,2020,34:101250.
参考文献 55
王豫跃,牛强,杨冠军,等.超高速激光熔覆技术绿色制造耐蚀抗磨涂层[J].材料研究与应用,2019,13(3):165-172.WANG Yuyue,NIU Qiang,YANG Guanjun,et al.lnvestigations on corrosion-resistant and wear-resistant coatings environmental-friendly manufactured by a novel super-high efficient laser cladding[J].Materials Research and Application,2019,13(3):165-172.(in Chinese)
参考文献 56
ZHANG Y,GAO X,LIANG X,et al.Effect of laser remelting on the microstructure and corrosion property of the arc-sprayed AlFeNbNi coatings[J].Surface and Coatings Technology,2020,398:126099.
参考文献 57
黄旭,张家诚,练国富,等.超高速激光熔覆研究现状及应用[J].机床与液压,2021,49(6):151-155,162.HUANG Xu,ZHANG Jiacheng,LIAN Guofu,et al.Research status and application of extreme high speed cladding[J].Machine Tool & Hydraulics,2021,49(6):151-155,162.(in Chinese)
参考文献 58
YE X,WANG J,YING Q,et al.Melting behavior of in-flight particles in ultra-high speed laser cladding[J].Journal of Materials Research and Technology,2023,24:7047-7057.
参考文献 59
WANG W,ZHANG Y,YUE C,et al.Processing defect,microstructure evolution and mechanical properties of laser powder bed fusion Al-12Si alloys[J].Journal of Materials Research and Technology,2023,26:681-696.
参考文献 60
WU Z,XU Z,FAN W.Online detection of powder spatters in the additive manufacturing process[J].Measurement,2022,194:111040.
参考文献 61
YE D,HONG G S,ZHANG Y,et al.Defect detection in selective laser melting technology by acoustic signals with deep belief networks[J].The International Journal of Advanced Manufacturing Technology,2018,96(5):2791-2801.
参考文献 62
CHEN L,YAO X,XU P,et al.Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning[J].Virtual and Physical Prototyping,2021,16(1):50-67.
参考文献 63
EVERTON S K,HIRSCH M,STRAVROULAKIS P,et al.Review of in-situ process monitoring and in-situ metrology for metal additive manufacturing[J].Materials & Design,2016,95:431-445.
参考文献 64
QIN L,ZHAO D,WANG W,et al.Geometric defects identification and deviation compensation in laser deposition manufacturing[J].Optics & Laser Technology,2022,155:108374.
参考文献 65
LI P,WARNER D H,FATEMI A,et al.Critical assessment of the fatigue performance of additively manufactured Ti-6Al-4V and perspective for future research[J].International Journal of Fatigue,2016,85:130-143.
参考文献 66
KAJI F,NGUYEN-HUU H,BUDHWANI A,et al.A deep-learning-based in-situ surface anomaly detection methodology for laser directed energy deposition via powder feeding[J].Journal of Manufacturing Processes,2022,81:624-637.
参考文献 67
MAZZARISI M,ERRICO V,ANGELASTRO A,et al.Influence of standoff distance and laser defocusing distance on direct laser metal deposition of a nickel-based superalloy[J].The International Journal of Advanced Manufacturing Technology,2022,120(3):2407-2428.
参考文献 68
LI K,LI T,MA M,et al.Laser cladding state recognition and crack defect diagnosis by acoustic emission signal and neural network[J].Optics & Laser Technology,2021,142:107161.
参考文献 69
CHEN L,YAO X,TAN C,et al.In-situ crack and keyhole pore detection in laser directed energy deposition through acoustic signal and deep learning[J].Additive Manufacturing,2023,69:103547.
参考文献 70
GAJA H,LIOU F.Defects monitoring of laser metal deposition using acoustic emission sensor[J].The International Journal of Advanced Manufacturing Technology,2017,90(1):561-574.
参考文献 71
LI Y,XIAO W,XIAO H,et al.Enhanced molten-pool boundary stability for microstructure control using quasi-continuous-wave laser additive manufacturing[J].Journal of Materials Research and Technology,2023,23:238-244.
参考文献 72
ASSELIN M,TOYSERKANI E,IRAVANITABRIZIPOUR M,et al.Development of trinocular CCD-based optical detector for real-time monitoring of laser cladding[C]//Mechatronics & Automation.IEEE International Conference,July 29-August 01,2005,Niagara Falls,ON,Canada.New York:IEEE,2005:1190-1196.
参考文献 73
HOJJATZADEH S M H,PARAB N D,GUO Q,et al.Direct observation of pore formation mechanisms during LPBF additive manufacturing process and high energy density laser welding[J].International Journal of Machine Tools and Manufacture,2020,153:103555.
参考文献 74
ZHANG K,LIU T,LIAO W,et al.Photodiode data collection and processing of molten pool of alumina parts produced through selective laser melting[J].Optik,2018,156:487-497.
参考文献 75
MUVVALA G,KARMAKAR D P,NATH A K.Online assessment of TiC decomposition in laser cladding of metal matrix composite coating[J].Materials & Design,2017,121:310-320.
参考文献 76
MISRA S,MOHANTY I,RAZA M S,et al.Investigation of IR pyrometer-captured thermal signatures and their role on microstructural evolution and properties of Inconel 625 tracks in DED-based additive manufacturing[J].Surface and Coatings Technology,2022,447:128818.
参考文献 77
MAZZARISI M,ANGELASTRO A,LATTE M,et al.Thermal monitoring of laser metal deposition strategies using infrared thermography[J].Journal of Manufacturing Processes,2023,85:594-611.
参考文献 78
D’ACCARDI E,CHIAPPINI F,GIANNASI A,et al.Online monitoring of direct laser metal deposition process by means of infrared thermography[J].Progress in Additive Manufacturing,2023,26:1-19.
参考文献 79
MAFFIA S,FURLAN V,PREVITALI B.Coaxial and synchronous monitoring of molten pool height,area,and temperature in laser metal deposition[J].Optics & Laser Technology,2023,163:109395.
参考文献 80
BI G,SCHÜRMANN B,GASSER A,et al.Development and qualification of a novel laser-cladding head with integrated sensors[J].International Journal of Machine Tools and Manufacture,2007,47(3):555-561.
参考文献 81
CHEN L,BI G,YAO X,et al.Multisensor fusion-based digital twin for localized quality prediction in robotic laser-directed energy deposition[J].Robotics and Computer-Integrated Manufacturing,2023,84:102581.
参考文献 82
GERON A.Hands-on machine learning with scikit-learn,keras,and tensorflow:concepts,tools,and techniques to build intelligent systems[M].California:O’ Reilly Media,Inc.,2019.
参考文献 83
SOOFI A,AWAN A.Classification techniques in machine learning:applications and issues[J].Journal of Basic & Applied Sciences,2017,13:459-465.
参考文献 84
KOTSIANTIS S B,ZAHARAKIS I D,PINTELAS P E.Machine learning:a review of classification and combining techniques[J].Artificial Intelligence Review,2006,26(3):159-190.
参考文献 85
GAJA H,LIOU F.Defect classification of laser metal deposition using logistic regression and artificial neural networks for pattern recognition[J].The International Journal of Advanced Manufacturing Technology,2018,94(1):315-326.
参考文献 86
DANG L,HE X,TANG D,et al.A fatigue life posterior analysis approach for laser-directed energy deposition Ti-6Al-4V alloy based on pore-induced failures by kernel ridge[J].Engineering Fracture Mechanics,2023,289:109433.
参考文献 87
KHANZADEH M,CHOWDHURY S,MARUFUZZAMAN M,et al.Porosity prediction:supervised-learning of thermal history for direct laser deposition[J].Journal of Manufacturing Systems,2018,47:69-82.
参考文献 88
LEE H,HEOGH W,YANG J,et al.Deep learning for in-situ powder stream fault detection in directed energy deposition process[J].Journal of Manufacturing Systems,2022,62:575-587.
参考文献 89
SEIFI S H,TIAN W,DOUDE H,et al.Layer-wise modeling and anomaly detection for laser-based additive manufacturing[J].Journal of Manufacturing Science and Engineering,2019,141(8):081013.
参考文献 90
CHEN T,WU W,LI W,et al.Laser cladding of nanoparticle TiC ceramic powder:effects of process parameters on the quality characteristics of the coatings and its prediction model[J].Optics & Laser Technology,2019,116:345-355.
参考文献 91
HAO J,YANG S,LE X,et al.Bead morphology prediction of coaxial laser cladding on inclined substrate using machine learning[J].Journal of Manufacturing Processes,2023,98:159-172.
参考文献 92
DANG L,HE X,TANG D,et al.A fatigue life prediction approach for laser-directed energy deposition titanium alloys by using support vector regression based on pore-induced failures[J].International Journal of Fatigue,2022,159:106748.
参考文献 93
FEENSTRA D R,MOLOTNIKOV A,BIRBILIS N.Utilisation of artificial neural networks to rationalise processing windows in directed energy deposition applications[J].Materials & Design,2021,198:109342.
参考文献 94
BHARDWAJ T,SHUKLA M.Laser additive manufacturing-direct energy deposition of ti-15mo biomedical alloy:artificial neural network based modeling of track dilution[J].Lasers in Manufacturing and Materials Processing,2020,7(3):245-258.
参考文献 95
LI J,SAGE M,GUAN X,et al.Machine Learning-enabled competitive grain growth behavior study in directed energy deposition fabricated Ti6Al4V[J].JOM,2020,72(1):458-464.
参考文献 96
CIAMPAGLIA A,TRIDELLO A,PAOLINO D S,et al.Data driven method for predicting the effect of process parameters on the fatigue response of additive manufactured AlSi10Mg parts[J].International Journal of Fatigue,2023,170:107500.
参考文献 97
GONZALEZ-VAL C,PALLAS A,PANADEIRO V,et al.A convolutional approach to quality monitoring for laser manufacturing[J].Journal of Intelligent Manufacturing,2020,31(3):789-795.
参考文献 98
ZHANG B,LIU S,SHIN Y C.In-process monitoring of porosity during laser additive manufacturing process[J].Additive Manufacturing,2019,28:497-505.
参考文献 99
TIAN Q,GUO S,MELDER E,et al.Deep learning-based data fusion method for in situ porosity detection in laser-based additive manufacturing[J].Journal of Manufacturing Science and Engineering,2020,143:1-38.
参考文献 100
HOSSAIN M S,TAHERI H.In-situ process monitoring for metal additive manufacturing through acoustic techniques using wavelet and convolutional neural network(CNN)[J].The International Journal of Advanced Manufacturing Technology,2021,116(11):3473-2488.
参考文献 101
PERANI M,BARALDO S,DECKER M,et al.Track geometry prediction for laser metal deposition based on on-line artificial vision and deep neural networks[J].Robotics and Computer-Integrated Manufacturing,2023,79:102445.
参考文献 102
FRANCIS J,BIAN L.Deep Learning for distortion prediction in laser-based additive manufacturing using big data[J].Manufacturing Letters,2019,20:10-14.
参考文献 103
XIE X,BENNETT J,SAHA S,et al.Mechanistic data-driven prediction of as-built mechanical properties in metal additive manufacturing[J].npj Computational Materials,2021,7(1):86-97.
参考文献 104
HU K,WANG Y,LI W,et al.CNN-BiLSTM enabled prediction on molten pool width for thin-walled part fabrication using laser directed energy deposition[J].Journal of Manufacturing Processes,2022,78:32-45.
参考文献 105
MOZAFFAR M,PAUL A,AL-BAHRANI R,et al.Data-driven prediction of the high-dimensional thermal history in directed energy deposition processes via recurrent neural networks[J].Manufacturing Letters,2018,18:35-39.
参考文献 106
REN K,CHEW Y,ZHANG Y F,et al.Thermal field prediction for laser scanning paths in laser aided additive manufacturing by physics-based machine learning[J].Computer Methods in Applied Mechanics and Engineering,2020,362:112734.
参考文献 107
ZHU X,JIANG F,GUO C,et al.Prediction of melt pool shape in additive manufacturing based on machine learning methods[J].Optics & Laser Technology,2023,159:108964.
参考文献 108
LI X,DAI R,CHEN S,et al.Offline planning optimization and formation prediction of laser directed energy deposition process[J].Optics & Laser Technology,2023,164:109510.
参考文献 109
GARCÍA-MORENO A I,ALVARADO-OROZCO J M,IBARRA-MEDINA J,et al.Ex-situ porosity classification in metallic components by laser metal deposition:a machine learning-based approach[J].Journal of Manufacturing Processes,2021,62:523-534.
参考文献 110
WANG Y,HU K,LI W,et al.Prediction of melt pool width and layer height for laser directed energy deposition enabled by physics-driven temporal convolutional network[J].Journal of Manufacturing Systems,2023,69:1-17.
参考文献 111
ZHU Q,LIU Z,YAN J.Machine learning for metal additive manufacturing:predicting temperature and melt pool fluid dynamics using physics-informed neural networks[J].Computational Mechanics,2021,67(2):619-635.
参考文献 112
TANG Y,RAHMANI D M,WANG G G.Review of transfer learning in modeling additive manufacturing processes[J].Additive Manufacturing,2023,61:103357.
参考文献 113
TAHERI H,KOESTER L W,BIGELOW T A,et al.In situ additive manufacturing process monitoring with an acoustic technique:clustering performance evaluation using K-means algorithm[J].Journal of Manufacturing Science and Engineering,2019,141(4):041011.
参考文献 114
OUIDADI H,GUO S,ZAMIELA C,et al.Real-time defect detection using online learning for laser metal deposition[J].Journal of Manufacturing Processes,2023,99:898-910.
参考文献 115
KHANZADEH M,CHOWDHURY S,TSCHOPP M A,et al.In-situ monitoring of melt pool images for porosity prediction in directed energy deposition processes[J].IISE Transactions,2019,51(5):437-455.
参考文献 116
REN W,WEN G,ZHANG Z,et al.Quality monitoring in additive manufacturing using emission spectroscopy and unsupervised deep learning[J].Materials and Manufacturing Processes,2022,37(11):1339-1346.
参考文献 117
YADAV P,SINGH V K,JOFFRE T,et al.Inline drift detection using monitoring systems and machine learning in selective laser melting[J].Advanced Engineering Materials,2020,22(12):2000660.
参考文献 118
PANDIYAN V,CUI D,LE-QUANG T,et al.In situ quality monitoring in direct energy deposition process using co-axial process zone imaging and deep contrastive learning[J].Journal of Manufacturing Processes,2022,81:1064-1075.
参考文献 119
YUAN B,GIERA B,GUSS G,et al.Semi-supervised convolutional neural networks for in-situ video monitoring of selective laser melting[C]//2019 IEEE winter conference on applications of computer vision(WACV),January 07-11,2019,Waikoloa,USA.New York:IEEE,2019:744-753.
参考文献 120
JAFARI-MARANDI R,KHANZADEH M,TIAN W,et al.From in-situ monitoring toward high-throughput process control:cost-driven decision-making framework for laser-based additive manufacturing[J].Journal of Manufacturing Systems,2019,51:29-41.
目录contents

    摘要

    机器学习作为人工智能领域的核心分支,通过算法分析数据,从中发现规律和模式,进而做出预测和决策,近年来在激光熔覆领域得到广泛应用。激光熔覆过程中形成的各类缺陷严重影响熔覆层的质量与性能,熔覆质量的可靠性与可重复性是激光熔覆技术面临的最大挑战。数据驱动的机器学习算法可用于激光熔覆过程监测与缺陷检测、反馈调控熔覆工艺、优化抑制熔覆缺陷,已成为本领域的研究热点。综述激光熔覆过程中产生的缺陷类型与成形机制,概述激光熔覆过程中产生的信号特征及其监测原理与手段,总结机器学习方法在激光熔覆过程中信号特征提取、缺陷分类识别与预测的研究进展,归纳缺陷检测的典型机器学习模型与算法。结果表明,机器学习算法可有效用于激光熔覆涂层缺陷检测,构建特征信号与涂层缺陷及熔覆工艺间的关系。目前研究采用的机器学习算法以监督学习算法为主,无监督和半监督学习算法对数据标注要求低,已在激光熔覆过程监测领域获得关注,并展现出巨大的潜力。研究结果为机器学习方法在激光熔覆领域的研究指出了热点与方向。

    Abstract

    The rapid development of artificial intelligence technology has led to significant changes and opportunities across various sectors. Machine learning, an important branch of artificial intelligence, can discover laws and patterns from data to make predictions and decisions. Furthermore, it has been widely used in the field of laser cladding in recent years. Laser cladding technology has emerged as a transformative method with numerous advantages, positioning it as a key player in various industrial applications. Its advantages, including high fusion efficiency, optimal material utilization, robust bonding, and extensive design flexibility, render it indispensable for repairing complex surface defects in metal parts. The occurrence of defects during the cladding process can significantly affect the quality and performance of the cladding layer. Ensuring the reliability and repeatability of cladding quality remains a significant challenge in the field of laser cladding technology. In this study, the application of machine learning algorithms in the field of laser cladding defect assessment is explored. A comprehensive and in-depth analysis of common defects and their formation mechanisms in the laser cladding process is provided. The acoustic, optical, and thermal signals generated during the cladding process are summarized, and the corresponding relationships between these signals and the cladding defects are described. Commonly used methods, sensors, and signal characteristics for monitoring the laser cladding process are summarized. Additionally, the classification and features of machine learning algorithms are organized and their use in signal processing is reviewed during the laser cladding process. The classification and characteristics of machine learning algorithms and their applications in laser cladding signal processing are summarized. Machine learning algorithms have been employed in detecting defects in laser cladding, typically by constructing datasets from features extracted from collected signals, the cladding process, and defect characteristics. These algorithms are used to establish relationships between the signals, defects, and the process. However, most current studies on laser cladding monitoring focus on a single pass or a small area of the cladding layer. The use of such small datasets can lead to model overfitting, thereby reducing the accuracy of defect detection. Nevertheless, the application of these algorithms facilitates the introduction of a dynamic feedback control mechanism that optimizes the cladding process and effectively mitigates defects. The convergence of laser cladding and machine learning has emerged as a vibrant area of research, tackling crucial issues and expanding the limits of quality assurance and process optimization. Researchers, both domestically and internationally, have examined pores, cracks, and other defects at various scales through experiments and simulations. However, the mechanisms behind these defects and their impact on the quality of cladding are not yet fully understood. There is a need for more comprehensive methods to study the laser cladding process. Developing a quantitative evaluation system that links the laser cladding process, signal data, and defect quality is a critical challenge in ensuring the reliability of laser cladding quality. Currently, various sensors, including acoustic, optical, and thermal types, are utilized to monitor the laser cladding process. These sensors aid in examining the relationship between the process signals, defects, and quality. However, the limitations in sensor accuracy and the efficiency of defect feature extraction pose challenges in establishing a precise process-signal-defect relationship. The predominant machine learning algorithms used in current research are supervised learning algorithms. However, unsupervised and semi-supervised learning algorithms, which require less data labeling, are drawing attention in the fields of laser melting and cladding process monitoring, demonstrating significant potential. This review emphasizes the current research hotspots and directions for applying machine learning methods in laser cladding.

  • 0 前言

  • 激光熔覆是一种高效的表面工程技术[1],通过在机械零件表面制备先进功能涂层,可有效提升机械零件在极端环境下的服役性能,在航空航天、能源、汽车和发电等工业领域得到了广泛应用,典型零部件包括叶片、涡轮盘、阀门、活塞、过热器管及各种轴类零部件等[2-4]。激光熔覆技术近年来发展迅速[5],其以高能激光束为热源,将金属粉末或线材熔化,与基材表面薄层形成熔凝,获得具有冶金结合的熔覆层[6],具有熔覆效率高、粉末利用率高、热影响区小、涂层稀释率小、冶金结合好、显微硬度高、耐腐蚀性好、可修复薄壁与小尺寸构件等优势[7-9]。然而激光熔覆涂层质量受工艺规划、材料选择及加工环境的影响,选择不当则会在涂层中产生孔隙、裂纹及变形等各类缺陷,降低熔覆层的力学性能[10-11]。如何优化激光熔覆过程,调节激光功率、送粉速率、扫描速度、熔覆材料等参量,减少熔覆层中的缺陷,确保熔覆层质量的可靠性与可重复性,已成为激光熔覆技术面临的最大挑战[12-13]

  • 通过对激光熔覆过程进行监测,可实现熔覆缺陷的发现与预测,为激光熔覆工艺提供优化反馈,降低熔覆加工失败的概率,缩短熔覆层研制周期。激光熔覆过程以复杂的物理过程为主,包括熔凝过程非平衡动力学、热力学和组织演变等[14-16]。为实现对激光熔覆过程的有效监测,须对熔覆过程中发生的物理化学反应进行深入理解。熔覆缺陷的产生是工艺参数的耦合结果,须掌握熔覆层中不同类型缺陷的成形机制与分布规律,总结熔覆缺陷的调控方法[17]。在此基础上,选择合适的传感器采集激光熔覆过程中产生的声、光、热等过程信号,获取熔池流动与凝固过程的状态特征[18]。目前常用的信号采集方法包括声发射传感器、高速摄像机、高温计等,均可实现熔覆过程信号的无损在线采集,许多研究者已通过熔覆过程的声信号、熔池流动图像及熔池温度场分布等信息,建立了缺陷生成与熔覆状态及工艺参数的关联关系[19-22]。然而,监测过程中产生的海量数据及工艺-信号-缺陷-质量之间的复杂关系使得基于传统统计方法的信号处理手段面临巨大挑战[23]。因此,基于机器学习的激光熔覆过程监测与缺陷检测已成为激光熔覆技术领域的研究前沿与热点。

  • 机器学习(Machine learning,ML)在缺陷检测、故障诊断及寿命预测领域得到了广泛应用[24-27],利用机器学习算法,可以从激光熔覆过程中采集到的信号里提取特征并进行分类与识别,在此基础上实现熔覆层中缺陷的分类、定位、预测及优化等。图1 归纳了应用于激光熔覆过程监测的典型机器学习算法及主要的缺陷类型。熔覆层中的缺陷主要分为孔隙、裂纹、欠熔合及变形等,用于过程监测的机器学习算法主要包括监督学习、无监督学习、半监督学习及强化学习几类。不同的机器学习算法具有各自的优缺点,须根据具体问题的性质、数据集的特征进行算法优选[28-30]。目前,国内外学者关于激光熔覆过程监测与缺陷检测相关的研究主要围绕激光熔覆过程中激光能量场与材料的作用关系、熔池热行为及原位监测方法、监测信号的处理与特征提取方法以及基于机器学习算法的分类预测相关工作[31-33],尚缺少对激光熔覆缺陷成形机制、监测信号特征提取与基于机器学习的分类预测相关的全面综述。

  • 图1 激光熔覆缺陷检测中的机器学习应用

  • Fig.1 Application of machine learning in laser cladding defect detection

  • 本文调研了近年来机器学习方法在激光熔覆过程监测与缺陷检测领域应用的国内外文献,对激光熔覆过程中缺陷类型与其成形机制进行了归纳,梳理了激光熔覆过程中产生信号的特征以及主要的监测原理和方法,探讨总结了不同机器学习算法在激光熔覆过程监测与缺陷检测中的应用特点,剖析了当前机器学习算法在涂层缺陷检测研究中存在的问题,并对未来机器学习在激光熔覆技术研究中的应用进行了总结与展望。

  • 1 熔覆缺陷与成形机理

  • 激光熔覆是一种动态物理冶金过程,其中高能束激光作用于合金粉末与基体,涉及粉末输送、激光能量吸收、材料熔化以及熔覆层凝固成形等多个环节,包含传热、对流、传质及结晶等多种复杂过程相互作用,受熔覆工艺参数、粉末与基体材料特性等多种因素的影响。熔覆过程中能量集中,熔覆层凝固时间极短,常因热源特性、加工工艺、材料性能等因素的匹配调控不当,熔覆层中产生孔隙、裂纹、欠熔合、夹渣和几何变形等缺陷[34],进而影响熔覆层的质量与组织性能,降低熔覆零件的服役寿命。

  • 1.1 孔隙生成与分布

  • 孔隙是激光熔覆涂层中最常见的缺陷,其大小、数量、形状及分布对熔覆层的硬度、强度、各向异性和疲劳性能都有直接影响[35-39]。孔隙作为熔覆层中的薄弱区,易产生应力集中,诱导生成裂纹,影响熔覆层的致密度、结合性能、疲劳强度以及服役寿命。熔覆层中形成孔隙的机理各不相同,气孔是激光熔覆过程最易产生的孔隙缺陷,粉末扩散不足、欠熔合及锁孔也会导致熔池中形成孔隙[40-41]。气孔根据生成机制主要分为卷入型与反应型两类。

  • 激光熔覆过程中,熔覆粉末与基材表面都会吸附空气及水分,高温作用下,挥发卷入熔池中形成气孔。气孔的逸出行为受表面张力、Marangoni 对流效应及反冲压力等影响[42],多力耦合下在熔池内运动,若气体未能在熔池凝固闭合前从固液表面逸出,便会被熔池捕获形成气孔缺陷[43]。熔融态金属与逸出的水汽可反应生成金属氧化物和氢气,氢气进一步卷入熔池形成气孔[44]。熔覆层内气孔的产生由熔池凝固过程中气体的逸出行为决定,如图2 所示,激光熔覆过程中熔池内箭头所示方向发生Marangoni对流,流动方向从底部向上到熔池中心表面,然后从中心向熔池边界流动。这种对流将气体转移到熔池底部,促进孔隙的运动,增加了小孔合并成大孔的碰撞机会,气孔沿流体流动方向移动,呈链状分布。

  • 图2 激光熔覆层中孔隙成形机制

  • Fig.2 Formation mechanism of pores within the laser cladding layer

  • 1.2 裂纹萌生与扩展

  • 熔覆层裂纹是内应力与冶金特性耦合作用下产生的缺陷,裂纹的成形原因、产生位置受熔覆参数影响。裂纹缺陷直接影响熔覆层的耐磨、耐腐蚀性能,甚至会作为应力集中点诱发熔覆件断裂失效。因此,基于机器学习方法,结合裂纹特征及分布,探索材料-工艺-缺陷-性能间的内在关联机制,调控熔覆材料及工艺是目前激光熔覆涂层缺陷调控领域的研究热点。

  • 激光能场作用下,粉末与基体熔化形成液态金属熔池,熔池内熔融金属受表面张力、重力、粘性切应力和保护气压力的共同作用,使得熔池对流凝固过程中各部分材料的热膨胀系数不同,进而在熔覆层中诱导产生热应力、相变应力、约束应力等内应力。当局部内应力积累超过材料的应力极限时,熔覆层中萌生出裂纹[45-46]。热应力在熔覆层中裂纹缺陷生成过程起主导作用[47-48],熔覆层材料热膨胀系数大于基体材料热膨胀系数时,热应力表现为拉应力,反之为压应力。熔覆层中的气孔、晶间液膜、偏聚硬质相易成为局部应力集中区[49],裂纹往往在这些位置萌生,并沿熔覆层中材料薄弱位置扩展。

  • 如图3 所示,裂纹类型可根据取向和位置进行分类。根据裂纹取向可将裂纹分为横向裂纹、纵向裂纹和涂层内的网状裂纹[50]。横向裂纹通常始于涂层与基材的接缝处,然后垂直于扫描方向延伸到涂层表面,表现为整个涂层垂直于扫描方向的断裂,横向裂纹形状比较平直,呈现跨晶断裂特征。纵向裂纹的萌生主要受涂层截面上的拉伸应力的影响,涂层较厚时,纵向裂纹主要从涂层表面沿冶金结合区延伸到底部;涂层变薄时,沿冶金结合区的界面裂纹在涂层两侧变为近似垂直的纵向裂纹。网状裂纹通常萌生于涂层中,由热应力驱动在三维空间中沿多方向发展,可延伸至涂层表面或界面,具有晶间和跨晶扩展的特点[50]

  • 图3 熔覆层中裂纹成形机制及主要分布区域

  • Fig.3 Formation mechanism of cracks and their main distribution areas in the cladding layer

  • 根据裂纹萌生位置可将裂纹分为熔覆层裂纹、搭接区裂纹和冶金结合区裂纹。激光熔覆过程中,熔池温度分布由中心至边缘逐渐降低,金属熔体在不平衡表面张力梯度的驱动下,从低张力区(中心) 向高张力区(边缘)流动,引发 Marangoni 对流效应[51],从而导致应力集中,使得熔覆层中萌生裂纹。熔覆层搭接区域因相邻焊道反复熔化,产生显著的温度梯度,热应力积累使得搭接区产生裂纹。由于熔覆层与基体具有不同的激光能量吸收特性、导热系数及热膨胀系数,在熔池流动凝固过程中,热传导会逐渐过渡到热对流阶段,使熔池中心的热应力明显小于熔池边界,最大的热应力出现在熔池底部位置,导致冶金结合区形成裂纹。

  • 1.3 欠熔合与夹渣

  • 粉末流从喷嘴喷出后与激光束耦合,粉末吸收激光辐射能量后被熔池捕获或形成飞溅[52]。如图4 所示,飞溅的粉末按物理性质可分为完全熔化和部分熔化状态,部分熔化的粉末会引起欠熔合。激光功率较低时,熔覆层表面欠熔合粉末颗粒会增大表面粗糙度[53]。此外,当粉末进给量较少时,熔池宽度较窄,引起涂层搭接率降低,导致多道焊缝间残留大量未熔粉末,焊缝中残留的熔渣称为夹渣。夹渣会降低焊缝的塑性和韧性,其尖角易造成应力集中,导致裂纹萌生[54]。多层熔覆制备过程中的重熔效应能改善层间粉末欠熔合状态[55],但层粗糙度会影响熔覆层整体的粗糙程度,引起层间间隙,诱使孔隙的产生,影响层间结合[56-57]。部分熔化的粉末会在熔池中形成欠熔合孔,对熔池的粘附起阻碍作用,一旦流动的熔池被阻断,熔池中的温度梯度在熔滴表面引起的表面张力使熔滴产生收卷成球的趋势,从而导致球化现象[5058]。熔滴间形成的空腔区域难以被完全填充,也会导致孔隙等缺陷的产生[5059]

  • 图4 粒子飞溅引起表面粗糙与球化等欠熔合缺陷[59-61]

  • Fig.4 Defects caused by particle spattering such as rough surface and spheroidization [59-61]

  • 1.4 熔覆层几何变形

  • 常见的激光熔覆几何缺陷包括平面度缺陷、熔融塌陷、变形、开裂和分层等。几何变形作为一种宏观缺陷,其成因主要是熔覆过程中应力和误差的积累。几何变形直接影响熔覆成形的精度,严重时会导致零件直接报废[34]。如图5 所示,熔覆层中的应力积累会导致翘曲变形,形成平面度缺陷[62],大型的熔覆件通常都存在平面度缺陷。熔覆过程中热量散布不均时,会在表面形成不稳定的熔池,常常在熔覆层边缘发生熔融塌陷[28]

  • 图5 激光熔覆表面几何变形[62]

  • Fig.5 Geometric deformation of laser cladded sample[62]

  • 引起熔覆层变形的残余应力主要包括热应力、结构应力、约束应力和凝固收缩力等[63],这些应力主要是因为熔覆过程中熔池快速冷却形成的温度梯度引起的。熔池快速加热和冷却产生了不均匀的热膨胀和收缩,从而形成了热应力[64-65]。熔池凝固过程中晶粒生长速率不同,引起的晶粒尺寸差异使得相变的时间和程度不同,从而导致结构应力的产生。熔覆层的变形程度和位置分布与约束模式有关,约束应力主要与材料属性、温度和零件尺寸相关。凝固收缩是指结晶和冷却过程中由于熔体体积减小和温度分布不均匀而导致的体积收缩,产生残余应力[66]。周期性多次重熔使得激光熔覆件中出现开裂与分层缺陷,分层通常出现在基体与熔覆层之间或连续的熔覆层之间,层间裂纹通常是由层间温度梯度不同导致的收缩率不同引起的。防止开裂和分层的有效措施包括有效散热、适当工艺参数和材料的适配性[67]

  • 2 熔覆过程信号及检测

  • 如图6 所示,激光熔覆涂层制备过程中,激光与粉末、基材作用会产生声、光、热等信号,这些信号包含丰富的熔覆过程信息,对监测熔覆过程异常和熔覆层中缺陷、预测熔覆层性能及优化工艺参数具有重要作用,高效准确地获得信号数据是实现熔覆层质量控制与机器学习预测的前提。

  • 图6 激光熔覆过程中熔池信号采集方式

  • Fig.6 Molten pool signal acquisition method during laser cladding process

  • 2.1 声信号

  • 粉末与基材熔化及熔池流动过程中会产生弹性波,其携带许多与内部特征相关的声波信号,包括熔覆层中的缺陷类型、缺陷位置及熔覆层质量等信息。选用合适的传感器采集并处理声信号特征,可实现缺陷监测和熔覆层质量预测。

  • 如图7 所示,LI 等[68]研究表明,正常熔覆过程中声发射(Acoustic emission,AE)信号的幅值在 0.05 dB 以内,而有裂纹的熔覆过程中声发射信号的幅值最高可达 1.5 dB,是正常状态的 30 倍,说明声信号可有效用于裂纹缺陷的监测。熔覆过程中的噪声信号复杂,须对采集信号进行降噪与特征提取。CHEN 等[69]利用麦克风传感器监测了马氏体时效钢 C300 粉末的激光熔覆过程,研究了声发射信号和裂纹、孔隙等缺陷的关系,通过自动化原位声学降噪、特征提取,建立了激光-材料相互作用的声信号分类模型,实现了基于声学信号的熔覆缺陷检测。识别熔覆过程中的特征信号是建立缺陷与声发射信号关系的基础,GAJA 等[70]研究了钛合金与工具钢复合粉体激光熔覆过程中的声发射信号,并从中提取了七种特征来分析激光熔覆过程,并使用机器学习算法对特征数据进行聚类分析,有效区分了裂纹与孔隙两种缺陷的声发射信号。

  • 图7 激光熔覆过程的原位声学信号采集装置[68]

  • Fig.7 In-situ acoustic signal acquisition device for the laser cladding process[68]

  • 声信号可以对熔覆过程中熔池状态及缺陷进行监测,通过提取、分类等处理获得的声学特征,可建立声信号和缺陷的对应关系。然而,声学传感器的类型、监测位置与角度等设置对声信号采集有直接影响。其难点主要在于声发射信号的准确采集以及信号处理过程中的特征提取,利用机器学习方法能够高效地建立信号特征与缺陷间的关系。目前,激光熔覆声发射信号监测研究主要针对单道熔覆,须进一步开发适用于多道、多层及整个加工过程的声信号监测方法。

  • 2.2 光信号

  • 熔覆过程中的图像信息可用来研究粉末、熔池、飞溅、孔隙、裂纹等现象。常用的光信号采集装置包括工业相机、高速摄像机、光谱仪及光电二极管等设备。通过对采集信号的特征提取,能够实现对熔覆涂层的质量和缺陷的监测。工业相机图像采集成本低、分辨率高,常用于评价涂层表面质量及缺陷。高速摄像机采样频率高,可捕捉瞬态熔池变化,监测飞溅、气孔等特征,可在线识别缺陷。光电二极管可将采集的光信号转为电信号,使得熔池、飞溅等产生的辐射信号转变为模拟电信号,丰富涂层质量的评价信息。X 射线光谱仪可穿透样品,反映熔覆层内部缺陷的数量、形貌与位置,可结合裂纹扩展模型,有效评价涂层的疲劳寿命。

  • 如图8 所示,LI 等[71]利用高速摄像机监控激光熔覆过程中的熔池动态特性,研究不同激光模式对熔池边界稳定性和凝固微观结构的影响,发现准连续波激光在增强几何形貌稳定性和促进柱状枝晶连续外延生长方面的优势。ASSELIN 等[72]设计了基于三组工业相机的熔池视觉在线测量系统,并开发了相应的图像分析算法,从图像中提取熔池的几何形状,并将其作为闭环信号,实现复杂曲面激光熔覆的过程控制。HOJJATZADEH 等[73]采用原位 X 射线方法研究了粉末熔化形成熔池的过程,通过将 X 射线信号转变为图像数据,实现气孔生成过程的原位观测,并提出新的气孔成形机制,为减少气孔缺陷提供依据。ZHANG 等[74]将熔池辐射的光信号通过衰减片和滤光片过滤后,通过光电二极管转化为电流信号,研究不同激光功率下采集信号的稳定性,发现熔池与光电二极管的相对距离和入射角对数据精度影响较大。

  • 图8 激光熔覆过程图像采集系统[71]

  • Fig.8 Image acquisition system for laser cladding process[71]

  • 光信号测量技术具有非接触、高效率、高准确度及易于实现自动化的特点,目前已广泛应用于激光熔覆过程监测,可作为闭环反馈控制信号,减少缺陷的生成,对激光熔覆层质量提升与灵活制造具有重要意义。然而,激光熔覆过程速度快,对反馈控制要求高,须提升光信号识别精度与速度,优化熔覆过程识别算法。

  • 2.3 热信号

  • 高能激光束能熔化粉末并在基材表面形成熔池,熔池的形成、运动与凝固均为热传递过程,激光熔覆过程中复杂的热传导对熔覆层的微观组织、残余应力、缺陷及变形等都有直接影响。因此,研究激光熔覆过程中的温度分布对优化熔覆层质量具有重要意义。激光熔覆过程中,用于温度监测的传感器主要有高温计和热成像仪两类,高温计通常只能测量局部区域的温度,而热成像仪的测温范围更大。对比传统的接触式热电偶测温传感器,高温计与热成像仪均为非接触式,可在线检测运动熔池的温度分布。

  • 如图9 所示,MUVVALA 等[75]采用高温计监测 TiC 颗粒增强金属基复合涂层在激光熔覆过程中的反应特性,发现低扫描速度、缓慢冷却过程会导致 TiC 颗粒分解,生成枝晶结构,证明通过研究熔池温度场可实现陶瓷增强相分解状态的无损分析。MISRA 等[76] 利用红外高温计监测熔池温度场分布,探索工艺参数与熔池特性的关系。通过采集温度场分布、熔覆层微观结构及沉积特性,实现了晶粒形态、相结构及涂层缺陷的预测。MAZZARISI 等[77]利用热成像仪研究了单层多道熔覆层制备过程中熔池平均温度、最高温度、热循环、冷却速率和热梯度特征,在此基础上分析了裂纹、孔隙缺陷的成形机制及硬度变化规律,提出熔覆工艺优化策略。D’ ACCARDI 等[78]采用红外热成像仪对激光熔覆过程中熔池温度进行跟踪和监测,从熔池选定区提取热特征,通过方差分析统计方法将温度数据与激光功率、扫描速度及粉末流量等熔覆参数进行相关性分析,并验证了红外热像仪在线监测激光熔覆过程中缺陷与质量异常的可行性。

  • 图9 激光熔覆过程热信号采集过程与分析处理方法[75]

  • Fig.9 Thermal history acquisition settings for laser cladding process, and thermal signal analysis and processing methods[75]

  • 温度场的分布与材料的形态、热辐射率密切相关。熔覆过程中存在液态的熔池、固态的粉末、凝固的金属及气态的蒸发金属,流动的熔池也使得不同位置的温度差异明显。因此,熔覆层的热辐射率并非常数,而是随材料状态、空间分布及温度变化。

  • 2.4 多信号融合

  • 采用单一传感器测得的信号难以全面反映激光熔覆过程中加工状态与缺陷信息。随着监测技术的发展,近年来出现许多采用多传感器集成的激光熔覆过程监测的研究,通过同时采集多种信号,能够更全面地反映熔覆特征,提高监测的准确性。

  • 如图10 所示,MAFFIA 等[79]构建了多传感器融合熔池特性监测系统,同时在线监测熔覆过程中熔池高度、面积与温度特征。结果表明,熔池高度主要受扫描速度的影响,而熔池面积和温度主要受激光功率影响,熔池面积与温度线性相关。BI 等[80]开发了一种集成相机和光电二极管的新型激光熔覆头,实现熔池特征的在线监测,利用控制器实现熔覆过程中距离与温度的闭环控制,避免熔覆堆积,实现熔覆壁厚控制。CHEN 等[81]设计了基于视觉图像、声发射信号以及温度场信号的多传感器时空信息融合方法,采用视觉信号标签缺陷区域,结合声发射信号与熔池热特征信号训练缺陷预测模型,获得了比传统单传感器缺陷预测方法更高的精度。

  • 图10 激光熔覆过程多传感器信号采集方案[79]

  • Fig.10 Multi-sensor signal acquisition scheme for laser cladding process [79]

  • 3 缺陷评估机器学习算法

  • 激光熔覆过程中,熔池中包含着丰富的熔覆信息,熔池形貌、温度场分布、光谱与声发射信号等特征均与凝固后熔覆层的内部组织结构、缺陷和几何精度密切相关。通过关联这些特征变量,优化工艺参数,可实现熔覆层质量的提升。机器学习旨在利用算法通过数据样本寻找规律,从而构建出具有 “举一反三”泛化能力的模型。如图11 所示,机器学习方法可用于激光熔覆层缺陷检测、性能评价、工艺优化及工艺参数-微观结构-宏观性能间的复杂相互关系探索。机器学习过程主要包括数据采集、数据预处理、特征工程、数据集划分、模型的训练与验证等。

  • 应用于激光熔覆过程研究的机器学习方法主要有监督学习、无监督学习、半监督学习三类。监督学习利用有标签的特征数据进行训练,可建立工艺参数、熔池特性、熔覆层缺陷特征及熔覆层性能的映射关系,以预测、分类缺陷类别。无监督学习利用无标签的特征数据进行训练,对具有相似特征的熔池与缺陷信息进行聚类或降维,挖掘工艺参数和熔覆层性能间的内在关系。半监督学习通过将监督学习的预测与半监督学习的聚类结合,利用少量有标签数据和大量无标签数据进行训练,以推断未知数据的缺陷类别。

  • 图11 激光熔覆缺陷检测、工艺性能优化的机器学习框架

  • Fig.11 Machine Learning Framework for Laser Cladding Defect Detection and Process Performance Optimization

  • 3.1 监督学习

  • 监督学习是目前应用最广泛的机器学习技术,常用监督学习算法包括逻辑回归( Logistic regression,LR)、决策树、支持向量机、K-近邻学习、人工神经网络和卷积神经网络等。通过组合多个基本监督学习模型可实现更好的预测,集成学习方法主要包括引导聚集(Bagging)算法、提升(Boosting) 方法和堆垛集成学习(Stacking)方法等[82]。根据数据的连续性或离散性,可将监督学习任务分为分类问题和回归问题,其中分类问题的标签是离散值,回归问题的标签是连续值[83-84]。表1 列出了近年来激光熔覆涂层缺陷分析预测相关的文献,内容包括监督学习算法、材料与缺陷类型、数据集类型及预测结果等。

  • 表1 用于激光熔覆涂层缺陷检测的监督学习算法

  • Table1 Supervised machine learning algorithms utilized in defect detection for laser cladding

  • (续)

  • Where MSE is mean square error, FN is false negative, MPE is mean percentage error, MAPE is mean absolute percentage error, RMSE is root mean square error, MWIR is medium wavelength infrared, MRE is Mean relative error.

  • GAJA 等[85]使用逻辑回归模型检测 Ti6Al4V 与 H13 混合粉末熔覆涂层中产生的缺陷。通过采集熔覆过程中的声发射信号,提取了峰值幅度、上升时间、持续时间、能量和计数等声发射特征,有效标记监测了裂纹和孔隙两种缺陷。DANG 等[86]采用岭回归模型进行钛合金疲劳寿命后验分析研究,提取显微组织指标和应力强度特征,建立疲劳寿命后验分析模型,获得了较好的泛化预测结果。回归模型通过分析变量间的数学关系预测结果,按数学关系可分为线性回归和非线性回归等。根据输入的激光熔覆参数、材料特性与缺陷类型,回归模型可用来预测熔覆层的关键性能指标。

  • 决策树代表的是对象属性与对象值之间的一种映射关系,其基本流程遵循“分而治之”策略。喷嘴内部流路堵塞引起的异常粉末流以及熔池区域的发光和飞溅物喷射使得异常送粉难以监测,LEE 等[88] 提出了基于同轴相机图像的熔覆异常和送粉异常识别技术,如图12 所示。利用同轴相机采集正常与喷嘴堵塞条件下的熔池特征,用于识别熔覆过程中的异常送粉。通过模型的训练与验证,发现随着采集图像曝光时间的增加,决策树方法的预测精度提高,在 300 μs 曝光时间下,预测精度达到 93%。 KHANZADEH 等[87]基于决策树模型建立了熔池图像与孔隙位置间的对应关系,对比了采用熔池热特征与熔池简单指标(长度、宽度、峰值温度、面积等)预测涂层孔隙率精度,发现包含熔池热特征的形态学模型结合监督学习模型表现出更好的预测精度,决策树模型将正常熔池识别为孔隙的误差值低至 0.03%。决策树既可以处理数值型特征,也可以处理分类型特征,对于原始数据的处理较为宽松,不需要大量的数据预处理工作,也无需对数据进行特殊转换,处理不同类型的数据时更为灵活。

  • 图12 基于图像识别的异常送粉检测流程[88]

  • Fig.12 Abnormal powder feeding detection process based on image recognition [88]

  • 支持向量机(Support vector machine,SVM)是基于训练集对样本空间进行分类的模型,其决策边界是对学习样本求解的最大边距超平面,也可通过核方法进行非线性分类。如图13 所示,SEIFI 等[89] 采用多线性主成分分析从熔池图像中提取逐层特征,然后使用 SVM 模型进行逐层质量预测,并采用两种交叉验证技术评估所提出方法的预测性能,证明该技术可实现零件质量在线监测。CHEN 等[90] 建立了基于 SVM 的陶瓷基激光熔覆涂层质量性能预测模型,探究熔覆工艺参数与涂层质量特性之间的关系。结果表明,预置粉末厚度、激光光斑直径和激光功率是影响熔覆层性能的关键工艺参数。支持向量回归(Support vector regression,SVR)模型使用 SVM 模型进行回归分析,寻找最优拟合曲线。HAO 等[91]采用 SVR 模型预测倾斜基板上 SS316L 激光熔覆涂层的高度、宽度和峰值移位点。模型输入包括倾斜角度、激光功率、送粉速率和扫描速度。训练模型后发现,通过调节模型超参数(惩罚系数、核系数),可提高 SVR 模型预测精度。DANG 等[92]建立了基于 SVR 模型的钛合金零件疲劳寿命预测模型,将应力强度因子与微观孔隙特征作为输入变量,验证得到的预测结果与实测的疲劳寿命数据接近。SVM 模型对小样本数据的分类效果较好,通过最大化分类边界的间隔,具有很强的泛化能力,能够在面对新数据时具备良好的性能。

  • 图13 熔池热成像数据特征提取与激光熔覆薄壁件缺陷检测[89]

  • Fig.13 Feature extraction of molten pool thermal imaging data and defect detection of laser cladding thin-walled parts[89]

  • K 近邻(K-nearest neighbors,KNN)模型通过测量不同特征值之间的距离对样本进行分类,该方法在定类决策上只依据最邻近的一个或者几个样本的类别来决定待分样本所属的类别。如图14 所示, CHEN 等[62]开展了基于熔覆层表面点云轮廓信息的表面缺陷预测研究,将表面缺陷分为凸起、凹痕和波浪形缺陷,采用 KNN 模型获得了 93.15%的准确率。WU 等[60]利用 KNN 模型研究激光熔覆过程中粉末飞溅现象。通过高速相机采集熔池上方 5 mm 处的飞溅粉末,按未熔、半熔和全熔三种状态对飞溅的粉末进行标签,依据熔覆涂层的孔隙率将涂层质量分为四个水平。在此基础上,通过分配符合质量水平的反距离权重,获得的涂层质量预测准确率在 95%以上。KHANZADEH 等[87]从激光熔覆熔池温度场分布中提取特征信息,应用 KNN 模型预测单道薄壁 Ti6Al4V 试样的孔隙率,经过参数调优和 K 折交叉验证方法的 KNN 模型分类准确率最高为 98.44%。KNN 模型直观简单,适用于小样本量的多类别分类问题,但对异常值不敏感。

  • 图14 激光熔覆层表面轮廓缺陷聚类分析方法[62]

  • Fig.14 Cluster analysis method for surface profile defects of laser cladding layer[62]

  • 人工神经网络(Artificial neural network,ANN) 模型由多个节点层组成,通常包含输入层、一个或多个隐藏层和输出层。节点通过相关的权重和阈值连接到另一个节点,依靠训练数据改变权重,降低训练集上的累计误差,基于梯度下降策略的误差逆传播(Back propagation,BP)算法就是典型代表。在涂层缺陷检测中,ANN 模型可以提高聚类以及分类的效率和精度。

  • 如图15 所示,FEENSTRA 等[93]采用 ANN 模型研究了激光熔覆工艺对熔覆焊道几何形状与稀释率的影响,输入层包括激光功率、光束直径、扫描速度等七项加工参数,输出层为熔覆层的几何尺寸。训练得到的模型对熔覆层的高度、深度及稀释率的预测精度分别达到 91%、95.5%和 92.7%,有效揭示了熔覆工艺参数与熔池凝固几何特性之间的关系。 BHARDWAJ 等[94]为研究单道 Ti15Mo 材料熔覆层的可重复性,通过采集熔覆层截面几何尺寸,建立了工艺参数耦合的 ANN 模型,优化模型参数后实现了熔覆层稀释率的高精度预测。LI 等[95]研究了 Ti6Al4V 激光熔覆熔池凝固过程中微观结构的形成机制,建立了基于晶界倾斜角与热梯度、晶体取向和 Marangoni 效应等因素的 ANN 模型,有效地描述竞争性晶粒生长行为及微观结构生成定量模拟。 ANN 模型在大规模和高维数据集上表现出色,能够学习和建模复杂的非线性关系,具有自适应性,能够通过学习从数据中提取特征,无需手工设计特征。 CIAMPAGLIA 等[96]训练了前馈神经网络和物理信息神经网络两种模型,用来研究熔覆层微观组织结构与缺陷对样品疲劳寿命的影响。针对 AlSi10Mg 数据集的疲劳强度预测结果与试验结果的平均误差为 4%,最大误差为 17%。在这项研究中,基于物理模型和数据驱动模型的优势被融合为一种新型的物理驱动的机器学习方法。利用从温度场和熔池流动中推导出的物理特征,推断熔池的峰值温度,使该方法能够有效地学习熔池动力学,并将马兰戈尼指数设计为熔池流动形态变化的指标,建立工艺参数与熔池宽度和层高之间的复杂非线性关系。 WANG 等[110]建立了基于物理驱动的时域卷积网络 (Temporal convolutional network,TCN)模型,设计了基于熔池动力学和熔池流动机理的物理模型,提取熔池宽度和熔池层高密切相关的固有物理特征,实现了熔池宽度和层高预测。传统的机器学习方法难以从温度场和熔池流动中挖掘隐藏的物理信息,使得对激光熔覆复杂的热机械现象缺乏物理可解释性,可能会导致输入和输出之间的错误关系。通过将物理模型中提取的物理知识或规则整合到机器学习驱动模型的输入中,可提高模型的透明度、可解释性和分析能力[111]

  • 图15 基于工艺参数对涂层形状和稀释率的 ANN 模型预测流程[93]

  • Fig.15 ANN model prediction process for coating shape and dilution rate based on process parameters[93]

  • 卷积神经网络(Convolutional neural network,CNN)模型具有优异的图像处理和模式识别性能,在激光熔覆熔池特征分类与熔覆层缺陷检测领域得到广泛应用。CNN 模型主要包括输入层、卷积层、激活层、池化层、全连接层和输出层,通过局部连接和参数共享,建立输入图像与输出目标之间的映射,实现图像的分类。目前,CNN 模型已应用于激光熔覆孔隙、裂纹、欠熔合及几何变形等缺陷的检测与分析。

  • GONZALEZ-VAL 等[97]利用 CNN 模型研究了三种钢材的激光熔覆层质量。模型输入为从熔池同轴红外图像中提取的特征参数和质量指标,缺陷数据集包括从 50 条焊道中获得的 24 444 张图片,人工标注缺陷类型,训练后模型预测气孔、欠熔合等缺陷的 F1(准确率和召回率的调和平均值)得分达到 0.975。如图16 所示,ZHANG 等[98]使用同轴高速相机监测海绵钛粉激光熔覆过程中的熔池特征,采用 X 射线横截面层析成像法提取熔覆层中孔隙尺寸,将熔池参数与孔隙尺寸作为 CNN 模型的输入参数,模型可实现对 100 μm 以下微孔的预测,对熔覆层中出现孔隙的预测准确率达到 91.2%。TIAN 等[99]开发了基于 CNN 的 PyroNet 模型,用于研究熔池温度场分布与层间孔隙率间的关系。采用断层扫描将孔径大于 0.05mm 的样品标记为缺陷,构建的数据集包含 840 张照片,通过交叉验证方式避免过拟合问题,获得的算法对 Ti6Al4V 薄壁件孔隙缺陷的预测准确率接近 100%。

  • 图16 基于 CNN 方法的熔池特征提取与孔隙缺陷预测模型[98]

  • Fig.16 Melt pool feature extraction and pore defect prediction model based on CNN method[98]

  • 如图17 所示,CHEN 等[69]开发了基于声发射信号的激光熔覆层缺陷识别技术。通过光学显微镜采集熔覆层中的孔隙与裂纹位置,并将其与采集的声发射信号进行时空对应,从中提取无缺陷、孔隙及裂纹特征对应的声学特征。通过在对应的噪声数据集上训练 CNN、自适应增强(Adaptive boosting,AdaBoost)与梯度提升(Gradient boosting,GB)等模型,对比其他经典机器学习模型,CNN 模型对熔覆层中缺陷的预测总体准确率达到 89%。HOSSAIN 等[100]采用声发射传感器采集设备空闲、仅粉末喷射、最佳熔覆工艺、低激光功率及低送粉率等五种状态下的声发射信号,并利用小波变换获取各工艺条件下声发射信号的时频谱。在此基础上,利用 CNN 模型关联声发射信号与熔覆层截面扫描电镜图像中观测到的缺陷,模型对 Ti6Al4V 熔覆层中的缺陷预测准确率达到 96%。

  • 图17 基于声发射信号的激光熔覆层孔隙与裂纹深度学习预测方法[69]

  • Fig.17 Deep learning prediction method of pores and cracks of laser cladding layer based on acoustic emission signal[69]

  • XIE 等[103]利用激光熔覆技术制备了多层镍基薄壁样品,通过调节熔覆间隔时间、优化壁面几何形态,研究制备工艺与样品力学性能之间的关系。如图18 所示,数据集包括壁面成形过程中采集的热历史及选定区域试样的极限抗拉强度。在此基础上,采用 CNN 模型预测不同几何位置处样品的抗拉强度,预测误差低于 3%,结果表明熔覆间隔有助于提升薄壁的力学性能。PERANI 等[101]为避免熔覆涂层的大变形,开展了激光熔覆层几何变形在线监测与预测的研究。利用同轴熔池图像以及工艺参数作为 CNN 模型训练数据,通过模型优化,发现具有三个以上卷积层的模型表现出更好的预测能力,模型为调节激光熔覆工艺,优化熔覆焊道的宽度、高度及几何形状提供支撑。FRANCIS 等[102]指出熔覆层的几何变形主要是由热循环产生的残余应力引起的,研究熔覆过程中局部区域的热历史可实现对熔覆层几何形状的畸变预测。通过采集 Ti6Al4V 圆盘熔覆制备过程中不同位置温度场与形变数据,建立包含熔覆工艺参数的 CNN 模型,模型对圆盘制造形变误差的预测可控制在 56 μm 内。

  • 图18 基于 CNN 方法的薄壁样品力学性能预测框架[103]

  • Fig.18 CNN-based framework for predicting mechanical performance of thin-walled samples[103]

  • CNN 模型侧重于从输入数据中提取空间特征,擅长处理图像分类、目标检测和图像分割等任务,其中像素或特征之间的空间关系至关重要。不同于循环神经网络(Recurrent neural networks,RNN)或长短期记忆(Long short term memory,LSTM)模型,CNN 模型没有内在的时序性概念。为解决熔覆过程监测时间上的梯度消失以及优化实时过程控制,需要建立混合机器学习模型[104-106],以提高模型的预测精度。

  • 集成学习通过将多个基本学习器组合,以提高整体预测性能和泛化能力,根据生成方式与学习器间的依赖关系,可分为串行方法和并行方法。串行方法的代表为 Boosting 方法和 Stacking 方法,并行方法的代表是 Bagging 算法和随机森林(Random forest,RF)。如图19 所示,GARCÍA-MORENO 等[109] 利用 RF 模型进行了激光金属熔覆(Laser metal deposition,LMD)涂层中孔隙缺陷的评估研究。通过图像降噪、孔隙分割,从图像中提取了 15 种孔隙缺陷相关的特征,依据孔隙尺寸将其分为微孔、大孔和细长孔三类,并人工标注,建立了 6 552 个样本的数据集,训练得到的模型分类准确率大于 94%。 RF 模型通过组合多个决策树的预测结果,可以减小过拟合风险,提高模型的泛化能力,对于大规模数据集的处理能力较强。ZHU 等[107]研究了激光熔覆工艺参数对 304 不锈钢熔覆层形状的影响。利用光学显微镜采集了 210 个熔覆层的截面形貌,并提取了熔池的高度和宽度用于构建数据集,使用极限梯度提升(XGBoost)算法构建预测模型,模型对熔池高度和宽度的预测准确率分别达到 97.0%和 96.3%。XGBoost 通过在每轮迭代中引入正则化项,降低了模型的过拟合风险,能够提供特征对于模型的重要性评估,这有助于理解模型是如何做出预测决策的。LI 等[108]针对激光熔覆过程中拐角处材料堆叠问题,采用 Stacking 方法建立熔覆层高度预测模型,以减少拐角处热量积聚导致的塌陷、鼓包与变形等缺陷。经过熔池特征提取和熔覆层轮廓处理后,以加工工艺参数和熔池特征等作为模型输入,熔覆层厚度作为输出,训练结果表明,Stacking 模型的预测准确性要高于单学习器 SVR 模型,其平均绝对百分比误差为 2.369 7%。Stacking 方法适用于具有多样性的基本模型,通过选择性能差异较大的基本模型,可以更好地捕捉数据的不同方面,提高集成模型的泛化能力。

  • 图19 基于 RF 方法的激光熔覆涂层缺陷采集、特征提取与模型训练验证过程[109]

  • Fig.19 Defect acquisition, feature extraction and model training verification process for laser cladding coating based on random forest method[109]

  • 监督学习的缺点在于对数据集质量的要求极高,由于激光熔覆成本相对较高,获得的数据有限,这给构建基于物理或数据的高质量模型带来了挑战。迁移学习(Transfer learning,TL)是一种新型的机器学习方法,可以将各种数据源与有限的新数据连接起来,并构建具有高度可转移性的模型。如图20 所示,基于实例和基于特征的 TL 方法仅应用于质量预测和过程优化,而基于模型的 TL 方法和多任务学习则广泛应用于质量预测与检测、缺陷检测和过程监控。对于激光熔覆数据集,大多数研究旨在探索从过程到过程、从涂层到涂层以及从材料到材料的可转移性。目前,激光熔覆过程的研究主要采用离线训练模式,并使用有限的数据验证构建的模型,然而训练好的模型对目标过程的预测准确度受到训练样本量的限制。在线 TL 方法辅助建模框架可能利用原位数据在线提高模型的预测性能[112]

  • 图20 迁移学习在激光熔覆中的应用[112]

  • Fig.20 Application of transfer learning in laser cladding[112]

  • 3.2 无监督学习

  • 无监督学习通常在对未标记数据无先验知识的情况下,探究数据的内在结构、关系及规律,主要包括聚类(Clustering)、自组织映射(Self-organizing map,SOM)及深度置信网络(Deep belief network,DBN)等算法,表2 总结了基于上述算法的激光熔覆缺陷检测文献。K-means 是最常用的聚类算法,其将样本数据分为 K 组,随机选取聚类中心,通过计算距离函数将数据进行分类,可用于研究数据分布结构或对数据对象进行分类。SOM 是一种竞争学习型的无监督学习算法,其通过计算输入空间中的数据,降维生成一个低维、离散的映射(Map),同时保持输入数据在高维空间的拓扑结构,进而实现对数据的分类。

  • 表2 用于激光熔覆工艺缺陷检测的无监督机器学习算法

  • Table2 Unsupervised machine learning algorithms utilized in defect detection for laser cladding processes

  • 如图21 所示,REN 等[116]采用 LSTM-自动编码器提取 7075 铝合金熔覆过程中光谱信号的特征,并从熔覆层截面的 SEM 照片中采集孔隙缺陷特征,在此基础上,利用 K-means 聚类对熔覆层的质量进行分类预测。该模型能够依据表面粗糙度和高孔隙率有效区分不合格熔覆层和合格熔覆层。LSTM 模型能够捕捉序列特征,提高整体预测性能;DBN 模型能够捕捉数据中的非线性关系,对于复杂的数据结构有更好的拟合能力;SVM 模型能在特征空间中构建超平面,对于线性不可分的问题具有良好的处理能力。结合 DBN 和 SVM 模型可以在更高维度的空间中进行分类,从而更好地处理声发射信号。

  • 图21 基于 LSTM 自编码器和 K-means 聚类的无监督深度学习模型实现框架(L 为偏差函数,S 为光谱函数, x 为原始光谱信号,x’为重建光谱信号)[116]

  • Fig.21 Implementation framework of unsupervised deep learning model based on LSTM autoencoder and K-means clustering (L is deviation function, S is spectral function, x is original spectral signals, x’is reconstructed spectral signals) [116]

  • 3.3 半监督学习

  • 监督学习需要大量标注数据训练模型,数据标记耗时且昂贵。无监督学习对数据质量要求高,聚类或降维后数据缺乏解释性,缺少目标和标签,使得模型的性能和准确度主观且难以量化,存在过拟合风险。半监督学习( Semi-supervised learning)可在部分标记数据的基础上,通过未标记数据进行辅助学习,未标记数据通常更易获取,因此可以扩大训练集的规模,提高模型性能与泛化能力。

  • 表3 总结了基于半监督学习算法的激光熔覆缺陷检测文献,YADAV 等[117]利用断层扫描技术研究了激光熔覆过程中的漂移现象,使用无监督的 K-means 聚类算法,借助少量标记数据对未标记数据进行标记,并帮助选择最合适的距离度量,利用标记的数据集训练了 KNN 模型,利用半监督方法成功地将熔覆层按“漂移”和“无漂移”分类。 K-means 聚类可以看作是对数据进行降维的一种方式,通过选择簇中心作为特征,减少输入维度,有助于 KNN 模型更有效地进行分类。如图22 所示, PANDIYAN 等[118]使用同轴相机进行激光熔覆过程原位质量监测,依据熔覆工艺将熔覆层质量分为六个等级,采用基于对比学习的 CNN 模型对采集的熔池图像进行聚类,在此基础上采用逻辑回归算法对低维数据进行监督分类,获得的半监督模型对熔覆层中缺陷类型的预测平均准确率为 97%。YUAN 等[119]设计了用于监测 316L 不锈钢粉末激光熔覆过程的半监督 CNN 模型,通过采集单道熔覆层的几何特征与熔池特征,采用部分标记数据和大量未标记数据训练了半监督 CNN 模型。结果表明,半监督方法的回归和分类性能都优于全监督方法。 JAFARI-MARANDI 等[120]采用红外热像仪提取了激光熔覆 Ti6Al4V 熔池热数据,并利用 X 射线捕获熔覆层中的微观结构与缺陷信息,应用多层感知机 (Multi layer perceptron,MLP)预测数据在 SOM 算法映射上的位置,构建的半监督自组织误差驱动神经网络可实现对熔覆层中孔隙缺陷空间分布的有效预测,为优化调控熔覆层的力学性能提供支撑。MLP 在学习过程中能够自动进行特征学习,提取输入数据中的有用信息,通过在 SOM 算法上映射 MLP 的输出,有助于降低数据维度并可视化数据结构,使模型更好地适应不同的数据特征。但结合 MLP 和 SOM 算法需要调整的超参数相对较多,可能会增加整体计算的复杂性。

  • 表3 用于超高速激光熔覆工艺缺陷检测的半监督机器学习算法

  • Table3 Semi-supervised machine learning algorithms utilized in defect detection for ultra-high speed laser cladding processes

  • 图22 基于半监督学习方法的激光熔覆熔池监测分类框架[118]

  • Fig.22 Classification framework for melt pools based on semi-supervised learning method [118]

  • 4 结论与展望

  • 综述了机器学习算法在激光熔覆缺陷评估领域的应用,对激光熔覆过程中常见的缺陷及其成形机制进行了全面深入的分析;对熔覆过程中产生的声、光、热信号进行了归纳,阐述了信号与熔覆缺陷之间的对应关系,总结了常用的激光熔覆过程监测方法、传感器与信号特征;梳理了机器学习算法的分类与特点,总结了其在激光熔覆过程信号处理中的应用。通过对文献的分析,总结概括如下:

  • (1)激光熔覆过程复杂,产生的缺陷直接影响熔覆层的质量,缺陷的成形机制与分布规律受多种因素影响,缺陷之间也相互影响与演化。目前,国内外研究人员已通过试验与模拟等手段从多尺度开展了孔隙、裂纹等缺陷的研究,但对相关缺陷的生成机理及缺陷对熔覆质量的作用机制仍不够深入,须采用更丰富的手段研究激光熔覆过程。

  • (2)构建激光熔覆过程工艺-信号-缺陷-质量定量评价体系是保证激光熔覆质量可靠性的关键挑战。目前,声、光与热等多种传感器都已应用于激光熔覆过程监测中,用来研究信号与工艺、缺陷及质量的关系,然而受传感器精度、缺陷特征提取效率等限制,建立工艺-信号-缺陷定量关系仍具挑战。须开发多传感器、多信号融合的在线激光熔覆过程监测技术与缺陷特征提取技术,从而获得全面、可靠、精确的熔覆信息与缺陷状态,实现全过程实时质量监控,是激光熔覆过程监测的重要发展方向。

  • (3)机器学习算法已经在激光熔覆缺陷检测中得到应用,通常基于采集信号中提取的特征、熔覆工艺及缺陷特征构建数据集,进而采用机器学习算法建立信号与缺陷及工艺间的关系。然而,目前激光熔覆过程监测研究多针对单道或小区域熔覆层,采集的小型数据集会引起模型的过拟合,导致实际缺陷检测精度降低,须针对激光熔覆过程设计通用标准缺陷检测数据库。另外,选择合适的机器学习算法对激光熔覆过程中的不同缺陷检测至关重要,不同的算法在处理图像数据或传感器信号时各有优势。 CNN 是各种缺陷图像数据处理的首选方法,SVM 方法适用于传感器信号或图像的多类分类问题, K-means 聚类在无监督和半监督学习中广泛使用。

  • 为推动激光熔覆技术成为机械制造与再制造领域的新质生产力,对机器学习方法在激光熔覆技术中的应用做如下展望:

  • (1)激光熔覆技术在制造业中具有广泛的应用前景,而机器学习技术的引入能有效提升激光熔覆的效率、降低熔覆涂层中的缺陷。目前报道的文献以监督学习算法为主,然而监督学习对数据标注要求高,需耗费大量时间与成本。因此,无监督和半监督学习算法已在激光熔覆过程监测领域获得关注,新型的模型层出不穷,并展现出巨大的潜力。

  • (2)通过机器学习技术,可以实现对激光熔覆设备的自动化控制和在线监控,通过对大量的激光熔覆数据进行分析和挖掘,优化调节激光功率、扫描速度和粉末喷射等参数,可实现激光熔覆过程的自动化与智能化,提高生产效率和降低成本。机器学习技术将为激光熔覆领域带来更多的创新和发展机遇,助力该技术在制造业中的广泛应用。

  • 参考文献

    • [1] SIDDIQUI A A,DUBEY A K.Recent trends in laser cladding and surface alloying[J].Optics & Laser Technology,2021,134:106619.

    • [2] WENG F,CHEN C,YU H.Research status of laser cladding on titanium and its alloys:A review[J].Materials & Design,2014,58:412-425.

    • [3] LIU J,YU H,CHEN C,et al.Research and development status of laser cladding on magnesium alloys:A review[J].Optics and Lasers in Engineering,2017,93:195-210.

    • [4] 高亚丽,路鹏勇,刘宇,等.镁合金表面激光熔覆研究现状[J].中国表面工程,2023,36(3):22-39.GAO Yali,LU Pengyong,LIU Yu,et al.Research status of laser cladding on magnesium alloy[J].China Surface Engineering,2023,36(3):22-39.(in Chinese)

    • [5] YANG Q,ZHANG P,LU Q,et al.Application and development of blue and green laser in industrial manufacturing:A review[J].Optics & Laser Technology,2024,170:110202.

    • [6] ZHANG H,LIU Y,BAI X,et al.Laser cladding highly corrosion-resistant nano/submicron ultrafine-grained Fe-based composite layers[J].Surface and Coatings Technology,2021,424:127636.

    • [7] ARIF Z U,KHALID M Y,UR REHMAN E,et al.A review on laser cladding of high-entropy alloys,their recent trends and potential applications[J].Journal of Manufacturing Processes,2021,68:225-273.

    • [8] ZHANG Q,WANG Q,HAN B,et al.Comparative studies on microstructure and properties of CoCrFeMnNi high entropy alloy coatings fabricated by high-speed laser cladding and normal laser cladding[J].Journal of Alloys and Compounds,2023,947:169517.

    • [9] ARIF Z U,KHALID M Y,AL RASHID A,et al.Laser deposition of high-entropy alloys:A comprehensive review[J].Optics & Laser Technology,2022,145:107447.

    • [10] HALDAR B,SAHA P.Identifying defects and problems in laser cladding and suggestions of some remedies for the same[J].Materials Today:Proceedings,2018,5(5):90-101.

    • [11] 曹佳俊,常成,邱兆国,等.AISI1045 钢表面激光熔覆 FeCoCrNiAl0.5Ti0.5 涂层的界面特性及摩擦性能[J].中国表面工程,2023,36(2):54-64.CAO Jiajun,CHANG Cheng,QIU Zhaoguo,et al.Interface characteristics and tribological properties of laser cladded FeCoCrNiAl0.5Ti0.5 coating on AISI 1045 steel[J] China Surface Engineering,2023,36(2):54-64.(in Chinese)

    • [12] OCELÍK V,EEKMA M,HEMMATI I,et al.Elimination of start/stop defects in laser cladding[J].Surface and Coatings Technology,2012,206(8):2403-2409.

    • [13] LI R,FENG A,ZHAO J,et al.Study on process optimization of WC-Ni60A cermet composite coating by laser cladding[J].Materials Today Communications,2023,37:107400.

    • [14] LI G,WANG Z,YAO L,et al.Concentration mixing and melt pool solidification behavior during the magnetic field assisted laser cladding of Fe-Cr-based alloy on 45 steel surface[J].Surface and Coatings Technology,2022,445:128732.

    • [15] SONG B,YU T,JIANG X,et al.Evolution and convection mechanism of the melt pool formed by V-groove laser cladding[J].Optics & Laser Technology,2021,144:107443.

    • [16] 何志远,贺文雄,杨海峰,等.铝合金表面激光熔覆研究进展[J].中国表面工程,2021,34(6):33-44.HE Zhiyuan,HE Wenxiong,YANG Haifeng,et al.Research progess in laser cladding on aluminum alloy surface[J].China Surface Engineering,2021,34(6):33-44.(in Chinese)

    • [17] 郭星星,帅美荣,王建梅,等.基于 NSGA-II 算法的激光熔覆单道成形工艺参数多目标优化[J].中国表面工程,2023,36(3):87-100.GUO Xingxing,SHUAI Meirong,WANG Jianmei,et al.Multi-objective optimization of laser cladding single-pass forming process parameters based on NSGA-II algorithm[J].China Surface Engineering,2023,36(3):87-100.(in Chinese)

    • [18] THAWARI N,GULLIPALLI C,CHANDAK A,et al.Influence of laser cladding parameters on distortion,thermal history and melt pool behaviour in multi-layer deposition of stellite 6:in-situ measurement[J].Journal of Alloys and Compounds,2021,860:157894.

    • [19] FALLAH V,ALIMARDANI M,CORBIN S F,et al.Temporal development of melt-pool morphology and clad geometry in laser powder deposition[J].Computational Materials Science,2011,50(7):2124-2134.

    • [20] WIRTH F,ARPAGAUS S,WEGENER K.Analysis of melt pool dynamics in laser cladding and direct metal deposition by automated high-speed camera image evaluation[J].Additive Manufacturing,2018,21:369-382.

    • [21] ZHANG Y M,LIM C W J,TANG C,et al.Numerical investigation on heat transfer of melt pool and clad generation in directed energy deposition of stainless steel[J].International Journal of Thermal Sciences,2021,165:106954.

    • [22] LI Y,XU F.Acoustic emission sources localization of laser cladding metallic panels using improved fruit fly optimization algorithm-based independent variational mode decomposition[J].Mechanical Systems and Signal Processing,2022,166:108514.

    • [23] 郭永明,叶福兴,祁航.超高速激光熔覆技术研究现状及发展趋势[J].中国表面工程,2022,35(6):39-50.GUO Yongming,YE Fuxing,QI Hang,et al.Research status and development of ultra-high speed laser cladding[J].China Surface Engineering,2022,35(6):39-50.(in Chinese)

    • [24] JINLONG W,YUXIN M,WENJIE P,et al.Evaluation of the effect of surface roughness parameters on fatigue of TC17 titanium alloy impeller using machine learning algorithm and finite element analysis[J].Engineering Failure Analysis,2023,153:107586.

    • [25] HAO W Q,TAN L,YANG X G,et al.A physics-informed machine learning approach for notch fatigue evaluation of alloys used in aerospace[J].International Journal of Fatigue,2023,170:107536.

    • [26] ZHAO S,SUN H,PENG F,et al.Feature fusion and distillation embedded sparse Bayesian learning model for in-situ foreknowledge of robotic machining errors[J].Journal of Manufacturing Systems,2023,71:546-564.

    • [27] LEE J A,SAGONG M J,JUNG J,et al.Explainable machine learning for understanding and predicting geometry and defect types in Fe-Ni alloys fabricated by laser metal deposition additive manufacturing[J].Journal of Materials Research and Technology,2023,22:413-423.

    • [28] SVETLIZKY D,DAS M,ZHENG B,et al.Directed energy deposition(DED)additive manufacturing:physical characteristics,defects,challenges and applications[J].Materials Today,2021,49:271-295.

    • [29] ZHOU L,MA G,ZHAO H,et al.Research status and prospect of extreme high-speed laser cladding technology[J].Optics & Laser Technology,2024,168:109800.

    • [30] LIN X,ZHU K,FUH J Y H,et al.Metal-based additive manufacturing condition monitoring methods:from measurement to control[J].ISA Transactions,2022,120:147-166.

    • [31] CAI Y,XIONG J,CHEN H,et al.A review of in-situ monitoring and process control system in metal-based laser additive manufacturing[J].Journal of Manufacturing Systems,2023,70:309-326.

    • [32] FU Y,DOWNEY A R J,YUAN L,et al.Machine learning algorithms for defect detection in metal laser-based additive manufacturing:A review[J].Journal of Manufacturing Processes,2022,75:693-710.

    • [33] WANG C,TAN X P,TOR S B,et al.Machine learning in additive manufacturing:state-of-the-art and perspectives[J].Additive Manufacturing,2020,36:101538.

    • [34] QIN L,WANG K,LI X,et al.Review of the formation mechanisms and control methods of geometrical defects in laser deposition manufacturing[J].Chinese Journal of Mechanical Engineering:Additive Manufacturing Frontiers,2022,1(4):100052.

    • [35] ZHAO S,YUAN K,GUO W,et al.A comparative study of laser metal deposited and forged Ti-6Al-4V alloy:uniaxial mechanical response and vibration fatigue properties[J].International Journal of Fatigue,2020,136:105629.

    • [36] WAN H,WANG Q,JIA C,et al.Multi-scale damage mechanics method for fatigue life prediction of additive manufacture structures of Ti-6Al-4V[J].Materials Science and Engineering:A,2016,669:269-278.

    • [37] LI J,CHENG X,LI Z,et al.Improving the mechanical properties of Al-5Si-1Cu-Mg aluminum alloy produced by laser additive manufacturing with post-process heat treatments[J].Materials Science and Engineering:A,2018,735:408-417.

    • [38] YU X,LIN X,TAN H,et al.Microstructure and fatigue crack growth behavior of Inconel 718 superalloy manufactured by laser directed energy deposition[J].International Journal of Fatigue,2021,143:106005.

    • [39] STERLING A J,TORRIES B,SHAMSAEI N,et al.Fatigue behavior and failure mechanisms of direct laser deposited Ti-6Al-4V[J].Materials Science and Engineering:A,2016,655:100-112.

    • [40] LEUNG C L A,MARUSSI S,ATWOOD R C,et al.In situ X-ray imaging of defect and molten pool dynamics in laser additive manufacturing[J].Nature Communications,2018,9(1):1355.

    • [41] LI S,CHEN B,TAN C,et al.In situ identification of laser directed energy deposition condition based on acoustic emission[J].Optics & Laser Technology,2024,169:110152.

    • [42] SONG B,YU T,JIANG X,et al.The relationship between convection mechanism and solidification structure of the iron-based molten pool in metal laser direct deposition[J].International Journal of Mechanical Sciences,2020,165:105207.

    • [43] KHAIRALLAH S A,ANDERSON A T,RUBENCHIK A,et al.Laser powder-bed fusion additive manufacturing:physics of complex melt flow and formation mechanisms of pores,spatter,and denudation zones[J].Acta Materialia,2016,108:36-45.

    • [44] YANG Z,WANG A,WENG Z,et al.Porosity elimination and heat treatment of diode laser-clad homogeneous coating on cast aluminum-copper alloy[J].Surface and Coatings Technology,2017,321:26-35.

    • [45] 操龙飞.金属材料的热膨胀特性研究[D].武汉:武汉科技大学,2013.CAO Longfei.Study on thermal expansion properties of steels[D].Wuhan:Wuhan University of Science and Technology,2013.(in Chinese)

    • [46] 李春彦,张松,康煜平,等.综述激光熔覆材料的若干问题[J].激光杂志,2002,3:5-9.LI Chunyan,ZHANG Song,KANG Yuping,et al.Comment on material system for laser cladding[J].Laser Journal,2002,3:5-9.(in Chinese)

    • [47] ZHOU S,ZENG X,HU Q,et al.Analysis of crack behavior for Ni-based WC composite coatings by laser cladding and crack-free realization[J].Applied Surface Science,2008,255(5):1646-1653.

    • [48] 申发明.超高速激光熔覆AISI431不锈钢涂层组织与耐蚀机理研究[D].哈尔滨:哈尔滨工业大学,2022.SHEN Faming.Research on microstructure and corrosion resistance mechanism of AISI431 stainless steel coating prepared by extra high speed laser cladding[D].Harbin:Harbin Institute of Technology,2022.(in Chinese)

    • [49] AUCOTT L,HUANG D,DONG H B,et al.A three-stage mechanistic model for solidification cracking during welding of steel[J].Metallurgical and Materials Transactions A,2018,49(5):1674-1682.

    • [50] GAO Z,WANG L,WANG Y,et al.Crack defects and formation mechanism of FeCoCrNi high entropy alloy coating on TC4 titanium alloy prepared by laser cladding[J].Journal of Alloys and Compounds,2022,903:163905.

    • [51] JIN K,YANG Z,CHEN P,et al.Dynamic solidification process during laser cladding of IN718:multi-physics model,solute suppressed nucleation and microstructure evolution[J].International Journal of Heat and Mass Transfer,2022,192:122907.

    • [52] SCHWERZ C,BIRCHER B A,KÜNG A,et al.In-situ detection of stochastic spatter-driven lack of fusion:application of optical tomography and validation via ex-situ X-ray computed tomography[J].Additive Manufacturing,2023,72:103631.

    • [53] XU X,DU J L,LUO K Y,et al.Microstructural features and corrosion behavior of Fe-based coatings prepared by an integrated process of extreme high-speed laser additive manufacturing[J].Surface and Coatings Technology,2021,422:127500.

    • [54] FONSECA E B,GABRIEL A H G,ARAÚJO L C,et al.Assessment of laser power and scan speed influence on microstructural features and consolidation of AISI H13 tool steel processed by additive manufacturing[J].Additive Manufacturing,2020,34:101250.

    • [55] 王豫跃,牛强,杨冠军,等.超高速激光熔覆技术绿色制造耐蚀抗磨涂层[J].材料研究与应用,2019,13(3):165-172.WANG Yuyue,NIU Qiang,YANG Guanjun,et al.lnvestigations on corrosion-resistant and wear-resistant coatings environmental-friendly manufactured by a novel super-high efficient laser cladding[J].Materials Research and Application,2019,13(3):165-172.(in Chinese)

    • [56] ZHANG Y,GAO X,LIANG X,et al.Effect of laser remelting on the microstructure and corrosion property of the arc-sprayed AlFeNbNi coatings[J].Surface and Coatings Technology,2020,398:126099.

    • [57] 黄旭,张家诚,练国富,等.超高速激光熔覆研究现状及应用[J].机床与液压,2021,49(6):151-155,162.HUANG Xu,ZHANG Jiacheng,LIAN Guofu,et al.Research status and application of extreme high speed cladding[J].Machine Tool & Hydraulics,2021,49(6):151-155,162.(in Chinese)

    • [58] YE X,WANG J,YING Q,et al.Melting behavior of in-flight particles in ultra-high speed laser cladding[J].Journal of Materials Research and Technology,2023,24:7047-7057.

    • [59] WANG W,ZHANG Y,YUE C,et al.Processing defect,microstructure evolution and mechanical properties of laser powder bed fusion Al-12Si alloys[J].Journal of Materials Research and Technology,2023,26:681-696.

    • [60] WU Z,XU Z,FAN W.Online detection of powder spatters in the additive manufacturing process[J].Measurement,2022,194:111040.

    • [61] YE D,HONG G S,ZHANG Y,et al.Defect detection in selective laser melting technology by acoustic signals with deep belief networks[J].The International Journal of Advanced Manufacturing Technology,2018,96(5):2791-2801.

    • [62] CHEN L,YAO X,XU P,et al.Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning[J].Virtual and Physical Prototyping,2021,16(1):50-67.

    • [63] EVERTON S K,HIRSCH M,STRAVROULAKIS P,et al.Review of in-situ process monitoring and in-situ metrology for metal additive manufacturing[J].Materials & Design,2016,95:431-445.

    • [64] QIN L,ZHAO D,WANG W,et al.Geometric defects identification and deviation compensation in laser deposition manufacturing[J].Optics & Laser Technology,2022,155:108374.

    • [65] LI P,WARNER D H,FATEMI A,et al.Critical assessment of the fatigue performance of additively manufactured Ti-6Al-4V and perspective for future research[J].International Journal of Fatigue,2016,85:130-143.

    • [66] KAJI F,NGUYEN-HUU H,BUDHWANI A,et al.A deep-learning-based in-situ surface anomaly detection methodology for laser directed energy deposition via powder feeding[J].Journal of Manufacturing Processes,2022,81:624-637.

    • [67] MAZZARISI M,ERRICO V,ANGELASTRO A,et al.Influence of standoff distance and laser defocusing distance on direct laser metal deposition of a nickel-based superalloy[J].The International Journal of Advanced Manufacturing Technology,2022,120(3):2407-2428.

    • [68] LI K,LI T,MA M,et al.Laser cladding state recognition and crack defect diagnosis by acoustic emission signal and neural network[J].Optics & Laser Technology,2021,142:107161.

    • [69] CHEN L,YAO X,TAN C,et al.In-situ crack and keyhole pore detection in laser directed energy deposition through acoustic signal and deep learning[J].Additive Manufacturing,2023,69:103547.

    • [70] GAJA H,LIOU F.Defects monitoring of laser metal deposition using acoustic emission sensor[J].The International Journal of Advanced Manufacturing Technology,2017,90(1):561-574.

    • [71] LI Y,XIAO W,XIAO H,et al.Enhanced molten-pool boundary stability for microstructure control using quasi-continuous-wave laser additive manufacturing[J].Journal of Materials Research and Technology,2023,23:238-244.

    • [72] ASSELIN M,TOYSERKANI E,IRAVANITABRIZIPOUR M,et al.Development of trinocular CCD-based optical detector for real-time monitoring of laser cladding[C]//Mechatronics & Automation.IEEE International Conference,July 29-August 01,2005,Niagara Falls,ON,Canada.New York:IEEE,2005:1190-1196.

    • [73] HOJJATZADEH S M H,PARAB N D,GUO Q,et al.Direct observation of pore formation mechanisms during LPBF additive manufacturing process and high energy density laser welding[J].International Journal of Machine Tools and Manufacture,2020,153:103555.

    • [74] ZHANG K,LIU T,LIAO W,et al.Photodiode data collection and processing of molten pool of alumina parts produced through selective laser melting[J].Optik,2018,156:487-497.

    • [75] MUVVALA G,KARMAKAR D P,NATH A K.Online assessment of TiC decomposition in laser cladding of metal matrix composite coating[J].Materials & Design,2017,121:310-320.

    • [76] MISRA S,MOHANTY I,RAZA M S,et al.Investigation of IR pyrometer-captured thermal signatures and their role on microstructural evolution and properties of Inconel 625 tracks in DED-based additive manufacturing[J].Surface and Coatings Technology,2022,447:128818.

    • [77] MAZZARISI M,ANGELASTRO A,LATTE M,et al.Thermal monitoring of laser metal deposition strategies using infrared thermography[J].Journal of Manufacturing Processes,2023,85:594-611.

    • [78] D’ACCARDI E,CHIAPPINI F,GIANNASI A,et al.Online monitoring of direct laser metal deposition process by means of infrared thermography[J].Progress in Additive Manufacturing,2023,26:1-19.

    • [79] MAFFIA S,FURLAN V,PREVITALI B.Coaxial and synchronous monitoring of molten pool height,area,and temperature in laser metal deposition[J].Optics & Laser Technology,2023,163:109395.

    • [80] BI G,SCHÜRMANN B,GASSER A,et al.Development and qualification of a novel laser-cladding head with integrated sensors[J].International Journal of Machine Tools and Manufacture,2007,47(3):555-561.

    • [81] CHEN L,BI G,YAO X,et al.Multisensor fusion-based digital twin for localized quality prediction in robotic laser-directed energy deposition[J].Robotics and Computer-Integrated Manufacturing,2023,84:102581.

    • [82] GERON A.Hands-on machine learning with scikit-learn,keras,and tensorflow:concepts,tools,and techniques to build intelligent systems[M].California:O’ Reilly Media,Inc.,2019.

    • [83] SOOFI A,AWAN A.Classification techniques in machine learning:applications and issues[J].Journal of Basic & Applied Sciences,2017,13:459-465.

    • [84] KOTSIANTIS S B,ZAHARAKIS I D,PINTELAS P E.Machine learning:a review of classification and combining techniques[J].Artificial Intelligence Review,2006,26(3):159-190.

    • [85] GAJA H,LIOU F.Defect classification of laser metal deposition using logistic regression and artificial neural networks for pattern recognition[J].The International Journal of Advanced Manufacturing Technology,2018,94(1):315-326.

    • [86] DANG L,HE X,TANG D,et al.A fatigue life posterior analysis approach for laser-directed energy deposition Ti-6Al-4V alloy based on pore-induced failures by kernel ridge[J].Engineering Fracture Mechanics,2023,289:109433.

    • [87] KHANZADEH M,CHOWDHURY S,MARUFUZZAMAN M,et al.Porosity prediction:supervised-learning of thermal history for direct laser deposition[J].Journal of Manufacturing Systems,2018,47:69-82.

    • [88] LEE H,HEOGH W,YANG J,et al.Deep learning for in-situ powder stream fault detection in directed energy deposition process[J].Journal of Manufacturing Systems,2022,62:575-587.

    • [89] SEIFI S H,TIAN W,DOUDE H,et al.Layer-wise modeling and anomaly detection for laser-based additive manufacturing[J].Journal of Manufacturing Science and Engineering,2019,141(8):081013.

    • [90] CHEN T,WU W,LI W,et al.Laser cladding of nanoparticle TiC ceramic powder:effects of process parameters on the quality characteristics of the coatings and its prediction model[J].Optics & Laser Technology,2019,116:345-355.

    • [91] HAO J,YANG S,LE X,et al.Bead morphology prediction of coaxial laser cladding on inclined substrate using machine learning[J].Journal of Manufacturing Processes,2023,98:159-172.

    • [92] DANG L,HE X,TANG D,et al.A fatigue life prediction approach for laser-directed energy deposition titanium alloys by using support vector regression based on pore-induced failures[J].International Journal of Fatigue,2022,159:106748.

    • [93] FEENSTRA D R,MOLOTNIKOV A,BIRBILIS N.Utilisation of artificial neural networks to rationalise processing windows in directed energy deposition applications[J].Materials & Design,2021,198:109342.

    • [94] BHARDWAJ T,SHUKLA M.Laser additive manufacturing-direct energy deposition of ti-15mo biomedical alloy:artificial neural network based modeling of track dilution[J].Lasers in Manufacturing and Materials Processing,2020,7(3):245-258.

    • [95] LI J,SAGE M,GUAN X,et al.Machine Learning-enabled competitive grain growth behavior study in directed energy deposition fabricated Ti6Al4V[J].JOM,2020,72(1):458-464.

    • [96] CIAMPAGLIA A,TRIDELLO A,PAOLINO D S,et al.Data driven method for predicting the effect of process parameters on the fatigue response of additive manufactured AlSi10Mg parts[J].International Journal of Fatigue,2023,170:107500.

    • [97] GONZALEZ-VAL C,PALLAS A,PANADEIRO V,et al.A convolutional approach to quality monitoring for laser manufacturing[J].Journal of Intelligent Manufacturing,2020,31(3):789-795.

    • [98] ZHANG B,LIU S,SHIN Y C.In-process monitoring of porosity during laser additive manufacturing process[J].Additive Manufacturing,2019,28:497-505.

    • [99] TIAN Q,GUO S,MELDER E,et al.Deep learning-based data fusion method for in situ porosity detection in laser-based additive manufacturing[J].Journal of Manufacturing Science and Engineering,2020,143:1-38.

    • [100] HOSSAIN M S,TAHERI H.In-situ process monitoring for metal additive manufacturing through acoustic techniques using wavelet and convolutional neural network(CNN)[J].The International Journal of Advanced Manufacturing Technology,2021,116(11):3473-2488.

    • [101] PERANI M,BARALDO S,DECKER M,et al.Track geometry prediction for laser metal deposition based on on-line artificial vision and deep neural networks[J].Robotics and Computer-Integrated Manufacturing,2023,79:102445.

    • [102] FRANCIS J,BIAN L.Deep Learning for distortion prediction in laser-based additive manufacturing using big data[J].Manufacturing Letters,2019,20:10-14.

    • [103] XIE X,BENNETT J,SAHA S,et al.Mechanistic data-driven prediction of as-built mechanical properties in metal additive manufacturing[J].npj Computational Materials,2021,7(1):86-97.

    • [104] HU K,WANG Y,LI W,et al.CNN-BiLSTM enabled prediction on molten pool width for thin-walled part fabrication using laser directed energy deposition[J].Journal of Manufacturing Processes,2022,78:32-45.

    • [105] MOZAFFAR M,PAUL A,AL-BAHRANI R,et al.Data-driven prediction of the high-dimensional thermal history in directed energy deposition processes via recurrent neural networks[J].Manufacturing Letters,2018,18:35-39.

    • [106] REN K,CHEW Y,ZHANG Y F,et al.Thermal field prediction for laser scanning paths in laser aided additive manufacturing by physics-based machine learning[J].Computer Methods in Applied Mechanics and Engineering,2020,362:112734.

    • [107] ZHU X,JIANG F,GUO C,et al.Prediction of melt pool shape in additive manufacturing based on machine learning methods[J].Optics & Laser Technology,2023,159:108964.

    • [108] LI X,DAI R,CHEN S,et al.Offline planning optimization and formation prediction of laser directed energy deposition process[J].Optics & Laser Technology,2023,164:109510.

    • [109] GARCÍA-MORENO A I,ALVARADO-OROZCO J M,IBARRA-MEDINA J,et al.Ex-situ porosity classification in metallic components by laser metal deposition:a machine learning-based approach[J].Journal of Manufacturing Processes,2021,62:523-534.

    • [110] WANG Y,HU K,LI W,et al.Prediction of melt pool width and layer height for laser directed energy deposition enabled by physics-driven temporal convolutional network[J].Journal of Manufacturing Systems,2023,69:1-17.

    • [111] ZHU Q,LIU Z,YAN J.Machine learning for metal additive manufacturing:predicting temperature and melt pool fluid dynamics using physics-informed neural networks[J].Computational Mechanics,2021,67(2):619-635.

    • [112] TANG Y,RAHMANI D M,WANG G G.Review of transfer learning in modeling additive manufacturing processes[J].Additive Manufacturing,2023,61:103357.

    • [113] TAHERI H,KOESTER L W,BIGELOW T A,et al.In situ additive manufacturing process monitoring with an acoustic technique:clustering performance evaluation using K-means algorithm[J].Journal of Manufacturing Science and Engineering,2019,141(4):041011.

    • [114] OUIDADI H,GUO S,ZAMIELA C,et al.Real-time defect detection using online learning for laser metal deposition[J].Journal of Manufacturing Processes,2023,99:898-910.

    • [115] KHANZADEH M,CHOWDHURY S,TSCHOPP M A,et al.In-situ monitoring of melt pool images for porosity prediction in directed energy deposition processes[J].IISE Transactions,2019,51(5):437-455.

    • [116] REN W,WEN G,ZHANG Z,et al.Quality monitoring in additive manufacturing using emission spectroscopy and unsupervised deep learning[J].Materials and Manufacturing Processes,2022,37(11):1339-1346.

    • [117] YADAV P,SINGH V K,JOFFRE T,et al.Inline drift detection using monitoring systems and machine learning in selective laser melting[J].Advanced Engineering Materials,2020,22(12):2000660.

    • [118] PANDIYAN V,CUI D,LE-QUANG T,et al.In situ quality monitoring in direct energy deposition process using co-axial process zone imaging and deep contrastive learning[J].Journal of Manufacturing Processes,2022,81:1064-1075.

    • [119] YUAN B,GIERA B,GUSS G,et al.Semi-supervised convolutional neural networks for in-situ video monitoring of selective laser melting[C]//2019 IEEE winter conference on applications of computer vision(WACV),January 07-11,2019,Waikoloa,USA.New York:IEEE,2019:744-753.

    • [120] JAFARI-MARANDI R,KHANZADEH M,TIAN W,et al.From in-situ monitoring toward high-throughput process control:cost-driven decision-making framework for laser-based additive manufacturing[J].Journal of Manufacturing Systems,2019,51:29-41.

  • 参考文献

    • [1] SIDDIQUI A A,DUBEY A K.Recent trends in laser cladding and surface alloying[J].Optics & Laser Technology,2021,134:106619.

    • [2] WENG F,CHEN C,YU H.Research status of laser cladding on titanium and its alloys:A review[J].Materials & Design,2014,58:412-425.

    • [3] LIU J,YU H,CHEN C,et al.Research and development status of laser cladding on magnesium alloys:A review[J].Optics and Lasers in Engineering,2017,93:195-210.

    • [4] 高亚丽,路鹏勇,刘宇,等.镁合金表面激光熔覆研究现状[J].中国表面工程,2023,36(3):22-39.GAO Yali,LU Pengyong,LIU Yu,et al.Research status of laser cladding on magnesium alloy[J].China Surface Engineering,2023,36(3):22-39.(in Chinese)

    • [5] YANG Q,ZHANG P,LU Q,et al.Application and development of blue and green laser in industrial manufacturing:A review[J].Optics & Laser Technology,2024,170:110202.

    • [6] ZHANG H,LIU Y,BAI X,et al.Laser cladding highly corrosion-resistant nano/submicron ultrafine-grained Fe-based composite layers[J].Surface and Coatings Technology,2021,424:127636.

    • [7] ARIF Z U,KHALID M Y,UR REHMAN E,et al.A review on laser cladding of high-entropy alloys,their recent trends and potential applications[J].Journal of Manufacturing Processes,2021,68:225-273.

    • [8] ZHANG Q,WANG Q,HAN B,et al.Comparative studies on microstructure and properties of CoCrFeMnNi high entropy alloy coatings fabricated by high-speed laser cladding and normal laser cladding[J].Journal of Alloys and Compounds,2023,947:169517.

    • [9] ARIF Z U,KHALID M Y,AL RASHID A,et al.Laser deposition of high-entropy alloys:A comprehensive review[J].Optics & Laser Technology,2022,145:107447.

    • [10] HALDAR B,SAHA P.Identifying defects and problems in laser cladding and suggestions of some remedies for the same[J].Materials Today:Proceedings,2018,5(5):90-101.

    • [11] 曹佳俊,常成,邱兆国,等.AISI1045 钢表面激光熔覆 FeCoCrNiAl0.5Ti0.5 涂层的界面特性及摩擦性能[J].中国表面工程,2023,36(2):54-64.CAO Jiajun,CHANG Cheng,QIU Zhaoguo,et al.Interface characteristics and tribological properties of laser cladded FeCoCrNiAl0.5Ti0.5 coating on AISI 1045 steel[J] China Surface Engineering,2023,36(2):54-64.(in Chinese)

    • [12] OCELÍK V,EEKMA M,HEMMATI I,et al.Elimination of start/stop defects in laser cladding[J].Surface and Coatings Technology,2012,206(8):2403-2409.

    • [13] LI R,FENG A,ZHAO J,et al.Study on process optimization of WC-Ni60A cermet composite coating by laser cladding[J].Materials Today Communications,2023,37:107400.

    • [14] LI G,WANG Z,YAO L,et al.Concentration mixing and melt pool solidification behavior during the magnetic field assisted laser cladding of Fe-Cr-based alloy on 45 steel surface[J].Surface and Coatings Technology,2022,445:128732.

    • [15] SONG B,YU T,JIANG X,et al.Evolution and convection mechanism of the melt pool formed by V-groove laser cladding[J].Optics & Laser Technology,2021,144:107443.

    • [16] 何志远,贺文雄,杨海峰,等.铝合金表面激光熔覆研究进展[J].中国表面工程,2021,34(6):33-44.HE Zhiyuan,HE Wenxiong,YANG Haifeng,et al.Research progess in laser cladding on aluminum alloy surface[J].China Surface Engineering,2021,34(6):33-44.(in Chinese)

    • [17] 郭星星,帅美荣,王建梅,等.基于 NSGA-II 算法的激光熔覆单道成形工艺参数多目标优化[J].中国表面工程,2023,36(3):87-100.GUO Xingxing,SHUAI Meirong,WANG Jianmei,et al.Multi-objective optimization of laser cladding single-pass forming process parameters based on NSGA-II algorithm[J].China Surface Engineering,2023,36(3):87-100.(in Chinese)

    • [18] THAWARI N,GULLIPALLI C,CHANDAK A,et al.Influence of laser cladding parameters on distortion,thermal history and melt pool behaviour in multi-layer deposition of stellite 6:in-situ measurement[J].Journal of Alloys and Compounds,2021,860:157894.

    • [19] FALLAH V,ALIMARDANI M,CORBIN S F,et al.Temporal development of melt-pool morphology and clad geometry in laser powder deposition[J].Computational Materials Science,2011,50(7):2124-2134.

    • [20] WIRTH F,ARPAGAUS S,WEGENER K.Analysis of melt pool dynamics in laser cladding and direct metal deposition by automated high-speed camera image evaluation[J].Additive Manufacturing,2018,21:369-382.

    • [21] ZHANG Y M,LIM C W J,TANG C,et al.Numerical investigation on heat transfer of melt pool and clad generation in directed energy deposition of stainless steel[J].International Journal of Thermal Sciences,2021,165:106954.

    • [22] LI Y,XU F.Acoustic emission sources localization of laser cladding metallic panels using improved fruit fly optimization algorithm-based independent variational mode decomposition[J].Mechanical Systems and Signal Processing,2022,166:108514.

    • [23] 郭永明,叶福兴,祁航.超高速激光熔覆技术研究现状及发展趋势[J].中国表面工程,2022,35(6):39-50.GUO Yongming,YE Fuxing,QI Hang,et al.Research status and development of ultra-high speed laser cladding[J].China Surface Engineering,2022,35(6):39-50.(in Chinese)

    • [24] JINLONG W,YUXIN M,WENJIE P,et al.Evaluation of the effect of surface roughness parameters on fatigue of TC17 titanium alloy impeller using machine learning algorithm and finite element analysis[J].Engineering Failure Analysis,2023,153:107586.

    • [25] HAO W Q,TAN L,YANG X G,et al.A physics-informed machine learning approach for notch fatigue evaluation of alloys used in aerospace[J].International Journal of Fatigue,2023,170:107536.

    • [26] ZHAO S,SUN H,PENG F,et al.Feature fusion and distillation embedded sparse Bayesian learning model for in-situ foreknowledge of robotic machining errors[J].Journal of Manufacturing Systems,2023,71:546-564.

    • [27] LEE J A,SAGONG M J,JUNG J,et al.Explainable machine learning for understanding and predicting geometry and defect types in Fe-Ni alloys fabricated by laser metal deposition additive manufacturing[J].Journal of Materials Research and Technology,2023,22:413-423.

    • [28] SVETLIZKY D,DAS M,ZHENG B,et al.Directed energy deposition(DED)additive manufacturing:physical characteristics,defects,challenges and applications[J].Materials Today,2021,49:271-295.

    • [29] ZHOU L,MA G,ZHAO H,et al.Research status and prospect of extreme high-speed laser cladding technology[J].Optics & Laser Technology,2024,168:109800.

    • [30] LIN X,ZHU K,FUH J Y H,et al.Metal-based additive manufacturing condition monitoring methods:from measurement to control[J].ISA Transactions,2022,120:147-166.

    • [31] CAI Y,XIONG J,CHEN H,et al.A review of in-situ monitoring and process control system in metal-based laser additive manufacturing[J].Journal of Manufacturing Systems,2023,70:309-326.

    • [32] FU Y,DOWNEY A R J,YUAN L,et al.Machine learning algorithms for defect detection in metal laser-based additive manufacturing:A review[J].Journal of Manufacturing Processes,2022,75:693-710.

    • [33] WANG C,TAN X P,TOR S B,et al.Machine learning in additive manufacturing:state-of-the-art and perspectives[J].Additive Manufacturing,2020,36:101538.

    • [34] QIN L,WANG K,LI X,et al.Review of the formation mechanisms and control methods of geometrical defects in laser deposition manufacturing[J].Chinese Journal of Mechanical Engineering:Additive Manufacturing Frontiers,2022,1(4):100052.

    • [35] ZHAO S,YUAN K,GUO W,et al.A comparative study of laser metal deposited and forged Ti-6Al-4V alloy:uniaxial mechanical response and vibration fatigue properties[J].International Journal of Fatigue,2020,136:105629.

    • [36] WAN H,WANG Q,JIA C,et al.Multi-scale damage mechanics method for fatigue life prediction of additive manufacture structures of Ti-6Al-4V[J].Materials Science and Engineering:A,2016,669:269-278.

    • [37] LI J,CHENG X,LI Z,et al.Improving the mechanical properties of Al-5Si-1Cu-Mg aluminum alloy produced by laser additive manufacturing with post-process heat treatments[J].Materials Science and Engineering:A,2018,735:408-417.

    • [38] YU X,LIN X,TAN H,et al.Microstructure and fatigue crack growth behavior of Inconel 718 superalloy manufactured by laser directed energy deposition[J].International Journal of Fatigue,2021,143:106005.

    • [39] STERLING A J,TORRIES B,SHAMSAEI N,et al.Fatigue behavior and failure mechanisms of direct laser deposited Ti-6Al-4V[J].Materials Science and Engineering:A,2016,655:100-112.

    • [40] LEUNG C L A,MARUSSI S,ATWOOD R C,et al.In situ X-ray imaging of defect and molten pool dynamics in laser additive manufacturing[J].Nature Communications,2018,9(1):1355.

    • [41] LI S,CHEN B,TAN C,et al.In situ identification of laser directed energy deposition condition based on acoustic emission[J].Optics & Laser Technology,2024,169:110152.

    • [42] SONG B,YU T,JIANG X,et al.The relationship between convection mechanism and solidification structure of the iron-based molten pool in metal laser direct deposition[J].International Journal of Mechanical Sciences,2020,165:105207.

    • [43] KHAIRALLAH S A,ANDERSON A T,RUBENCHIK A,et al.Laser powder-bed fusion additive manufacturing:physics of complex melt flow and formation mechanisms of pores,spatter,and denudation zones[J].Acta Materialia,2016,108:36-45.

    • [44] YANG Z,WANG A,WENG Z,et al.Porosity elimination and heat treatment of diode laser-clad homogeneous coating on cast aluminum-copper alloy[J].Surface and Coatings Technology,2017,321:26-35.

    • [45] 操龙飞.金属材料的热膨胀特性研究[D].武汉:武汉科技大学,2013.CAO Longfei.Study on thermal expansion properties of steels[D].Wuhan:Wuhan University of Science and Technology,2013.(in Chinese)

    • [46] 李春彦,张松,康煜平,等.综述激光熔覆材料的若干问题[J].激光杂志,2002,3:5-9.LI Chunyan,ZHANG Song,KANG Yuping,et al.Comment on material system for laser cladding[J].Laser Journal,2002,3:5-9.(in Chinese)

    • [47] ZHOU S,ZENG X,HU Q,et al.Analysis of crack behavior for Ni-based WC composite coatings by laser cladding and crack-free realization[J].Applied Surface Science,2008,255(5):1646-1653.

    • [48] 申发明.超高速激光熔覆AISI431不锈钢涂层组织与耐蚀机理研究[D].哈尔滨:哈尔滨工业大学,2022.SHEN Faming.Research on microstructure and corrosion resistance mechanism of AISI431 stainless steel coating prepared by extra high speed laser cladding[D].Harbin:Harbin Institute of Technology,2022.(in Chinese)

    • [49] AUCOTT L,HUANG D,DONG H B,et al.A three-stage mechanistic model for solidification cracking during welding of steel[J].Metallurgical and Materials Transactions A,2018,49(5):1674-1682.

    • [50] GAO Z,WANG L,WANG Y,et al.Crack defects and formation mechanism of FeCoCrNi high entropy alloy coating on TC4 titanium alloy prepared by laser cladding[J].Journal of Alloys and Compounds,2022,903:163905.

    • [51] JIN K,YANG Z,CHEN P,et al.Dynamic solidification process during laser cladding of IN718:multi-physics model,solute suppressed nucleation and microstructure evolution[J].International Journal of Heat and Mass Transfer,2022,192:122907.

    • [52] SCHWERZ C,BIRCHER B A,KÜNG A,et al.In-situ detection of stochastic spatter-driven lack of fusion:application of optical tomography and validation via ex-situ X-ray computed tomography[J].Additive Manufacturing,2023,72:103631.

    • [53] XU X,DU J L,LUO K Y,et al.Microstructural features and corrosion behavior of Fe-based coatings prepared by an integrated process of extreme high-speed laser additive manufacturing[J].Surface and Coatings Technology,2021,422:127500.

    • [54] FONSECA E B,GABRIEL A H G,ARAÚJO L C,et al.Assessment of laser power and scan speed influence on microstructural features and consolidation of AISI H13 tool steel processed by additive manufacturing[J].Additive Manufacturing,2020,34:101250.

    • [55] 王豫跃,牛强,杨冠军,等.超高速激光熔覆技术绿色制造耐蚀抗磨涂层[J].材料研究与应用,2019,13(3):165-172.WANG Yuyue,NIU Qiang,YANG Guanjun,et al.lnvestigations on corrosion-resistant and wear-resistant coatings environmental-friendly manufactured by a novel super-high efficient laser cladding[J].Materials Research and Application,2019,13(3):165-172.(in Chinese)

    • [56] ZHANG Y,GAO X,LIANG X,et al.Effect of laser remelting on the microstructure and corrosion property of the arc-sprayed AlFeNbNi coatings[J].Surface and Coatings Technology,2020,398:126099.

    • [57] 黄旭,张家诚,练国富,等.超高速激光熔覆研究现状及应用[J].机床与液压,2021,49(6):151-155,162.HUANG Xu,ZHANG Jiacheng,LIAN Guofu,et al.Research status and application of extreme high speed cladding[J].Machine Tool & Hydraulics,2021,49(6):151-155,162.(in Chinese)

    • [58] YE X,WANG J,YING Q,et al.Melting behavior of in-flight particles in ultra-high speed laser cladding[J].Journal of Materials Research and Technology,2023,24:7047-7057.

    • [59] WANG W,ZHANG Y,YUE C,et al.Processing defect,microstructure evolution and mechanical properties of laser powder bed fusion Al-12Si alloys[J].Journal of Materials Research and Technology,2023,26:681-696.

    • [60] WU Z,XU Z,FAN W.Online detection of powder spatters in the additive manufacturing process[J].Measurement,2022,194:111040.

    • [61] YE D,HONG G S,ZHANG Y,et al.Defect detection in selective laser melting technology by acoustic signals with deep belief networks[J].The International Journal of Advanced Manufacturing Technology,2018,96(5):2791-2801.

    • [62] CHEN L,YAO X,XU P,et al.Rapid surface defect identification for additive manufacturing with in-situ point cloud processing and machine learning[J].Virtual and Physical Prototyping,2021,16(1):50-67.

    • [63] EVERTON S K,HIRSCH M,STRAVROULAKIS P,et al.Review of in-situ process monitoring and in-situ metrology for metal additive manufacturing[J].Materials & Design,2016,95:431-445.

    • [64] QIN L,ZHAO D,WANG W,et al.Geometric defects identification and deviation compensation in laser deposition manufacturing[J].Optics & Laser Technology,2022,155:108374.

    • [65] LI P,WARNER D H,FATEMI A,et al.Critical assessment of the fatigue performance of additively manufactured Ti-6Al-4V and perspective for future research[J].International Journal of Fatigue,2016,85:130-143.

    • [66] KAJI F,NGUYEN-HUU H,BUDHWANI A,et al.A deep-learning-based in-situ surface anomaly detection methodology for laser directed energy deposition via powder feeding[J].Journal of Manufacturing Processes,2022,81:624-637.

    • [67] MAZZARISI M,ERRICO V,ANGELASTRO A,et al.Influence of standoff distance and laser defocusing distance on direct laser metal deposition of a nickel-based superalloy[J].The International Journal of Advanced Manufacturing Technology,2022,120(3):2407-2428.

    • [68] LI K,LI T,MA M,et al.Laser cladding state recognition and crack defect diagnosis by acoustic emission signal and neural network[J].Optics & Laser Technology,2021,142:107161.

    • [69] CHEN L,YAO X,TAN C,et al.In-situ crack and keyhole pore detection in laser directed energy deposition through acoustic signal and deep learning[J].Additive Manufacturing,2023,69:103547.

    • [70] GAJA H,LIOU F.Defects monitoring of laser metal deposition using acoustic emission sensor[J].The International Journal of Advanced Manufacturing Technology,2017,90(1):561-574.

    • [71] LI Y,XIAO W,XIAO H,et al.Enhanced molten-pool boundary stability for microstructure control using quasi-continuous-wave laser additive manufacturing[J].Journal of Materials Research and Technology,2023,23:238-244.

    • [72] ASSELIN M,TOYSERKANI E,IRAVANITABRIZIPOUR M,et al.Development of trinocular CCD-based optical detector for real-time monitoring of laser cladding[C]//Mechatronics & Automation.IEEE International Conference,July 29-August 01,2005,Niagara Falls,ON,Canada.New York:IEEE,2005:1190-1196.

    • [73] HOJJATZADEH S M H,PARAB N D,GUO Q,et al.Direct observation of pore formation mechanisms during LPBF additive manufacturing process and high energy density laser welding[J].International Journal of Machine Tools and Manufacture,2020,153:103555.

    • [74] ZHANG K,LIU T,LIAO W,et al.Photodiode data collection and processing of molten pool of alumina parts produced through selective laser melting[J].Optik,2018,156:487-497.

    • [75] MUVVALA G,KARMAKAR D P,NATH A K.Online assessment of TiC decomposition in laser cladding of metal matrix composite coating[J].Materials & Design,2017,121:310-320.

    • [76] MISRA S,MOHANTY I,RAZA M S,et al.Investigation of IR pyrometer-captured thermal signatures and their role on microstructural evolution and properties of Inconel 625 tracks in DED-based additive manufacturing[J].Surface and Coatings Technology,2022,447:128818.

    • [77] MAZZARISI M,ANGELASTRO A,LATTE M,et al.Thermal monitoring of laser metal deposition strategies using infrared thermography[J].Journal of Manufacturing Processes,2023,85:594-611.

    • [78] D’ACCARDI E,CHIAPPINI F,GIANNASI A,et al.Online monitoring of direct laser metal deposition process by means of infrared thermography[J].Progress in Additive Manufacturing,2023,26:1-19.

    • [79] MAFFIA S,FURLAN V,PREVITALI B.Coaxial and synchronous monitoring of molten pool height,area,and temperature in laser metal deposition[J].Optics & Laser Technology,2023,163:109395.

    • [80] BI G,SCHÜRMANN B,GASSER A,et al.Development and qualification of a novel laser-cladding head with integrated sensors[J].International Journal of Machine Tools and Manufacture,2007,47(3):555-561.

    • [81] CHEN L,BI G,YAO X,et al.Multisensor fusion-based digital twin for localized quality prediction in robotic laser-directed energy deposition[J].Robotics and Computer-Integrated Manufacturing,2023,84:102581.

    • [82] GERON A.Hands-on machine learning with scikit-learn,keras,and tensorflow:concepts,tools,and techniques to build intelligent systems[M].California:O’ Reilly Media,Inc.,2019.

    • [83] SOOFI A,AWAN A.Classification techniques in machine learning:applications and issues[J].Journal of Basic & Applied Sciences,2017,13:459-465.

    • [84] KOTSIANTIS S B,ZAHARAKIS I D,PINTELAS P E.Machine learning:a review of classification and combining techniques[J].Artificial Intelligence Review,2006,26(3):159-190.

    • [85] GAJA H,LIOU F.Defect classification of laser metal deposition using logistic regression and artificial neural networks for pattern recognition[J].The International Journal of Advanced Manufacturing Technology,2018,94(1):315-326.

    • [86] DANG L,HE X,TANG D,et al.A fatigue life posterior analysis approach for laser-directed energy deposition Ti-6Al-4V alloy based on pore-induced failures by kernel ridge[J].Engineering Fracture Mechanics,2023,289:109433.

    • [87] KHANZADEH M,CHOWDHURY S,MARUFUZZAMAN M,et al.Porosity prediction:supervised-learning of thermal history for direct laser deposition[J].Journal of Manufacturing Systems,2018,47:69-82.

    • [88] LEE H,HEOGH W,YANG J,et al.Deep learning for in-situ powder stream fault detection in directed energy deposition process[J].Journal of Manufacturing Systems,2022,62:575-587.

    • [89] SEIFI S H,TIAN W,DOUDE H,et al.Layer-wise modeling and anomaly detection for laser-based additive manufacturing[J].Journal of Manufacturing Science and Engineering,2019,141(8):081013.

    • [90] CHEN T,WU W,LI W,et al.Laser cladding of nanoparticle TiC ceramic powder:effects of process parameters on the quality characteristics of the coatings and its prediction model[J].Optics & Laser Technology,2019,116:345-355.

    • [91] HAO J,YANG S,LE X,et al.Bead morphology prediction of coaxial laser cladding on inclined substrate using machine learning[J].Journal of Manufacturing Processes,2023,98:159-172.

    • [92] DANG L,HE X,TANG D,et al.A fatigue life prediction approach for laser-directed energy deposition titanium alloys by using support vector regression based on pore-induced failures[J].International Journal of Fatigue,2022,159:106748.

    • [93] FEENSTRA D R,MOLOTNIKOV A,BIRBILIS N.Utilisation of artificial neural networks to rationalise processing windows in directed energy deposition applications[J].Materials & Design,2021,198:109342.

    • [94] BHARDWAJ T,SHUKLA M.Laser additive manufacturing-direct energy deposition of ti-15mo biomedical alloy:artificial neural network based modeling of track dilution[J].Lasers in Manufacturing and Materials Processing,2020,7(3):245-258.

    • [95] LI J,SAGE M,GUAN X,et al.Machine Learning-enabled competitive grain growth behavior study in directed energy deposition fabricated Ti6Al4V[J].JOM,2020,72(1):458-464.

    • [96] CIAMPAGLIA A,TRIDELLO A,PAOLINO D S,et al.Data driven method for predicting the effect of process parameters on the fatigue response of additive manufactured AlSi10Mg parts[J].International Journal of Fatigue,2023,170:107500.

    • [97] GONZALEZ-VAL C,PALLAS A,PANADEIRO V,et al.A convolutional approach to quality monitoring for laser manufacturing[J].Journal of Intelligent Manufacturing,2020,31(3):789-795.

    • [98] ZHANG B,LIU S,SHIN Y C.In-process monitoring of porosity during laser additive manufacturing process[J].Additive Manufacturing,2019,28:497-505.

    • [99] TIAN Q,GUO S,MELDER E,et al.Deep learning-based data fusion method for in situ porosity detection in laser-based additive manufacturing[J].Journal of Manufacturing Science and Engineering,2020,143:1-38.

    • [100] HOSSAIN M S,TAHERI H.In-situ process monitoring for metal additive manufacturing through acoustic techniques using wavelet and convolutional neural network(CNN)[J].The International Journal of Advanced Manufacturing Technology,2021,116(11):3473-2488.

    • [101] PERANI M,BARALDO S,DECKER M,et al.Track geometry prediction for laser metal deposition based on on-line artificial vision and deep neural networks[J].Robotics and Computer-Integrated Manufacturing,2023,79:102445.

    • [102] FRANCIS J,BIAN L.Deep Learning for distortion prediction in laser-based additive manufacturing using big data[J].Manufacturing Letters,2019,20:10-14.

    • [103] XIE X,BENNETT J,SAHA S,et al.Mechanistic data-driven prediction of as-built mechanical properties in metal additive manufacturing[J].npj Computational Materials,2021,7(1):86-97.

    • [104] HU K,WANG Y,LI W,et al.CNN-BiLSTM enabled prediction on molten pool width for thin-walled part fabrication using laser directed energy deposition[J].Journal of Manufacturing Processes,2022,78:32-45.

    • [105] MOZAFFAR M,PAUL A,AL-BAHRANI R,et al.Data-driven prediction of the high-dimensional thermal history in directed energy deposition processes via recurrent neural networks[J].Manufacturing Letters,2018,18:35-39.

    • [106] REN K,CHEW Y,ZHANG Y F,et al.Thermal field prediction for laser scanning paths in laser aided additive manufacturing by physics-based machine learning[J].Computer Methods in Applied Mechanics and Engineering,2020,362:112734.

    • [107] ZHU X,JIANG F,GUO C,et al.Prediction of melt pool shape in additive manufacturing based on machine learning methods[J].Optics & Laser Technology,2023,159:108964.

    • [108] LI X,DAI R,CHEN S,et al.Offline planning optimization and formation prediction of laser directed energy deposition process[J].Optics & Laser Technology,2023,164:109510.

    • [109] GARCÍA-MORENO A I,ALVARADO-OROZCO J M,IBARRA-MEDINA J,et al.Ex-situ porosity classification in metallic components by laser metal deposition:a machine learning-based approach[J].Journal of Manufacturing Processes,2021,62:523-534.

    • [110] WANG Y,HU K,LI W,et al.Prediction of melt pool width and layer height for laser directed energy deposition enabled by physics-driven temporal convolutional network[J].Journal of Manufacturing Systems,2023,69:1-17.

    • [111] ZHU Q,LIU Z,YAN J.Machine learning for metal additive manufacturing:predicting temperature and melt pool fluid dynamics using physics-informed neural networks[J].Computational Mechanics,2021,67(2):619-635.

    • [112] TANG Y,RAHMANI D M,WANG G G.Review of transfer learning in modeling additive manufacturing processes[J].Additive Manufacturing,2023,61:103357.

    • [113] TAHERI H,KOESTER L W,BIGELOW T A,et al.In situ additive manufacturing process monitoring with an acoustic technique:clustering performance evaluation using K-means algorithm[J].Journal of Manufacturing Science and Engineering,2019,141(4):041011.

    • [114] OUIDADI H,GUO S,ZAMIELA C,et al.Real-time defect detection using online learning for laser metal deposition[J].Journal of Manufacturing Processes,2023,99:898-910.

    • [115] KHANZADEH M,CHOWDHURY S,TSCHOPP M A,et al.In-situ monitoring of melt pool images for porosity prediction in directed energy deposition processes[J].IISE Transactions,2019,51(5):437-455.

    • [116] REN W,WEN G,ZHANG Z,et al.Quality monitoring in additive manufacturing using emission spectroscopy and unsupervised deep learning[J].Materials and Manufacturing Processes,2022,37(11):1339-1346.

    • [117] YADAV P,SINGH V K,JOFFRE T,et al.Inline drift detection using monitoring systems and machine learning in selective laser melting[J].Advanced Engineering Materials,2020,22(12):2000660.

    • [118] PANDIYAN V,CUI D,LE-QUANG T,et al.In situ quality monitoring in direct energy deposition process using co-axial process zone imaging and deep contrastive learning[J].Journal of Manufacturing Processes,2022,81:1064-1075.

    • [119] YUAN B,GIERA B,GUSS G,et al.Semi-supervised convolutional neural networks for in-situ video monitoring of selective laser melting[C]//2019 IEEE winter conference on applications of computer vision(WACV),January 07-11,2019,Waikoloa,USA.New York:IEEE,2019:744-753.

    • [120] JAFARI-MARANDI R,KHANZADEH M,TIAN W,et al.From in-situ monitoring toward high-throughput process control:cost-driven decision-making framework for laser-based additive manufacturing[J].Journal of Manufacturing Systems,2019,51:29-41.

  • 手机扫一扫看