毛恺奕,杜彦斌,何国华,彭云川,李志强

• 1:重庆工商大学制造装备机构设计与控制重庆市重点实验室

摘要(Abstract)：

激光熔覆多道成形层形貌受激光熔覆过程中多个工艺参数的综合影响，为获得良好的熔覆层形貌，提出了一种基于灰狼优化(GWO)算法优化随机森林回归(RFR)算法(GWO-RFR)的激光熔覆多道成形层形貌预测方法。以12Cr13不锈钢为基体，Fe60为熔覆粉末，设计试错法结合中心复合实验，测量成形层宽高比和稀释率。基于多道激光熔覆实验数据，建立激光熔覆工艺参数与成形层形貌间的GWO-RFR回归预测模型，并与RFR模型、响应面模型(RSM)的预测结果进行比较。结果表明：与RFR模型和RSM模型相比，GWO-RFR模型的预测结果和评价指标均优于RFR模型和RSM模型，GWO-RFR预测模型能够更准确地预测熔覆层形貌，更接近实际值，可为获得优异的激光熔覆多道成形层形貌提供理论依据。

关键词(KeyWords)： 激光熔覆;形貌;灰狼优化算法;随机森林回归算法

Abstract:

Keywords:

基金项目(Foundation): 重庆市自然科学基金联合基金重点项目(CSTB2022NSCQ-LZX0011);; 重庆市高校创新研究群体资助项目(CXQT21024);; 重庆英才计划“包干制项目”(cstc2022ycjh-bgzxm0056);重庆英才计划(CQYC20210302226)

作者(Author): 毛恺奕,杜彦斌,何国华,彭云川,李志强

参考文献(References)：

• [1] 李方义，李振，王黎明，等.内燃机增材再制造修复技术综述[J].中国机械工程，2019,30(9):1119-1127.LI Fang-yi,LI Zhen,WANG Li-ming,et al.Review on ICE remanufacture with additive repair technology[J].China Mechanical Engineering,2019,30(9):1119-1127.
• [2] 陈妮，韦佳伟，张鑫磊，等.激光改性后氮化铝基板的可加工性研究[J].中国机械工程，2021,32(23):2817-2822.CHEN Ni,WEI Jia-wei,ZHANG Xin-lei,et al.Research on machinability of laser modified AlN substrates[J].China Mechanical Engineering,2021,32(23):2817-2822.
• [3] 练国富，姚明浦，陈昌荣，等.激光熔覆多道搭接成形质量与效率控制方法[J].表面技术，2018,47(9):229-239.LIAN Guo-fu,YAO Ming-pu,CHEN Chang-rong,et al.Control of the quality and efficiency of multi-track overlapping laser cladding[J].Surface Technology,2018,47(9):229-239.
• [4] 高玉龙，马国梁，高晓华，等.激光熔覆CoCrNiMnTix高熵合金涂层组织及耐磨性能研究[J].表面技术，2022,51(9):351-358.GAO Yu-long,MA Guo-liang,GAO Xiao-hua,et al.Microstructure and wear resistance of CoCrNiMnTix high-entropy alloy coating by laser cladding[J].Surface Technology,2022,51(9):351-358.
• [5] 练国富，曹强，郑颖，等.激光熔覆多道搭接成形及粉末利用率[J].材料热处理学报，2023,44(2):127-139.LIAN Guo-fu,CAO Qiang,ZHENG Ying,et al.Multi-channel lap forming and powder utilization ratio of laser cladding[J].Transactions of Materials and Heat Treatment,2023,44(2):127-139.
• [6] 张天刚，孙荣禄.TC4表面单道与多道搭接激光熔覆自润滑涂层微观组织对比研究[J].材料导报，2017,31(4):47-51.ZHANG Tian-gang,SUN Rong-lu.Comparative study of self-lubricating coatings microstructure on TC4 alloy surface prepared by single-pass laser cladding and multitrack laser cladding[J].Materials Review,2017,31(4):47-51.
• [7] 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 Xing-xing,SHUAI Mei-rong,WANG Jian-mei,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.
• [8] 李家豪，舒林森，衡钊，等.基于PCA与RSM-DE算法的激光熔覆参数多目标优化[J].焊接学报，2023,44(2):67-73.LI Jia-hao,SHU Lin-sen,HENG Zhao,et al.Multi-objective optimization of laser cladding parameters based on PCA and RSM-DE algorithm[J].Transactions of the China Welding Institution,2023,44(2):67-73.
• [9] 张杰，张群莉，姚建华，等.激光熔覆IN718合金工艺优化及界面组织性能分析[J].中国激光，2022,49(16):204-213.ZHANG Jie,ZHANG Qun-li,YAO Jian-hua,et al.Process optimization and interfacial microstructure and properties analysis of laser cladded IN718 alloy[J].Chinese Journal of Lasers,2022,49(16):204-213.
• [10] 吴嘉祾，祁文军，王旭祥，等.基于不同算法的BP神经网络预测激光熔覆层几何特征与性能[J].应用激光，2022,42(10):41-52.WU Jia-ling,QI Wen-jun,WANG Xu-xiang,et al.Prediction on geometrical characteristics and properties of cladding layer based on BP network improved by different algorithms[J].Applied Laser,2022,42(10):41-52.
• [11] 梁万旭，杨勇，金康，等.基于响应面法的同轴送粉多道激光熔覆层形貌预测[J].激光与光电子学进展，2022,59(1):1-11.LIANG Wan-xu,YANG Yong,JIN Kang,et al.Morphology prediction of coaxial powder feeding multichannel laser cladding layer based on response surface[J].Laser & Optoelectronics Progress,2022,59(1):1-11.
• [12] Meng G,Zhu L,Zhang J,et al.Statistical analysis and multi-objective process optimization of laser cladding TiC-Inconel718 composite coating[J].Optik,2021,240:166828.
• [13] 施晓帅，姜芙林，王玉玲，等.基于响应面法的激光熔覆3540Fe涂层形貌及质量预测研究[J].表面技术，2022,51(12):392-405.SHI Xiao-shuai,JIANG Fu-lin,WANG Yu-ling,et al.Morphology and quality prediction of laser cladding 3540Fe coating based on response surface method[J].Surface Technology,2022,51(12):392-405.
• [14] 王永菲，王成国.响应面法的理论与应用[J].中央民族大学学报(自然科学版),2005(3):236-240.WANG Yong-fei,WANG Cheng-guo.The application of response surface methodology[J].Journal of the CUN(Natural Science Edition),2005(3):236-240.
• [15] 李时春，莫彬，肖罡，等.激光增材制造成形质量表征与调控研究进展[J].材料导报，2022,36(10):141-149.LI Shi-chun,MO Bin,XIAO Gang,et al.Research progress of forming quality and its characterization and regulation in laser additive manufacturing[J].Materials Reports,2022,36(10):141-149.
• [16] 杨思瑞，白海清，鲍骏，等.基于回归分析和遗传算法优化的BP神经网络熔覆层形貌预测[J].激光与光电子学进展，2022,59(21):149-157.YANG Si-rui,BAI Hai-qing,BAO Jun,et al.Prediction of cladding layer morphology based on BP neural network optimized by regression analysis and genetic algorithm[J].Laser & Optoelectronics Progress,2022,59(21):149-157.
• [17] 梁旭东，王炜，赵凯，等.随机森林回归分析在激光熔覆形貌预测中的应用[J].中国有色金属学报，2020,30(7):1644-1652.LIANG Xu-dong,WANG Wei,ZHAO Kai,et al.Application of random forest regression analysis in trace geometry prediction of laser cladding[J].The Chinese Journal of Nonferrous Metals,2020,30(7):1644-1652.
• [18] Yaseer A,Chen H.Machine learning based layer roughness modeling in robotic additive manufacturing[J].Journal of Manufacturing Processes,2021,70:543-552.
• [19] He G H,Du Y B,Liang Q,et al.Modeling and optimization method of laser cladding based on GA-ACO-RFR and GNSGA-II[J].International Journal of Precision Engineering and Manufacturing-Green Technology,2023,10:1207-1222.
• [20] 马斌，梁强，贾艳艳，等.基于BPNN、SVR和RF模型的7050合金高温流动应力预测[J].材料热处理学报，2023,44(3):196-204.MA Bin,LIANG Qiang,JIA Yan-yan,et al.Prediction of high temperature flow stress of 7050 aluminum alloy based on BPNN,SVR and RF models[J].Transactions of Materials and Heat Treatment,2023,44(3):196-204.
• [21] 杜彦斌，胡言峰，许磊，等.基于PSO-SVR的15-5PH合金粉末激光熔覆层稀释率预测方法[J].计算机集成制造系统，2022,28(7):2169-2178.DU Yan-bin,HU Yan-feng,XU Lei,et al.Prediction method for dilution rate of laser cladding layer of 15-5PH alloy powder based on PSO-SVR[J].Computer Integrated Manufacturing System,2022,28(7):2169-2178.
• [22] Breiman L.Random forests[J].Machine Learning,2001,45(1):5-32.
• [23] 尹儒，门昌骞，王文剑.一种模型决策森林算法[J].计算机科学与探索，2020,14(1):108-116.YIN Ru,MEN Chang-qian,WANG Wen-jian.Model decision forest algorithm[J].Journal of Fronticrs of Computer Science and Technology,2020,14(1):108-116.
• [24] Zhao B,Zhai H,Shao H,et al.Potential of vibrational spectroscopy coupled with machine learning as a non-invasive diagnostic method for COVID-19[J].Computer Methods and Programs in Biomedicine,2023,229:107295.
• [25] 庞祎帆，傅戈雁，王明雨，等.基于响应面法和遗传神经网络模型的高沉积率激光熔覆参数优化[J].中国激光，2021,48(6):152-162.PANG Yi-fan,FU Ge-yan,WANG Ming-yu,et al.Parameter optimization of high deposition rate laser cladding based on the response surface method and genetic neural network model[J].Chinese Journal of Lasers,2021,48(6):152-162.
• [26] 胡言峰，杜彦斌，许磊，等.响应面法与粒子群算法集成的激光熔覆工艺参数优化方法[J].机械科学与技术，2023,42(4):629-637.HU Yan-feng,DU Yan-bin,XU Lei,et al.Optimization method of processing parameters in laser cladding by integrating response surface methodology and particle swarm optimization[J].Mechanical Science and Technology for Aerospace Engineering,2023,42(4):629-637.

文章评论(Comment)：