Post Weld Heat Treatment Optimization of Dissimilar Friction Stir Welded AA2024-T3 and AA7075-T651 Using Machine Learning and Metaheuristics

被引:12
作者
Insua, Pinmanee [1 ,2 ]
Nakkiew, Wasawat [2 ,3 ]
Wisittipanich, Warisa [2 ]
机构
[1] Chiang Mai Univ, Fac Engn, Grad Program Ind Engn, Chiang Mai 50200, Thailand
[2] Chiang Mai Univ, Fac Engn, Dept Ind Engn, Chiang Mai 50200, Thailand
[3] Chiang Mai Univ, Fac Engn, Adv Mfg & Management Technol Res Ctr AM2Tech, Dept Ind Engn, Chiang Mai 50200, Thailand
关键词
dissimilar friction stir welding; post weld heat treatment; machine learning; metaheuristics; process optimization; ultimate tensile strength; elongation percentage; MECHANICAL-PROPERTIES; ALUMINUM-ALLOY; MICROSTRUCTURE; JOINTS; STRENGTH; 2024-T4;
D O I
10.3390/ma16052081
中图分类号
O64 [物理化学(理论化学)、化学物理学];
学科分类号
070304 ; 081704 ;
摘要
Post weld heat treatment, or PWHT, is often used to improve the mechanical properties of materials that have been welded. Several publications have investigated the effects of the PWHT process using experimental designs. However, the modeling and optimization using the integration of machine learning (ML) and metaheuristics have yet to be reported, which are fundamental steps toward intelligent manufacturing applications. This research proposes a novel approach using ML techniques and metaheuristics to optimize PWHT process parameters. The goal is to determine the optimal PWHT parameters for both single and multiple objective perspectives. In this research, support vector regression (SVR), K-nearest neighbors (KNN), decision tree (DT), and random forest (RF) were ML techniques employed to obtain a relationship model between PWHT parameters and mechanical properties: ultimate tensile strength (UTS) and elongation percentage (EL). The results show that the SVR demonstrated superior performance among ML techniques for both UTS and EL models. Then, SVR is used with metaheuristics such as differential evolution (DE), particle swarm optimization (PSO), and genetic algorithms (GA). SVR-PSO shows the fastest convergence among other combinations. The final solutions of single-objective and Pareto solutions were also suggested in this research.
引用
收藏
页数:17
相关论文
共 50 条
[1]   Modelling of soil moisture retention curve using machine learning techniques: Artificial and deep neural networks vs support vector regression models [J].
Achieng, Kevin O. .
COMPUTERS & GEOSCIENCES, 2019, 133
[2]   Optimization of friction stir welding process parameters using soft computing techniques [J].
Alkayem, Nizar Faisal ;
Parida, Biswajit ;
Pal, Sukhomay .
SOFT COMPUTING, 2017, 21 (23) :7083-7098
[3]   Experimental investigation of kerf of Ti6Al4V exposed to different heat treatment processes in WEDM and optimization of parameters using genetic algorithm [J].
Altug, Mehmet ;
Erdem, Mehmet ;
Ozay, Cetin .
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY, 2015, 78 (9-12) :1573-1583
[4]  
[Anonymous], 2009, P 10 ACM SIGEVO WORK
[5]   CASE-STUDIES IN HIGH-DIMENSIONAL CLASSIFICATION [J].
APTE, C ;
SASISEKHARAN, R ;
SESHADRI, V ;
WEISS, SM .
APPLIED INTELLIGENCE, 1994, 4 (03) :269-281
[6]   The effect of post-weld heat treatment on the mechanical properties of 2024-T4 friction stir-welded joints [J].
Aydin, Hakan ;
Bayram, Ali ;
Durgun, Ismail .
MATERIALS & DESIGN, 2010, 31 (05) :2568-2577
[7]  
Bader J., 2010, Hypervolume-based search for multi-objective optimization: theory and methods
[8]  
Bajic B, 2018, Machine learning techniques for smart manufacturing: applications and challenges in industry 4.0, P29
[9]  
Chang C.C., 2010, ADV MATER RES-SWITZ
[10]   Adaptive control optimization in micro-milling of hardened steels-evaluation of optimization approaches [J].
Coppel, Ricardo ;
Abellan-Nebot, Jose V. ;
Siller, Hector R. ;
Rodriguez, Ciro A. ;
Guedea, Federico .
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY, 2016, 84 (9-12) :2219-2238