文献类型：期刊文献

英文题名：Multiobjective Path Optimization for Arc Welding Robot Based on DMOEA/D-ET Algorithm and Proxy Model

作者：Wang, Xuewu[1];Xia, Zelong[1];Zhou, Xin[1];Guo, Yinan[2];Gu, Xingsheng[1];Yan, Huaicheng[1,3]

机构：[1]East China Univ Sci & Technol, Key Lab Smart Mfg Energy Chem Proc, Minist Educ, Shanghai 200237, Peoples R China;[2]China Univ Min & Technol, Sch Electromech & Informat Engn, Beijing 100084, Peoples R China;[3]Chengdu Univ, Sch Informat Sci & Engn, Chengdu 610106, Peoples R China

年份：2021

卷号：70

外文期刊名：IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT

收录：;EI(收录号：20211110072813);WOS:【SCI-EXPANDED(收录号:WOS:000749872900010)】；

基金：This work was supported by the National Natural Science Foundation of China under Grant 62076095, Grant 61973120, and Grant 61973305. The Associate Editor coordinating the review process was Shutao Li.

语种：英文

外文关键词：Adaptive strategy; arc welding robot; decomposition multiobjective evolutionary algorithm based on the event triggering (DMOEA/D-ET); event triggering strategy; path optimization; welding deformation

摘要：In the arc welding process, a reasonable welding path can improve welding efficiency and quality, especially when a large number of welding seams exist. Hence, this article introduces an intelligent path optimization strategy to optimize the sequence of welding seams. The shortest path length, power consumption, and welding deformation are considered optimization objectives, and welding deformation is studied based on the proxy model to improve optimization efficiency. Then, the improved discrete multiobjective decomposition algorithm based on event-triggering strategy (DMOEA/D-ET) is proposed. Here, the grid method and decomposition multiobjective algorithm (MOEA/D) are used to improve search performance. The adaptive neighborhood strategy is used to improve the quality and distribution of noninferior solutions. After the comparison with NSGA-II, NSGA-III, and CGMOPSO algorithms, the DMOEA/D-ET algorithm shows better performance on both convergence and diversity. Finally, the proposed strategy is used for welding robot path planning, and the simulation results show its effectiveness.