{"title":"基于变形和残余应力的焊接顺序优化多目标遗传算法","authors":"Jesus Romero-Hdz, G. Toledo-Ramirez, B. Saha","doi":"10.1109/MICAI-2016.2016.00021","DOIUrl":null,"url":null,"abstract":"Compared to deformation, residual stress has not been taken into account in the literature when it comes to welding process optimization. It also plays an important role to measure the weld quality. This paper reports the implementation of a multi-objective based Genetic Algorithm (GA) for welding sequence optimization, in which both structural deformation and residual stress are offered equal importance. The optimal weights between them are dynamically selected through optimizing a multi-objective fitness function in an iterative manner. A thermomechanical finite element analysis (FEA) was used to predict both deformation and residual stress. We chose the elitism selection approach to ensure that the three best individuals are copied over once into the next generation to facilitate convergence by preserving good candidates which can offer an optimal solution. We exploited a sequential string searching algorithm into single point crossover method to avoid the repetition of single beads into the sequence. We utilized a bit string mutation operator by changing the direction of the welding from one bead chosen randomly from the sequence. Welding simulation experiments were conducted on a typical widely used mounting bracket which has eight seams. Multi-objective based GA effectively reduces the computational complexity over exhaustive search with significant reduction of both structural deformation (~80%) and residual stress (~15%)","PeriodicalId":405503,"journal":{"name":"2016 Fifteenth Mexican International Conference on Artificial Intelligence (MICAI)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":"{\"title\":\"Deformation and Residual Stress Based Multi-Objective Genetic Algorithm for Welding Sequence Optimization\",\"authors\":\"Jesus Romero-Hdz, G. Toledo-Ramirez, B. Saha\",\"doi\":\"10.1109/MICAI-2016.2016.00021\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Compared to deformation, residual stress has not been taken into account in the literature when it comes to welding process optimization. It also plays an important role to measure the weld quality. This paper reports the implementation of a multi-objective based Genetic Algorithm (GA) for welding sequence optimization, in which both structural deformation and residual stress are offered equal importance. The optimal weights between them are dynamically selected through optimizing a multi-objective fitness function in an iterative manner. A thermomechanical finite element analysis (FEA) was used to predict both deformation and residual stress. We chose the elitism selection approach to ensure that the three best individuals are copied over once into the next generation to facilitate convergence by preserving good candidates which can offer an optimal solution. We exploited a sequential string searching algorithm into single point crossover method to avoid the repetition of single beads into the sequence. We utilized a bit string mutation operator by changing the direction of the welding from one bead chosen randomly from the sequence. Welding simulation experiments were conducted on a typical widely used mounting bracket which has eight seams. Multi-objective based GA effectively reduces the computational complexity over exhaustive search with significant reduction of both structural deformation (~80%) and residual stress (~15%)\",\"PeriodicalId\":405503,\"journal\":{\"name\":\"2016 Fifteenth Mexican International Conference on Artificial Intelligence (MICAI)\",\"volume\":\"62 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2016-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"5\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2016 Fifteenth Mexican International Conference on Artificial Intelligence (MICAI)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/MICAI-2016.2016.00021\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 Fifteenth Mexican International Conference on Artificial Intelligence (MICAI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MICAI-2016.2016.00021","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Deformation and Residual Stress Based Multi-Objective Genetic Algorithm for Welding Sequence Optimization
Compared to deformation, residual stress has not been taken into account in the literature when it comes to welding process optimization. It also plays an important role to measure the weld quality. This paper reports the implementation of a multi-objective based Genetic Algorithm (GA) for welding sequence optimization, in which both structural deformation and residual stress are offered equal importance. The optimal weights between them are dynamically selected through optimizing a multi-objective fitness function in an iterative manner. A thermomechanical finite element analysis (FEA) was used to predict both deformation and residual stress. We chose the elitism selection approach to ensure that the three best individuals are copied over once into the next generation to facilitate convergence by preserving good candidates which can offer an optimal solution. We exploited a sequential string searching algorithm into single point crossover method to avoid the repetition of single beads into the sequence. We utilized a bit string mutation operator by changing the direction of the welding from one bead chosen randomly from the sequence. Welding simulation experiments were conducted on a typical widely used mounting bracket which has eight seams. Multi-objective based GA effectively reduces the computational complexity over exhaustive search with significant reduction of both structural deformation (~80%) and residual stress (~15%)