焊接
材料科学
差异进化
算法
电磁线圈
叠加原理
渡线
进化算法
遗传算法
机械工程
计算机科学
物理
数学分析
数学优化
复合材料
数学
人工智能
工程类
量子力学
作者
Sen Lin,Nengzhuo Chou,Yujia Zhao,Hao Jiang,Junjia Cui,Yong Cai,Guangyao Li,Yi Min Xie,Guoping Wang
标识
DOI:10.1016/j.jmrt.2023.06.129
摘要
A topologically constrained optimization methodology associated with a discretized differential evolution (DE) is employed for the structure design of a magnetic pulse welding (MPW) coil. To reveal the explicit expression for the effective welding area produced by the MPW process, the impact angle and collision velocity are innovatively considered for determination of the objective function. As several disconnected regions are generated by the Rand-to-Best/2/exp mutation strategy and subsequent superposition-style crossover operation, the algorithm is combined with geometrical correction to form a simply connected conductor. The proposed algorithm is effective and robust as it have been verified based on the evolutionary history and numerical trials with coupled electromagnetic–mechanical simulations. According to physical validation results, a wider effective welding area is produced between aluminum alloy and steel planar plates by the optimized coil, which results in better performance in a tensile test. Under a 23 kJ discharge current, the maximum loading is significantly enhanced by 46.21% compared with the original coil.
科研通智能强力驱动
Strongly Powered by AbleSci AI