摘要
焊接参数的设置对电阻点焊质量有着至关重要的作用,将有限元数值模拟技术与BP神经网络及遗传算法相结合,对不锈钢薄板电阻点焊过程的工艺参数进行优化。对点焊过程进行有限元分析,将模拟结果作为网络的数据样本,以焊接电流,电极压力,焊接时间这三个主要的焊接工艺参数作为优化参数,以焊接熔核尺寸作为优化目标,建立了优化参数与目标函数之间的BP神经网络模型。结合遗传算法的全局寻优能力,获得使熔核尺寸最大的三大主要工艺参数的最优搭配。为电阻点焊工艺参数的选取提供了一条合理途径,对提高点焊质量有一定的意义。
Welding parameters setting is of great importance to the quality of resistance spot welding （ RSW）. The welding parameter optimization of stainless steel sheets is studied by finite element method （FEM） simulation technology, BP neural network and genetic algorithm （GA）. The spot welding process is analyzed by FEM and the simulation results are used as the network data sample. In the es- tablished BP neural network model, the three main welding process parameters including welding cur- rent, electrode pressure and welding time are design variables and weld nugget size is considered as the objective. With the global optimization ability of GA, the best parameter combination is obtained as the goal of getting the largest nugget. A reasonable method is offered for parameter selection of RSW process. To improve the quality of welding is also of certain significance.
出处
《组合机床与自动化加工技术》
北大核心
2013年第6期139-141,共3页
Modular Machine Tool & Automatic Manufacturing Technique
基金
国家自然科学基金项目(51175348)
广东省教育部产学研结合项目(2009B09300356)
深港创新圈计划项目(ZYB20097090135A)
关键词
电阻点焊
神经网络
遗传算法
数值模拟
参数优化
resistance spot welding （RSW）
neural network
genetic algorithm （GA）
numerical simulation
parameter optimization
