论文部分内容阅读
采用连续光纤激光器对304不锈钢薄板进行了焊接工艺研究;为了提高激光焊接质量,引入BP神经网络对激光焊接工艺参数进行了优化,建立了焊接质量指标与焊接工艺参数之间的神经网络预测模型,并利用神经网络模型选择了较优的工艺参数。实验结果表明采用该工艺参数进行激光焊接可获得成形良好、无缺陷的焊缝。神经网络的性能预测指标与实际值间的偏差小于5%,可用于激光焊接工艺设计。
A continuous fiber laser was used to study the welding process of 304 stainless steel sheet. In order to improve the quality of laser welding, the BP neural network was introduced to optimize the parameters of the laser welding. The neural network prediction model between welding quality index and welding process parameters was established, The neural network model was used to select the optimal process parameters. The experimental results show that using the process parameters for laser welding can get a well-formed, weld-free weld. The deviation between the performance prediction index of neural network and the actual value is less than 5%, which can be used in laser welding process design.