摘要
本文基于BP神经网络优良的逼近拟合特性,将BP神经网络算法引进电机壳体加工工艺过程中,使用BP神经网络模型预测电机壳体拉伸成形质量,结合电机壳体拉伸成形实验以验证BP神经网络模拟的准确性,并结合实验结果优选出电机壳体拉伸成形的最佳工艺参数,并得到最优的毛坯形状尺寸,实验结果表明电机壳体拉深成形效果良好,进一步验证了参数优化结果的可靠性。
Based on the excellent approximation fitting characteristics of BP neural network, the BP neural network algorithm to the processing process of motor shell was introduced. The quality of the motor shell was predicted by using the BP neural network model. Combining with the tensile test of motor shell, the accuracy of BP neural network simulation was verified. Combining the experimental results, the technological parameters of the motor shell drawing was optimized. The optimum shape size of blank was obtained. The experimental results show that the effect of the deep drawing of the motor shell was good, and the reliability of the parameter optimization results was verified by the progress.
作者
杨树财
王贤良
杨松涛
YANG Shu-cai;WANG Xian-liang;YANG Song-tao(Mechanical ＆ Power Engineering College,Harbin Univ.Sci.Tech.,Harbin 150080)
出处
《航空精密制造技术》
2018年第5期19-22,共4页
Aviation Precision Manufacturing Technology
基金
黑龙江省教育厅科学技术研究项目(11541048)
关键词
BP神经网络
拉伸成形工艺
优选工艺参数
神经网络模型
BP neural network
stretch forming process
optimization of process parameter
neural network model
