摘要
为了更好地监控焊接周围区域温度从而提高搅拌摩擦焊焊接质量,利用遗传BP神经网络算法建立了前进方向温度预测模型和前进侧方向温度预测模型。以40组数据为测试样本,对所建两个模型的预测结果与实测值进行了分析研究,其平均相对误差分别为0.41%和0.73%,预测结果与实测值的相关系数为0.9727和0.9585。结果表明:两个模型预测精度高,对于搅拌摩擦焊接周围区域温度的研究有重要的理论参考价值。
In order to better monitor the temperature around the welding area and improve the quality of friction stir welding, the forward direction temperature prediction model and the forward side temperature prediction model were established by using genetic BP neural network algorithm. Taking 40 groups of data as test samples, the predicted values and measured values of the two models were analyzed. The average relative errors are 0.41% and 0.73%, respectively. The correlation coefficients between predicted results and measured values are 0.9727 and 0.9585. The results show that the two models have high prediction accuracy, which has important theoretical reference value for the study on temperature around the friction stir welding area.
作者
张喆
张永林
陈书锦
ZHANG Zhe;ZHANG Yonglin;CHEN Shujin(School of Electronic Information,Jiangsu University of Science and Technology,Zhenjiang 212000,China;School of Materials Science and Engineering,Jiangsu University of Science and Technology,Zhenjiang 212000,China)
出处
《热加工工艺》
北大核心
2020年第3期142-145,共4页
Hot Working Technology
基金
国家自然科学基金项目(51675248)
江苏省研究生科研与实践创新计划项目(SJCX18_0774).
关键词
BP神经网络
遗传算法
搅拌摩擦焊接
区域温度
BP neural network
genetic algorithm
friction stir welding
zone temperature
作者简介
张喆(1994-),男,江苏常州人,硕士,主要研究方向:光机电一体化,电话:18362883158,E-mail:zhangzhe2217@163.com;通讯作者:张永林(1972-),男,江苏泗阳人,副教授,博士,主要研究方向:机器人控制、工业自动化技术,电话:0511-84449658,E-mail:zhangyonglin@just.edu.cn
