摘要
文中研究了微电阻点焊条件下TC2钛合金焊点质量监测问题.首先对典型焊点的微观金相组织进行了分析,发现在熔核区域形成了针状马氏体;接着选取典型的电极电压曲线,根据其变化趋势划分为四个阶段,结合焊点的形成过程进行了相应解释,并分析了电极电压曲线特征量与熔核直径之间的相关性;最后应用BP神经网络,采用相关性较强的电极电压特征量及焊接电流作为神经网络输入,选取熔核直径作为神经网络输出,结合试验数据.结果表明,通过神经网络训练及测试,有效地实现了对焊点质量的可靠预测.
The quality monitoring in small scale resistance spot welding of TC2 titanium alloy is conducted in this paper.The microstructure of a typical spot weld is analyzed. Needlelike martensite can be found in the weld nugget. The typical electrode voltage curve can be divided into four stages,and interpreted in combination with the nugget development. Correlation analysis is made between nugget diameter and features extracted from the electrode voltage curve. A back propagation neural network model is developed using extracted features and nugget diameter as network input and output,respectively. The results show that the quality monitoring can be finally achieved after model training and testing.
出处
《焊接学报》
EI
CAS
CSCD
北大核心
2017年第9期51-54,共4页
Transactions of The China Welding Institution
基金
国家自然科学基金资助项目(11072083)
关键词
钛合金
微电阻点焊
质量监测
电极电压信号
神经网络
titanium alloy
small scale resistance spot welding
quality monitoring
electrode voltage signal
neural network
