摘要
针对铝合金变极性等离子弧焊特点,建立了背面小孔视觉传感系统,实现小孔图像的实时采集和处理,并对小孔行为与焊缝成形关系开展了工艺研究,开发了基于小孔熔池与熔透状态的极限学习机预测模型,模型预测精度达到96.7%.研究表明,该模型能够较为可靠地预测变极性等离子弧焊的熔透状态,最终实现焊接质量实时监控的目标.
In the variable polarity plasma arc welding,a simple-flexible vision system was established to acquire and process the keyhole images in real-time,and the process study on the relationship between the keyhole behavior and weld formation was conducted.A novel extreme learning machine(ELM)prediction model was proposed based on keyhole puddle and penetration state,the prediction accuracy of the ELM model is up to 96.6%.The research shows that ELM model can predict the penetration state of variable polarity plasma arc welding credibly and achieve real time monitoring for welding quality.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2016年第S1期79-82,共4页
Journal of Shanghai Jiaotong University
基金
国家自然科学基金面上项目(51275301)
关键词
变极性等离子弧焊
小孔特征
熔透预测模型
极限学习机
variable polarity plasma arc welding
keyhole characteristics
penetration prediction model
extreme learning machine
