一种镀锌板电阻点焊质量的无损评判方法  

One nondestructive evaluating method of galvanized sheet quality in resistance spot welding

作　　者：魏雷[1] 张鹏贤[1] 陈剑虹[1] 柳勤兵[1] 

机构地区：[1]兰州理工大学有色金属合金省部共建教育部重点实验室,甘肃兰州730050

出　　处：《电焊机》2009年第7期73-76,共4页Electric Welding Machine

基　　金：国家自然科学基金资助项目(50275028)

摘　　要：探索了一种以点焊表面数字图像为信息源的质量评判方法。获取镀锌板点焊接头表面的图像,通过对不同焊接规范下焊点的表面图像分析,发现焊点表面图像可以被分为几个不同的特征环区;研究焊点表面图像特征环区与接头质量的相关性,根据特征区域面积与焊点抗剪强度的相关性分析结果,选择了相关性显著的特征参数作为输入向量,接头的抗剪强度作为输出参量,建立了一种以剪切强度为评判指标的BP神经网络模型。试验结果表明:基于图像信息处理实现镀锌板焊点质量无损监测的方法切实可行。A new method was explored to monitoring joint quality based on information processing in digital image of welding spot surface in resistance spot welding.First of all, a few characteristic zones related to welding processing were mined from the image of welding spot surface through analyzing the image collected from the spot welding joint surface in different norms.Secondly though the correlation analysis between characteristic zones areas and tensile share strength of spot welded joint, a few characteristic parameters were selected as input vectors,the shear strength was selected as target vectors.A neural network model was set up to evaluate spot welded quality.The results showed that it is feasible that of galvanized sheet can be monitored based on image information of welding spot surface.

关 键 词：电阻点焊 图像处理 BP网络 

分 类 号：TG453.9[金属学及工艺—焊接]