摘要
以裂纹、气孔、夹渣、未熔合和未焊透五类不锈钢焊缝缺陷为研究对象,通过Gabor小波变换对焊缝缺陷的超声信号进行特征提取,利用粒子群优化算法(Particle Swarm Optimization,PSO)对支持向量机(Support Vector Machine,SVM)优化后进行焊缝超声缺陷的识别,提高了对焊缝缺陷的分类精度。经过实验,证明所用模型对裂纹、气孔、夹渣、未熔合和未焊透五类缺陷的分类精度达到了96.36%。相比现有的方法,本方法识别面更广,精度更高,具有一定的工程价值。
Taking cracks,pores,slag inclusions,incomplete fusion and incomplete penetration of five types of stainless steel welded seam defects as the research objects,the ultrasonic signals of weld defects are extracted by Gabor wavelet transform.Particle Swarm Optimization(PSO)is used to optimize the Support Vector Machine(SVM)to identify the ultrasonic defects of the welded seam,which improves the classification accuracy of welded seam defects.After experiments,the model used in this paper has a classification accuracy of 96.36%for the five types of defects:cracks,pores,slag inclusions,incomplete fusion and incomplete penetration.Compared with the existing methods,this method has wider recognition area and higher accuracy,and has certain engineering value.
作者
傅留虎
赵娜
闫耀东
张睿
柴钰杰
李宇琪
FU Liu-hu;ZHAO Na;YAN Yao-dong;ZHANG Rui;CHAI Yu-jie;LI Yu-qi(College of Computer Science and Technology,Taiyuan University of Science and Technology,Taiyuan 030024,China;Shanxi Design and Research Institute of Mechanical and Electrical Engineering Co.,Ltd.,Taiyuan 030009,China;College of Information and Technology,Taiyuan University of Science and Technology,Jincheng 048011,China)
出处
《机械工程与自动化》
2020年第5期12-14,共3页
Mechanical Engineering & Automation
基金
山西省应用基础研究基金项目(201801D221179)
先进控制与装备智能化山西省重点实验室开放课题(ACEI202002)
山西省高等学校科技创新项目(2019L0653)
2019年教育部产学合作协同育人项目(201902016003)
2019年度研究生教育改革研究课题项目(2019JG171)
太原科技大学教学改革与研究项目(201937)
太原科技大学校级大学生创新创业训练项目(XJ2019072,XJ2020159)。
