Eddy Current Detection And Identification Of Defect Types For Thin-walled Stainless Steel Welded Pipe Welds Based On COMSOL Simulation

Thin-walled stainless steel welded pipes are widely used in nuclear power generation,hydropower generation,automobiles,oil transportation and other industries.Welded pipe non-destructive testing is one of the important processes in welded pipe production.Accurate,fast and effective identification of welded pipe defects has attracted more and more attention from scholars,and various non-destructive testing methods have been applied to the defect detection of welded pipes.Eddy current testing,as an important nondestructive testing method,is widely used in shallow surface flaw detection of metal components,so it is suitable for nondestructive testing of thin-walled stainless steel welded pipes in this study.The defects of welded pipes are concentrated in the welding seam.The causes of different defects are different,and the damage to the pipeline is also different.The commonly used eddy current impedance detection method focuses on the detection of defects,and it is difficult to achieve qualitative identification of defect types.judge.Achieving accurate identification of defect types will be helpful for product quality control and failure analysis of thin-walled stainless steel welded pipes.This subject identifies and analyzes the defects of thin-walled stainless steel welded pipes from three aspects: theory,simulation and experiment.The first chapter introduces the research significance of weld defects and type identification and the research status at home and abroad,and introduces the technical characteristics and development trends of eddy current testing;the second chapter introduces the principle of eddy current testing for welded pipes,using an equivalent circuit The impedance of the eddy current coil is analyzed by the method of this paper,and the classification and selection of the eddy current detection coil is introduced.In the third chapter,the finite element simulation model of the thin-walled stainless steel welded pipe based on COMSOL is established,and the best eddy current probe parameters are selected for the defect.Eddy current testing test;Chapter 4 introduces the feature extraction of eddy current signal defect based on EMD-HHT.Obtain the complete time-frequency feature matrix for different defect types.The fifth chapter introduces the defect type recognition based on PCA-RBF neural network.PCA dimensionality reduction was performed on the time-frequency characteristic matrix of weld defects of thin-walled stainless steel welded pipes,two RBF neural networks were used to identify the types of weld defects of welded pipes,and three error evaluation indexes were selected to evaluate the classification model.The results of this study show the validity and reliability of the method for identifying the defect types of thin-walled stainless steel welded pipes proposed in this paper,and provide a feasible method for the identification of welded pipe defect types.