Research On Detection Method Of Weld Defects In Different Structures Based On Guided Wave

As a high incidence of defects in the pipeline structure,the weld has a threat to the safe operation of the project.In order to ensure the safe operation of the pipeline,periodic inspection is required.In this thesis,,the identification,quantitative and location of weld defects of double weld line pipe,single weld bend pipe and double weld cross section pipe were studied based on the guided wave technique through the combination of Abaqus numerical simulation and experiment.The symmetric asymmetric modal separation algorithm and the time reversal wave field focusing method are applied in this study.The main contents and conclusions are as follows:(1)The symmetrical and asymmetric modal separation algorithm was applied to identify the weld defects according to the asymmetry of the wave packet.The numerical results show that there is no significant difference in the influence of the reflection coefficient between the internal and external defects of the pipe weld under the same section defect rate.According to the quantity relation between the reflected coefficient of weld defect and the defect rate of the cross section,the defect quantitation is realized.(2)The time reversal wave field focusing method was applied to locate the weld defect.The numerical results show that compared with pre-time reversal,the time reversal method effectively improves the recognition rate of the defects in bend pipe and the front weld defects of double weld pipe,and realizes the circumferential location of the rear weld defects in double weld pipe.(3)An experimental research on weld defect detection of double weld straight pipe was also carried out.Experiments were implemented on the pipeline with different degrees of defect at 20°,45°,60°,90°,120°,and 150°,respectively,and four sets of frequency tests were performed on the defects of the same size.The experimental results show that the weld defect with a cross-section defect rate of great than 8%can be identified based on the symmetric asymmetric modal separation algorithm.