Research On Defect Detection Method Of High Voltage Cable Based On Dual Frequency Eddy Current

As a new eddy current non-destructive testing technology,Dual-Frequency Eddy Current Testing(DFECT)has the advantages of realizing interference suppression in detection process and multi-parameter detection of detected objects.In this paper,the DFECT is applied to the defect detection of high-voltage cables.The detection methods,experimental data processing,electromagnetic field simulation analysis,defect detection of multi-core copper cable and single-core aluminum cable,defect detection GUI program design are studied.Main research contents are as follows:Based on the principle of traditional eddy current testing,a new detection method is proposed to solve the existing problems of the complex hardware circuit.The calculation method of penetration depth in eddy current testing and the numerical calculation theory of sinusoidal electromagnetic field are discussed.Some data processing methods in the detection are summarized,including wavelet denoising,spectrum analysis,filtering processing,and polynomial fitting,which lays the theoretical foundation for the experimental data processing in the following chapters.Based on Pro/E software and COMSOL Multiphysics software,the process of simulation analysis of eddy current current testing of high-voltage cable is discussed in detail.Under the conditions of AC frequency,current value,lift-off value,number of coil turns and coil structure,the distribution law of the induced eddy currents,the distribution law of the nearby magnetic field when the excitation coil is at a certain position,and the distribution law of the x-axis,y-axis and z-axis magnetic field signals collected by the TRM sensor when the excitation coil is scanning along the pipeline.The influence of AC frequency,current value,lift-off value,number of coil turns and coil structure on eddy current detection signal of high-voltage cable is simulated based on finite element method,which provides some references for the design of highvoltage cable defect detection experiment.The DFECT scheme for multi-core copper cable with a defect is designed.The distribution laws of x-axis,y-axis and z-axis magnetic field signals under this parameter are summarized,and the defect of the multi-core copper cable is identified by the abnormality of the magnetic field signals,verifying the feasibility and correctness of the proposed new detection method.The spectrum analysis and polynomial fitting method are used to process the data under the parameters of low frequency 20 Hz,high frequency 1000 Hz,and supply voltage 10 V.The rusults show that the processed data can better identify the defects.By changing high-frequency parameter,multiple experiments are performed on single-core aluminum cables with two defects.The effects of high-frequency parameter on the three axes magnetic field signals along the scanning trajectory and the effect of defect identification are studied and the distribution law of three axes magnetic field signals are summarized.According to the ideal degree of defect detection,several sets of detection parameters are preferred.Wavelet analysis,filtering processing and other methods are used to post-process the x-axis magnetic field data under these sets of detection parameters.Based on MATLAB software,the GUI program of DFECT system is designed.A small defect is made and then detected on a single-core aluminum cable,and the collected signals of the TMR sensors are post-processed using the GUI program.The location and number of defects are identified.Some explanations are given for the setting of the defect detection threshold.A method based on neural networt and polynomial fitting to determine the threshold size is proposed.