Research On Pulsed Eddy Current Nondestructive Testing Technology

Pulsed eddy current testing technology does not belong to production technology,but it is of great significance. Based on the extensive literature，we summarized thedevelopment status of pulsed eddy current testing technology.After in-depth understanding of the working principle and technical features，wemade a thorough analysis of the skin effect and the lift-off effect, then did simulationexperiments using non-destructive testing software CIVA. The simulation resultsshowed that with the increase of the excitation frequency, the smaller the depth ofpenetration in the metal, the larger the surface eddy current density, provided areference for the latter part of the experiment. With the lift-off distance decreases, thegreater the density of the metal surface eddy current and eddy current distribution ismore concentrated, make it easier to detect defects fixed.On the basis of these studies, we designed a pulsed eddy current nondestructivetesting system. We optimized the design parameters of the excitation coil, and thendid simulation experiments to study the influence between the excitation coil core andinduced eddy current in metal using CIVA software. After comparing the differencebetween the wire coil and the Hall sensor, we determined the wire coil program andmanufactured the sensor probe finally. After comparing the different power amplifiersolutions, we chose solid state relays. Finally, NI data acquisition card and LabVIEWsoftware completed the signal acquisition and preservation.Signal processing and analysis is the focus of this paper. After using pulsed eddycurrent nondestructive testing systems for metal defect detection exper iments, we gettransient signal and one-time scan waveforms. However, the high frequency noise intransient signal and peak value scanning curve jitter make an adverse effect on the testresults. We take Matlab as a tool to study the application of wavelet thresholdde-noising in transient signals. Then we used wavelet multiscale analysis ideas, studythe application of trend extraction of wavelet in one-time scan waveforms.