Correlation Processing Method For Steel Plate Pulsed Eddy Current Thickness Measurement Signals And Its Realization

Pulsed Eddy Current(Pulsed Eddy Current,PEC)technology is more and more widely used in the detection of wall thickness reduction of metal components due to its advantages such as no cladding removal,no shutdown,and rapid scanning detection.This dissertation is funded by the National Key R & D Program Projects "Research on Non-ferromagnetic Materials Pulsed Eddy Current and Micromagnetic Detection Technology and Instrument Development" and "Electromagnetic-Magnetic-Acoustic Compound of Tank Corrosion in Oil Detection Module Development".Thickness measurement processing methods of pulsed eddy current detection signals for ferromagnetic and ferromagnetic steel plates were studied.First,by building a pulse eddy current testing experimental platform to obtain the typical signal of steel plate pulse eddy current detection,combined with the characteristics of steel plate thickness mainly reflected in the characteristics of the pulse eddy current signal attenuation section,a piecewise cross-correlation processing method of steel plate PEC thickness measurement signal is proposed.The method can identify the signal attenuation process and realize the parameterized characterization of the PEC thickness measurement signal with variable basis function,which lays the foundation for the subsequent search for the characteristic quantity that characterizes the wall thickness of the specimen.Secondly,on the basis of comparative analysis using Laplace wavelet real part,Laplace wavelet imaginary part and unilateral exponential function as basis functions to process typical stainless steel PEC detection signals,select Laplace wavelet imaginary part as the basis function for analyzing stainless steel PEC detection signals;then based on The piecewise crosscorrelation method calculates the distribution of the correlation coefficient,characteristic frequency and damping ratio in the time domain of the 304 stainless steel plate PEC detection signal at different wall thicknesses and different lift heights,and then selects the average of the characteristic frequency with the correlation coefficient greater than a certain threshold The value is a characteristic quantity that characterizes the wall thickness of the non-ferromagnetic material specimen,and it is a power function relationship with the wall thickness of the specimen,which provides a new method for solving the problem of stainless steel plate thickness detection under large lift.Thirdly,in view of the characteristics of the noisy signal of the pulsed eddy current detection of the ferromagnetic steel plate,the double logarithm domain median filter with the order of 501 is selected to complete the original signal filtering process;the unilateral exponential function is used as the basis function to process the filtered signal to obtain correlation coefficient Method and index items change with time;according to the processing results,the average value of the index items whose correlation coefficient is greater than a certain threshold is selected as the characteristic quantity that characterizes the wall thickness of the ferromagnetic material specimen.The feature quantity and the wall thickness of the test piece have a good fitting effect when fitted by a power function,and are not affected by lift-off.Finally,complete the PEC testing software development.Based on the aforementioned research work,according to the requirements of pulse eddy current testing,the software is divided into four modules: parameter setting,data acquisition,algorithm and interface,and then the module implementation process based on Visual C ++ language and Windows platform is discussed in detail,and the module works Normal and stable,can meet the requirements of use.