Research On The Inversion Algorithm For Layered Medium Parameters In Ground Penetrating Radar

Ground penetrating radar is a rapidly developed non-destructive detection technology with high resolution,high efficiency and convenient and flexible use in recent years,and the use of ground penetrating radar to obtain layered medium parameter information has been widely used in airports,urban roads,highways,tunnels and water conservancy projects and other infrastructure.However,when the existing parameter inversion algorithms use waveform inversion to obtain layered media parameters,they generally have problems such as poor stability,low accuracy of inversion results and high time complexity,which cannot meet the needs of actual detection.Therefore,this paper focuses on the research on the inversion algorithm of stratiform medium parameters of ground penetrating radar with good stability,high accuracy of inversion results and low time complexity,and the main research contents are as follows:(1)According to the working principle of ground penetrating radar,the relevant theories of the layered medium parameter inversion algorithm are analyzed and studied,mainly including the basic steps of inversion,the forward model of electromagnetic wave propagation in the layered medium,the parameter representation and simplification of the layered medium model,the solution of the objective function,and finally the stochastic gradient descent method is used to invert the layered medium,and it is pointed out that the existing parameter inversion methods generally have problems such as low accuracy,poor stability and high time complexity,which paves the way for the in-depth study of the in-depth study of the layered medium parameter inversion algorithm.(2)Aiming at the problems of low accuracy and high time complexity of the general inversion results of the existing layered medium parameter inversion algorithms,a resampling particle swarm optimization(ARAWPSO)algorithm that automatically adjusts the inertia weight according to the different particle fitness is proposed for the inversion of the parameters of the layered medium.Firstly,the proportional selection method is used to resample the particle swarm optimization(PSO)algorithm to reduce invalid iterations.Then,the inertia weight,which is automatically adjusted according to the fitness value,is introduced into the resampling PSO algorithm,and the resampled particle swarm is traction to improve the optimization efficiency and accuracy of the algorithm.Through simulation experiments and measured experiments,it is verified that the ARAWPSO algorithm can improve the accuracy of the inversion results and reduce the time complexity of the algorithm.(3)In view of the interference of noise,clutter and other factors in the measured data,the research on ground penetrating radar data processing was carried out.Firstly,the preprocessing process of the actual echo data is analyzed,including data reading,bad sector removal,DC component removal and median filtering.Then,the transmitted signal of the radar is obtained through experiments and the round-trip time of the electromagnetic wave in each layer of the medium is calculated.Finally,based on MATLAB,a layered medium parameter inversion software platform is written to analyze the processing process of actual echo data,which has the advantages of flexible,efficient and accurate processing of actual echo data.