Research On Signal Recognition Technology Of Ground Penetrating Radar About Subgrade Disease

Road collapse events caused by roadbed diseases often occur.Periodic nondestructive testing and evaluation of roadbeds can effectively identify roadbed diseases at an early stage and provide necessary technical information for timely treatment to prevent road collapse accidents.Ground penetrating radar has achieved good application results in roadbed based disease detection,but the response characteristics of ground penetrating radar in the early stage of roadbed disease development are not easy to identify.In this paper,theoretical analysis,numerical simulation and physical model experiments are used to identify the radar signals of typical roadbed diseases such as roadbed looseness,voiding,roadbed voids,soil water enrichment and roadbridge transitional section voids,which are subgrade diseases.Early governance provides the necessary geological data.In the process of ground penetrating radar signal recognition and processing,firstly,the wavelet transform threshold denoising method is used to filter out the original radar signal,and the signal-to-noise ratio of the radar signal is improved,the signal can reflect the distribution of the formation well and improve the accuracy of data interpretation of ground penetrating radar.Secondly,based on the finite difference time domain algorithm,the roadbed disease model of roadbed looseness,void,subgrade void,soil water enrichment and road-bridge transition section is established,and Hilbert transform is performed on the ground penetrating radar forward signal.The instantaneous amplitude profile and the instantaneous phase profile can well identify the size of the anomalous body and its spatial position.Thirdly,for the same type of subgrade diseases with different developmental degrees and different dielectric constants and subgrades,the differential sensing analysis of ground penetrating radar signals in data processing and image recognition process shows that the development of subgrade diseases is higher.The rate of change of the differential curve is stronger.Fourthly,the spectrum characteristics of radar signal of roadbed roadbed disease are studied by ARMA power spectrum analysis.The results show that the ARMA power spectrum can highlight the recognition characteristics of high-frequency components of the signal compared with the amplitude spectrum,and the smaller the sliding average spectrum is,The smoother the curve,the more prominent the main frequency parameters of the signal;the higher the order,the more the spectrum value of the weak signal can be reflected,and the moving average of the signal is closer to the true value.Through the ARMA spectrum analysis of the anomalous area signal,the ability of the ground penetrating radar signal to identify the medium type of the roadbed disease target can be improved.Finally,through the physical model experiment method,the radar signal response characteristics of roadbed looseness,void,subgrade void and soil water-rich disease model are obtained.At the same time,the radar response characteristics are identified by the above method,and the interpretation accuracy of the measured data is improved.The effectiveness of the method for detecting groundbased radar ground penetrating radar signals in this paper is further verified.