Three-Dimensional S-Wave Velocity Structure Imaging Of Shallow Ground Surface Based On Frequency-Bessel Transform Method

With the accelerated modernization of human society accelerates and people’s demand for a better lifestyle grows,surface resources can no longer meet these demands.Therefore,it is also becoming more and more important to develop and utilize urban underground space.Especially the shallow surface layers within 100 meters underground,they can meet various needs in our daily life,including rail transportation,water engineering,cable communication engineering,oil and gas transmission and living entertainment.In recent years,in order to meet the geotechnical engineering needs of shallow surface,surface wave exploration methods have been widely used in detecting urban underground space structures.Compared with field exploration where human activities are less,the conventional active source surface wave exploration faces great challenges in densely populated cities and towns,where there is serious noise interference and the operation is required to be efficient and non-destructive without affecting the living order.Therefore,passive source surface wave exploration methods based on environmental noise have emerged,and they are widely used in shallow surface exploration.The noise signal received by the detector is the result of the joint action of the fundamental and higher-order surface waves,while the traditional background noise dispersion imaging method has limited resolution of the higher-order surface waves,and the obtained dispersion spectrum is easy to mix the surface waves of each order,which is not conducive to the extraction of higher-order dispersion curves.For complex geological structures,especially those containing soft and weak layers,the introduction of higher-order dispersion curves when inverting the transverse velocity structure of the subsurface medium can not only reduce the multiresolution of the inversion,but also significantly improve the convergence speed and accuracy of the inversion.For this purpose,we adopt the frequency-Bessel transform method(F-J Method)proposed by Wang to obtain higher-order dispersion curves with higher resolution and apply it to environmental noise data at shallow ground surface.In this paper,three dispersion imaging methods such as spatial autocorrelation method,inter-correlation phase shift method and frequency-Bessel transform method are introduced in detail,and the imaging effects of F-J method and phase shift method for extracting higher-order dispersion curves are compared and analyzed by simulation experiments to verify the feasibility of frequency-Bessel transform method for dispersion imaging of shallow ground surface waves.Based on this method,the effect of geometric parameters on the F-J method is further discussed in this paper through numerical simulations.Theoretically: the F-J method requires less arrangement of the observation system,and the linear,L-shaped,rectangular and randomly distributed observation systems can obtain clear dispersion curves of all orders,among which the dispersion imaging effect of the geophone randomly distributed observation system is better than other arranged observation systems,which makes the F-J method applicable to more site environments.The larger the deployment area of the observation system,the higher the resolution of the dispersion spectrum obtained by the F-J method.For the directional noise sources existing near the receiving array,the F-J method has good noise immunity and can also clearly calculate the dispersion curves of each order.In order to verify the reliability of the conclusions obtained from the numerical simulation experiments,the author applied the frequency-Bessel transform method to two sets of measured cases,both of which were able to extract the higher-order dispersion curves,and finally obtained the shallow surface 3D S-wave velocity structure by inversion.After comparing the calculated results of F-J method with Chen’s MAPS method,laminar imaging method and borehole data,it is found that the F-J method is closer to the logging data in the deep inversion.This indicates that,unlike the MAPS method,the F-J method is more effective in constraining the deep formation inversion and improving the accuracy of the inversion results due to the extraction of higher-order surface wave information.Based on the above theoretical simulations and the validation of the actual logging cases,it is demonstrated that the F-J method has a greater application prospect for imaging the 3D S-wave velocity structure at the shallow surface.