Abstract
The development and utilization of urban underground space depend heavily on an understanding of urban geological conditions. The microtremor survey method is essential in urban geological surveys due to its quickness, convenience, non-destructiveness, and interference resistance. Since only the fundamental dispersion curves of Rayleigh waves can be obtained by utilizing the spatial autocorrelation method, the inversion results have multiple solutions. To improve the accuracy of the microtremor survey, this study employed the frequency-Bessel transform to extract the fundamental and higher modes of dispersion information of Rayleigh waves from the microtremor data array and verified the effectiveness of this method by synthesizing theoretical microtremor signals. Additionally, taking into account the order identification challenges brought on by mode jumps or missing modes in the dispersion curve, this study processed a multi-mode dispersion curve based on the newly proposed inversion objective function coupled with a genetic algorithm to obtain a shallow surface S-wave velocity structure. Compared to the traditional inversion objective function, the new function presented in this study could address mode misidentification more effectively and improved the accuracy of inversion calculations. Finally, the applicability and dependability of the frequency-Bessel-transform-based microtremor survey method were evaluated in a practical engineering case.