Model-Based Inversion of Rayleigh Wave Dispersion Curves Via Linear and Nonlinear Methods

Abstract

The use of Rayleigh wave dispersion curve for estimating shear wave velocity (Vs) is a common task in the field of engineering geophysics. However, because of the nonlinear nature of Rayleigh wave dispersion curves, using a proper technique in the inversion procedure in order to find adequate model parameters is a challenging problem. In this study, a comparative study is performed between linear and nonlinear inversion methods. Between different techniques, a new nonlinear searching algorithm, i.e., the modified cuckoo search (MCS) algorithm was introduced to invert the Rayleigh wave dispersion curves. Also, the generalized inverse (GI) method was used as a linear searching algorithm applied to the construction of the Vs profile. The proposed algorithms were tested by different synthetic (noisy and noise free) datasets. The results show that the MCS algorithm has more exploration ability in comparison with the original cuckoo search. Then, an actual dataset was inverted by the introduced inversion methods. The results show that the performance of the MCS is better than that of the GI method. MCS is a fast, powerful and user-friendly technique in the field of dispersion curve inversion.