Inversion of multi-channel surface wave data using a sequential hybrid approach

Abstract

A sequential hybrid scheme combining a genetic algorithm (GA) with the Levenberg-Marquardt (LM) method is proposed as a means to invert dispersion curves obtained via multi-channel analysis of surface waves for shear (S)-wave velocity profile. Such a hybrid optimization algorithm has the potential to exploit the advantages of both methods. In the hybrid scheme, the GA was implemented to construct an initial estimate for the LM method. The main objective of the paper was to compare the results obtained from the proposed hybrid scheme with those of both GA and LM methods. The accuracy of the proposed hybrid algorithm solution was validated through using both theoretical and field data. The inverted models produced by the hybrid approach fitted the true models well when the synthetic dispersion curve had no noise or with 5% noise. Furthermore, a very good agreement was obtained between field data obtained from borehole measurements and the inverted results produced by the proposed approach.