Backtracking search algorithm for effective and efficient surface wave analysis

摘要: Abstract Surface wave dispersion analysis is widely used in geophysics to infer near-surface shear (S)-wave velocity profiles for a wide variety of applications. However, inversion surface data challenging most local-search methods due its high nonlinearity and multimodality. In this work, we proposed implemented new Rayleigh curve scheme based on backtracking search algorithm (BSA), novel powerful evolutionary (EA). Development BSA motivated by studies that attempt develop an possesses desirable features different optimization problems which include the ability reach problem's global minimum more quickly successfully with small number control parameters low computational cost, as well robustness ease application problem models. The inverse procedure applied nonlinear fundamental-mode curves S-wave profiles. To evaluate calculation efficiency effectiveness BSA, four noise-free noisy synthetic sets are firstly inverted. Then, performance compared genetic algorithms (GA) two sets. Finally, real-world example from waste disposal site NE Italy inverted examine applicability approach real data. Furthermore, against GA further scores BSA. Results both actual demonstrate should be considered good not only terms accuracy but also convergence speed. great advantages simple, robust easy implement. Also there fewer tune.