Introducing Particle Swarm Optimization (PSO) to Invert Refraction Seismic Data

Abstract

Summary Seismic refraction method is a powerful geophysical technique in near surface study. In order to achieve reliable results, processing of refraction seismic data in particular inversion stage should be done accurately. In this study refraction travel times inversion considered by Particle Swarm Optimization (PSO) algorithm. In this year PSO algorithm, as a metaheuristic optimization method, is used in many fields of studies. Application of this method in different geophysical data inversion shows that this algorithm is powerful, fast and easy. We developed PSO code in Matlab for inversion of seismic refraction data. For efficiency evaluation of developed code, different synthetic models are inverted. Finally PSO inversion code investigated in a case study at the part of Tabriz city in north-West of Iran for hazard assessment. Then field dataset inverted by PSO code. Obtained model was comprised by the geological information of study area. The results emphasize the reliability of the PSO code to invert refraction seismic data with an acceptable misfit and convergence speed.