A comparative analysis of particle swarm optimization (PSO) and very fast simulated annealing (VFSA) inversion techniques for self‐potential (SP) anomalies

Abstract

This paper compares the particle swarm optimization (PSO) and very fast simulated annealing (VFSA), which are both non-linear optimization techniques, results for selfpotential (SP) anomalies due to electrochemical nature. PSO is a simple global optimization strategy that simulates the social behavior observed in a flock (swarm) of birds searching for food. VFSA is based on the application of a non-linear optimization algorithm. These two methods have been tested over several theoretical models and field data. The paper uses only one general equation for the inversion of SP anomalies due to vertical cylinder, horizontal cylinder and sphere. The comparison of the results over horizontal cylindrical and spherical targets and three field data from Turkey, India and Germany are presented in this paper. The results show that both PSO and VFSA inversion schemes give similar results. However, tuning a VFSA algorithm was difficult as compared to PSO. Further, PSO contains merit of both local and global search algorithms.