Robust oriented particle swarm optimization algorithm applied to inverse problems

This paper proposes a particle swarm optimization algorithm to find the robust solution of an inverse problem in which uncertainties or perturbations are inevitable. To reduce the heavy computational burden for expected fitness evaluations, a strategy to assign expected fitness only to the best solution searched by a particle in every iterative cycle is proposed, a repository is introduced to memory the searched history and then employed to determine the expected fitness value of a specific solution. The simplex method is used to find the worst case solution of a potential candidate to consider hard constraint functions. The numerical results serve to demonstrate the pros and cons of the proposed algorithm and the necessity to devote efforts in the development of robust oriented optimal algorithms.