Particle Swarm Optimization Method Based on Chaotic Local Search and Roulette Wheel Mechanism

Abstract

Combining the particle swarm optimization (PSO) technique with the chaotic local search (CLS) and roulette wheel mechanism (RWM), an efficient optimization method solving the constrained nonlinear optimization problems is presented in this paper. PSO can be viewed as the global optimizer while the CLS and RWM are employed for the local search. Thus, the possibility of exploring a global minimum in problems with many local optima is increased. The search will continue until a termination criterion is satisfied. Benefit from the fast globally converging characteristics of PSO and the effective local search ability of CLS and RWM, the proposed method can obtain the global optimal results quickly which was tested for six benchmark optimization problems. And the improved performance comparing with the standard PSO and genetic algorithm (GA) testified its validity.