Adaptive Accelerated Exploration Particle Swarm Optimizer for global multimodal functions

This paper presents a novel variant of particle swarm optimization (PSO) called adaptive accelerated exploration particle swarm optimizer (AAEPSO). AAEPSO algorithm identifies the particles which are far away from the goal and accelerate them towards goal with an exploration power. These strategies particularly avoid the premature convergence and improve the quality of solution. The performance comparisons of search efficiency, quality of solution and stability of the proposed algorithm are provided against (differential evolution) DE, evolutionary strategy (ES), artificial bee colony optimization (ABC) and particle swarm optimization (PSO) algorithms. The comparison is carried out on the set of 10, 30 and 50 dimension complex multimodal benchmark functions. Simulation results indicate the superiority of the proposed AAEPSO over existing algorithms in terms of efficiency, quality solution and stability.