A Improved Particle Swarm Optimization Algorithm with Dynamic Acceleration Coefficients

Abstract

Particle swarm optimization (PSO) is one of the famous heuristic methods. However, this method may suffer to trap at local minima especially for multimodal problem. This paper proposes a modified particle swarm optimization with dynamic acceleration coefficients (ACPSO). To efficiently control the local search and convergence to the global optimum solution, dynamic acceleration coefficients are introduced to PSO. To improve the solution quality and robustness of PSO algorithm, a new best mutation method is proposed to enhance the diversity of particle swarm and avoid premature convergence. The effectiveness of ACPSO algorithm is tested on different benchmarks. Simulation results found that the proposed ACPSO algorithm has good solution quality and more robust than other methods reported in previous work.