The parameter selection of PSO using Lyapunov theory
Qun Jia, Yongxin Li · International Journal of Applied Mathematics & Statistics/International journal of applied mathematics and statistics · 2014
The particle swarm optimization（PSO） is a random intelligence algorithm with the characteristic of multidimensional search in the space. Its stability is a necessary precondition for its convergence. To improve the performance of the PSO algorithm, this paper adopts the Lyapunov theory to analyze the stability of the spatial state equation transformed from the standard PSO algorithm, and determines the stability constraint between parameter and parameter , which is used as the guidance of the stable flight of the particles. Thus the constraint PSO is proposed as the improvement of the standard PSO. Through the test function, this paper verifies that the algorithm with the selection of parameters constrained by the Lyapunov theory is able to improve the performance of the standard PSO model. In other words, the constraint PSO has faster convergence and better accuracy than the standard PSO.