Study on an Improved PSO Algorithm and its Application for Solving Function Problem

Particle swarm optimization(PSO) algorithm has the advantages of simplicity and easy implementation, but it exits the weaknesses of the being easy to fall into local minimum and premature convergence. In order to overcome these weaknesses of PSO algorithm, the inertia weight and learning factor are improved and the PSO algorithm is initialized by using chaotic optimization in order to propose an improved PSO(WLFCPSO) algorithm. In the proposed WLFCPSO algorithm, chaotic optimization strategy is used to initialize the parameters of PSO algorithm in order to obtain the more reasonable initialization parameters. The adaptive inertia weight adjustment strategy is used to control the adjustment ability of inertia weight in order to keep the diversity of the inertia weight. The dynamic linear adjustment strategy for learning factors is used to gradually reduce the cognitive ability of the individual and improve the global search ability of particles. In order to prove the effectiveness of the proposed WLFCPSO algorithm, several benchmark functions are selected. The experiment results show that the proposed WLFCPSO algorithm has the rapider convergence speed and higher solution precision for solving high-dimension function optimization problems.