Author(s)

Fang Yiqiu, Cheng Yuan, Ge Junwei

Corresponding Author

Fang Yiqiu

ABSTRACT

As one of the representatives of intelligent algorithm, Particle Swarm Optimization (PSO) has been widely concerned and applied since it was proposed. However, the traditional Particle Swarm Optimization (PSO) algorithm has some disadvantages, such as premature convergence, local optimization and lo resolution accuracy. In order to solve the problems in the algorithm, this paper proposes a dynamic inertia weight Particle Swarm Optimization algorithm based on Gaussian Disturbance. Through testing experiments with 5 benchmark functions, the improved algorithm has significantly improved its global search ability and optimization accuracy, and also overcomes the shortcoming of traditional Particle swarm Optimization (PSO).

KEYWORDS

Dynamic inertia weight, Gaussian Disturbance, Particle Swarm Optimization