Particle swarm algorithm based on normal cloud

This paper presents a novel parameter automation strategy for the particle swarm optimization algorithm; the normal cloud model is used to improve the performance of the particle swarm optimization algorithm. First, the normal cloud model is used to initialize the population; particles are no longer uniformly distributed throughout the search space. Second, one and the same normal cloud is used to nonlinearly, dynamically adjust inertia weight and update two random numbers in velocity update equation. Therefore, three components in the velocity update equation do interact in the PSO search process, which maintains the diversity of the population, provides balance between the global and local search abilities and makes the convergence faster. Experimental results are provided to show that the improved particle swarm optimization algorithm can successfully locate all optima on a small set of benchmark functions. A comparison of the improve algorithm with the standard particle swarm optimization algorithm is also made.