Enforced mutation to enhancing the capability of particle swarm optimization algorithms

Abstract

Particle Swarm Optimization (PSO), proposed by Professor Kennedy and Eberhart in 1995, attracts many attentions to solve for a lot of optimization problems nowadays. Due to its simplicity of setting-parameters and computational efficiency, it becomes one of the most popular algorithms in optimizations. However, the discrepancy of PSO is the low dimensionality of the problem can be solved. Once the optimized function becomes complicated, the efficiency gained in PSO degradates rapidly. More complex algorithms on PSO required. Therefore, different algorithms will be applied to different problems with difficulties. Three different algorithms are suggested to solve different problems accordinately. In summary, proposed PSO algorithms apply well to problems with different difficulties in the final simulations. © 2011 Springer-Verlag.