An Adaptive Simple Particle Swarm Optimization Algorithm

The particle swarm optimization algorithm with constriction factor (CFPSO) has some demerits, such as relapsing into local extremum, slow convergence velocity and low convergence precision in the late evolutionary. An adaptive simple particle swarm optimization with constriction factor (AsCFPSO) is combined with chaotic optimization, then a new CFPSO is developed, i.e., a chaotic optimization-based adaptive simple particle swarm optimization equation with constriction factor (CAsCFPSO). Distribution vector of particles is defined as constriction factor in optimization process. Furthermore, piecewise linear chaotic map is employed to perform chaotic optimization due to its ergodicity and stochasticity. Consequently, the particles are accelerated to overstep the local extremum in AsCFPSO algorithm. The experiment results of six classic benchmark functions show that the proposed algorithm improves extraordinarily the convergence velocity and precision in evolutionary optimization, and can break away efficiently from the local extremum. Furthermore, the algorithm obtains better optimization results with smaller populations and evolutionary generations. Therefore, the proposed algorithm improves the practicality of the particle swarm optimization.