【摘要】 针对基本粒子群优化算法稳定性较差和易陷入局部收敛的缺点,提出了一种基于群体适应度方差的自适应粒子群优化算法。一方面,在可行域中采用混沌初始化生成均匀分布的粒群,提高了初始解的质量;另一方面,构造了基于群体适应度方差的惯性权重的自适应变换公式,增强了算法跳出局部最优解的能力。仿真实验结果表明了该算法的可行性和有效性。更多还原

【Abstract】 Particle Swarm Optimization(PSO) is a novel stochastic global optimization evolutionary algorithm.To overcome the poor stability and local convergence of PSO,an adaptive particle swarm optimization based on colony fitness variance(FV-APSO) is proposed.On the one hand,the uniformly distributed particles are generated in the feasible region by chaos so as to improve the quality of the initial solutions.On the other hand,the adaptive transformation formula of the inertia weight,which is based on fitness variance of swarm is constructed in order to enhance the ability of the PSO to get away from the local optimum.Simulation results show that the FV-APSO is feasible and effective.更多还原