基于质心和自适应指数惯性权重改进的粒子群算法

针对粒子群优化(PSO)算法易出现早熟收敛及寻优精度低等问题,为提高粒子群优化算法寻优能力,提出了一种基于质心和自适应指数惯性权重改进的粒子群优化算法(CEPSO)。首先,使用各粒子的适应度计算权重系数;然后,分别使用各粒子当前位置和迄今为止最优位置构造了加权的种群质心和最优个体质心,使用平均粒距来度量群体状态,并依据群体状态设计了分段指数惯性权重;最后,结合使用分段指数惯性权重和双质心调整了粒子速度更新公式。仿真结果表明,CEPSO能增强寻优能力,并具有较强的稳定性。

Improved particle swarm optimization algorithm based on centroid and self-adaptive exponential inertia weight

Aiming at the problem that Particle Swarm Optimization (PSO) algorithm is easily trapped into local optima and has low accuracy in convergence, in order to improve the optimization capability of PSO algorithm, an improved particle swarm optimization algorithm-Centroids combined with self-adaptive Exponential inertia weight PSO (CEPSO) was proposed. Firstly, weighting coefficients were calculated by the fitness of each particle. Secondly, double centroids, the population centroid and the best individual centroid were constructed, which were the weighted combination of each particle's current position and its by far best position. Finally, the proposed algorithm worked on the centroids and the self-adaptive exponential inertia weight designed by the swarm diversity correspondingly to the different working stages of the swarm to adjust its velocity updating formula. The experimental results show that CEPSO can enhance the search ability, and it has strong stability.