一种改进的粒子群优化算法

An improved Particle Swarm Optimization algorithm

粒子群优化算法(PSO)是一种群体智能进化计算方法,但在搜索过程中粒子紧跟最优粒子运动降低了粒子多样性和全局搜索能力,从而易陷入局部极值。本文提出一种新的粒子群优化算法(PSO-EWD),主要改进体现在2个方面:将惯性权重与进化因子相关联,根据种群的进化状态而改变权重大小,以平衡全局搜索能力与局部搜索能力;将时变的分布式时延引入速度更新公式中,以增加粒子的多样性。本文通过5种算法在9个基准函数上的实验对比,证明了新提出的算法相较于另外4种算法具有更优的适应度值、稳定性和收敛速度。

Particle Swarm Optimization algorithm(PSO)is a kind of evolutionary calculation method with swarm intelligence.In the search process,all particles closely follow the optimal particle's movement,which reduces the particles'diversity and global search ability.So it is easy to fall into local optima.In this paper,a new swarm optimization algorithm(PSO-EWD)has been proposed which is mainly improved in two aspects:the inertia weight is associated with the evolution factor,and the weight is changed according to the evolution state of the population to balance the global search ability and the local search ability;the distributed time-varying delays are introduced into the velocity update formula to increase diversity of the particles.In this paper,the experimental comparison of five algorithms on nine benchmark functions shows that the proposed algorithm has better fitness value,stability and convergence speed than the other four algorithms.