一种改进的混沌粒子群优化算法

粒子群优化算法（PSO）自提出以来,已经被广泛地应用于求解各类复杂的优化问题,过去对粒子群算法的研究主要集中在融入新的优化方法或对其相关参数进行调整,但这样只会使得PSO更加复杂.针对这一问题,文中提出一种改进的混沌粒子群优化算法（ICPSO）,ICPSO从粒子群优化算法的时间与寻优实时角度出发（即在较短的时间内获得较好的解）,对粒子速度更新算子进行了简化,每隔一定代数后,在最优解邻近区域引入混沌扰动以避免种群陷入局部最优解.数值实验结果表明：提出的算法相对于文献给出的PSO改进算法,不仅能够获得较好的最优解,而且还具有较快的收敛速度和较好的稳定性.

A fast chaos particle swarm optimization algorithm

Particle swarm optimization （PSO） is a population-based stochastic global optimization method. It has been applied to many practical complex optimization problems. The focus of past research has been with making the PSO method more complex by incorporating new optimization techniques into PSO or adjusting relative parameters. To solve this problem, this paper proposes an improved chaos particle swarm optimization algorithm （ICPSO）. Form the viewpoints of time and optimization, it simplifies the velocity equation, and introduces chaos perturbation to jump into local optimum. The numerical experiments show that the proposed algorithm, as compared with the reported algorithm in literature, can not only obtain better optimum, but also get faster speed of convergence and good stability.