摘要： 粒子群优化算法在求解复杂函数时，存在收敛速度慢、求解精度不高、易陷入局部最优点等问题。为此，提出一种自适应混沌粒子群优化算法。在基本粒子群算法中引入混沌变量，当算法陷入早熟收敛时进行混沌搜索，同时引入非线性递减的惯性权重。实验结果表明，该算法具有较快的收敛速度和较高的收敛精度，能有效避免早熟收敛问题。

关键词: 粒子群优化算法, 混沌, 自适应惯性权重, 早熟收敛, 全局优化

Abstract: The Particle Swarm Optimization(PSO) algorithm has a few disadvantages in solving complex functions, including slow convergence rates, low solving precisions and high possibilities of being trapped in local optimum. An Adaptive Chaos Particle Swarm Optimization algorithm (ACPSO) is presented based on several improvements in original PSO. ACPSO improves the performances of the standard PSO by applying chaos searching mechanism to avoid premature convergence. The dynamically decreasing inertia weight is employed to enhance the balance of global and local search of algorithm. Experimental results show that the proposed algorithm not only has great advantages of convergence property over standard PSO, but also avoids effectively being trapped in local optimum.

Key words: Particle Swarm Optimization(PSO) algorithm, chaos, adaptive inertia weight, premature convergence, global optimization

中图分类号: 

• TP18