混沌量子粒子群优化算法

Chaos quantum-behaved particle swarm optimization algorithm

针对量子粒子群优化算法在处理高维复杂函数时存在的收敛速度慢、易陷入局部极小等问题,提出了混沌量子粒子群优化算法。采用了基于群体适应值方差的早熟判断机制,同时提出了一种基于混沌搜索的新方法,提高了搜索效率。数值实验结果表明,混沌量子粒子群算法效率高、优化性能好,且具有很强的避免陷入局部最优的能力,其性能远远优于一般的粒子群算法和量子粒子群算法。

Using quantum-behaved particle swarm optimization （QPSO） to handle complex functions with high-dimension has the prob- lems of low convergence speed and sensitivity to local convergence. The chaos quantttm-behaved particle swarm optimization algorithm （CQPSO） is proposed. The method of judging the local convergence by the variance of the population＇s fitness is proposed, which enhances searching efficiency. Numerical simulation results show that CQPSO is of high efficiency, and of excellent optimum perfor- mance, Especially it＇s of strong ability to avoid running into local optima. It is of much better performance to PSO and QPSO.