一种基于生物趋化的改进粒子群算法

针对标准粒子群算法进行多极点函数优化时易导致早熟收敛及陷入局部最优的问题,把生物趋化原理引入到粒子群优化算法中,改变传统粒子群优化算法只存在吸引操作而没有排斥操作的单向性,提出一种保持种群多样性的改进算法,并对其关键参数的选择进行了研究。仿真实验结果表明,与传统粒子群优化算法相比,基于生物趋化的粒子群算法对于处理复杂的多峰函数或优化问题,可显著提高算法的全局寻优性能。

An Improved Particle Swarm Optimizing Algorithm Based on Chemotaxis Principle

Aiming at the resulting in premature convergence and plunging into local optimum for standard particle swarm optimization in solving multiple-order pole functions, chemotaxis principle in biology was introduced into particle swarm optimization algorithm to change the single direction characteristic of the traditional algorithm which only had attracting operation instead of repulsing operation. The ameliorated algorithm to maintain population diversity was proposed and the selection of key parameters were studied. Simulation experiment results indicated that the improved particle swarm optimization based on chemotaxis principle could prominently improve the global optimization ability of the algorithm when dealing with complicated multimodal functions or other problems.