基于健康度的人工蜂群粒子群算法

针对标准粒子群算法存在收敛速度慢和易陷入局部最优等问题，提出了一种基于健康度的人工蜂群粒子群算法。通过动态地对各个粒子的健康状况进行评价，对正常粒子和病态粒子分别进行处理，避免无效搜索，提高算法的收敛速度；在处理病态粒子时，一方面以大概率借鉴人工蜂群的搜索策略提高算法的探索能力，另一方面以小概率增加粒子群的多样性，避免陷入局部最优。实验结果表明，与标准粒子群算法和其他改进算法相比，该算法收敛速度快、寻优精度高。

Particle health degree based artificial bee colony particle swarm optimization

To the problems of slow convergence and easy to fall into local optimum appearing in standard particle swarm optimization, a particle Health degree based Artificial Bee Colony Particle Swarm Optimization（HABCPSO） algorithm is proposed. It can dynamically evaluate the particles by using health degree. In order to enhance the speed of convergence, the proposed algorithm deals with normal and ill particles respectively. To avoid falling into local optimum, on one hand, using artificial colony algorithm of search strategy to improve exploration ability by large probability；on the other hand, increasing the diversity of particle swarm by small probability. Experimental result shows that, compared with SPSO and other improved algorithm, the new algorithm is much faster and more accurate.