An adaptive niche particle swarm optimization algorithm by evolution grads

作者：Yang Jing^*; Yao Songping; Mang Jianpei

来源：International Conference on Information Computing and Automation, 2007-12-20 to 2007-12-22.

A new particle swarm optimizer, called evolutionary grad-included niche PSO, is presented based on the "prematurity" in global optimization. The new algorithm maintains particle swarm';s diversity by dividing it to small niches. The convergence is speeded up by evolutionary gradient, and the capability of jumping out of the local optimal solution is enhanced. The experimental results show that the improved algorithm not only has more advantage in multi-modal function optimization problem than genetic algorithm and PSO, but also can avoid the premature convergence problem effectively.

• 出版日期2008
• 单位哈尔滨工程大学