基于混沌的多目标粒子群优化算法

针对多目标优化问题,提出了一种改进的粒子群算法.该算法为了寻找新解,引入了混沌搜索技术,同时采用了一种新的方法--拥挤距离法定义解的适应度.并采取了精英保留策略,在提高非劣解集多样性的同时,使解集更加趋近于Pareto集.最后,把算法应用到4个典型的多目标测试函数.数值结果表明,该算法能够有效的收敛到Pareto非劣最优目标域,并沿着Pareto非劣目标域有很好的分散性.

Particle swarm optimization algorithm based on chaotic series for multi-objective optimization problems

An improved particle swarm algorithm for multiobjective optimization problems is proposed.In this algorithm,chaotic series is used for finding new solutions and a new method,crowding distance method,is adopted to define fitness value of the solutions.Mean-while,the strategy of elite holding is quoted.This algorithm not only improves the diversity of solution set but also makes the nondomi-nated solutions approach the Pareto set as close as possible.At last,the algorithm is applied to four classical test fun...