摘要
为了进一步提高多目标粒子群优化算法的收敛性和多样性,提出一种多策略改进的多目标粒子群优化算法.建立具有精英粒子领导的异构更新模式并设置个体学习增强因子项,促使种群能够快速寻找真实Pareto最优解.引入外部档案冗余机制,利用其变异及对种群的干扰策略增强解的多样性,避免算法早熟现象的发生.仿真实验结果表明,与其他几种优化算法相比,所提出的算法表现出较好的收敛性和多样性.
An improved multi-objective particle swarm optimization algorithm based on multiple strategies（MIMOPSO）is proposed for the optimization mechanism of particle swarm optimization, which can further improve the convergence and distribution.First, a heterogeneous learning mode and a leadership of the elite particles to individual learning pattern are established to prompt the population to find true Pareto optimal solutions quickly.Then, the variation and disturbance to the population strategy of the external file redundancy mechanism is used to enhance the diversity of the solution and avoid the premature phenomenon of the algorithm.Experimental results show that, comparing with other several kinds of optimization algorithm, the proposed algorithm has better convergence and diversity.
出处
《控制与决策》
EI
CSCD
北大核心
2017年第3期435-442,共8页
Control and Decision
基金
河北省高等学校创新团队领军人才培育计划项目(LJRC013)
国家冷轧板带及装备工程研究中心开放课题项目(2012006)
河北省自然科学基金面上项目(F2016203249)
关键词
多目标优化
粒子群算法
多策略
增强因子
冗余机制
multi-objective optimization
particle swarm optimization algorithm
multiple strategies
intensifying factor
redundancy mechanism
