摘要
为了提高多目标优化算法解集的分布性和收敛性,提出一种基于分解和差分进化的多目标粒子群优化算法(d MOPSO-DE).该算法通过提出方向角产生一组均匀的方向向量,确保粒子分布的均匀性;引入隐式精英保持策略和差分进化修正机制选择全局最优粒子,避免种群陷入局部最优Pareto前沿;采用粒子重置策略保证群体的多样性.与非支配排序(NSGA-II)算法、多目标粒子群优化(MOPSO)算法、分解多目标粒子群优化(d MOPSO)算法和分解多目标进化-差分进化(MOEA/D-DE)算法进行比较,实验结果表明,所提出算法在求解多目标优化问题时具有良好的收敛性和多样性.
In order to improve the convergence and diversity of the Pareto optimal set in multi-objective optimization algorithms, a multi-objective particle swarm optimization algorithm based on decomposition and differential evolution（d MOPSO-DE） is proposed, in which the direction angle is presented to generate a set of direction vectors for maintaining the uniform distribution of the swarm.To avoid getting trapped into a local Pareto optimal front,decomposition-based strategy and differential evolution operator are used to generate the global best leader.Moreover,particle memory re-initialization is applied to enhance the diversity of the swarm.The preliminary results show that,compared with Non-dominated sorting genetic algorithm-II（NSGA-II）, multi-objective particle swarm optimizer（MOPSO）, multi-objective particle swarm optimizer based on decomposition（d MOPSO） and multi-objective evolutionary algorithm based on decomposition and differential evolution（MOEA/D-DE）, the proposed algorithm has good performance on convergence and diversity.
出处
《控制与决策》
EI
CSCD
北大核心
2017年第3期403-410,共8页
Control and Decision
基金
国家自然科学基金项目(61374137)
关键词
分解
差分进化算法
多目标优化
粒子群优化算法
方向角
decomposition
differential evolution
multi-objective optimization
particle swarm optimization algorithm
direction angles
