摘要
针对电力系统多目标最优潮流计算问题,提出一种基于(非劣最优)Pareto解集的改进粒子群算法AL iPSO。用最优值评估选取法求取粒子和全局最优位置,解决目标函数间可能存在的冲突。并将关联度自适应学习应用于多目标优化,提出适合Pareto解特点的适应度设计和随机惯性权策略,克服PSO算法容易早熟而陷入局部最优解的缺点。通过对IEEE 6、IEEE 14节点系统多目标最优潮流计算,验证了该算法的有效性。
An improved particle swarm optimization algorithm called ALiPSO was presented based on pareto optimal set to solve multi-objective OPF problem. The algorithm obtained the particle and the group＇s best previous position by evaluating and selecting optimal value, and resolved the conflicts among multiple objective functions. The method of adaptive linkage learning was developed. Fitness assignment and random inertia weight strategy were used to avoid getting trapped in local optimal solution arose by prematurity. Case studied on IEEE 6-bus system and 14-bus system showed the effectiveness of the proposed algorithm.
出处
《电力系统及其自动化学报》
CSCD
北大核心
2007年第3期51-57,共7页
Proceedings of the CSU-EPSA
关键词
粒子群优化算法
非劣最优解集
多目标
最优潮流计算
关联度自适应学习
适应度设计
随机惯性权策略
particle swarm optimization
pareto optimal set
multi-objective
optimal power flow （OPF）
adaptive linkage learning
fitness assignment
random inertia weight strategy
