摘要
文章提出一种多目标混沌粒子群算法(multi-objective chaos particle swarm optimization,MOCPSO)-内点法(interior point method,IPM)联合算法(MOCPSO-IPM)来求解考虑经济环境状况的发电调度问题(economic environmental dispatching,EED),同时将负荷节点电压偏移量纳入规划。该算法包括3个阶段:1)用混沌初始化、混沌搜索和外部存档交叉变异改进的粒子群算法进行全局搜索,获取初步优化的非支配解集;2)采用基于前后代支配度比较的粒子群进化收敛判据。当满足判据时,对当前非支配解集进行自适应K-means聚类,在每个类中选取与该聚类中心归一化距离最近的可行粒子作为该类的代表粒子,用以概括非支配解集曲面的分布情况;3)对每个代表粒子,针对目标函数的取值情况,建立ε-constriant化模型,用内点法进行局部搜索和深度寻优。通过IEEE 30节点和2736节点标准算例进行仿真实验,从多角度来评估和证实本算法的有效性。
A multi-objective chaos particle swarm optimization cooperated with an interior point method(MOCPSO-IPM) is used to solve the economic-environmental dispatching with node voltage deviation as the third objective function. The algorithm consists of three stages: 1. A prelim Pareto set is acquired through a general searching by an improved PSO with chaos initialization, chaos search and cross mutation operation;2. A convergence criterion of PSO based on the ranking of dominance between two generations is adopted. When the particle swarm is evolved maturely, an adaptive K-means clustering is used to select those representative particles which are the feasible solutions closest to the cluster centers in each class to describe the distribution of the surface of nondominated solution set;3. For each class’s representative particle in a class, the ε-constraint model is established according to the value of the objective function, and a further optimization is carried out in IPM. With a standard IEEE 30-bus and 2736-bus test system, the effectiveness of the algorithm is verified when solving the EED problem.
作者
张吉昂
王萍
程泽
ZHANG Ji’ang;WANG Ping;CHENG Ze(School of Electrical and Information Engineering,Tianjin University,Nankai District,Tianjin 300072,China)
出处
《电网技术》
EI
CSCD
北大核心
2021年第2期613-621,共9页
Power System Technology
关键词
发电调度
帕累托最优
多目标混沌粒子群
K-MEANS聚类
内点法
generation dispatching
Pareto optimality
multiobjective chaotic particle swarm optimization
K-means clustering
interior point method
