基于改进Hopfield神经网络算法的变电站出线间隔优化分配  

Optimal distribution of substation outlet intervals based on improved hopfield neural network algorithm

作　　者：卢家欢 叶新[1] 黄民翔[1] LU Jia-huan;YE Xin;HUANG Min-xiang(College of Electrical Engineering,Zhejiang University,Hangzhou 310027,China)

机构地区：[1]浙江大学电气工程学院,浙江杭州310027

出　　处：《能源工程》2019年第1期25-31,共7页Energy Engineering

基　　金：国家电网公司科技项目(5211JY150016)

摘　　要：以负载率、备用间隔以及非公用间隔这三个指标对变电站进行聚类作为优化序列参照,综合考虑出线间隔不平衡率以及出线间隔优化经济成本等多个指标对出线间隔优化分配问题进行数学建模;通过引入拉格朗日乘子以及粒子群算法的全局记忆性,提出了基于改进Hopfield神经网络算法的变电站出线间隔优化方法,并且将其应用于解决多目标、多约束的变电站出线间隔优化分配;最后通过算例验证了算法的正确性及有效性。The three indexes of load rate, standby interval and non- common interval were clustered to be the sequence reference of the optimization. Considering the unbalanced rate of outlet line intervals and the cost of intervals optimization and some other indicators, a mathematical model was proposed to solve the optimal allocation problem. By introducing Lagrange multipliers and Particle Swarm Optimization (PSO) Algorithm s global memory, a method which was based on improved Hopfield neural network algorithm was offered to deal with the problem of substation line intervals optimization, and applied to solve multiple-target and multiple-constraint problems such as substation outlet intervals optimization. Finally, result of a numerical example was given to verify the correctness and validity of the proposed algorithm.

关 键 词：出线间隔 优化分配 Hopfield神经网络算法 拉格朗日乘子 全局记忆性 

分 类 号：TM715[电气工程—电力系统及自动化]