基于改进莱维蛇优化的主动配电网孤岛划分方法  

An Island Partitioning Method of Active Distribution Networks Based on Improved Levy Snake Optimizer Algorithm

作　　者：邓佳鑫 熊炜[1] 邹晓松[1] 袁旭峰[1] DENG Jiaxin;XIONG Wei;ZOU Xiaosong;YUAN Xufeng(School of Electrical Engineering,Guizhou University,Guiyang 550025,Guizhou,China)

机构地区：[1]贵州大学电气工程学院,贵州贵阳550025

出　　处：《电网与清洁能源》2024年第12期18-26,共9页Power System and Clean Energy

基　　金：国家重点研发计划项目(2022YFE0205300);国家自然科学基金项目(52367005)。

摘　　要：孤岛隔离是配电系统发生重大故障时的1种弹性解决方案。为了提高用户供电可靠性,降低源荷波动对孤岛恢复的影响,减少能源浪费,提出了一种计及源荷不确定性的配电网孤岛划分方法。以多场景分析的方法刻画场景消除不确定性,基于长短时记忆神经网络的场景生成方法形成分布式电源出力和负荷的典型场景,采用快速前代消除的方法削减场景;建立了基于莱维蛇优化的孤岛划分模型,根据源荷数据,在各时段进行孤岛划分,获得最佳恢复方案。以IEEE 33配电网模型为例进行仿真分析,仿真结果表明,加入莱维飞行策略后,所提方法能够快速有效地求解出孤岛最优方案,提高了系统的稳定性。Island partitioning is a resilient solution for used for significant faults in the power distribution system.In order to improve the reliability of power supply to customers,reduce the impact of source-load fluctuations on islanding restoration,and minimize energy waste,this paper proposes an island partitioning method in the active distribution network that considers the source-load uncertainty.Firstly,the uncertainty is eliminated by portraying the scenarios with the method of multi-scenario analysis(MA).The typical scenarios of Distributed Generation(DG)output and load are formed by the scenario generation method based on the long short-term memory(LSTM)network,and the scenarios are cut down by the method of fast antecedent elimination.Secondly,the islanding model is established based on Levi snake optimization(LSO)and island partitioning in muti-time for the optimal recovery scheme.Finally,simulation experiments are conducted on the modified IEEE33 system to verify the effectiveness of the proposed method.

关 键 词：主动配电网 孤岛划分 长短时记忆 蛇优化算法 莱维飞行策略 

分 类 号：TM732[电气工程—电力系统及自动化]