配电网故障区间定位的仿电磁学算法  

Electromagnetism-like Mechanism Based Fault Section Diagnosis for Distribution Network

作　　者：郭壮志[1] 吴杰康[1] 

机构地区：[1]广西大学电气工程学院,广西壮族自治区南宁市530004

出　　处：《中国电机工程学报》2010年第13期34-40,共7页Proceedings of the CSEE

基　　金：国家自然科学基金项目(50767001);广西自然科学基金项目(桂科自0640028);广西壮族自治区研究生教育创新计划项目(105930901001;105930904068)~~

摘　　要：采用间接方法实现配电网故障定位在本质上是一个具有0-1离散约束条件的最优化问题。提出基于仿电磁学算法(electromagnetism-like mechanism,ELM)的高容错性配电网故障定位方法,该方法对实时信息的模拟和模型的构建采用0-1编码方式,以此为基础建立环网开环运行配电网的故障定位统一数学模型,能一次性实现多区段的故障定位。针对配电网运行灵活多变的特点,采用分支选择法来提高算法的运行效率。ELM在结构上具有独特的优越性,通过对一个具有3个独立配电区域的典型配电网进行仿真分析表明,采用该方法进行配电网故障定位时具有很高的容错性能,能够实现单一故障和复故障下的准确定位。通过与遗传算法相比较,ELM在快速性、准确性和高容错性方面有明显优势。The indirect method to locate the distribution network fault is constraint optimization problem with 0-1 discrete constraints in essence. To enhance the efficiency of the fault location and solve the problem efficiently, a novel high-tolerance fault location indirect method for distribution network was presented based on electromagnetism-like mechanism （ELM）. A new unified mathematical model for open-loop operating mode was constructed with 0-1 codes tor constraints to realize fault location once for all. Branch- selection method was also applied to improve the efficiency of the distribution network fault location. The simulation for distribution network with three independent distribution areas testifies the correctness and effectiveness of the improved mathematical model as well as the fastness, accuracy and high tolerant performance of the ELM based fault location algorithm by comparing with Genetic Algorithm.

关 键 词：配电网 故障定位 仿电磁学算法 高容错性 分支选择法 

分 类 号：TM72[电气工程—电力系统及自动化]