继电保护系统故障的智能定位方法研究  

An intelligent fault location method for a relay protection system

作　　者：郦阳 王宝华[1] LI Yang;WANG Baohua(School of Automation,Nanjing University of Science and Technology,Nanjing 210018,China)

机构地区：[1]南京理工大学自动化学院,江苏南京210018

出　　处：《电力系统保护与控制》2022年第2期69-76,共8页Power System Protection and Control

基　　金：国家自然科学基金项目资助(51607036)。

摘　　要：为了解决传统继电保护系统故障定位依赖人工且耗时耗力的问题,开展了继电保护智能定位方法研究工作。首先分析了智能变电站继电保护系统故障的主要类型,为模拟故障实验提供理论支撑。随即使用PCS-978装置模拟继电保护故障,采集故障实验数据组成数据库,为继电保护智能定位方法提供数据支撑。然后根据故障实验设计了继电保护故障定位矩阵和故障定位表。最后提出了一种基于深度神经网络,并结合合成少数类过采样技术和随机森林算法的智能定位方法。验证结果表明,智能定位方法可以提高智能变电站保护系统故障的诊断效率,帮助继电保护人员快速准确定位故障,具有较好的实用性。To solve the problems of time consuming and manual work in fault location of the traditional relay protection system,the paper presents research on an intelligent positioning method for relay protection.First,the main types of faults in the relay protection system of a smart substation are analyzed,providing theoretical support for fault simulation experiments.Secondly,the faults are simulated using PCS-978,and the data from experiments are collected to build a database,providing data support for the intelligent positioning method.Then,based on the experiments,the fault location matrix and table are designed.Finally,an intelligent positioning method based on a deep neural network is proposed,incorporating SMOTE and a random forest algorithm.The empirical results demonstrate that the intelligent location method can improve the efficiency of fault diagnosis in a smart substation protection system,can help relay protection technicians locate faults quickly and accurately,and is practical.

关 键 词：继电保护故障主要类型 继电保护系统故障模拟 故障定位矩阵 深度神经网络 智能定位 

分 类 号：TM77[电气工程—电力系统及自动化]