深度学习在输电线路短路故障辨识的应用研究  

Research on Application of Deep Learning in Short-circuit Fault Identification of Transmission Line

作　　者：周自强 范鹏[1,2] 赵淳 ZHOU Ziqiang;FAN Peng;ZHAO Chun(NARI Group Corporation(State Grid Electric Power Research Institute),Nanjing 211106,China;Wuhan NARI Limited Liability Company of State Grid Electric Power Research Institute,Wuhan 430074,China)

机构地区：[1]南瑞集团有限公司(国网电力科学研究院),江苏南京211106 [2]国网电力科学研究院武汉南瑞有限责任公司,湖北武汉430074

出　　处：《电工技术》2020年第23期94-95,98,共3页Electric Engineering

摘　　要：随着分布式故障测距技术的快速发展,输电线路故障行波数据量呈现指数性增长,传统的短路故障辨识研究方法效率低下。为此,以典型的雷击故障(绕击和反击)和非雷击故障图像库为基础,提出一种基于深度学习的短路故障辨识方法,对行波数据的时域信息进行特征识别,以波头和波尾为输入量,构建输电线路短路故障辨识模型。该方法为电力公司输电线路运维提供了理论依据。With the rapid development of distributed fault location technology,the traveling wave data of transmission line fault shows exponential growth,but the traditional short-circuit fault identification method is inefficient.Therefore,based on typical lightning faults(shielding failure and back flashover)and non-lightning faults image base,a short-circuit fault identification method based on deep learning was proposed.The time-domain information of traveling wave data was characterized,and the wave head and wave tail were used as input to construct a short-circuit fault identification model for transmission lines.This method provided a theoretical basis for power company transmission line operation and maintenance.

关 键 词：深度学习 行波 故障辨识 雷击 非雷击 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]