基于联合神经网络的配电网线损异常检测  

Distribution Network Line Loss Anomaly Detection Based on Joint Neural Network

作　　者：黄文栋 王德辉 刘明昊 许卓佳 杨溢儒 HUANG Wendong;WANG Dehui;LIU Minghao;XU Zhuojia;YANG Yiru(Guangzhou Power Supply Bureau of Guangdong Power Grid Co.,Ltd.,Guangzhou 510000,China)

机构地区：[1]广东电网有限责任公司广州供电局,广东广州510000

出　　处：《微型电脑应用》2024年第12期264-267,280,共5页Microcomputer Applications

摘　　要：随着能源互联网战略的不断推进,配电网在能源互联网中的作用越来越重要,配电网线损率是电力公司的重要评价指标之一,异常线损诊断不准确是电力公司亟待解决的问题。提出一种基于联合神经网络的配电网异常线损检测方法。将采集到的配电网数据进行预处理,利用联合神经网络预测线损和负荷,实现对所辖变电站异常线路的检测。利用某市供电公司管辖的配电网运行数据进行模拟试验,并设计多组比较实验来验证所提方法的有效性。实验结果表明,基于联合神经网络的方法能够快速、准确地诊断异常线损,具有实际应用价值。With the continuous promotion of energy internet strategy,distribution network plays an increasingly important role in energy internet.The line loss rate of distribution network is one of the important evaluation indexes of electric power companies.The inaccurate diagnosis of abnormal line loss is an urgent problem for electric power companies to solve.This paper presents a method of abnormal line loss detection in distribution network based on joint neural network.The collected data of the distribution network are preprocessed,and the combined neural network is used to predict the line loss and load to realize the detection of abnormal lines in the substations under the control.In this paper,simulation experiments are carried out with the operating data of the distribution network administered by a city power supply company,and several groups of comparison experiments are designed to verify the effectiveness of the proposed method.The experimental results show that the method based on the joint neural network can diagnose the abnormal line loss quickly and accurately,and has practical application value.

关 键 词：联合神经网络 长短期记忆 异常检测 线损 

分 类 号：TM281[一般工业技术—材料科学与工程]