基于分布式并行计算的电力系统不良数据识别方法  

A Method for Identifying Bad Data in Power System Based on Distributed Parallel Computing

作　　者：冷迪 邱子良 黄建华 秦思远 LENG Di;QIU Ziliang;HUANG Jianhua;QIN Siyuan(Shenzhen Power Supply Bureau Co.,Ltd.,Shenzhen 518000,China)

机构地区：[1]深圳供电局有限公司,广东深圳518000

出　　处：《电工技术》2022年第20期149-151,共3页Electric Engineering

摘　　要：电力系统不良数据识别方法以单一误差为标准多次比对、多次循环,难以同时处理大量数据,导致不良数据识别误差大、速率低。为改进不良数据识别方法存在的缺陷,设计了基于分布式并行计算的电力系统不良数据识别方法。采用MapReduce模型搭建分布式并行计算框架;设定不良数据判断标准,预处理电力数据;利用标准残差向量和残差灵敏度,识别电力系统不良数据。通过试验验证识别方法的应用效果,结果表明所提方法的平均识别相对误差为12.51%,多种类不良数据漏检率较低,证实了该识别方法的应用效果良好。The identification method of bad data in power system uses single error as the standard for multiple comparisons and cycles,which makes it difficult to process a large number of data at the same time,resulting in large error and low rate of bad data identification.In order to improve the defect of bad data identification method,a method of bad data identification in power system based on distributed parallel computing is designed.MapReduce model is used to build a distributed parallel computing framework.Judgment criteria is set for bad data and preprocess power data.The standard residual vector and residual sensitivity are used to identify the bad data of power system.The experiment results show that the average recognition relative error of the proposed method is 12.51%,and the rate of missed detection of various kinds of bad data is low,which proves that the application effect of the identification method is good.

关 键 词：分布式并行计算 电力系统 不良数据 数据识别 MAPREDUCE模型 标准残差向量 

分 类 号：TM743[电气工程—电力系统及自动化]