云计算环境下基于C4.5决策树算法的智能变电站电力设备状态监测方法  

Condition monitoring method of power equipment in intelligent substation based on C4.5 decision tree algorithm in cloud computing environment

作　　者：和定繁 蒋羽鹏 杨珊 陈贺 HE Dingfan;JIANG Yupeng;YANG Shan;CHEN He(No.2 Substation Operation Office,Kunming Power Supply Bureau,Kunming 650000,China;Zhejiang Dali Technology Co.,Ltd.,Hangzhou 310053,China)

机构地区：[1]昆明供电局变电运行二所,云南昆明650000 [2]浙江大立科技股份有限公司,浙江杭州310053

出　　处：《电子设计工程》2021年第8期135-139,144,共6页Electronic Design Engineering

基　　金：国网公司科技项目(JL71-15-042)。

摘　　要：为解决现有的智能变电站难以及时、有效地实现电力设备实时监测的问题,提出了一种面向智能变电站的电力设备状态监测方法。考虑到现有智能变电站电力设备数据量呈现指数增长,引入云计算运算模式,构建了基于Hadoop平台的MapReduce分布式处理系统。通过采用C4.5决策树算法实现了云计算,并基于MapReduce对算法进行改进。将并行算法作为核心优化Ha⁃doop平台的并行计算模式,实现数据的并行处理。实验结果表明,文中所提出的方法能够提升数据处理效率,增加电力设备故障诊断与识别的总体准确率,有效实现智能变电站电力设备的实时监测。In order to solve the problem that the existing intelligent substation is difficult to realize the real⁃time monitoring of power equipment timely and effectively,this paper proposes a power equipment condition monitoring method for intelligent substation.Considering the exponential growth of the data volume of the existing intelligent substation power equipment,the cloud computing operation mode is introduced to construct the MapReduce distributed processing system based on Hadoop platform.In this paper,C4.5 decision tree algorithm is used to realize cloud computing,and the algorithm is improved based on MapReduce.The parallel algorithm is used as the core optimization mode of Hadoop platform to realize the parallel processing of data.The experimental results show that the proposed method can improve the efficiency of data processing,increase the overall accuracy of fault diagnosis and identification of power equipment,and effectively realize the real⁃time monitoring of power equipment in intelligent substation.

关 键 词：智能变电站 电力设备 状态监测 HADOOP平台 MapReduce分布式处理系统 C4.5决策树算法 

分 类 号：TP911[自动化与计算机技术]