基于PCA-ELM的日同期线损检测系统的研发  

Research and development of the daily line loss prediction system based on PCA-ELM

作　　者：夏翔[1] 董大伟 方建亮[1] 姜巍[1] 何欣 胡剑地 Xia Xiang;Dong Dawei;Fang Jianliang;Jiang Wei;He Kin;Hu Jiandi(Zhejiang Electric Power Corporation, Hangzhou 310007, China;SGIT-Greatpower, Fuzhou 350003,China)

机构地区：[1]国网浙江省电力公司,杭州310007 [2]国网信通亿力科技有限责任公司,福州350003

出　　处：《电子测量技术》2018年第7期99-104,共6页Electronic Measurement Technology

摘　　要：针对配电网线损给能源及经济带来严重损失,且电力设备采集到的数据不同期的问题,提出了基于主成分分析-极限学习机（PCA-ELM）的日同期线损预测系统。首先,采用PCA方法对影响配电网线损的参数进行提取,将综合贡献值大于85%的参数选择出来;其次,应用ELM算法将PCA提取的参数作为输入数据,线损值作为输出;基于PCA-ELM方法开发了一套配电网线损实时预测监控系统,用于对当前线损值的实时显示;最后采用算例仿真,验证了PCA-ELM算法在线损预测上的可靠性,在电力公司的运行效果也验证了该线损检测系统的有效性能。In view of the distribution network line loss brought serious damage to energy and economy,and power equipment problems of different period,the data collected is proposed in this paper,based on the same line loss prediction system of PCA-ELM.Firstly,the parameters of the distribution network line loss were extracted by PCA method,and the integrated contribution value was greater than 85%.Secondly,the parameters of PCA were extracted by ELM algorithm as input data,and the line loss was output.Then based on the PCA-ELM method,a real-time prediction monitoring system for distribution network wire loss is developed for real-time display of the current line loss.Finally,an example simulation is used to verify the reliability of the PCA-ELM algorithm in line loss prediction,and the effectiveness of the line loss prediction system is also verified in the operation effect of the power company.

关 键 词：线损 检测 主成分分析 极限学习机 

分 类 号：TM744[电气工程—电力系统及自动化] TP183[自动化与计算机技术—控制理论与控制工程]