关联驱动下配电网同期线损异常数据辨识  

Abnormal data identification of synchronous line loss in distribution network driven by correlation

作　　者：陆海波 尹建兵 张志鹏 李飞 翁理胜 LU Haibo;YIN Jianbing;ZHANG Zhipeng;LI Fei;WENG Lisheng(Hangzhou Power Supply Company of State Grid Zhejiang Electric Power Co.,Ltd.,Hangzhou 310020,China;Guodian Huayan Electric Power Technology Co.,Ltd.,Guangzhou 510000,China)

机构地区：[1]国网浙江省电力有限公司杭州供电公司,浙江杭州310020 [2]国电华研电力科技有限公司,广东广州510000

出　　处：《电子设计工程》2024年第16期102-105,110,共5页Electronic Design Engineering

摘　　要：配电网同期线损数据的可靠性对于有效实现电网降损与节能是非常关键的,辨识异常数据能够提升配电网同期线损数据的可靠性。为此,设计了关联驱动下配电网同期线损异常数据辨识方法。采用基于多值属性的关联规则挖掘算法,挖掘配电网同期线损数据。利用改进小波阈值去噪算法,对挖掘的配电网同期线损数据实施去噪处理。基于K-means聚类算法、改进型萤火虫算法与聚类可靠性评估指标,设计线损异常数据辨识模型,实现配电网同期线损异常数据辨识。测试结果表明,设计方法的平均误辨识点数和漏辨识点数分别低于10个和5个,平均相对辨识误差保持在1.0以下,具有较好的同期线损异常数据辨识性能。The reliability of synchronous line loss data of distribution network is very important for the effective realization of power network loss reduction and energy saving.Identifying abnormal data can improve the reliability of synchronous line loss data in distribution networks.Therefore,the abnormal data identification of synchronous line loss in distribution network driven by correlation is designed.The association rule mining algorithm based on multi⁃valued attributes is used to mine the line loss data of distribution network in the same period.The improved wavelet threshold denoising algorithm is used to de⁃noise the mined line loss data of the distribution network in the same period.Based on K⁃means clustering algorithm,improved firefly algorithm and clustering reliability evaluation index,the identification model of abnormal line loss data is designed to realize the identification of abnormal line loss data in the distribution network at the same time.The test results show that the average number of false identification points and missing identification points of the design method are less than 10 and 5,respectively,and the average relative identification error is kept below 1.0,which has a good performance of identifying abnormal data of line loss at the same time.

关 键 词：关联规则挖掘算法 配电网同期线损 异常数据辨识 改进型萤火虫算法 

分 类 号：TN911.73[电子电信—通信与信息系统]