期刊文献+

基于LSSVM的用户异常用电行为辨识研究 被引量:3

Research on Recognition of Abnormal Power Consumption Behavior of Users Based on LSSVM
下载PDF
导出
摘要 针对用电检查过程中低压用户窃电和电表故障等异常用电行为难以准确和高效辨识的问题,研究了一种最小二乘支持向量机(LSSVM)的用电异常识别方法。首先分析了用电信息采集系统采集的数据特征,建立了电量突变,同期差距,同类用户差距和“候鸟”特征的用户四种异常用电指标模型。然后,针对原始数据中的零电量异常数据,从五个方面分析了其成因并设计出适用的数据清洗方案。接着根据训练数据集,对辨识模型进行训练,获取最优分类函数并区分出异常数据,并将该结果与异常用电指标进行匹配,最后通过案例分析从识别准确度和算法执行效率方面验证了本文算法的可行性。 Aiming at the problem that it is difficult to accurately and efficiently identify abnormal electricityconsumption behaviors such as theft of electricity by low voltage users and meter failures during electricity inspection,a least square support vector machine(LSSVM)method for identifying abnormal electricity consumption is studied.Firstly,the characteristics of the data collected by the electricity consumption information collection system areanalyzed,and four abnormal electricity consumption index models for users with sudden changes in electricityconsumption,synchronization gap,gap between similar users and"migratory birds"characteristics are established.Then,aiming at the zero battery abnormal data in the original data,the causes were analyzed from five aspects and an applicable data cleaning program was designed.Then according to the training data set,the identification model is trained to obtain the optimal classification function and distinguish abnormal data,and match the result with theabnormal electricity consumption index.Finally,it is verified from the recognition accuracy and algorithm execution efficiency through case analysis.The feasibility of the algorithm in this paper.
作者 梁捷 梁广明 黄水莲 LIANG Jie;LIANG Guangming;HUANG Shuilian
出处 《青海电力》 2021年第2期27-31,65,共6页 Qinghai Electric Power
关键词 支持向量机 “候鸟”用户 最优分类函数 异常辨识 Support Vector Machine “Migratory Bird”User Optimal Classification Function Anomaly Identification
  • 相关文献

参考文献6

二级参考文献76

共引文献72

同被引文献23

引证文献3

二级引证文献5

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部