电力信息与通信技术2025，Vol.23Issue(8)：67-74,8.DOI:10.16543/j.2095-641x.electric.power.ict.2025.08.09
基于RLS-SC的电动自行车户内充电在线检测方法
Online Detection Method of Indoor Electric Bicycle Charging Based on RLS-SC
摘要
Abstract
Residential electric bicycle charging fire hazards,high super-vision costs,in urgent need of a real-time and accurate electric bicycle charging online detection technology to assist urban community governance.In this paper,an online identification method of electric bicycle based on recursive least squares filter and subtraction clustering is proposed.Firstly,considering the small power of electric bicycle in the opening stage is easy to be confused with other household equipment,a two-window event detection model is proposed to realize the extraction of electric vehicle-like opening events under complex aliases.Secondly,in order to solve the problem that the constant voltage charging stage of electric bicycles takes a long time and is easy to be superimposed by other electrical equipment,a recursive least square(RLS)filter based on electric bicycles constant voltage charging stage end event extraction model is constructed.Then,based on the two-stage event features,the user electric bicycle data sample is updated and a user-specific classifier model is generated to achieve accurate online detection of user electric bicycles.Finally,using two different environments in Gulou District of Nanjing as data sources,the experimental results show that the detection rate of the method is more than 92%and the accuracy is more than 96%,which indicates that the method has practical application value.关键词
非介入式负荷辨识/电动自行车/RLS-SC/事件检测/减法聚类Key words
non-intrusive load identification/electric bicycle/event detection/subtractive clustering分类
信息技术与安全科学引用本文复制引用
伍栋文,朱亮,陈忠敏,胡涵天,胡琛,周麟云..基于RLS-SC的电动自行车户内充电在线检测方法[J].电力信息与通信技术,2025,23(8):67-74,8.基金项目
国网江西省电力有限公司科技项目"电动自行车户内充电行为特征识别技术研究及应用"(521852230001). （521852230001）
