基于多突变点与模板匹配的用电设备在线状态监测方法  

An Online Condition Monitoring Method of Electrical Equipment Based on Multiple Change Points and Template Matching

作　　者：贾灿 齐金鹏[1] 袁傲 薛宇鑫 戴理 JIA Can;QI Jinpeng;YUAN Ao;XUE Yuxin;DAI Li(College of Information Science and Technology,Donghua University,Shanghai 201620,China)

机构地区：[1]东华大学信息科学与技术学院,上海201620

出　　处：《电子科技》2024年第6期69-76,共8页Electronic Science and Technology

基　　金：国家自然科学基金(61305081,61104154);上海市自然科学基金(16ZR1401300,16ZR1401200)。

摘　　要：针对现阶段用电设备状态监测技术存在的处理速度较慢、准确率较低等问题,文中基于多突变点检测和模板匹配策略提出了一种用电设备在线状态监测方法。该方法在缓冲区模型和滑动窗口模型的基础上,利用多路搜索树突变点检测(Ternary Search Tree and Kolmogorov-Smirnov,TSTKS)算法形成窗口维度和缓冲区维度的特征向量,通过两种维度的模板匹配实现用电设备的运行状态匹配和状态切换时刻定位。基于家用电冰箱的仿真实验结果表明,所提方法具有检测速度快、准确率高等优点,可为用电设备状态监测领域提供参考。In view of the problems such as slow processing speed and low accuracy of condition monitoring technology for electrical equipment at this stage,an online status monitoring method of electrical equipment based on the multi-change point detection technology and template matching strategy is proposed.Based on the buffer model and the sliding window model,this method uses the TSTKS(Ternary Search Tree and Kolmogorov-Smirnov)algorithm to form the feature vector of the window dimension and the buffer dimension.The running state matching and status switching time positioning of the electrical equipment are realized by template matching of the two dimensions.Simulation experiments results of the household refrigerator show that the proposed method has the advantages of fast detection speed and high accuracy,and can provide a reference for the state monitoring field of electrical equipment.

关 键 词：大数据分析 时序数据 用电设备 状态监测 缓冲区模型 多突变点检测 滑动窗口 模板匹配 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]