基于MESCM算法的电网用电行为智能识别  

Intelligent recognition of power grid electricity consumption behavior based on MESCM algorithm

作　　者：练琳 卢万平 黄家宝 LIAN Lin;LU Wanping;HUANG Jiabao(Hechi Power Supply Bureau of Guangxi Power Grid Co.,Ltd.,Hechi 547000,China)

机构地区：[1]广西电网有限责任公司河池供电局,广西河池547000

出　　处：《电子设计工程》2024年第5期127-130,135,共5页Electronic Design Engineering

摘　　要：针对电网负荷纵向特性分析精度较差,导致用电行为识别结果不精准的问题,提出了一种基于MESCM算法的电网用电行为智能识别方法。M-P律是MESCM样本协方差矩阵的特征值,基于动态特征值分析电网在不同信噪比情况下用电负荷的随机特性。使用MESCM算法,计算归一化处理的日用电量和电网一天用电量序列,提取电网静态与动态用电特征。设置滑动窗口,计算滑动窗口到数据点空间中心的距离,以此为依据构建智能识别函数,由此完成用电行为类别判定。实验结果表明,该方法与标准电网负荷数据存在5 kW的最大误差,且用电行为识别结果与实际类别一致,具有较强的抗干扰能力。Aiming at the problem that the analysis accuracy of longitudinal characteristics of power grid load is poor,resulting in inaccurate recognition results of power consumption behavior,an intelligent identification method of power grid electricity consumption behavior based on MESCM algorithm is proposed.The M-P law is the eigenvalue of the covariance matrix of the MESCM samples,and the random characteristics of the electricity load of the power grid under different signal-to-noise ratios are analyzed based on the dynamic eigenvalues.Using the MESCM algorithm,the normalized daily electricity consumption and the grid’s one-day electricity consumption sequence are calculated,and the static and dynamic electricity consumption characteristics of the grid are extracted.Set the sliding window,calculate the distance from the sliding window to the spatial center of the data point,and build an intelligent recognition function based on this,thereby completing the classification of electricity consumption behavior.The experimental results show that the method has a maximum error of 5 kW with the standard power grid load data,and the identification results of electricity consumption are consistent with the actual categories,and it has strong anti-interference ability.

关 键 词：MESCM算法 电网负荷 用电行为 智能识别 

分 类 号：TN102[电子电信—物理电子学]