基于线性回归的精准识别窃电用户的研究

Research on accurately identifying electricity stealing users based on linear regression

为了实现低压嵌入式设备精准识别窃电用户,该文针对窃电用户的线损特点,根据量测开关窃电检测实际应用场景,基于智能量测开关所采集的电表功率数据以及自身计量功率(电量),结合相似性原理,构建了窃电检测模型与相似性测量模型。制定量测开关窃电检测流程,基于实验采集数据计算得到回归参数以及相似性距离,并对实验结果进行了分析比较。实例应用表明,智能量测开关基于距离的序列相似性无法定量判断出窃电用户,而基于线性回归算法的窃电检测模型能够精准识别窃电用户。

In order to realize the accurate identification of electricity stealing users by low⁃voltage embedded devices,this paper aims at the characteristics of line loss of electricity stealing users,according to the actual application scenario of electricity stealing detection of measurement switches,based on the meter power data collected by the intelligent measurement switch and its own metered power(electricity),and combined with the similarity principle,the electricity stealing detection model and the similarity measurement model are constructed.At the same time,the detection process of measuring switch electricity stealing is formulated,the regression parameters and similarity distance are calculated based on the experimental data,and the experimental results are analyzed and compared.The example application shows that the distance⁃based sequence similarity of the intelligent measurement switch cannot quantitatively determine the electricity stealing users,while the electricity stealing detection model based on the linear regression algorithm can accurately identify the electricity stealing users.