基于高维随机矩阵的用户异常用电行为判定

Judgment of abnormal power consumption behavior of users based on high-dimensional random matrix

目前研究的用户异常用电行为判定方法存在判定结果不准确的问题,难以精准地确定用户异常行为的时间段。为此,提出了一种基于高维随机矩阵的用户异常用电行为判定方法。通过筛选公式获得异常用电数据与正常用电数据,完成数据存储;根据筛选出的用户异常用电数据与正常用电数据构建了高维随机矩阵,计算矩阵特征值;利用特征值的波动程度确定了异常用电线路与异常用电用户,进而对用户的异常用电行为进行判定。通过实验研究验证了提出的判定方法具有较高的可靠性与判定精确性,相比于传统判定方法,基于高维随机矩阵的用户异常用电行为判定方法能够准确地确定用户异常线路,同时检测出用户异常行为时间段。

The current research on the judgment method of users′ abnormal power consumption behavior has the problem of inaccurate judgment results,which is difficult to accurately determine the time period of users′ abnormal behavior. Therefore,a method to determine the abnormal power consumption behavior of users based on high-dimensional random matrix is proposed. The abnormal power consumption data and normal power consumption data are obtained through the screening formula to complete the data storage. According to the screened abnormal power consumption data and normal power consumption data,a high-dimensional random matrix is constructed and the eigenvalues of the matrix are calculated.The abnormal power lines and abnormal power users are determined by using the fluctuation degree of eigenvalues,and then the abnormal power consumption behavior of users is determined. Through experimental research,it is verified that the proposed judgment method has high reliability and judgment accuracy. Compared with the traditional judgment method,the user abnormal power consumption behavior judgment method based on high-dimensional random matrix can accurately determine the user abnormal line and detect the time period of user abnormal behavior.