Electricity Abnormal Based On Production And Operation Status Identification Research On Detection Methods

Abnormal user-side electricity consumption directly causes the loss of net profit of power enterprises,which is a traditional problem faced by power supply enterprises in all countries.With the rapid development of smart power grid and the continuous progress of power metering technology,the user-side data of power grid presents a situation of high complexity and high redundancy,which lays an important foundation for data-driven abnormal detection of power consumption.Various data-driven intelligent abnormal power consumption detection methods studied in recent years generally have the defect of high false positive rate.Under the condition of general structural shortage of power grid enterprises and difficulty in ensuring effective and effective implementation of universal electricity consumption inspection,abnormal power consumption detection has always been a long-standing problem for power supply enterprises.According to the above problem,this thesis established a production and business operation state recognition based on users of common electrical detection model,in the fully compared the user under normal condition and many kinds of abnormal state electricity the differences of various electrical parameters,the user of the three-phase power as a characteristic index,identifies the user the power of the daily behavior pattern and the production and business operation mode,Then the daily load characteristics were analyzed by clustering.When the load characteristics of the abnormal period of low electric quantity converge with the normal production and operation state of low electric quantity,it is considered to be the abnormal result of the normal transition of the user state,and the suspicion of abnormal power consumption can be excluded.The proposed method carries out the secondary screening of production and operation status on the basis of the user’s abnormal electric quantity,which avoids the defect of the abnormal electricity consumption detection method based on the abnormal electric quantity,which is difficult to determine whether it is the stolen power or the abnormal and easy to misreport caused by the low electric quantity production and operation status.In order to verify the feasibility of the proposed method,the actual measured data in a certain area of Zhejiang Province were used for simulation analysis,and 40 actual users of electric theft were tested and verified.The test results show that the proposed method can accurately identify users of electric theft in most cases,and the effectiveness of the proposed method is verified.