Research on Identification and Processing Method for Abnormal Data of Residential Electric Power Consumption

摘要

This paper studies the abnormal data generated by technical problems during power data collection. Due to the power consumption behavior of residential power users is greatly affected by some factors such as atmospheric temperature and holidays, the paper presents the criterion of factors affecting electric power consumption, and proposes a method for identifying and processing abnormal data of residential electric power consumption based on a comprehensive study on the above factors on the abnormal data. The identified abnormal data is analyzed by the proposed criterion, and the combination of the mean and interpolation method is used to correct and restore the real effective value in the abnormal data, which greatly preserves the feature of original data. By verifying the one-year electric power consumption data of a community in Henan Province, the results show that the proposed method provides theoretical support for retaining payload data and has prospects for development.