Gated Recurrent Units Based Abnormal Detection Method for Imbalanced Electricity Consumption Data

Timely detection of abnormal electricity consumption behaviors plays an important role in saving energy. But the detection of abnormal electricity consumption is faced with many problems. Imbalanced data are important challenges in abnormal electricity detection. When normal data are much more than abnormal data, the model can hardly learn the features of the minority class data, which leads to low detection efficiency. Therefore, in this paper, we employ adaptive synthetic sampling (ADASYN) to achieve effective expansion of the minority class data. In addition, we adopt gated recurrent units to realize the classification of electricity consumption data.We conduct detailed experiments to verify our proposed method. Experimental results show that our method is more effective than other methods.