Power Consumption Anomaly Detection of Smart Grid Based on CAE-GRU

With the growth of global energy demand, the application of smart grid technology has become widespread. Anomaly detection in power systems is crucial for ensuring the stability and economy of power supply. Deep learning technologies offer new opportunities in this field. This paper proposes a deep learning approach based on Convolutional Autoencoders (CAEs) and Gated Recurrent Units (GRUs) for anomaly detection in smart grid power data. This method integrates three types of feature data, namely user power consumption, line loss correlation, and meter error, and combines the moving window technology to construct a CAE-GRU network model. Experimental results show that, compared with traditional methods, this method has higher accuracy in anomaly detection, which can effectively identify potential problems in the power grid and provide strong support for the optimized operation of the smart grid.