A Smart Grid Traffic Anomaly Detector Based on Deep Learning

The information network of the power system is an important part of the long- term continuous and effective operation in power industry. The complex network structure between the power grid and the information network in the smart grid brings a great challenge to the abnormal detection of network flow in the security of information communication network. However, traditional machine learning algorithms often have shortcomings such as low detection accuracy and poor real-time performance in solving the problem of network flow anomaly detection, and they are not possible to provide a high-precision early warning function for network security events in advance. In this paper, we introduce the principle of traffic anomaly detection technology based on intrusion detection, and propose a deep learning predictor for abnormal traffic in grid system. By learning the temporal and spatial features of traffic data, deep learning predictor can predict traffic features in the next period and classify traffic security events for the predicted traffic. To a certain extent, it can meet the demand of abnormal traffic detection of smart grid servers, so as to achieve the purpose of improving power grid information security.