Real-time Monitoring of Smart Grid Terminals based on Multi-dimensional Information Fusion

With the development of ubiquitous electric Internet of things(UEIoT), the number and types of terminal devices in the smart grid have reached an unprecedented level. In order to ensure the stable operation of the smart grid, we need to monitor the operation status of each terminal, detect various software-level or hardware-level attacks in time, and make corresponding decisions. In this paper, we propose a novel detection method based on multi-dimensional information fusion and computer vision, in which power consumption, traffic, message data and security logs generated by terminal equipments are drawn as characteristic graphs, and then we use a convolution neural network(CNN) to recognize them. According to the detection results, we can know whether there is an attack and if so, which kind of attack has occurred. We also use actual measurement data of an DTU to test our method, and it achieves real-time detection and has high precision.