Deep learning-based intrusion detection system for electric vehicle charging station

The integration of the open communication layer to the physical layer of the power grids facilitates bidirectional communication, automation, remote control, distributed, and embedded intelligence, and smart resource management, in the grids. However, cybersecurity threats are inherent with the open communication layer, which can violate the confidentiality, integrity, and availability (CIA) of the grid resources. The soaring usage and popularity of electric vehicles (EVs) demand the robust deployment of trustworthy electric vehicle charging station (EVCS). We propose the novel deep learning-based intrusion detection systems (IDS) to detect the denial of service (DoS) attacks in the EVCS. The deep neural network (DNN) and long-short term memory (LSTM) algorithms are implemented (in python 3.7.8) to detect and classify DoS attacks in the EVCS. Results show that both the DNN and LSTM based IDS achieved more than 99% detection accuracy. On top, the LSTM method is superior to the DNN method in terms of accuracy, precision, recall, and measure