Decentralized Coordinated Cyberattack Detection and Mitigation Strategy in DC Microgrids Based on Artificial Neural Networks

Abstract

DC microgrids can be considered as cyber-physical systems (CPSs) and they are vulnerable to cyberattacks. Therefore, it is highly recommended to have effective plans to detect and remove cyberattacks in dc microgrids. This article shows how artificial neural networks can help to detect and mitigate coordinated false data injection attacks (FDIAs) on current measurements as a type of cyberattacks in dc microgrids. FDIAs try to inject the false data into the system to disrupt the control application, which can make the dc microgrid shutdown. The proposed method to mitigate FDIAs is a decentralized approach and it has the capability to estimate the value of the false injected data. In addition, the proposed strategy can remove the FDIAs even for unfair attacks with high domains on all units at the same time. The proposed method is tested on a detailed simulated dc microgrid using the MATLAB/Simulink environment. Finally, real-time simulations by OPAL-RT on the simulated dc microgrid are implemented to evaluate the proposed strategy.