A Dynamic Watermarking Integrated With Kalman Filter‐Based Cyber Attack Detection Strategy in Cyber‐Physical Microgrid

ABSTRACT Cyber‐physical power system (CPPS) has emerged due to the integration of communication, and information advancement in the electrical grid. Owing to the extensive growth of the DC‐microgrid (DC‐MG) network, and even with the implementation of advanced control, monitoring, and operating methods, these networks remain vulnerable to a range of cyber attacks in the near future. Thus, it is crucial to defend against assaults on these networked cyber‐physical systems (CPS). In this work, false data injection (FDI) attacks, time delay attacks, replay attacks, and malware‐based attacks are created in the test system and the system performance is measured. These attacks can be identified and recognized by the proposed cyber attack detection methodology based on dynamic watermarking with a Kalman filter for analyzing the terminal current, and load current using smart sensors, actuators, and controllers to eliminate the state of the system. The dynamic signals with watermarking can identify malicious misbehavior from sensors and actuators. This paper also demonstrates how networked systems of sensors and actuators can be secured using the proposed cyber detection approach. The aforementioned cyber attacks are created in the DC‐MG test system with four distributed generation units representing four physical nodes in a networked cyber‐physical power system created in MATLAB environment version 2023b. The purpose of examining the simulation results for various attacks is to assess the performance of the proposed methodology. The results shows the performance of the proposed strategy can give a more reliable and accurate decision method.