Enhancing Security in Islanded AC Microgrid: Detecting and Mitigating FDI Attacks in Secondary Consensus Control through AI-Based Method

This paper presents a successfully developed secondary consensus control system to ensure accurate voltage regulation among distributed generators in an islanded AC microgrid. The communication network is built using Graph theory and Laplacian matrix techniques. The impact of False Data Injection (FDI) attacks on the communication link, particularly on consensus control measurement data, is analyzed to observe the resiliency of the consensus controller. Timely mitigation of FDI attacks is crucial for maintaining the system’s stability. To address this, we propose a novel Artificial Neural Network (ANN) based observer, capable of estimating the system’s state under load changes and attack scenarios. The proposed AI-based detection method effectively identifies and mitigates FDI attacks swiftly, preventing further deterioration of the system. The whole system is modeled in MATLAB to validate the effectiveness of the proposed method, showcasing its ability to safeguard the microgrid against cyber threats. The study highlights the resilience of the consensus controller and emphasizes the significance of quick FDI attack detection and mitigation for ensuring the stability and reliability of the entire microgrid system.