Resilient detection and adaptive control against FDI attacks in AC microgrids

Aiming at the problem of false data injection attacks in AC microgrids, this study proposes a collaborative detection and mitigation scheme. We constructed a detection model based on an adaptive neuro fuzzy inference system by extracting key operational state features from the microgrid. The attack vector was estimated in real time by combining the optimal state observer, and the power allocation strategy was dynamically modified to eliminate the impact of the attack. After simulation of 4-distributed generation system and multi-scenario test, the scheme shows strong robustness under cyber attack and uncertain environments, significantly improves detection efficiency and system resilience compared with traditional methods and provides a scalable solution for smart grid security protection.