As distributed generation units and renewable energy sources become increasingly integrated into modern power systems, ensuring the stability of the DC-link voltage—especially in the presence of potential cyber threats—poses a significant challenge. Traditional control methods, such as proportional–integral (PI) controllers, and even many contemporary intelligent algorithms, often struggle to maintain performance under unexpected cyberattacks or falsified data due to their reliance on accurate system models or extensive retraining. This study proposes an Adaptive Neuro-Fuzzy Inference System (ANFIS)-based control strategy that merges the adaptive learning ability of neural networks with the robustness of fuzzy logic, enabling real-time adjustment of control parameters. The primary contribution of this approach is its capacity to autonomously detect and mitigate sophisticated cyber threats—including False Data Injection Attacks (FDIA), Denial-of-Service (DoS) attacks, and cyber-induced load fluctuations—without the need for predefined system models or extensive retraining. Simulation results on the IEEE 13-bus network with integrated solar and wind generation, implemented in MATLAB/Simulink, show that the proposed controller significantly improves DC-link voltage stability, shortens recovery time, and enhances overall network resilience compared to conventional PI and other intelligent controllers. These findings highlight that the ANFIS-based controller effectively addresses the limitations of traditional methods, offering a practical and robust solution for modern smart and resilient power grids.
