فهرست نشریات علمی دسترسی آزاد ایران

This paper presents a deep learning-based adaptive robust control strategy for smart microgrids, aiming to simultaneously improve power quality, reduce active power losses, and enhance resistance against cyber-attacks. The proposed controller integrates error-estimation-based robust control with an adaptive deep neural network that dynamically updates control coefficients in response to uncertain operating conditions. In addition, an embedded attack detection and mitigation mechanism safeguards the system against threats such as false data injection, denial-of-service, and replay attacks. The effectiveness of the proposed approach is evaluated on a three-phase multi-bus microgrid under diverse load variations, disturbances, and cyberattack scenarios. Comparative results against classical PI, sliding mode control, H∞, and model predictive control schemes demonstrate that the proposed controller achieves lower total harmonic distortion, faster settling times, reduced active power losses, and higher reliability indices. These findings confirm the potential of the proposed method as a practical and efficient solution for securing and optimizing next-generation smart microgrids.