Enhanced load frequency regulation in cyberattack triggered grids with renewable energy and electric vehicles

This study investigates the vulnerability of frequency regulation (FR) in communication-dependent smart grids to cybersecurity threats, focusing on their impact on system stability and introducing a novel detection and defense strategy. A modified IEEE 39-bus system, featuring three distinct test zones that incorporate intermittent renewable energy sources, thermal power plants, and electric vehicles, serves as the testing framework. The research presents a three-degree-of-freedom TID-LADRC (3DoF-TID-LADRC) cascade controller optimized using the quasi-opposition-based reptile search algorithm (QORSA). This controller significantly outperforms existing methods, achieving over 27.56%, 23.68%, and 25% improvements in area-1 frequency deviation, area-2 frequency deviation, and tie-line power deviation, respectively, compared to literature benchmarks. The proposed control strategy demonstrates enhanced stability, with a gain margin of 47 dB, surpassing the 45.9 dB achieved by the 2DoF-PID controller. To counter cybersecurity threats, the study proposes a deep-learning-based detection and defense mechanism leveraging a Conditional Generative Adversarial Network (cGAN). Real-time validation on the modified IEEE 39-bus system, using the OPAL-RT platform, confirms that the deviation between simulated and real-time results is less than 6%. The findings validate the proposed approach's effectiveness in detecting and mitigating cyberattacks, ensuring that power system frequency and tie-line power deviations remain within acceptable limits even under adverse conditions.