Efficient Fusion-Based Optimization of Fractional-Order PID Plus Double-Derivative Controller for Automatic Voltage Regulation

Abstract Automatic voltage regulators (AVRs) play a vital role in maintaining voltage stability and improving the overall reliability of power systems. However, they are frequently affected by load fluctuations, steady-state faults, and oscillatory transient responses, which degrade voltage regulation performance. Conventional control strategies struggle to adequately address these challenges, as they often exhibit limited robustness and insufficient dynamic performance under varying operating conditions. To overcome these limitations, this paper proposes a fractional-order PID controller with double derivative (FOPIDD 2 ), optimized using a novel triple-hybrid metaheuristic algorithm known as FGWO-WaOA-NM. The proposed fusion algorithm leverages gray wolf optimizer (GWO) to enhance the exploration capability of walrus optimization algorithm (WaOA), while Nelder–Mead (NM) is incorporated to refine local exploitation. The robustness and effectiveness of the proposed optimization framework are first validated using complex benchmark problems from the CEC2019 test suite. The performance of FGWO-WaOA-NM is benchmarked against cutting-edge algorithms, such as gray wolf, walrus, whale, reptile search, zebra, tunicate swarm, seagull, and prairie dog optimizations. Results demonstrate that FGWO-WaOA-NM consistently achieves superior solution quality and faster convergence. A comprehensive comparative analysis is conducted against various controllers, including PID, FOPID, PIDD 2 and FOPIDD 2 optimized using advanced algorithms to validate the effectiveness of the proposed control scheme. The proposed FGWO-WaOA-NM-optimized FOPIDD 2 controller demonstrates faster convergence, enhanced robustness, reduced overshoot, and improved transient performance, confirming its suitability for modern AVR applications.