Adaptive Fuzzy-PID Control with Grey Wolf Optimization for Enhanced Performance in Hybrid Renewable Systems

Grid-connected PV-wind systems experience strongly varying operating conditions that challenge fixed-gain control. This study evaluates a Fuzzy-PID controller with offline Grey Wolf Optimizer (GWO) tuning using two explicitly separated simulation layers: thesis-level MATLAB/Simulink system indicators and a reproducible reduced-order DC-link benchmark. The reproducible layer uses common disturbance traces, five synthetic stochastic weather profiles, independent-seed validation, and an equal-optimization-budget comparison between GWO-PID and GWO-Fuzzy-PID. With 30 wolves and 15 iterations for both architectures, the optimized PID and optimized fuzzy controller achieved nearly identical training objectives (0.08349 and 0.08401). Across 10 unseen seeds, GWO-PID produced slightly lower mean RMSE than GWO-Fuzzy-PID in all five scenarios; under Stormy operation, the values were 7.746 and 7.873 V, respectively, compared with 9.517 V for Fixed PID. Extended 30 s and 60 s tests confirmed that GWO-based tuning retained its principal benefit under severe disturbances, while the fuzzy layer did not provide a systematic RMSE advantage over an equally optimized PID. Rainy/Low-Wind diagnostics showed a persistent source-power deficit and near-continuous actuator current-limit operation, explaining the approximately 22 V short-horizon RMSE. Controller-only MATLAB R2017a timing was below 8 microseconds/update for the fuzzy implementations. The results therefore support GWO-based severe-disturbance robustness, while separating the contribution of optimization from that of fuzzy adaptation and avoiding a claim of universal controller superiority.