Energy Valley Optimization-Based Adaptive Fuzzy Logic Control for Grid-Connected Photovoltaic Systems: Simulation and Experimental Validation

This paper presents a novel energy valley optimization algorithm (EVOA) to achieve maximum power point tracking (MPPT) and dynamic regulation of grid-connected photovoltaic (PV) systems under varying irradiance conditions. The proposed approach optimizes four Adaptive Fuzzy Logic Controllers (AFLCs) to enhance tracking accuracy, transient response, and disturbance rejection. The proposed strategy is evaluated against AFLC based on genetic algorithm (GA), AFLC based on Arithmetic Optimization Algorithm (AOA), and Marine Predator Algorithm-based proportional-integral (MPA-PI) control using the MATLAB/Simulink 2026a package and real-time implementation on a dSPACE DS1104 platform under practical operating conditions in Ismailia, Egypt. Under step irradiance variations, AFLC-EVOA achieves faster maximum power convergence, reduced oscillations, improved DC-link voltage regulation, and enhanced PV current tracking. Quantitatively, the proposed controller reduces the average maximum power tracking error by 13.25%, 30.26%, and 52.69% compared with AFLC-AOA, AFLC-GA, and MPA-PI, respectively. Moreover, the mean Integral Absolute Error (IAE) is reduced by 11.92%, 30.16%, and 38.58%, respectively. The proposed controller also maintains reactive power close to zero, with deviations limited to approximately ±0.05 pu. Experimental results further confirm reduced power and voltage ripple and improved transient recovery. These findings demonstrate that EVOA-based adaptive tuning significantly enhances AFLC performance, establishing AFLC-EVOA as a robust and effective MPPT solution for grid-connected PV systems.