A Uniform Robust Exact Differentiator Based Neuro-Fuzzy Fractional Order Sliding Mode Control for Optimal Standalone Solar Photovoltaic System

Abstract

This study introduces an innovative neurofuzzy fractional-order sliding mode control approach for standalone photovoltaic systems, designed to mitigate uncertainties and disturbances caused by fluctuating environmental conditions. The method combines a fuzzy logic neural network, uniform robust exact differentiator, and fractional-order sliding mode control. The neural network accurately predicts nonlinear reference voltage trajectories, whereas the differentiator estimates unmeasurable states and external disturbances. The inclusion of fractional-order control improved the adaptability and robustness of the system. The stability of the proposed approach is rigorously validated using the Lyapunov theory. MATLAB simulations and experimental results significantly improve tracking accuracy and overall system performance, providing a robust and efficient solution to optimize energy extraction in standalone PV systems.