Model and design of a fuzzy-based Hopfield NN tracking controller for standalone PV applications

Harnessing energy from abundant, free sunlight is currently a hot topic within the research community. The availability of inexpensive solar modules has made it possible to harvest solar energy at higher efficiency. Photovoltaic (PV) modules have nonlinear characteristics, and hence, the process of impedance matching is obligatory. Proper impedance matching ensures extraction of the maximum amount of power in a PV scheme. Several algorithms that are used to operate DC to DC converters around the Maximum Power Point (MPP) are reported in the literature. Amongst those algorithms, Fuzzy Logic Control (FLC) coupled with other controllers performs well under partial shading conditions. This paper designs a new 5 × 7 optimized FLC-coupled Hopfield Neural Network (NN) maximum tracking technique. A Hopfield NN is used to routinely tune the fuzzy membership function. Entire components of a PV array, a DC-DC buck-boost zeta converter and a designed MPP tracking controller are implemented in a Matlab-Simulink tool to validate the Hopfield NN. The results validate the effectiveness and execution of the Hopfield NN using the optimized fuzzy system. The designed system was successfully tested on an experimental prototype. The experimental values demonstrate the feasibility and improved functionality of the scheme.