Artificial intelligence (AI) has proven effective in the optimization and control of photovoltaic (PV) systems MPPT, especially under nonlinear operating conditions such as partial shading, by providing alternatives to traditional methods. Fuzzy logic controllers (FLCs) are one of the many AI techniques that are used MPPT because of their ability to control uncertainty. Most FLC-based MPPT systems, however, operate under the conventional 49-rule architecture, which promotes complicated computation and hinders practical implementation in real time constraining systems. The primary objective of this work is to present simplified MPPT applications for PV systems. The minimum number of fuzzy rules required to achieve a reasonable control performance level was determined using an offline, data-driven clustering technique. From this analysis, an FLC of the Mamdani type with five rules was selected and implemented using a zone-based simplification method. The proposed controller is used for MPPT of the PV array and achieves a minimal structure and less computational complexity and still maintains tracking accuracy and steady-state stability. The simulation results, obtained in MATLAB/Simulink, under different irradiance conditions, proved that the proposed five-rule FLC is reliable and robust in achieving MPPT with almost no overshoot and improved suitability for low-cost digital control platforms.
ABBAS et al. (Thu,) studied this question.