Abstract
During recent decades, the power system has experienced the ever-increasing penetration level of renewable energy sources (RESs), both in a centralized and a distributed manner. Solar energy in the form of solar photovoltaic (PV) panels has captured attention in power systems, particularly at the distribution level and it is being used all around the world with acceptable solar irradiance. However, such technology is associated with severe technical shortfall in harnessing the maximum power. Accordingly, several techniques have been introduced for the maximum power point tracking (MPPT) of PV systems. In this respect, an MPPT technique, augmented by the incremental conductance (INC) and hybrid shuffled frog-leaping and pattern search algorithm (HSFLA-PS) based adaptive neuro-fuzzy inference system (ANFIS) has been presented in this paper for the solar PV systems applications. The proposed framework is comprised of two stages. The optimal values of the voltages for different values of temperatures and solar irradiances are derived in the first stage by utilizing the SFLA-PS method. After implementing the training process, the ANFIS would give an optimal voltage, taking into account different values of solar irradiance. The INC method will be initialized from this point to search for the maximum power point known as "MPP". The merit of the combinatorial ANFIS and INC method is that it would need a lower number of samples for the training process. The results, obtained from simulating the proposed framework indicate that the combinatorial HSFLA-PS-ANFIS-INC technique would result in the global maxima in various climate conditions at a higher convergence rate and efficiency. (C) 2021 Elsevier Ltd. All rights reserved.