Enhancing power generation via an adaptive neuro-fuzzy and backstepping maximum-power-point tracking approach for photovoltaic single-ended primary inductor converter system

The distinct characteristics of photovoltaic generators related to power and current present a complex problem in terms of optimizing their power output. To tackle this, maximum-power-point tracking techniques such as the adaptive neuro-fuzzy inference system are frequently utilized for their swift adaptability and reduced fluctuations. In addition, the backstepping controller is often selected to handle both linear and non-linear systems due to its exceptional reliability. The purpose of this research is to propose an innovative method that merges the adaptive neuro-fuzzy inference system and backstepping controller to refine the tracking of the optimal power point and to bolster the stability of the photovoltaic system in the face of unpredictable scenarios, such as those presented by the Ropp irradiance examination, which utilizes a single-ended primary inductor converter as a stage for power electronics adaptation. Simulations conducted using MATLAB^®/Simulink^® demonstrate that the combination of adaptive neuro-fuzzy inference system and backstepping controller achieves an impressive efficiency of 99.6% and exhibits fast, robust, and accurate responses compared with other algorithms such as artificial neural networks combined with the backstepping controller and conventional perturb and observe algorithm.