A new framework for improving MPPT algorithms through search space reduction
Ambe Harrison, Cyrille Feudjio, Christophe Raoul Fotso Mbobda, Njimboh Henry Alombah
Abstract
In this article, a novel framework for improving maximum power point tracking (MPPT) is presented. A robust initialization of the MPPT algorithm is implemented, and the search space is reduced by rapidly predicting the maximum power point (MPP) voltage. In order to accomplish this, unknown parameters required to establish an empirical irradiance-free relationship between the open circuit voltage and MPP voltage were derived using three most prevalent PV panel models. The P&O MPPT algorithm implements this framework, which establishes a systematic boundary to confine the MPP voltage to a reduced search space by virtue of this relationship. The evaluations performed on three distinct varieties of commercial PV panels demonstrated that the PV panels modeled in this study exhibit a high degree of conformity with the data provided by the manufacturers. Furthermore, the estimation of MPP voltage allows for a reduction of the search space by as much as 2% of its initial area. The efficacy of the proposed MPPT algorithm is found to be 96.64%, higher than that of the P&O and other implemented contemporary algorithms, when evaluated against a common platform and rapidly changing climatic conditions. Real-world climatic experiments are performed in order to assess the feasibility of the established research. The experimental results indicate that the established irradiance-free MPP voltage has a mean error of merely 0.0128V. Furthermore, the combined output power generated by the MPPT proposed algorithm under experimental conditions is comparatively superior in performance.