Ranking teaching–learning-based optimization algorithm to estimate the parameters of solar models

Abstract

As one of the most promising renewable energies, solar energy can be converted to electricity through photovoltaic (PV) systems. It is indispensable to identify the parameters of PV systems with the aim of controlling and simulating. Thanks to the complexity of PV systems, parameter identification is still a challenging task. In this paper, we develop a Ranking Teaching–Learning-Based​ Optimization (RTLBO) to solve the problem, in which Teaching–Learning-Based​ Optimization (TLBO) is a population-based swarm algorithm and mimics the learning process in a classroom. RTLBO ranks learners into superior and inferior groups, in which the outstanding learners learn from the top three agents to boost the local search. In contrast, the low learners learn from each other by guidance. The two phases are in parallel to balance the local and global search. The proposed RTLBO is used to extract parameters of different models, including the single diode model, double diode model and three PV module models. TLBO, four TLBO variants, and fifteen meta-heuristic algorithms are selected as the rivals of RTLBO. Several experiments have shown that our method is a reliable and effective algorithm when addressing the parameters of PV systems.