Parameter Estimation of Photovoltaic Systems Based on Artificial Intelligence Algorithm

Abstract

The optimum operation of Photovoltaic (PV) systems mainly depends on appropriate modeling of solar cells and optimum approximation of parameters associated with them. Traditionally, mathematical equations based analytical techniques were utilized for parameter extraction, but they involve complex mathematical computations and huge convergence time. To overcome these issues, recently, various Artificial Intelligent (AI) algorithms have been proposed for optimum parameter extraction of PV cells and modules. This chapter provides a comprehensive overview of the implementation of AI techniques and their variants for estimating the parameters of PV cells and modules. Three different categories of AI methodologies, i.e., Machine Learning (ML), Artificial Neural Networks (ANN), and Nature Inspired Optimization Algorithms (NIOAs), are discussed for parameter estimation problems. Due to the vast application of NIOAs in PV cell parameter estimation, they are classified into physics-based, sociology-based, mathematics-based, and biology-based algorithms. For comparison and evaluation of different AI approaches, statistical error values, output of simulated P-V or I-V curves and extracted parameters of PV cell are mentioned against each algorithm. Furthermore, these approaches are summarized comprehensively to guide readers for their proper and optimal implementation in PV cell extraction. Lastly, the study concludes with some perspectives on AI algorithms and recommendations for their future scope in PV systems.