Ensemble of constraint handling techniques for PV parameter extraction using differential evolutionary algorithms

Abstract

The depletion of fossil fuels and rising environmental concerns have paved the way for the development of clean renewable energy sources. Photovoltaic (PV) cells are represented by electrical equivalent circuits. Finding the right circuit model parameters for PV cells is critical task. Estimating accurate parameters helps in better performance assessment, control, efficiency calculation and maximum power point tracking. This manuscript describes a new approach for obtaining PV system parameters using ensemble of constraint handling techniques (ECHT) with evolutionary algorithms (EA). Four distinguished technologies of solar PV cells are considered to estimate the parameters with best accuracy. Experiments reveal that ECHT outperforms each individual constraint handling approach by competing with state-of-the-art algorithms. The experimental data for these Kyocera cells is compared with estimated values obtained from the proposed algorithm using MATLAB 2021B for different irradiation. The performance plots show excellent match between the real and simulated values. The root mean square error (RMSE) values for research tax credit RTC France were found to be 7.325513*10-4 and Kyocera processing the normalize RMSE of 0.414%. On comparison with recent algorithms the proposed method achieves the lowest root mean square error (RMSE) meeting the main objective of proposed work.