Novel single diode solar cell nonlinear model: Optimization and validation

Abstract

This paper focuses on nonlinear modeling of solar cells. In contrast to traditional methods that assume that the series and parallel resistances of solar cells are either constant or linearly dependent on the supply voltage, this paper introduces five nonlinear forms of this dependence. For each proposed form, the paper derives an analytical formula for the current expressed through the Lambert W function. To estimate the parameters of solar cells, a hybrid variant of the Pelican Optimization Algorithm (POA) and chaotic sequences, marked as C-POA, is proposed. The efficiency and applicability of the proposed algorithm, as well as the rationale for using the proposed nonlinear models of solar cells, were tested by observing the RTC France solar cell and two modules, the MSX 60 and the Photowat PWP 201 solar module. The justification for applying the proposed approach (new algorithm and proposed solar cell models) was also tested on the KC200GT solar module for different weather conditions. The numerical results indicate that the accuracy of solar cell modeling, as expressed through the root mean square error (RMSE) value, can be significantly improved by applying any of the five proposed models, sometimes by more than 50%. The calculation accuracy is significantly better than that of the most accurate model, i.e., the three-diode model of solar cells. Therefore, the presented research offers a new perspective in solar cell modeling.