A solar PV model parameters estimation based on an improved manta foraging algorithm with dynamic fitness distance balance

Abstract

Accurately simulating and operating photovoltaic (PV) modules or solar cells requires determining specific model parameters based on experimental data. Extracting these parameters is crucial for analyzing system performance under various conditions such as temperature and sunlight variations. However, modeling solar photovoltaic systems is inherently nonlinear, which calls for an efficient algorithm. In this study, we employ the MRFO-dFDB (Manta Ray Foraging Optimization with dynamic Fitness Distance Balance) algorithm, which utilizes fitness distance balance to balance the exploration and exploitation of the search area when assessing parameters in solar PV models. By applying MRFO-dFDB to extract parameters from the STP6-120/36 and Photowatt-PWP201 solar modules, we observe exceptional predictive performance for both single diode (SDM) and double diode (DDM) models. MRFO-dFDB exhibits superior performance compared to state-of-the-art methods. It achieves lower Root-Mean-Square Error (RMSE) values, specifically < 15.3 mA for the STP6-120/36 module and <2.4 mA for the Photowatt-PWP201 module. Additionally, it demonstrates lower maximum errors of 39.02 mA and 5.33 mA, as well as lower power errors of 155.42 mW and 14.122 mW, for the STP6-120/36 and Photowatt-PWP201 solar modules, respectively. Furthermore, it exhibits excellent performance with faster computation speed (< 30.1 seconds in all tests), further emphasizing its superiority.