Extracting photovoltaic cells parameters for three diode model using HSDA algorithm

Developing the metaheuristic algorithms directly for extracting the photovoltaic cells and panels (PV) parameters or the implementation of algorithms for PV already used for other problems conduct to accurately obtain parameters and, generally, in a short time. The objective of the paper is to accurately extract the parameters of the photovoltaic cells and panels. To obtain this the hybrid successive discretization algorithm (HSDA) is applied to the triple diode model (TDM) for the extraction of photovoltaic cells and panels parameters. Using HSDA, nine parameters are extracted for two photovoltaic cells and four photovoltaic panels. The HSDA algorithm results for root mean square errors (RMSE) and mean absolute errors (MAE) are compared with other ones obtained with the best algorithms from research literature for prove the performance of the algorithm considered. The comparison proves that the HSDA algorithm has lower or equal RMSE. Therefore, the decrease in RMSE varies from 0.4 % to 38.5 %, in the case of the RTC France silicon photovoltaic cell (La Radiotechnique Compele) and for the Photowatt - PWP201 photovoltaic panel the variation is from 0.01 % to 813 %. The RMSE for STM6–40 is around 4 times lower. The root mean square error decreases 4 times for Kyocera KC200GT and for PVM 752 GaAs the decrease varies between 33 % and 6 times. The comparison between the models used for photovoltaic cells shows that the TDM has given lower values for RMSE and MAE proving it is the best solution to analyze them.