Farias-Basulto, G. A.; Reyes-Figueroa, P.; Ulbrich, C.; Szyszka, B.; Schlatmann, R.; Klenk, R.: PERFORMANCE EVALUATION AND PARAMETRIZATION OF CIGS THIN FILM SOLAR MODULES THROUGH MULTIPLE LINEAR REGRESSIONS. In: Proceedings of the 37th EU-PVSECMünchen: WIP, 2020. - ISBN 3-936338-73-6, p. 627-632
Abstract:
For optimum design and accurate yield prediction, it is required to know the parameters of a photovoltaic module as a function of module temperature and irradiation. The commonly available values within the data sheet, such as parameters under STC, temperature coefficients at STC and low light performance at a fixed irradiation, will give a first but not very precise estimation of module parameters at arbitrary irradiation and temperature. The accuracy can be improved with algorithms where the module is characterized (indoors or outdoors) at several different combinations of irradiation and temperature and where the data is then represented by a set of parametrized equations (Rosell and Ibáñez, Heydenreich, Sandia). We propose a new algorithm that, in contrast to some of the existing approaches, provides equations not only for the maximum power point (Pmpp) output, but also for open circuit voltage, short circuit current, and current at MPP. It uses only a minimum of initial assumptions concerning the module behavior (black box model) which should make it universally applicable for different solar cell technologies. On the other hand, by comparing the black box equations to analytical descriptions of CIGSSe solar cells, we can cross-link both models and derive device physics parameters from the black box representation. Sets of jV curves have been generated from the analytical description to verify the extraction procedure. In addition, these allow us to study the influence of series and shunt resistance on the accuracy of the black box model. Comparison of model based prediction and actual outdoor data collected over several months in Berlin confirms that the black box model offers a simple and intuitive method to predict location/season specific module performance ratios and to analyze them in depth.