Validation of a multiple linear regression model for CIGSSe photovoltaic module performance and Pmpp prediction

This work presents the validation of a heuristic model, which predicts the electrical characteristics of CIGSSe thin film solar modules. This model is based on four-coefficient equations, used to determine electrical parameters from photovoltaic devices such as open circuit voltage, short circuit current, current at maximum power point and maximum power point. The coefficients are obtained numerically by fitting these equations to measured datasets related to various irradiances and module temperatures. These four coefficients or predictors per parameter can then be used to calculate a parameter at different conditions. The datasets employed in this work were obtained from thin film CIGSSe modules, measured under both controlled laboratory and operating outdoor conditions. The validation of the model is performed by comparing the presented approach to well-known established models and methods for module power rating including the international standards IEC and SAPM. The comparison is performed using statistical analysis, comparing the deviation between the predicted and the measured output power. Furthermore, the possibility of evaluating the temperature coefficients through this model is also explored. The proposed model has been applied and validated yielding high correlation coefficients for CIGSSe modules for energy rating, power output forecasting and temperature coefficient calculation.