Approaches for Developing a Regression Model for Sizing a Stand-Alone Photovoltaic System

This paper proposes a methodology to develop a regression model for sizing a stand-alone photovoltaic (SAPV) system, thereby predicting the operational requirements of an SAPV system for a given location and the load conditions. The application of this methodology is demonstrated by developing a pair of regression models for a particular case of solar irradiation, load, and efficiency, which serve as main inputs to this model. The proposed model is based on the simulated long-term performance of an SAPV system. The validation of the simulation methodology is done by comparing the simulated and experimental values of current distribution pattern of an SAPV system. The comparative results of simulation and experiment results match a confidence level of 92%. The proposed regression model was also compared with a real-time example. This model explicitly helps a designer to evaluate the system requirements for given input data within a predefined tolerance.