Performance Predictions Accuracy of a Smart Sun Tracking System for Photovoltaic Applications

Abstract

Despite technological innovations, solar energy efficiency is still low nowadays. It thus needs to be optimized. To improve solar photovoltaic (PV) modules efficiency, several methods have been designed and implemented. Such methods are the Maximum Power Point Tracking (MPPT) and Solar Tracking (ST). In this decade, the solar tracking method is facing a great challenge: the minimization of the electrical energy consumption of electromechanical systems used to track the sun’s position. Manufacturers must be able to predict the net electrical power output obtained from a tracking PV panel to save energy and time. The purpose of the present work is to simulate the performance of a smart solar tracker and assess the accuracy of predictions. To achieve this objective, the one diode model of solar PV cell has been used. In an innovative approach, the one diode model has been simulated using experimental data of temperature and global solar irradiance which are easily available on the internet. The accuracy of predicted values was analysed using reliable statistical indicators such as the Mean Absolute Error (MAE), the Root Mean Square Error (RMSE), the Coefficient of determination (R-square) and the Pearson Coefficient of Correlation (PCC). Simulation results of the power output of the tracking PV panel is better than the one of the fixed panel with 0.25, 0.32, 0.66, and 0.81 respectively for the MAE, the RMSE, the R-square and the PCC. These results show that predictions are in good agreement with experimental measures.