Fast and Efficient Way of PV Parameters Estimation Based on Combined Analytical and Numerical Approaches
Aissa Hali, Yamina Khlifi
Abstract
Abstract— In this work, a new simpler and more efficient method is proposed to estimate the unknown photovoltaic (PV) parameters of solar cells and PV modules. The proposed method is based on combined analytical and numerical (CAN) approaches. Since the theory and structure of this method are succinct, it can be applied easily. For this study, a single-diode model known as a five parameters model was chosen for modeling the solar cells and PV modules. The proposed technique of PV parameters determination aims to minimize the absolute error between experimental and calculated output current while increasing the speed of convergence to the optimum solution. The accuracy of the suggested approach is tested on a commercial monocrystalline silicon (R.T.C France) solar cell at 33°C and 1000 W/m2 and two PV modules: a commercial Photowatt-PWP 201 in which 36 polycrystalline silicon cells are connected in series with experimental current-voltage characteristic given at 45°C and 1000 W/m2, and an amorphous module referred to as Cocoa aSiMicro03036 with varying environmental conditions. Comprehensive results and statistical analysis indicate that the proposed CAN method is more accurate than most of the published techniques. This accuracy has been proven by lowest statistical errors for all treated data. This good agreement is identified also by the lowest root mean squared error obtained with a value of 7.920179 × 10−4 and 2.083999 × 10−3 for the case RTC France solar cell and Photowatt-PWP 201 PV module respectively, while being lower than 3.948 × 10−4 for the case of Cocoa aSiMicro03036 PV module.