Applied Sciences, Vol. 13, Pages 10013: Solar Power Prediction Modeling Based on Artificial Neural Networks under Partial Shading

Applied Sciences, Vol. 13, Pages 10013: Solar Power Prediction Modeling Based on Artificial Neural Networks under Partial Shading

Applied Sciences doi: 10.3390/app131810013

Authors: Younghyun Lee Jonghwan Lee

Photovoltaic systems are emerging as an important device to address the environmental pollution generated from conventional energy production. The objectives of this study are to accurately predict the power of photovoltaic systems under partial shading conditions and to model high-efficiency photovoltaic systems. First, the power loss under partial shading conditions was addressed using a bypass diode. In previous studies, for the power prediction, one or two parameters were trained through artificial neural networks. In this study, we employ five main parameters to improve the accuracy: the photo-current (Iph), diode saturation current (I0 ), diode idealization factor (n), series resistance (Rs), and shunt resistance (Rsh). Compared to the results of previous studies, the proposed model yielded consistent results. As a result, more accurate power predictions are possible with variations in temperature and irradiation.