The Potential of Artificial Neural Network (ANN) Technique in Photovoltaic Power Output Modeling Based on Meteorological Data

Rusul S. Hadi†, Osamah F. Abdulateef‡*

†College of Engineering, Energy Engineering Department, University of Baghdad, Baghdad, Iraq.

‡AlKhwarizmi College of Engineering, Automated Manufacturing Engineering Department, University of Baghdad, Baghdad, Iraq.

*Corresponding Author Email: drosamah@kecbu.uobaghdad.edu.iq, drosamah65@gmail.com

This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

It has become critical to estimate the effect of ambient conditions on the photovoltaic power output for optimal designing of the PV systems. For this purpose, several techniques have already been used. In the present study, we have presented a solar power modeling technique using the general regression neural network (GRNN) to model a PV output power and to approximate the generated power on the basis of the ambient conditions such as the ambient and cell temperature, wind speed, humidity, and irradiance. We considered the effect of each climate factor on the PV power output estimation using the feature-selection process with 4 combinations of input variables. All data used in this study was sourced from the experimental research conducted in Baghdad city between January 2018 and May 2018. The data of 5 months were used for training and testing the neural network. The simulation results indicated that the 4 combinations of input variables were valid and showed good modeling performance, albeit the accuracy of the GRNNmodel with the 5 and 4 input factors (ignoring the wind speed factor), achieved maximum performance for the estimation of the PV power output with the coefficients of determination (R-Squared) of 0.98533 and 0.98796, respectively.