A Day Ahead Hourly Solar Radiation Forecasting by Artificial Neural Networks: A Case Study for Trabzon Province


CEVIK S. , Cakmak R., Altas I. H.

2017 International Artificial Intelligence and Data Processing Symposium (IDAP), Malatya, Turkey, 16 - 17 September 2017 identifier identifier

Abstract

The integration of renewable energy sources to the electrical grid leads to some issues in the grid because of intermittent and variable characteristics of renewable energy sources such as wind and solar. It has been predicted that the solar based electricity production will has highest annual increase rate among the other renewable sources. Due to the intermittent and volatile nature of the solar energy in an electricity grid where photovoltaic systems are intensive, load planning is essential. Therefore, day ahead solar radiation forecasting will contribute to the load planning studies. Artificial neural networks are one of the methods that are applied frequently and successfully in forecasting studies. In this study, a cause effect based artificial neural network (ANN) has been designed and performed for day ahead hourly solar radiation forecasting in Trabzon province. A similar day selection algorithm has been utilized to get more accurate forecasting by the ANN. The designed ANN has been trained and tested in MATLAB simulation environment without using ready codes of MATLAB ANN toolbox. The obtained results reveal that the designed ANN forecasts the solar radiation with acceptable error for a place such as Trabzon which has rainy and cloudy weather conditions.