Forecasting Wind Power Generation Using Artificial Neural Network

Abstract

Today, among renewable energy sources, wind energy is used effectively as a clean and sustainable energy source in electricity generation. The uncertain nature of renewable energy sources and the smart ability of the neural network approach to process complex time series inputs have allowed the use of artificial neural network (ANN) methods in the prediction of renewable energy generation. In this study, the speed and power of wind turbines and electricity generation were estimated from wind speed data using artificial neural networks. In our calculations, the real wind speed data were used in the test phase, and the speed-power data of six different types of wind turbines were used in the training phase. It has been shown that the predictions made by our ANN model from the regression curves of the training, validation, and test data obtained are quite successful and reliable. According to our results, it has been understood that the wind potential of our selected region is good enough and that the electrical energy need for this region can be met from wind energy by using the appropriate wind turbine type, so it is quite appropriate to invest in wind energy.