Prediction the Power Generated Low-Power Wind Turbine Based on Multilayer Perceptron

The trend towards the development of renewable energy sources, including wind energy and their integration into energy systems leads to the emergence of various problems related to the dependence of such energy sources on weather conditions. Therefore, successful management of energy systems with a high share of wind power in the generating capacity requires the ability to predict their output with sufficiently high accuracy. Currently, various power output prediction methods are being studied in order to achieve the best possible performance, and artificial neural networks are among the most popular in that regard. This paper addresses the task of generation forecasting of an individual wind turbine using real data. The shallow feed forward neural network was created using the Neural Network Toolbox in MATLAB Software. The power generation forecasting results of the considered low- power wind turbine are presented. To evaluate proposed forecasting model performance, the mean squared error metric was used.