Neural Network a Way to Forecast the Generation of Electricity From a Wind Turbine

Abstract

This report reviews the main factors influencing the performance and productivity of wind energy sources. These sources function mainly as mains-connected, which affects both the quality of the voltage and the reliability of the electrical networks to which they are connected. It is essential for the realization of uninterrupted power supply to perform an accurate forecast analysis of the electricity produced by these sources. For the purposes of forecasting, an artificial intelligence device is presented. The purpose of the report is defined, namely to create a neural network through which to forecast the electricity produced by a wind generator. The operability of the developed model was assessed. To achieve this goal, an interactive system for numerical calculations, analysis and graphs - MATLAB / Toolbox / NNtool was used. The article is structured as follows: The first and second parts of the report present the problem related to forecasting the production of electricity from wind sources. A modern, possible solution to the problem by using a neural network is presented. In the third part of the report, a neural network for predicting the production of electricity from a wind turbine has been set up. The main steps in creating the network are described, the accuracy of its work is assessed. In the last part, based on the research, conclusions are drawn.