Using neural network for the predicition of electricity consumption

Abstract

This thesis deals with prediction of electricity consumption bz using neural networks. The first part deals with the theory of neural networks, in which it refers to the history of neural networks, meaning, use in various other fields and there are some examples of the structure of individual types of neural networks. It further outlines the problems of the current state of prediction of electricity consumption in the Czech Republic and also in general. The third part of this work is a practical example of an application for forecasting electricity consumption for short-time periods. At the end of the work is a global conclusion and an avaluation of the achieved results.