WIT Press


The Prediction Of Electric Energy Consumption Using An Artificial Neural Network

Price

Free (open access)

Paper DOI

10.2495/EQ140121

Volume

190

Pages

9

Page Range

109 - 117

Published

2014

Size

382 kb

Author(s)

I. Chernykh, D. Chechushkov & T. Panikovskaya

Abstract

This paper presents the results of the studies on forecasting the electrical loads for a megapolis district with the use of artificial neural networks (ANN) as one of the most accomplished and promising solutions to this challenge. A theoretical approach to the issue is combined with the results of experimental studies using real schedules. Keywords: artificial neural network, electrical loads, updated input data. 1 Introduction The transition to the wholesale electricity market [1] has exacerbated the problems associated with the accuracy of forecasting electricity demand by tightening the requirements for compiling speed and the reliability of forecasts. New conditions associated with the introduction of heavy penalties require the introduction of modern electricity consumers of software and hardware for the collection of information on power consumption, for making accurate forecasts using flexible models and for reacting appropriately to changing trends. The relevance of research in this area is shown in the application of the results to the operational and tactical management of power consumption that can reduce production costs and, as a consequence, increase the company's competitiveness by reducing production costs. Thus for the entity acquiring electricity on the wholesale and/or retail market, there is the task of compiling reliable applications for electricity consumption for a certain period ahead. The relevance of this research is supported by the peculiarities of distributed Energy Production and Management in the 21st Century, Vol. 1 109 www.witpress.com, ISSN 1743-3541 (on-line) WIT Transactions on Ecology and The Environment, Vol 190, © 2014 WIT Press doi:10.2495/EQ140121

Keywords

artificial neural network, electrical loads, updated input data