• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.86.2009.tde-12082010-213202
Document
Author
Full name
Alcantaro Lemes Rodrigues
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2009
Supervisor
Committee
Grimoni, Jose Aquiles Baesso (President)
Barros, Virginia Parente de
Ramos, Dorel Soares
Title in Portuguese
Redes neurais artificiais aplicadas na previsão de preços do mercado spot de energia elétrica
Keywords in Portuguese
Comercialização de Energia Elétrica
Heurística Computacional
Inteligência Artificial
Redes Neurais Artificiais
Sistema Interligado Nacional
Sistemas de Informação para Apoio a Decisão
Abstract in Portuguese
A comercialização de energia elétrica no Brasil e no mundo sofreu diversas modificações nos últimos 20 anos. Com o objetivo de alcançar o equilíbrio econômico entre oferta e demanda do bem chamado eletricidade, os agentes deste mercado seguem as regras definidas pela sociedade (governo, empresas e consumidores) e também as leis da natureza (hidrologia). Para tratar de problemas tão complexos, estudos são realizados na área da heurística computacional. O objetivo deste trabalho é elaborar um software de previsão de preços do mercado spot utilizando redes neurais artificiais (RNA). As RNA são muito utilizadas em diversas aplicações, principalmente em heurística computacional, nas quais sistemas não lineares apresentam desafios computacionais difíceis de serem superados devido ao efeito da maldição da dimensionalidade. Tal maldição se deve pelo fato do poder computacional atual não ser suficiente para processar problemas com elevada combinação de variáveis. O problema de prever os preços do mercado spot depende de fatores como: (a) a previsão de demanda (carga); (b) a previsão da oferta (reservatórios, regime de chuvas e clima), fator de capacidade; e (c) o equilíbrio da economia (precificação, leilões, influência de mercados externos, política econômica, orçamento governamental, política governamental). Estes fatores são utilizados na construção do sistema de previsão e os resultados de sua eficácia são testados e apresentados.
Title in English
Artificial neural networks applied on the forecast of the spot market prices for electricity.
Keywords in English
Artificial Intelligence
Artificial Neural Networks
Computational Heuristics.
Electric Power Trade
Information Systems for Decision Support
the National Interconnected System
Abstract in English
The commercialization of electricity in Brazil as well as in the world has undergone several changes over the past 20 years. In order to achieve an economic balance between supply and demand of the good called electricity, stakeholders in this market follow both rules set by society (government, companies and consumers) and set by the laws of nature (hydrology). To deal with such complex issues, various studies have been conducted in the area of computational heuristics. This work aims to develop a software to forecast spot market prices in using artificial neural networks (ANN). ANNs are widely used in various applications especially in computational heuristics, where non-linear systems have computational challenges difficult to overcome because of the effect named curse of dimensionality. This effect is due to the fact that the current computational power is not enough to handle problems with such a high combination of variables. The challenge of forecasting prices depends on factors such as: (a) foresee the demand evolution (electric load); (b) the forecast of supply (reservoirs, hydrology and climate), capacity factor; and (c) the balance of the economy (pricing, auctions, foreign markets influence, economic policy, government budget and government policy). These factors are considered be used in the forecasting model for spot market prices and the results of its effectiveness are tested and huge presented.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2010-08-30
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.