Utilize este identificador para referenciar este registo: http://hdl.handle.net/10400.22/1395
Título: Adaptive learning in multiagent systems: a forecasting methodology based on error analysis
Autor: Sousa, Tiago
Pinto, Tiago
Vale, Zita
Praça, Isabel
Morais, H.
Palavras-chave: Adaptive learning
Electricity markets
Error analysis
Forecasting methods
Information theory
Multiagent systems
Data: 2012
Editora: Springer Berlin Heidelberg
Relatório da Série N.º: Advances in Intelligent and Soft Computing; Vol. 156
Resumo: Electricity markets are complex environments, involving a large number of different entities, playing in a dynamic scene to obtain the best advantages and profits. MASCEM is a multi-agent electricity market simu-lator to model market players and simulate their operation in the market. Market players are entities with specific characteristics and objectives, making their decisions and interacting with other players. MASCEM pro-vides several dynamic strategies for agents’ behaviour. This paper presents a method that aims to provide market players strategic bidding capabilities, allowing them to obtain the higher possible gains out of the market. This method uses an auxiliary forecasting tool, e.g. an Artificial Neural Net-work, to predict the electricity market prices, and analyses its forecasting error patterns. Through the recognition of such patterns occurrence, the method predicts the expected error for the next forecast, and uses it to adapt the actual forecast. The goal is to approximate the forecast to the real value, reducing the forecasting error.
URI: http://hdl.handle.net/10400.22/1395
ISBN: 978-3-642-28761-9
978-3-642-28762-6
ISSN: 1867-5662
Versão do Editor: http://link.springer.com/chapter/10.1007/978-3-642-28762-6_42
Aparece nas colecções:ISEP – GECAD – Livro, parte de livro, ou capítulo de livro

Ficheiros deste registo:
Ficheiro Descrição TamanhoFormato 
CAPL_TiagoSousa_2012_GECAD826,96 kBAdobe PDFVer/Abrir    Acesso Restrito. Solicitar cópia ao autor!


FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpace
Formato BibTex MendeleyEndnote Degois 

Todos os registos no repositório estão protegidos por leis de copyright, com todos os direitos reservados.