Please use this identifier to cite or link to this item:
Author(s): Maria Helena Vasconcelos
João Peças Lopes
Title: An hybrid approach based on neural networks and regression Tree Models for fast dynamic security assessment
Issue Date: 2003
Abstract: This paper presents a new hybrid automatic learningapproach, which combines artificial neural networks (ANN) andregression trees (RT), to perform on-line dynamic securityassessment of power systems. In the proposed method, the RT isfirstly used to split the vast amount of knowledge data that describesa security problem into several less spread and disjoint problems.Then, an ANN is trained for each of these new smaller problems,resulting in a tree structure with an ANN predicting functionassociated to each leaf. Moreover, the capability of the RT to performfeature subset selection before ANN training is also tested. With thisnew method, the advantages of the two techniques are exploited inorder to obtained a more accurate model without compromisingprediction time. The quality of the approach is illustrated through itsapplication to a major security problem of the power system ofMadeira Island (Portugal).
Subject: Engenharia
Source: Proceedings of the International Conference on Intelligent Systems Applications to Power Systems, ISAP2003
Document Type: Artigo em Livro de Atas de Conferência Internacional
Rights: openAccess
Appears in Collections:FEUP - Artigo em Livro de Atas de Conferência Internacional

Files in This Item:
File Description SizeFormat 
57930.pdf349.97 kBAdobe PDFThumbnail

This item is licensed under a Creative Commons License Creative Commons