Wander Valente
IFET - sudeste de minas gerais, Engenharia Mecatronica, Faculty Member
ABSTRACT This paper proposes a new approach to address the complex problem of the internal faults classification in power transformers. The fault classification is made using an algorithm based on a geometric approach to construct an... more
ABSTRACT This paper proposes a new approach to address the complex problem of the internal faults classification in power transformers. The fault classification is made using an algorithm based on a geometric approach to construct an artificial neural network. The input parameters are related to the concentrations of various gases dissolved in insulating oil. The proposed technique is able to generate both the topology and the weight for each neuron without specifying initial network parameters. Datasets of dissolved gas concentration in high-voltage transformers that have been found in the scientific literature are used to validate the methodology. The results indicate a high accuracy rate compared to other important computational intelligence methods used to high-voltage transformers faults classification.
ABSTRACT This paper presents an architecture for a provenance considering the domain of scientific experiments performed by computer simulation. The objective of the model is to capture and manage information from collabora-tive research... more
ABSTRACT This paper presents an architecture for a provenance considering the domain of scientific experiments performed by computer simulation. The objective of the model is to capture and manage information from collabora-tive research environments interconnected by a grid computing. Furthermore, the model must operate independently of any technology that provides support for workflow execution. The proposed architecture is based on standard fea-tures such as OPM, web services and ontologies. Resumo. Este artigo apresenta uma proposta de arquitetura para um sistema de proveniência no domínio dos experimentos científicos executados através de simulação por computador. O objetivo do modelo projetado consiste em capturar e gerenciar as informações de proveniência em ambientes colabora-tivos de pesquisa interconectados através de uma grade computacional. Além disso, o modelo deve atuar de forma independente de qualquer tecnologia que provê suporte à execução de workflows. A arquitetura proposta baseia-se em recursos como o padrão OPM, serviços web e ontologias.