Detecção de falhas com PCA e PLS aplicados a uma planta didática

• Ano de defesa: 2014
• Autor(a) principal: Britto, Rodrigo da Silva
• Orientador(a): Lima, Leocarlos Bezerra da Silva
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Programa de Pós-Graduação: Pós-Graduação em Engenharia Elétrica
• Palavras-chave em Português: Engenharia elétrica; Automação industrial; Controle de processo; Sistema de monitoramento de falhas; Análise de Componentes Principais (ACP); Mínimos Quadrados Parciais (PLS); Principal Component Analysis (PCA); Métodos estatísticos
• Palavras-chave em Inglês: Electric engineering; Process control
• Área do conhecimento CNPq: CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://ri.ufs.br/handle/riufs/5012

A fault monitoring system in general, which, beside the detection, isolation, diagnosis and fault recuperation steps is a research area of great interest, since the fault occurrence may lead to negative consequences on different levels on social, economical and environmental bases. With the increasing complexity of the industrial process, it is often necessary a quick detection leading to an optimized fault management system and therefore avoiding the loss of material and human resources. This work develops a study on fault detection statistical techniques applied to a didactic plant. The didactic plant deployed in this study comprises a controlled simple industrial process. For the fault detection in this process it were applied the main statistical methods: Principal Component Analysis (PCA) and Partial Least Squares (PLS). Those methods were implemented and applied on the process aiming a comparative analysis concerning themselves. As a result, the methods were able to detect every kind of emulated fault, with little or none detection delay and with similar performances.