Portal de Periódicos da CAPES

Capítulo de livro Produção Nacional Revisado por pares

BIBTEX RIS

Turbo Machinery Failure Prognostics

2014; Springer Science+Business Media; Linguagem: Inglês

10.1007/978-3-319-07455-9_37

1611-3349

Inhaúma Neves Ferraz, Ana Cristina Bicharra García,

Machine Fault Diagnosis Techniques

The turbomachinery systems on oil platforms are extremely sensitive, as their failure can cause total shutdown of the platform's production activity. Scheduled stoppages for preventive maintenance are effective, since qualified personnel and replacement parts are available and ready. On the other hand, shutting down a turbo machine ceases petroleum production. Not rarely, maintenance people conclude the machine could have worked much longer. Monitoring the machines' behaviors to predict the best time to stop has been an industry trend. Failure prognosis has been performed using statistical methods and Artificial Intelligence techniques, such as Neural Networks, Fuzzy Systems and Expert Systems. This study presents the results obtained by developing an expert system based on case-based reasoning for turbomachinery failure prognostics.