Trabalho em anais de eventoOpen Access
Prognostics of Rolling Bearings based on Neuro-Fuzzy Evolving Models
SILVA FILHO, C. D.; VITORIO, A. E.; AYRES JUNIOR, F. A. C.; OTANI, M.; MEDEIROS, J. N.; NUNES, R. C.; SILVA, R.; VARELA, J. A.; MOURA, D. R.; SANTOS, K. V.; COSTA, J. B.; MENDONCA, R. S.; BESSA, I. V
Resumo
Resumo
This paper presents a novel approach to data-driven fault prognostics for rolling bearings. The proposed method is based on an evolving neuro-fuzzy model capable of incrementally learning representations of nonlinear and time-varying behaviors. The proposed evolving model is particularly well-suited for modeling the complex, nonlinear, and non-uniform nature of degradation processes. In addition, granular computing is incorporated to quantify uncertainty in the remaining useful life (RUL) predictions based on the principle of justifiable granularity. Experimental results demonstrate that the proposed method offers a promising solution for accurate and interpretable fault prognostics of rolling bearings.
palavras-chave
Palavras-chave
rolling bearingsfault prognosisremaining useful life predictiongranular computinguncertainty quantificationdata-driven degradation modelingnonlinear time-varying systemsjustifiable granularity
Status
Published
links
Links e arquivos
Trabalho em anais de evento