Feature extraction and evaluation for Health Assessment and Failure prognostics.

Kamal Medjaher (1) , Fatih Camci (2) , Noureddine Zerhouni (1)

1 FEMTO-ST - Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174)
2 IVHM - Integrated Vehicle Health Management Centre.

The estimation of Remaining Useful Life (RUL) of industrial equipments can be realized on their most critical components. Based on this assumption, the identified critical component must be monitored to track its health state during its operation. Then, the acquired data are processed to extract relevant features, which are used for RUL estimation. This paper presents an evaluation method for the goodness of the features, extracted from raw monitoring signals, for health assessment and prognostics of critical industrial components. The evaluation method is applied to several simulated datasets as well as features obtained from a particular application on bearings.

Domaines: Automatique / Robotique

Kamal Medjaher, Fatih Camci, Noureddine Zerhouni. Feature extraction and evaluation for Health Assessment and Failure prognostics.. Proceedings of First European Conference of the Prognostics and Health Management Society, PHM-E'12., Jul 2012, Dresden, Germany. pp.111-116. ⟨hal-00718330⟩