Review on Health Indices Extraction and Trend Modeling for Remaining Useful Life Estimation - Archive ouverte HAL Accéder directement au contenu
Chapitre D'ouvrage Artificial Intelligence Techniques for a Scalable Energy Transition Année : 2020

Review on Health Indices Extraction and Trend Modeling for Remaining Useful Life Estimation

Samir Benmoussa
  • Fonction : Auteur
  • PersonId : 1090800
Enrico Zio

Résumé

Because of the growing demands of equipment availability, performance, and maintenance, the scientific community has been developing methods for forecasting failures, and for the estimation of the Remaining Useful Life (RUL) for scheduling Condition-Based Maintenance (CBM) and Predictive Maintenance (PM). The National Aeronautics and Space Administration (NASA) was among the first to work on prognosis, because in the aerospace field, prognosis of failure can avoid catastrophes. Performance evaluation is a key element in fault prognosis and several methods have been proposed, based on different evaluation criteria. The work presented in [63, 100, 101] goes in the direction of a standardization of these criteria and proposes performance metrics applicable to different methods of fault prognosis. These metrics allow, on the one hand, establishing design requirements by quantifying acceptable performance limits and on the other hand, comparing different methods. In [101], a structured synthesis of the used metrics for the evaluation of the performance of fault prognosis methods that adapt to different application domains is presented, including Prognosis Horizon (PH), M. A. Djeziri ()
Fichier non déposé

Dates et versions

hal-03518990 , version 1 (02-03-2022)

Identifiants

Citer

Mohand Arab Djeziri, Samir Benmoussa, Enrico Zio. Review on Health Indices Extraction and Trend Modeling for Remaining Useful Life Estimation. Remaining Useful Life estimation without needing for prior knowledge of the degradation features, pp.183 - 223, 2020, ⟨10.1007/978-3-030-42726-9_8⟩. ⟨hal-03518990⟩
19 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More