A copula-based sampling method for data-driven prognostics

Zhimin Xi, Rong Jing, Pingfeng Wang, Chao Hu

Research output: Contribution to journalArticlepeer-review

Abstract

This paper develops a Copula-based sampling method for data-driven prognostics. The method essentially consists of an offline training process and an online prediction process: (i) the offline training process builds a statistical relationship between the failure time and the time realizations at specified degradation levels on the basis of off-line training data sets; and (ii) the online prediction process identifies probable failure times for online testing units based on the statistical model constructed in the offline process and the online testing data. Our contributions in this paper are three-fold, namely the definition of a generic health index system to quantify the health degradation of an engineering system, the construction of a Copula-based statistical model to learn the statistical relationship between the failure time and the time realizations at specified degradation levels, and the development of a simulation-based approach for the prediction of remaining useful life (RUL). Two engineering case studies, namely the electric cooling fan health prognostics and the 2008 IEEE PHM challenge problem, are employed to demonstrate the effectiveness of the proposed methodology.

Original languageEnglish (US)
Pages (from-to)72-82
Number of pages11
JournalReliability Engineering and System Safety
Volume132
DOIs
StatePublished - Dec 2014
Externally publishedYes

Keywords

  • COPULA
  • Data-driven prognostics
  • Prognostics and health management (PHM)
  • Reliability
  • Remaining useful life

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Industrial and Manufacturing Engineering

Fingerprint

Dive into the research topics of 'A copula-based sampling method for data-driven prognostics'. Together they form a unique fingerprint.

Cite this