Novel RUL prediction of assets based on the integration of auto-regressive models and an RUSBoost classifier

Georgios Fagogenis, David Flynn, David Michael Lane

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents a novel, data-driven algorithm for the computation of the Remaining Useful Life (RUL) of an asset. The algorithm utilizes the asset’s state history to learn a prognostic model from data. The prognostic model comprises an ensemble of Auto-Regressive (AR) models, together with a state- of-the-art classifier. The AR part of the algorithm is used to predict the system’s state evolution. The classifier discriminates between healthy and faulty operation, given the asset’s current state. The predicted state, as computed by the AR model, is fed to the classifier. The first time when the predicted state is classified as faulty is returned as the RUL of the system. The resulting prognostic algorithm was tested on the CMAPSS dataset as provided from NASA Ames Research Center. Cases of unknown future input trajectory as well as cases with multiple faults have been investigated.
Original languageEnglish
Title of host publication2014 IEEE Conference on Prognostics and Health Management (PHM)
PublisherIEEE
ISBN (Print)9781479959426
DOIs
Publication statusPublished - 9 Feb 2015
Event2014 International Conference on Prognostics and Health Management - Cheney, United States
Duration: 22 Jun 201425 Jun 2014

Conference

Conference2014 International Conference on Prognostics and Health Management
Abbreviated titlePHM 2014
CountryUnited States
CityCheney
Period22/06/1425/06/14

Fingerprint

Classifiers
United States National Aeronautics and Space Administration
NASA
History
Trajectories
Research

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Computer Science Applications
  • Software
  • Health Information Management

Cite this

Fagogenis, Georgios ; Flynn, David ; Lane, David Michael. / Novel RUL prediction of assets based on the integration of auto-regressive models and an RUSBoost classifier. 2014 IEEE Conference on Prognostics and Health Management (PHM). IEEE, 2015.
@inproceedings{41fcb87264fa48a28e09db01129db6de,
title = "Novel RUL prediction of assets based on the integration of auto-regressive models and an RUSBoost classifier",
abstract = "This paper presents a novel, data-driven algorithm for the computation of the Remaining Useful Life (RUL) of an asset. The algorithm utilizes the asset’s state history to learn a prognostic model from data. The prognostic model comprises an ensemble of Auto-Regressive (AR) models, together with a state- of-the-art classifier. The AR part of the algorithm is used to predict the system’s state evolution. The classifier discriminates between healthy and faulty operation, given the asset’s current state. The predicted state, as computed by the AR model, is fed to the classifier. The first time when the predicted state is classified as faulty is returned as the RUL of the system. The resulting prognostic algorithm was tested on the CMAPSS dataset as provided from NASA Ames Research Center. Cases of unknown future input trajectory as well as cases with multiple faults have been investigated.",
author = "Georgios Fagogenis and David Flynn and Lane, {David Michael}",
note = "INSPEC Accession Number: 14920730",
year = "2015",
month = "2",
day = "9",
doi = "10.1109/ICPHM.2014.7036373",
language = "English",
isbn = "9781479959426",
booktitle = "2014 IEEE Conference on Prognostics and Health Management (PHM)",
publisher = "IEEE",
address = "United States",

}

Fagogenis, G, Flynn, D & Lane, DM 2015, Novel RUL prediction of assets based on the integration of auto-regressive models and an RUSBoost classifier. in 2014 IEEE Conference on Prognostics and Health Management (PHM)., 7036373, IEEE, 2014 International Conference on Prognostics and Health Management, Cheney, United States, 22/06/14. https://doi.org/10.1109/ICPHM.2014.7036373

Novel RUL prediction of assets based on the integration of auto-regressive models and an RUSBoost classifier. / Fagogenis, Georgios; Flynn, David; Lane, David Michael.

2014 IEEE Conference on Prognostics and Health Management (PHM). IEEE, 2015. 7036373.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Novel RUL prediction of assets based on the integration of auto-regressive models and an RUSBoost classifier

AU - Fagogenis, Georgios

AU - Flynn, David

AU - Lane, David Michael

N1 - INSPEC Accession Number: 14920730

PY - 2015/2/9

Y1 - 2015/2/9

N2 - This paper presents a novel, data-driven algorithm for the computation of the Remaining Useful Life (RUL) of an asset. The algorithm utilizes the asset’s state history to learn a prognostic model from data. The prognostic model comprises an ensemble of Auto-Regressive (AR) models, together with a state- of-the-art classifier. The AR part of the algorithm is used to predict the system’s state evolution. The classifier discriminates between healthy and faulty operation, given the asset’s current state. The predicted state, as computed by the AR model, is fed to the classifier. The first time when the predicted state is classified as faulty is returned as the RUL of the system. The resulting prognostic algorithm was tested on the CMAPSS dataset as provided from NASA Ames Research Center. Cases of unknown future input trajectory as well as cases with multiple faults have been investigated.

AB - This paper presents a novel, data-driven algorithm for the computation of the Remaining Useful Life (RUL) of an asset. The algorithm utilizes the asset’s state history to learn a prognostic model from data. The prognostic model comprises an ensemble of Auto-Regressive (AR) models, together with a state- of-the-art classifier. The AR part of the algorithm is used to predict the system’s state evolution. The classifier discriminates between healthy and faulty operation, given the asset’s current state. The predicted state, as computed by the AR model, is fed to the classifier. The first time when the predicted state is classified as faulty is returned as the RUL of the system. The resulting prognostic algorithm was tested on the CMAPSS dataset as provided from NASA Ames Research Center. Cases of unknown future input trajectory as well as cases with multiple faults have been investigated.

UR - http://www.scopus.com/inward/record.url?scp=84929598476&partnerID=8YFLogxK

U2 - 10.1109/ICPHM.2014.7036373

DO - 10.1109/ICPHM.2014.7036373

M3 - Conference contribution

SN - 9781479959426

BT - 2014 IEEE Conference on Prognostics and Health Management (PHM)

PB - IEEE

ER -