Novelty detection methods for online health monitoring and post data analysis of turbopumps

Lei Hu, Niaoqing Hu, Xinpeng Zhang, Fengshou Gu, Ming Gao

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

As novelty detection works when only normal data are available, it is of considerable promise for health monitoring in cases lacking fault samples and prior knowledge. We present two novelty detection methods for health monitoring of turbopumps in large-scale liquid-propellant rocket engines. The first method is the adaptive Gaussian threshold model. This method is designed to monitor the vibration of the turbopumps online because it has minimal computational complexity and is easy for implementation in real time. The second method is the one-class support vector machine (OCSVM) which is developed for post analysis of historical vibration signals. Via post analysis the method not only confirms the online monitoring results but also provides diagnostic results so that faults from sensors are separated from those actually from the turbopumps. Both of these two methods are validated to be efficient for health monitoring of the turbopumps.

Original languageEnglish
Pages (from-to)1933-1942
Number of pages10
JournalJournal of Mechanical Science and Technology
Volume27
Issue number7
DOIs
Publication statusPublished - 1 Jul 2013

Fingerprint

Health
Monitoring
Liquid propellants
Rocket engines
Support vector machines
Computational complexity
Sensors

Cite this

Hu, Lei ; Hu, Niaoqing ; Zhang, Xinpeng ; Gu, Fengshou ; Gao, Ming. / Novelty detection methods for online health monitoring and post data analysis of turbopumps. In: Journal of Mechanical Science and Technology. 2013 ; Vol. 27, No. 7. pp. 1933-1942.
@article{735eae28adc340fea4de42cd3f4f76b1,
title = "Novelty detection methods for online health monitoring and post data analysis of turbopumps",
abstract = "As novelty detection works when only normal data are available, it is of considerable promise for health monitoring in cases lacking fault samples and prior knowledge. We present two novelty detection methods for health monitoring of turbopumps in large-scale liquid-propellant rocket engines. The first method is the adaptive Gaussian threshold model. This method is designed to monitor the vibration of the turbopumps online because it has minimal computational complexity and is easy for implementation in real time. The second method is the one-class support vector machine (OCSVM) which is developed for post analysis of historical vibration signals. Via post analysis the method not only confirms the online monitoring results but also provides diagnostic results so that faults from sensors are separated from those actually from the turbopumps. Both of these two methods are validated to be efficient for health monitoring of the turbopumps.",
keywords = "Gaussian threshold model, Health monitoring, Novelty detection, One-class support vector machine, Turbopump",
author = "Lei Hu and Niaoqing Hu and Xinpeng Zhang and Fengshou Gu and Ming Gao",
year = "2013",
month = "7",
day = "1",
doi = "10.1007/s12206-013-0508-x",
language = "English",
volume = "27",
pages = "1933--1942",
journal = "KSME Journal",
issn = "1738-494X",
publisher = "Korean Society of Mechanical Engineers",
number = "7",

}

Novelty detection methods for online health monitoring and post data analysis of turbopumps. / Hu, Lei; Hu, Niaoqing; Zhang, Xinpeng; Gu, Fengshou; Gao, Ming.

In: Journal of Mechanical Science and Technology, Vol. 27, No. 7, 01.07.2013, p. 1933-1942.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Novelty detection methods for online health monitoring and post data analysis of turbopumps

AU - Hu, Lei

AU - Hu, Niaoqing

AU - Zhang, Xinpeng

AU - Gu, Fengshou

AU - Gao, Ming

PY - 2013/7/1

Y1 - 2013/7/1

N2 - As novelty detection works when only normal data are available, it is of considerable promise for health monitoring in cases lacking fault samples and prior knowledge. We present two novelty detection methods for health monitoring of turbopumps in large-scale liquid-propellant rocket engines. The first method is the adaptive Gaussian threshold model. This method is designed to monitor the vibration of the turbopumps online because it has minimal computational complexity and is easy for implementation in real time. The second method is the one-class support vector machine (OCSVM) which is developed for post analysis of historical vibration signals. Via post analysis the method not only confirms the online monitoring results but also provides diagnostic results so that faults from sensors are separated from those actually from the turbopumps. Both of these two methods are validated to be efficient for health monitoring of the turbopumps.

AB - As novelty detection works when only normal data are available, it is of considerable promise for health monitoring in cases lacking fault samples and prior knowledge. We present two novelty detection methods for health monitoring of turbopumps in large-scale liquid-propellant rocket engines. The first method is the adaptive Gaussian threshold model. This method is designed to monitor the vibration of the turbopumps online because it has minimal computational complexity and is easy for implementation in real time. The second method is the one-class support vector machine (OCSVM) which is developed for post analysis of historical vibration signals. Via post analysis the method not only confirms the online monitoring results but also provides diagnostic results so that faults from sensors are separated from those actually from the turbopumps. Both of these two methods are validated to be efficient for health monitoring of the turbopumps.

KW - Gaussian threshold model

KW - Health monitoring

KW - Novelty detection

KW - One-class support vector machine

KW - Turbopump

UR - http://www.scopus.com/inward/record.url?scp=84880419750&partnerID=8YFLogxK

U2 - 10.1007/s12206-013-0508-x

DO - 10.1007/s12206-013-0508-x

M3 - Article

VL - 27

SP - 1933

EP - 1942

JO - KSME Journal

JF - KSME Journal

SN - 1738-494X

IS - 7

ER -