Skip to main navigation Skip to search Skip to main content

Parameter estimation in reliability modeling of distributed detection systems

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

Abstract

Most reliability models are associated with their own parameters which are typically estimated from the history data. For the widely used distributed detection system in fault detection, the system reliability depends on the number of normally working detectors and the accuracy of its local detectors. Parameters of the reliability model of distributed detection system are subject to random variation as the detection system may be used in different purposes and environments. Hence, to evaluate the reliability accurately, it is necessary to obtain the system parameters precisely from the test data we have. In this paper, we present a Bayesian approach to estimate the unknown parameters of distributed detection system from the scarce data and quantify the uncertainty on the system reliability by measure of variance. A simulation is conducted as well to calculate the effect on the system reliability from the uncertainty of the parameters. An example is applied to illustrate the parameter estimation by Bayesian approach. Copyright © 2007 IFAC.
Original languageEnglish
Title of host publicationIFAC Proceedings Volumes (IFAC-PapersOnline)
Pages19-24
Volume1
Publication statusPublished - 2007
Externally publishedYes
Event1st IFAC Workshop on Dependable Control of Discrete Systems, DCDS'07 - Cachan, France
Duration: 13 Jun 200715 Jun 2007

Publication series

Name
Volume1
ISSN (Print)1474-6670

Conference

Conference1st IFAC Workshop on Dependable Control of Discrete Systems, DCDS'07
PlaceFrance
CityCachan
Period13/06/0715/06/07

Research Keywords

  • Bayesian approach
  • Distributed detection system
  • Parameter estimation
  • Reliability analysis
  • Uncertainty

Fingerprint

Dive into the research topics of 'Parameter estimation in reliability modeling of distributed detection systems'. Together they form a unique fingerprint.

Cite this