Evolutionary Bayesian fusion for transformers fault detection

Cui, Yi, Naranpanawe, Lakshitha and Seo, Junhyuck (2016). Evolutionary Bayesian fusion for transformers fault detection. In: Australasian Universities Power Engineering Conference-2016 (AUPEC 2016), Brisbane, Australia, (). 25-28 September 2016.

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
Author Cui, Yi
Naranpanawe, Lakshitha
Seo, Junhyuck
Title of paper Evolutionary Bayesian fusion for transformers fault detection
Conference name Australasian Universities Power Engineering Conference-2016 (AUPEC 2016)
Conference location Brisbane, Australia
Conference dates 25-28 September 2016
Place of Publication Piscataway, NJ, United States
Publisher IEEE
Publication Year 2016
Sub-type Fully published paper
ISBN 9781509014057
Total pages 6
Collection year 2017
Language eng
Formatted Abstract/Summary
This paper presents an evolutionary Bayesian fusion method for transformer fault detection. It adopts Bayesian Network (BN) to explore the causal relationship between different potential faults inside transformer and fault symptoms; and then such knowledge is used to identify the fault types of transformer. Since Bayesian network acquires fault evidence gradually and transformer fault diagnosis is also an evolutionary process, the proposed method determines an optimal set of measurements, which need to be performed at each diagnostic step. This methodology can improve the accuracy of transformer fault identification while improve the efficiency of diagnostic process since the number of required measurements is minimized and only meaningful fault evidences are used in fault identification. Case studies are presented to verify the proposed method.
Keyword Bayesian network
Data and information fusion
Maximum posteriori probability estimation
Multiple source
Fault detection
Transformer
Q-Index Code EX
Q-Index Status Provisional Code
Institutional Status UQ

 
Versions
Version Filter Type
Citation counts: Google Scholar Search Google Scholar
Created: Sat, 23 Jul 2016, 03:40:41 EST by Yi Cui on behalf of School of Information Technol and Elec Engineering