Transformer fault diagnosis based on the improved D-S evidence theory

To overcome the bad performance of marine transformer fault diagnosis, which induced by the reliability of symptom and the correlation with each other. Based on improved D-S evidence theory, a novel marine transformer fault diagnosis method is proposed in this paper. Firstly, the fault diagnosis mod...

Full description

Saved in:
Bibliographic Details
Published in2009 International Conference on Machine Learning and Cybernetics Vol. 6; pp. 3413 - 3417
Main Authors Yan-Dong Hou, Xiao-Jun Wang, Cheng-Lin Wen, Wei-Xing Xu
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.07.2009
Subjects
Automation
Correlation
Cybernetics
D-S evidence theory
Fault diagnosis
Information entropy
Machine learning
Oil insulation
Power system reliability
Power transformer insulation
Power transformers
Reliability
Uncertainty
Online AccessGet full text

Cover

More Information
Summary:To overcome the bad performance of marine transformer fault diagnosis, which induced by the reliability of symptom and the correlation with each other. Based on improved D-S evidence theory, a novel marine transformer fault diagnosis method is proposed in this paper. Firstly, the fault diagnosis model is established based on traditional D-S evidence theory. Secondly, the effective measuring for the reliability of symptom and the correlation with each other is sloved by information entropy and information energy. Finally, a numerical example is given to illustrate the effectiveness of the proposed method.
ISBN:9781424437023
1424437024
ISSN:2160-133X
DOI:10.1109/ICMLC.2009.5212787