AN ARTIFICIAL NEURAL-NETWORK APPROACH TO TRANSFORMER FAULT-DIAGNOSIS

Authors
ZHANG Y DING X LIU Y GRIFFIN PJ
Citation
Y. Zhang et al., AN ARTIFICIAL NEURAL-NETWORK APPROACH TO TRANSFORMER FAULT-DIAGNOSIS, IEEE transactions on power delivery, 11(4), 1996, pp. 1836-1841
Citations number
8
Categorie Soggetti
Engineering, Eletrical & Electronic
Journal title
IEEE transactions on power delivery → ACNP
ISSN journal
08858977
Volume
11
Issue
4
Year of publication
1996
Pages
1836 - 1841
Database
ISI
SICI code
0885-8977(1996)11:4<1836:AANATT>2.0.ZU;2-0
Abstract
This paper presents an artificial neural network (ANN) approach to dia gnose and detect faults in oil-filled power transformers based on diss olved gas-in-oil analysis. A two-step ANN method is used to detect fau lts with or without cellulose involved. Good diagnosis accuracy is obt ained with the proposed approach.