Application Research on Deep Convolution Neural Network Based Fault Diagnosis Technology for Traction Converter

Converter is a key component of traction system in electric locomotive. The fault of converter can easily lead to the paralysis of train operation and is one of the most dangerous failures of electric locomotive. In order to avoid poor generalization of feature selection in expert experience and sim...

Full description

Saved in:
Bibliographic Details
Main Authors: LI Chen, ZHANG Huiyuan, LIU Yong, YANG Weifeng
Format: Article
Language:zho
Published: Editorial Office of Control and Information Technology 2021-01-01
Series:Kongzhi Yu Xinxi Jishu
Subjects:
Online Access:http://ctet.csrzic.com/thesisDetails#10.13889/j.issn.2096-5427.2021.05.010
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849224788200914944
author LI Chen
ZHANG Huiyuan
LIU Yong
YANG Weifeng
author_facet LI Chen
ZHANG Huiyuan
LIU Yong
YANG Weifeng
author_sort LI Chen
collection DOAJ
description Converter is a key component of traction system in electric locomotive. The fault of converter can easily lead to the paralysis of train operation and is one of the most dangerous failures of electric locomotive. In order to avoid poor generalization of feature selection in expert experience and simulation mode in traction converter fault diagnosis, this paper proposes a fault diagnosis method based on deep convolution neural network. Structure parameters in convolutional and pooling layer in Xception model are modified to match the fault data of the traction converter for model training. Experimental results show that the accuracy of Top-1 is 0.842 2 and the accuracy of Top-3 is 0.920 1, which indicates that the proposed method is robust and accurate for fault diagnosis of traction converter, and channels enhancing can provide better generalization for model and realize fault classification.
format Article
id doaj-art-3693382de7d54a85804bcd23b2cdeada
institution Kabale University
issn 2096-5427
language zho
publishDate 2021-01-01
publisher Editorial Office of Control and Information Technology
record_format Article
series Kongzhi Yu Xinxi Jishu
spelling doaj-art-3693382de7d54a85804bcd23b2cdeada2025-08-25T06:53:40ZzhoEditorial Office of Control and Information TechnologyKongzhi Yu Xinxi Jishu2096-54272021-01-0138606582324172Application Research on Deep Convolution Neural Network Based Fault Diagnosis Technology for Traction ConverterLI ChenZHANG HuiyuanLIU YongYANG WeifengConverter is a key component of traction system in electric locomotive. The fault of converter can easily lead to the paralysis of train operation and is one of the most dangerous failures of electric locomotive. In order to avoid poor generalization of feature selection in expert experience and simulation mode in traction converter fault diagnosis, this paper proposes a fault diagnosis method based on deep convolution neural network. Structure parameters in convolutional and pooling layer in Xception model are modified to match the fault data of the traction converter for model training. Experimental results show that the accuracy of Top-1 is 0.842 2 and the accuracy of Top-3 is 0.920 1, which indicates that the proposed method is robust and accurate for fault diagnosis of traction converter, and channels enhancing can provide better generalization for model and realize fault classification.http://ctet.csrzic.com/thesisDetails#10.13889/j.issn.2096-5427.2021.05.010traction converterdeep learningconvolution neural networkfault diagnosis
spellingShingle LI Chen
ZHANG Huiyuan
LIU Yong
YANG Weifeng
Application Research on Deep Convolution Neural Network Based Fault Diagnosis Technology for Traction Converter
Kongzhi Yu Xinxi Jishu
traction converter
deep learning
convolution neural network
fault diagnosis
title Application Research on Deep Convolution Neural Network Based Fault Diagnosis Technology for Traction Converter
title_full Application Research on Deep Convolution Neural Network Based Fault Diagnosis Technology for Traction Converter
title_fullStr Application Research on Deep Convolution Neural Network Based Fault Diagnosis Technology for Traction Converter
title_full_unstemmed Application Research on Deep Convolution Neural Network Based Fault Diagnosis Technology for Traction Converter
title_short Application Research on Deep Convolution Neural Network Based Fault Diagnosis Technology for Traction Converter
title_sort application research on deep convolution neural network based fault diagnosis technology for traction converter
topic traction converter
deep learning
convolution neural network
fault diagnosis
url http://ctet.csrzic.com/thesisDetails#10.13889/j.issn.2096-5427.2021.05.010
work_keys_str_mv AT lichen applicationresearchondeepconvolutionneuralnetworkbasedfaultdiagnosistechnologyfortractionconverter
AT zhanghuiyuan applicationresearchondeepconvolutionneuralnetworkbasedfaultdiagnosistechnologyfortractionconverter
AT liuyong applicationresearchondeepconvolutionneuralnetworkbasedfaultdiagnosistechnologyfortractionconverter
AT yangweifeng applicationresearchondeepconvolutionneuralnetworkbasedfaultdiagnosistechnologyfortractionconverter