Improved Deep Transfer Auto-Encoder for Fault Diagnosis of Gearbox Under Variable Working Conditions With Small Training Samples

It is considerable to solve practical fault diagnosis task of gearbox under variable working conditions by introducing sufficient auxiliary data. For this purpose, a new approach called improved deep transfer auto-encoder is proposed for intelligent diagnosis of gearbox faults under variable working...

Full description

Bibliographic Details
Main Authors: Zhiyi He, Haidong Shao, Xiaoyang Zhang, Junsheng Cheng, Yu Yang
Format: Article
Language:English
Published: IEEE 2019-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/8809383/
id doaj-01416544a76c40329302061c7a1facd1
record_format Article
spelling doaj-01416544a76c40329302061c7a1facd12021-03-30T00:23:26ZengIEEEIEEE Access2169-35362019-01-01711536811537710.1109/ACCESS.2019.29362438809383Improved Deep Transfer Auto-Encoder for Fault Diagnosis of Gearbox Under Variable Working Conditions With Small Training SamplesZhiyi He0Haidong Shao1https://orcid.org/0000-0001-7106-0009Xiaoyang Zhang2Junsheng Cheng3Yu Yang4State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, ChinaState Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, ChinaXi’an Aeronautics Computing Technique Research Institute, Xi’an, ChinaState Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, ChinaState Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, ChinaIt is considerable to solve practical fault diagnosis task of gearbox under variable working conditions by introducing sufficient auxiliary data. For this purpose, a new approach called improved deep transfer auto-encoder is proposed for intelligent diagnosis of gearbox faults under variable working conditions with small training samples. First, multi-wavelet is employed as activation function for effectively learning useful features hidden in the non-stationary vibration data. Second, correntropy is used to modify the cost function to enhance the reconstruction quality. Third, pre-train an improved deep auto-encoder using sufficient auxiliary data in the source domain, and transfer its parameters to the target model. Finally, the improved deep transfer should be fine-tuned by small training samples in the target domain to adapt to the characteristics of the rest testing data. The proposed approach is used to analyze two sets of experimental vibration data collected from gearbox under variable working conditions. The results show that the proposed approach can accurately diagnose different faults of gearbox even the working conditions have significant changes, which is superior to the existing methods.https://ieeexplore.ieee.org/document/8809383/Improved deep transfer auto-encodergearbox fault diagnosisvariable working conditionsmulti-wavelet activation functionmodified cost function
collection DOAJ
language English
format Article
sources DOAJ
author Zhiyi He
Haidong Shao
Xiaoyang Zhang
Junsheng Cheng
Yu Yang
spellingShingle Zhiyi He
Haidong Shao
Xiaoyang Zhang
Junsheng Cheng
Yu Yang
Improved Deep Transfer Auto-Encoder for Fault Diagnosis of Gearbox Under Variable Working Conditions With Small Training Samples
IEEE Access
Improved deep transfer auto-encoder
gearbox fault diagnosis
variable working conditions
multi-wavelet activation function
modified cost function
author_facet Zhiyi He
Haidong Shao
Xiaoyang Zhang
Junsheng Cheng
Yu Yang
author_sort Zhiyi He
title Improved Deep Transfer Auto-Encoder for Fault Diagnosis of Gearbox Under Variable Working Conditions With Small Training Samples
title_short Improved Deep Transfer Auto-Encoder for Fault Diagnosis of Gearbox Under Variable Working Conditions With Small Training Samples
title_full Improved Deep Transfer Auto-Encoder for Fault Diagnosis of Gearbox Under Variable Working Conditions With Small Training Samples
title_fullStr Improved Deep Transfer Auto-Encoder for Fault Diagnosis of Gearbox Under Variable Working Conditions With Small Training Samples
title_full_unstemmed Improved Deep Transfer Auto-Encoder for Fault Diagnosis of Gearbox Under Variable Working Conditions With Small Training Samples
title_sort improved deep transfer auto-encoder for fault diagnosis of gearbox under variable working conditions with small training samples
publisher IEEE
series IEEE Access
issn 2169-3536
publishDate 2019-01-01
description It is considerable to solve practical fault diagnosis task of gearbox under variable working conditions by introducing sufficient auxiliary data. For this purpose, a new approach called improved deep transfer auto-encoder is proposed for intelligent diagnosis of gearbox faults under variable working conditions with small training samples. First, multi-wavelet is employed as activation function for effectively learning useful features hidden in the non-stationary vibration data. Second, correntropy is used to modify the cost function to enhance the reconstruction quality. Third, pre-train an improved deep auto-encoder using sufficient auxiliary data in the source domain, and transfer its parameters to the target model. Finally, the improved deep transfer should be fine-tuned by small training samples in the target domain to adapt to the characteristics of the rest testing data. The proposed approach is used to analyze two sets of experimental vibration data collected from gearbox under variable working conditions. The results show that the proposed approach can accurately diagnose different faults of gearbox even the working conditions have significant changes, which is superior to the existing methods.
topic Improved deep transfer auto-encoder
gearbox fault diagnosis
variable working conditions
multi-wavelet activation function
modified cost function
url https://ieeexplore.ieee.org/document/8809383/
work_keys_str_mv AT zhiyihe improveddeeptransferautoencoderforfaultdiagnosisofgearboxundervariableworkingconditionswithsmalltrainingsamples
AT haidongshao improveddeeptransferautoencoderforfaultdiagnosisofgearboxundervariableworkingconditionswithsmalltrainingsamples
AT xiaoyangzhang improveddeeptransferautoencoderforfaultdiagnosisofgearboxundervariableworkingconditionswithsmalltrainingsamples
AT junshengcheng improveddeeptransferautoencoderforfaultdiagnosisofgearboxundervariableworkingconditionswithsmalltrainingsamples
AT yuyang improveddeeptransferautoencoderforfaultdiagnosisofgearboxundervariableworkingconditionswithsmalltrainingsamples
_version_ 1724188297938337792