A unified framework incorporating predictive generative denoising autoencoder and deep Coral network for rolling bearing fault diagnosis with unbalanced data

•A unified framework is proposed for fault diagnosis with unbalanced data.•Predictive generative denoising autoencoder is developed for data generating.•Deep Coral network is used for feature clustering and fault recognition. In practical engineering, data imbalance is an urgent problem to be solved...

Full description

Saved in:
Bibliographic Details
Published in:Measurement : journal of the International Measurement Confederation Vol. 178; p. 109345
Main Authors: Li, Xingqiu, Jiang, Hongkai, Liu, Shaowei, Zhang, Jianjun, Xu, Jun
Format: Journal Article
Language:English
Published: London Elsevier Ltd 01.06.2021
Elsevier Science Ltd
Subjects:
Alignment
Artificial neural networks
Bearings
Datasets
Deep Coral network
Fault diagnosis
Generative model
Neural networks
Noise reduction
Roller bearings
Rolling bearing
Unbalanced data
Deep Coral network
Fault diagnosis
Rolling bearing
Generative model
Unbalanced data
ISSN:0263-2241, 1873-412X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•A unified framework is proposed for fault diagnosis with unbalanced data.•Predictive generative denoising autoencoder is developed for data generating.•Deep Coral network is used for feature clustering and fault recognition. In practical engineering, data imbalance is an urgent problem to be solved for rolling bearing fault diagnosis. This paper proposes a unified framework incorporating predictive generative denoising autoencoder (PGDAE) and deep Coral network (DCN). The proposed framework mainly comprises two parts i.e. generative model and diagnosis model. The generative model PGDAE is used to generate extra fault data, and it is constructed by gated recurrent unit. By this way, unbalanced dataset can reach equilibrium. The diagnosis model DCN is used for fault recognition, which is constructed by a deep convolutional neural network with correlation alignment. Correlation alignment is used to adapts the class-specific features learned from real data and simulated data. Finally, experimental bearing data are used to evaluate the performance. The results show that the generative model can effectively generate workable fault data and the diagnosis model can accurately recognize the fault modes.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0263-2241
1873-412X
DOI:10.1016/j.measurement.2021.109345