Transfer Learning-Based Fault Diagnosis under Data Deficiency

被引:21
作者
Cho, Seong Hee [1 ]
Kim, Seokgoo [1 ,2 ]
Choi, Joo-Ho [3 ]
机构
[1] Korea Aerosp Univ, Dept Aerosp & Mech Engn, Goyang 10540, South Korea
[2] Univ Florida, Dept Mech & Aerosp Engn, Gainesville, FL 32611 USA
[3] Korea Aerosp Univ, Sch Aerosp & Mech Engn, Goyang 10540, South Korea
来源
APPLIED SCIENCES-BASEL | 2020年 / 10卷 / 21期
基金
新加坡国家研究基金会;
关键词
transfer learning; fault diagnosis; data deficiency; imbalanced data; linear motion guide; FEATURE-EXTRACTION;
D O I
10.3390/app10217768
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
In the fault diagnosis study, data deficiency, meaning that the fault data for the training are scarce, is often encountered, and it may deteriorate the performance of the fault diagnosis greatly. To solve this issue, the transfer learning (TL) approach is employed to exploit the neural network (NN) trained in another (source) domain where enough fault data are available in order to improve the NN performance of the real (target) domain. While there have been similar attempts of TL in the literature to solve the imbalance issue, they were about the sample imbalance between the source and target domain, whereas the present study considers the imbalance between the normal and fault data. To illustrate this, normal and fault datasets are acquired from the linear motion guide, in which the data at high and low speeds represent the real operation (target) and maintenance inspection (source), respectively. The effect of data deficiency is studied by reducing the number of fault data in the target domain, and comparing the performance of TL, which exploits the knowledge of the source domain and the ordinary machine learning (ML) approach without it. By examining the accuracy of the fault diagnosis as a function of imbalance ratio, it is found that the lower bound and interquartile range (IQR) of the accuracy are improved greatly by employing the TL approach. Therefore, it can be concluded that TL is truly more effective than the ordinary ML when there is a large imbalance between the fault and normal data, such as smaller than 0.1.
引用
收藏
页码:1 / 11
页数:11
相关论文
共 18 条
[1]  
[Anonymous], 2010, 19 INT C ELECT MACHI
[2]   Multisensor data fusion for gearbox fault diagnosis using 2-D convolutional neural network and motor current signature analysis [J].
Azamfar, Moslem ;
Singh, Jaskaran ;
Bravo-Imaz, Inaki ;
Lee, Jay .
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2020, 144
[3]   Early fault diagnosis of rotating machinery based on wavelet packets-Empirical mode decomposition feature extraction and neural network [J].
Bin, G. F. ;
Gao, J. J. ;
Li, X. J. ;
Dhillon, B. S. .
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2012, 27 :696-711
[4]  
Buch KD, 2014, ANNU IEEE IND CONF
[5]   Data fusion and machine learning for industrial prognosis: Trends and perspectives towards Industry 4.0 [J].
Diez-Olivan, Alberto ;
Del Ser, Javier ;
Galar, Diego ;
Sierra, Basilio .
INFORMATION FUSION, 2019, 50 :92-111
[6]   A Generic Intelligent Bearing Fault Diagnosis System Using Compact Adaptive 1D CNN Classifier [J].
Eren, Levent ;
Ince, Turker ;
Kiranyaz, Serkan .
JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY, 2019, 91 (02) :179-189
[7]   Deep Convolutional Transfer Learning Network: A New Method for Intelligent Fault Diagnosis of Machines With Unlabeled Data [J].
Guo, Liang ;
Lei, Yaguo ;
Xing, Saibo ;
Yan, Tao ;
Li, Naipeng .
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2019, 66 (09) :7316-7325
[8]   Fault diagnosis of rotating machinery based on improved wavelet package transform and SVMs ensemble [J].
Hu, Qiao ;
He, Zhengjia ;
Zhang, Zhousuo ;
Zi, Yanyang .
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2007, 21 (02) :688-705
[9]   Mechanical fault diagnosis and prediction in IoT based on multi-source sensing data fusion [J].
Huang, Min ;
Liu, Zhen ;
Tao, Yang .
SIMULATION MODELLING PRACTICE AND THEORY, 2020, 102
[10]   Prediction of remaining useful life by data augmentation technique based on dynamic time warping [J].
Kim, Seokgoo ;
Kim, Nam Ho ;
Choi, Joo-Ho .
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2020, 136