Novel Convolutional Neural Network (NCNN) for the Diagnosis of Bearing Defects in Rotary Machinery
Top Cited Papers
- 1 February 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 70 (00189456), 1-10
- https://doi.org/10.1109/tim.2021.3055802
Abstract
This work presents the development of novel convolutional neural network (NCNN) for effective identification of bearing defects from small samples. For effective feature learning from small training data, cost function of convolution neural network (CNN) is modified by adding additional sparsity cost in the existing cost function. A novel trigonometric cross-entropy function is developed to compute the sparsity cost. The proposed cost function introduces sparsity by avoiding unnecessary activation of neurons in the hidden layers of CNN. For identification of bearing defects from small training samples, NCNN-based transfer learning is applied in the following manner. First, the raw vibration signals as well as envelope signals from source domain machine are obtained. Thereafter, these envelope signals are applied to NCNN for the learning of features from the big training data acquitted from the source domain. After feature learning, knowledge gained from NCNN is transferred to do fine-tuning of NCNN from small training samples of target domain. Thereafter, defect identification is carried out by applying the test data of target domain to fine-tuned NCNN. The experimental result validates that the proposed cross-entropy function introduces sparsity in CNN and, hence, creates an effective deep learning which can even work under a situation when training data are not available in abundant.Keywords
Funding Information
- National Natural Science Foundation of China (U1909217, U1709208)
- Zhejiang Provincial Natural Science Foundation of China (LD21E050001)
- Zhejiang Special Support Program for High-level Personnel Recruitment of China (2018R52034)
This publication has 37 references indexed in Scilit:
- Hierarchical adaptive deep convolution neural network and its application to bearing fault diagnosisMeasurement, 2016
- Enhancing Weak Defect Features Using Undecimated and Adaptive Wavelet Transform for Estimation of Roller Defect Size in a BearingTribology Transactions, 2016
- A novel fault diagnosis technique for photovoltaic systems based on artificial neural networksRenewable Energy, 2016
- Gearbox Fault Identification and Classification with Convolutional Neural NetworksShock and Vibration, 2015
- Thrust bearing groove race defect measurement by wavelet decomposition of pre-processed vibration signalMeasurement, 2013
- Outer race defect width measurement in taper roller bearing using discrete wavelet transform of vibration signalMeasurement, 2013
- Fault Diagnoise of Rolling Bearing Based on EEMD and Instantaneous Energy Density SpectrumApplied Mechanics and Materials, 2011
- A Survey on Transfer LearningIEEE Transactions on Knowledge and Data Engineering, 2009
- Observations of changes in acoustic emission waveform for varying seeded defect sizes in a rolling element bearingApplied Acoustics, 2009
- A note on fuzzy information measuresPattern Recognition Letters, 1997