An improved deep convolutional neural network with multi-scale information for bearing fault diagnosis

Abstract In recent years, deep learning technique has been used in mechanical intelligent fault diagnosis and it has achieved much success. Among the deep learning models, convolutional neural network (CNN) is able to accomplish the feature learning without priori knowledge and pattern classification automatically, which makes it to be an end-to-end method. However, CNN may fall into local optimum when lack of useful information in the input signal. Diversity resolution expressions of signal in frequency domain can be obtained by using the filters with different scales (lengths) and more expressions may provide more useful information. Thus, in this paper, an improved CNN named multi-scale cascade convolutional neural network (MC-CNN) is proposed for the classification information enhancement of input. In MC-CNN, a new layer has been added before convolutional layers to construct a new signal of more distinguishable information. The composed signal is obtained by concatenating the signals convolved by original input and kernels of different lengths. To reduce the abundant neurons produced by the multi-scale signal, a convolutional layer with kernels of small size and a pooling layer are added after the multi-scale cascade layer. To verify the proposed method, the original CNNs and MC-CNN are applied to the pattern classification of bearing vibration signal with four conditions under normal and noise environments, respectively. The classification results show that the proposed MC-CNN is more effective than the commonly CNNs. In addition, the lower t-distributed stochastic neighbor embedding (t-SNE) clustered error verifies the effectiveness and necessity of MC layer further. By analyzing the kernels learned from the multi-scale cascade layer, it can be found that the kernels act as filters of different resolutions to make the frequency domain structure of different fault signals more distinguishable. By studying the influence of kernel scale in MC layer on fault diagnosis, it is found that the optimal scale does exist and will be a research emphasis in the future. Moreover, the effectiveness of MC-CNN is verified furthermore by analyzing the application of MC-CNN in bearing fault diagnosis under nonstationary working conditions.

[1]  Gaoliang Peng,et al.  A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load , 2018, Mechanical Systems and Signal Processing.

[2]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[3]  Haidong Shao,et al.  Rolling bearing fault detection using continuous deep belief network with locally linear embedding , 2018, Comput. Ind..

[4]  Minping Jia,et al.  Intelligent fault diagnosis of rotating machinery using improved multiscale dispersion entropy and mRMR feature selection , 2019, Knowl. Based Syst..

[5]  Xia Xu,et al.  R-VCANet: A New Deep-Learning-Based Hyperspectral Image Classification Method , 2017, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[6]  Chunhui Zhao,et al.  Sparse Exponential Discriminant Analysis and Its Application to Fault Diagnosis , 2018, IEEE Transactions on Industrial Electronics.

[7]  Chunhui Zhao,et al.  A full‐condition monitoring method for nonstationary dynamic chemical processes with cointegration and slow feature analysis , 2018 .

[8]  Haidong Shao,et al.  Rolling bearing fault feature learning using improved convolutional deep belief network with compressed sensing , 2018 .

[9]  Weiqiang Wang,et al.  A new perspective: Recognizing online handwritten Chinese characters via 1-dimensional CNN , 2019, Inf. Sci..

[10]  Binqiang Chen,et al.  Manifold learning-based subspace distance for machinery damage assessment , 2016 .

[11]  Carl Salvaggio,et al.  A Fully Convolutional Network for Semantic Labeling of 3D Point Clouds , 2017, ISPRS Journal of Photogrammetry and Remote Sensing.

[12]  S.A.V. Satya Murty,et al.  Roller element bearing fault diagnosis using singular spectrum analysis , 2013 .

[13]  Chunhui Zhao,et al.  Dynamic Distributed Monitoring Strategy for Large-Scale Nonstationary Processes Subject to Frequently Varying Conditions Under Closed-Loop Control , 2019, IEEE Transactions on Industrial Electronics.

[14]  Robert Jenssen,et al.  Automatic autonomous vision-based power line inspection: A review of current status and the potential role of deep learning , 2018, International Journal of Electrical Power & Energy Systems.

[15]  Mo-Yuen Chow,et al.  Neural-network-based motor rolling bearing fault diagnosis , 2000, IEEE Trans. Ind. Electron..

[16]  Lei Guo,et al.  Robust bearing performance degradation assessment method based on improved wavelet packet–support vector data description , 2009 .

[17]  Haidong Shao,et al.  A novel method for intelligent fault diagnosis of rolling bearings using ensemble deep auto-encoders , 2018 .

[18]  Moncef Gabbouj,et al.  Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks , 2017 .

[19]  Chunhui Zhao,et al.  Recursive Exponential Slow Feature Analysis for Fine-Scale Adaptive Processes Monitoring With Comprehensive Operation Status Identification , 2019, IEEE Transactions on Industrial Informatics.

[20]  Liang Gao,et al.  A new subset based deep feature learning method for intelligent fault diagnosis of bearing , 2018, Expert Syst. Appl..

[21]  Chunhui Zhao,et al.  Fault Subspace Selection Approach Combined With Analysis of Relative Changes for Reconstruction Modeling and Multifault Diagnosis , 2016, IEEE Transactions on Control Systems Technology.

[22]  Zhengjia He,et al.  Novel method for bearing performance degradation assessment – A kernel locality preserving projection-based approach , 2014 .

[23]  Boualem Boashash,et al.  1-D CNNs for structural damage detection: Verification on a structural health monitoring benchmark data , 2018, Neurocomputing.

[24]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[25]  Chunhui Zhao,et al.  Online Fault Diagnosis in Industrial Processes Using Multimodel Exponential Discriminant Analysis Algorithm , 2019, IEEE Transactions on Control Systems Technology.

[26]  Jonathan Cheung-Wai Chan,et al.  Semi-Supervised Deep Learning Classification for Hyperspectral Image Based on Dual-Strategy Sample Selection , 2018, Remote. Sens..

[27]  Gary K. L. Tam,et al.  3D Mesh Segmentation via Multi-branch 1D Convolutional Neural Networks , 2017, Graph. Model..

[28]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[29]  Fanrang Kong,et al.  Multiscale slope feature extraction for rotating machinery fault diagnosis using wavelet analysis , 2013 .

[30]  Yan Han,et al.  Multi-level wavelet packet fusion in dynamic ensemble convolutional neural network for fault diagnosis , 2018, Measurement.

[31]  Tara N. Sainath,et al.  FUNDAMENTAL TECHNOLOGIES IN MODERN SPEECH RECOGNITION Digital Object Identifier 10.1109/MSP.2012.2205597 , 2012 .

[32]  Xiang Li,et al.  Remaining useful life estimation in prognostics using deep convolution neural networks , 2018, Reliab. Eng. Syst. Saf..

[33]  Robert X. Gao,et al.  Mechanical Systems and Signal Processing Approximate Entropy as a Diagnostic Tool for Machine Health Monitoring , 2006 .

[34]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[35]  Yaguo Lei,et al.  Deep normalized convolutional neural network for imbalanced fault classification of machinery and its understanding via visualization , 2018, Mechanical Systems and Signal Processing.

[36]  Wentao Mao,et al.  A novel deep output kernel learning method for bearing fault structural diagnosis , 2019, Mechanical Systems and Signal Processing.

[37]  Chunhui Zhao,et al.  A Sparse Reconstruction Strategy for Online Fault Diagnosis in Nonstationary Processes with No a Priori Fault Information , 2017 .

[38]  Gianpaolo Francesco Trotta,et al.  Computer vision and deep learning techniques for pedestrian detection and tracking: A survey , 2018, Neurocomputing.

[39]  Haiyang Pan,et al.  Sigmoid-based refined composite multiscale fuzzy entropy and t-SNE based fault diagnosis approach for rolling bearing , 2018, Measurement.

[40]  Jianbo Yu,et al.  Bearing performance degradation assessment using locality preserving projections and Gaussian mixture models , 2011 .

[41]  Alexandros Iosifidis,et al.  Improving Efficiency in Convolutional Neural Network with Multilinear Filters , 2018, Neural Networks.

[42]  Jay Lee,et al.  Watchdog Agent - an infotronics-based prognostics approach for product performance degradation assessment and prediction , 2003, Adv. Eng. Informatics.