Mechanical fault intelligent diagnosis using attention-based dual-scale feature fusion capsule network

被引:23
作者
Zhang, Qingyu [1 ]
Li, Jimeng [1 ]
Ding, Wanmeng [1 ]
Ye, Zhangdi [1 ]
Meng, Zong [1 ]
机构
[1] Yanshan Univ, Coll Elect Engn, Qinhuangdao 066004, Peoples R China
基金
中国国家自然科学基金;
关键词
Mechanical fault diagnosis; Capsule networks; Dual -scale feature fusion; Attention mechanism; EMPIRICAL MODE DECOMPOSITION; ROTATING MACHINERY; NEURAL-NETWORK;
D O I
10.1016/j.measurement.2022.112345
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Capsule networks, which have achieved many achievements in machine vision, have also gotten wide attention in machinery fault diagnosis. However, due to the non-stationarity and diversity of mechanical system vibration signals, as well as the dual-scale characteristics of different fault features, it is often difficult for existing single -scale capsule networks to fully mine vital discriminative features in the data, which is not conducive to accurate identification of mechanical faults. This paper studies an attention-based dual-scale feature fusion capsule network for mechanical fault diagnosis. It first extracts dual-scale features from grayscale images obtained from vibration signals using convolutional layers composed of kernels of different sizes; secondly, an attention-based two-branch network is designed to calculate the weights of features at different scales, and accordingly dual-scale feature fusion is performed; finally, the obtained features are entered into the capsule layers, and the classifi-cation and identification of mechanical faults are realized by optimizing the model using the classification loss and reconstruction loss. A rolling bearing experimental dataset and a motor fault dataset are adopted to assess the performance of the proposed method, and the comparison results confirm its effectiveness and superiority, indicating that it has the potential to be a useful tool for detecting mechanical faults.
引用
收藏
页数:12
相关论文
共 40 条
[1]   Rolling Bearing Fault Diagnosis Based on an Improved Denoising Method Using the Complete Ensemble Empirical Mode Decomposition and the Optimized Thresholding Operation [J].
Abdelkader, Rabah ;
Kaddour, Abdelhafid ;
Bendiabdellah, Azeddine ;
Derouiche, Ziane .
IEEE SENSORS JOURNAL, 2018, 18 (17) :7166-7172
[2]   Generalization of deep neural network for bearing fault diagnosis under different working conditions using multiple kernel method [J].
An, Zenghui ;
Li, Shunming ;
Wang, Jinrui ;
Xin, Yu ;
Xu, Kun .
NEUROCOMPUTING, 2019, 352 :42-53
[3]   Mechanical model development of rolling bearing-rotor systems: A review [J].
Cao, Hongrui ;
Niu, Linkai ;
Xi, Songtao ;
Chen, Xuefeng .
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2018, 102 :37-58
[4]   A deep capsule neural network with stochastic delta rule for bearing fault diagnosis on raw vibration signals [J].
Chen, Tianyou ;
Wang, Zhihua ;
Yang, Xiang ;
Jiang, Kun .
MEASUREMENT, 2019, 148
[5]   Deep Focus Parallel Convolutional Neural Network for Imbalanced Classification of Machinery Fault Diagnostics [J].
Duan, Andongzhe ;
Guo, Liang ;
Gao, Hongli ;
Wu, Xiangdong ;
Dong, Xun .
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2020, 69 (11) :8680-8689
[6]   Margin Loss: Making Faces More Separable [J].
Gao, Riqiang ;
Yang, Fuwei ;
Yang, Wenming ;
Liao, Qingmin .
IEEE SIGNAL PROCESSING LETTERS, 2018, 25 (02) :308-312
[7]   Combination bidirectional long short-term memory and capsule network for rotating machinery fault diagnosis [J].
Han, Tian ;
Ma, Ruiyi ;
Zheng, Jigui .
MEASUREMENT, 2021, 176
[8]  
Hinton GE., 2018, ICLR POSTER
[9]   A Robust Weight-Shared Capsule Network for Intelligent Machinery Fault Diagnosis [J].
Huang, Ruyi ;
Li, Jipu ;
Wang, Shuhua ;
Li, Guanghui ;
Li, Weihua .
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2020, 16 (10) :6466-6475
[10]   Deep Ensemble Capsule Network for Intelligent Compound Fault Diagnosis Using Multisensory Data [J].
Huang, Ruyi ;
Li, Jipu ;
Li, Weihua ;
Cui, Lingli .
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2020, 69 (05) :2304-2314