Application of Interpretable Convolutional Neural Network Enabled by Bi-Damped Wavelets in Bearing Fault Diagnosis

ZHANG Long; XIAO Yiwen; ZHOU Shenci; WANG Chaobing; LIU Hesheng; HUANG Cong; QIAN Tongshuai

doi:10.7652/xjtuxb202503019

您当前的位置：

首页 >

文章列表页 >

Application of Interpretable Convolutional Neural Network Enabled by Bi-Damped Wavelets in Bearing Fault Diagnosis

更新时间：2026-01-06

- Application of Interpretable Convolutional Neural Network Enabled by Bi-Damped Wavelets in Bearing Fault Diagnosis
- Journal of Xi’an Jiaotong University Vol. 59, Issue 3, Pages: 210-221(2025)
- 作者机构：
  
  华东交通大学机电与车辆工程学院， 330013，南昌
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202503019
  CLC： TH17
- Received：07 June 2024，
  
  Online First：18 September 2024，
  
  Published：10 March 2025
- 稿件说明：
移动端阅览
ZHANG Long, XIAO Yiwen, ZHOU Shenci, et al. Application of Interpretable Convolutional Neural Network Enabled by Bi-Damped Wavelets in Bearing Fault Diagnosis[J]. Journal of Xi’an Jiaotong University, 2025, 59(3): 210-221.
DOI：

ZHANG Long, XIAO Yiwen, ZHOU Shenci, et al. Application of Interpretable Convolutional Neural Network Enabled by Bi-Damped Wavelets in Bearing Fault Diagnosis[J]. Journal of Xi’an Jiaotong University, 2025, 59(3): 210-221. DOI： 10.7652/xjtuxb202503019.

摘要

针对深度学习模型决策机理和分类依据无法得知以及Morlet、Laplace等小波不能很好地匹配轴承真实故障脉冲响应的问题，提出一种可解释卷积神经网络模型——Bdw-ResNet。首先，设计了一种sigmoid激活函数加权的双阻尼小波，并构建小波卷积层，同时嵌入即插即用的轻量级的局部注意力机制。然后，结合现有的残差神经网络，建立了既可物理解释又与神经网络兼容的Bdw-ResNet模型以及故障诊断的完整流程，并在南昌铁路局机车轴承数据集和青岛四方轮对轴承两个数据集上进行验证。最后，从先验赋能和归因解释两个方面阐述了Bdw-ResNet模型及其组件的映射关系，并进行了可解释性分析。结果表明：所提方法在两个数据集上分别取得了98.73%、99.46%的识别准确率，与基准模型和其他的组合相比性能提升明显，为可解释性深度学习的故障诊断提供了一种解决思路。

Abstract

To address the issues of opacity in decision-making mechanisms and classification criteria in deep learning models

as well as the mismatch between wavelets such as Morlet and Laplace and the real fault impulse response of bearings

an interpretable convolutional neural network model

Bdw-ResNet

is proposed. Firstly

a bi-damped wavelet

weighted by sigmoid activation function

is designed

and a wavelet convolution layer is constructed

incorporating a plug-and-play lightweight local attention mechanism. Then

by integrating existing residual neural networks

the Bdw-ResNet model is established

which is both physically interpretable and compatible with neural networks

along with a comprehensive fault diagnosis process. The model is validated using the Nanchang Railway Bureau locomotive bearing dataset and the Qingdao Sifang wheelset bearing dataset. Finally

the mapping relationships of the Bdw-ResNet model and its components are explored

discussing priori empowerment and attributional explanation

followed by an interpretability analysis. The results show that the proposed method achieves recognition accuracies of 98.73% and 99.46% on the two datasets

respectively

representing a significant performance improvement over the benchmark model and other combinations. It provides a viable solution for interpretable deep learning in fault diagnosis.

关键词

Keywords

references

顾晓辉 , 杨绍普 , 刘文朋 , 等 . 高速列车轴箱轴承健康监测与故障诊断研究综述 [J ] . 力学学报 , 2022 , 54 ( 7 ): 1780 - 1796 .

GU Xiaohui , YANG Shaopu , LIU Wenpeng , et al . Review of health monitoring and fault diagnosis of axle-box bearing of high-speed train [J ] . Chinese Journal of Theoretical and Applied Mechanics , 2022 , 54 ( 7 ): 1780 - 1796 .

黄梓幸 , 宋冬利 , 罗亚萍 , 等 . 车辆服役环境对高速列车轴箱轴承温度的影响分析 [J ] . 中南大学学报(自然科学版) , 2024 , 55 ( 3 ): 1252 - 1262 .

HUANG Zixing , SONG Dongli , LUO Yaping , et al . Analysis of influence of vehicle environment on temperature of high speed railways axle box bearing [J ] . Journal of Central South University (Science and Technology) , 2024 , 55 ( 3 ): 1252 - 1262 .

张龙 , 刘皓阳 , 张号 , 等 . 改进共空间模式与多源特征融合的轴承智能诊断方法 [J ] . 西安交通大学学报 , 2023 , 57 ( 8 ): 127 - 137 .

ZHANG Long , LIU Haoyang , ZHANG Hao , et al . Intelligent bearing fault diagnosis using modified common spatial pattern and multi-source feature fusion [J ] . Journal of Xi'an Jiaotong University , 2023 , 57 ( 8 ): 127 - 137 .

ZHAO Zhibin , LI Tianfu , WU Jingyao , et al . Deep learning algorithms for rotating machinery intelligent diagnosis: an open source benchmark study [J ] . ISA Transactions , 2020 , 107 : 224 - 255 .

ZHANG Yu , TINO P , LEONARDIS A , et al . A survey on neural network interpretability [J ] . IEEE Transactions on Emerging Topics in Computational Intelligence , 2021 , 5 ( 5 ): 726 - 742 .

李响 , 付春霖 , 雷亚国 , 等 . 保证数据隐私的装备协同智能故障诊断联邦迁移学习方法 [J ] . 机械工程学报 , 2023 , 59 ( 6 ): 1 - 9 .

LI Xiang , FU Chunlin , LEI Yaguo , et al . Federated transfer learning method for privacy-preserving collaborative intelligent machinery fault diagnostics [J ] . Journal of Mechanical Engineering , 2023 , 59 ( 6 ): 1 - 9 .

HUANG Wenyi , CHENG Junsheng , YANG Yu . Rolling bearing fault diagnosis and performance degradation assessment under variable operation conditions based on nuisance attribute projection [J ] . Mechanical Systems and Signal Processing , 2019 , 114 : 165 - 188 .

WU Hao , LI Jimeng , ZHANG Qingyu , et al . Intelligent fault diagnosis of rolling bearings under varying operating conditions based on domain-adversarial neural network and attention mechanism [J ] . ISA Transactions , 2022 , 130 : 477 - 489 .

李霖 , 张西宁 , 刘书语 , 等 . 基于二维经验小波纹理域特征自适应提取的轴承故障诊断方法 [J ] . 西安交通大学学报 , 2021 , 55 ( 12 ): 79 - 86 .

LI Lin , ZHANG Xining , LIU Shuyu , et al . Bearing fault diagnosis method based on 2D empirical wavelet transform texture domain feature adaptive extraction [J ] . Journal of Xi'an Jiaotong University , 2021 , 55 ( 12 ): 79 - 86 .

李晋斐 , 赵冬青 , 王栋民 , 等 . 一种基于组合赋权法的小波去噪质量评价方法 [J ] . 北京航空航天大学学报 , 2023 , 49 ( 3 ): 718 - 725 .

LI Jinfei , ZHAO Dongqing , WANG Dongmin , et al . A quality evaluation method for wavelet denoising based on combinatorial weighting method [J ] . Journal of Beijing University of Aeronautics and Astronautics , 2023 , 49 ( 3 ): 718 - 725 .

ALEKSEEV A , BOBE A . GaborNet: Gabor filters with learnable parameters in deep convolutional neural network [C ] // 2019 International Conference on Engineering and Telecommunication (EnT) . Piscataway, NJ, USA : IEEE , 2019 : 1 - 4 .

LIAO Mengliang , LIU Chang , WANG Cong , et al . Research on a rolling bearing fault detection method with wavelet convolution deep transfer learning [J ] . IEEE Access , 2021 , 9 : 45175 - 45188 .

LI Sinan , LI Tianfu , SUN Chuang , et al . WPConvNet: an interpretable wavelet packet kernel-constrained convolutional network for noise-robust fault diagnosis [J/OL ] . IEEE Transactions on Neural Networks and Learning Systems ( 2023-06-15 ) [ 2024-06-01 ] . https://doi.org/10.1109/TNNLS.2023.3282599 https://doi.org/10.1109/TNNLS.2023.3282599 .

张德祥 , 王俊 , 袁培成 . 基于注意力机制的多尺度全场景监控目标检测方法 [J ] . 电子与信息学报 , 2022 , 44 ( 9 ): 3249 - 3257 .

ZHANG Dexiang , WANG Jun , YUAN Peicheng . Object detection method for multi-scale full-scene surveillance based on attention mechanism [J ] . Journal of Electronics & Information Technology , 2022 , 44 ( 9 ): 3249 - 3257 .

施健聪 , 王兴龙 , 张俊 . 基于双重注意力机制的异步电机故障诊断方法 [J ] . 振动与冲击 , 2023 , 42 ( 21 ): 110 - 118 .

SHI Jiancong , WANG Xinglong , ZHANG Jun . Induction motor fault diagnosis method based on dual-attention mechanism [J ] . Journal of Vibration and Shock , 2023 , 42 ( 21 ): 110 - 118 .

HU Jie , SHEN Li , SUN Gang . Squeeze-and-excitation networks [C ] // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway, NJ, USA : IEEE , 2018 : 7132 - 7141 .

WANG Qilong , WU Banggu , ZHU Pengfei , et al . ECA-net: efficient channel attention for deep convolutional neural networks [C ] // 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway, NJ, USA : IEEE , 2020 : 11531 - 11539 .

刘方 , 翟涛涛 , 侯超强 , 等 . 基于多普勒调制时移Laplace小波的列车轴承故障声信号瞬态成分快速提取方法 [J ] . 仪器仪表学报 , 2022 , 43 ( 3 ): 40 - 48 .

LIU Fang , ZHAI Taotao , HOU Chaoqiang , et al . A fast transient component extraction method of train bearing fault acoustic signal based on Doppler modulated time-shifting Laplace wavelet [J ] . Chinese Journal of Scientific Instrument , 2022 , 43 ( 3 ): 40 - 48 .

LI Jimeng , TAO Jinxin , DING Wanmeng , et al . Period-assisted adaptive parameterized wavelet dictionary and its sparse representation for periodic transient features of rolling bearing faults [J ] . Mechanical Systems and Signal Processing , 2022 , 169 : 108796 .

王天宇 , 陈晗 , 王刚 , 等 . 采用小波变换和双向长短期记忆网络的脑电睡眠分期模型 [J ] . 西安交通大学学报 , 2022 , 56 ( 9 ): 104 - 111 .

WANG Tianyu , CHEN Han , WANG Gang , et al . EEG sleep staging model using wavelet transform and bidirectional long short-term memory network [J ] . Journal of Xi'an Jiaotong University , 2022 , 56 ( 9 ): 104 - 111 .

LI Tianfu , ZHAO Zhibin , SUN Chuang , et al . Wavelet KernelNet: an interpretable deep neural network for industrial intelligent diagnosis [J ] . IEEE Transactions on Systems, Man, and Cybernetics: Systems , 2022 , 52 ( 4 ): 2302 - 2312 .

HOU Qibin , ZHOU Daquan , FENG Jiashi . Coordinate attention for efficient mobile network design [C ] // 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway, NJ, USA : IEEE , 2021 : 13708 - 13717 .

WU Yuxin , HE Kaiming . Group normalization [C ] // Computer Vision-ECCV 2018 . Cham, Switzerland : Springer International Publishing , 2018 : 3 - 19 .

严如强 , 商佐港 , 王志颖 , 等 . 可解释人工智能在工业智能诊断中的挑战和机遇：先验赋能 [J ] . 机械工程学报 , 2024 , 60 ( 12 ): 1 - 20 .

YAN Ruqiang , SHANG Zuogang , WANG Zhiying , et al . Challenges and opportunities of XAI in industrial intelligent diagnosis: priori-empowered [J ] . Journal of Mechanical Engineering , 2024 , 60 ( 12 ): 1 - 20 .

严如强 , 周峥 , 杨远贵 , 等 . 可解释人工智能在工业智能诊断中的挑战和机遇：归因解释 [J ] . 机械工程学报 , 2024 , 60 ( 12 ): 21 - 40 .

YAN Ruqiang , ZHOU Zheng , YANG Yuangui , et al . Challenges and opportunities of XAI in industrial intelligent diagnosis: attribution interpretation [J ] . Journal of Mechanical Engineering , 2024 , 60 ( 12 ): 21 - 40 .

HE Chao , SHI Hongmei , LIU Xiaorong , et al . Interpretable physics-informed domain adaptation paradigm for cross-machine transfer diagnosis [J ] . Knowledge-Based Systems , 2024 , 288 : 111499 .

SHANG Zuogang , ZHAO Zhibin , YAN Ruqiang . Denoising fault-aware wavelet network: a signal processing informed neural network for fault diagnosis [J ] . Chinese Journal of Mechanical Engineering , 2023 , 36 ( 1 ): 9 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Fault Diagnosis Method for Asynchronous Motor Using Deep Learning

Fault Diagnosis of Rolling Bearings Using Deep Transfer Learning and Adaptive Weighting

Intelligent Quantitative Non-Destructive Evaluation Method for Delamination of Carbon Fiber Composite Tube with Mirror-Reflection Infrared Thermography

Model Predictive Control-Oriented Deep Koopman Global Linear Modeling Method for Wind Turbines

Construction and Application of Multi-Agent Large Language Model for Power Transformer Fault Diagnosis

Related Author

王丽华 1

谢阳阳 1

张永宏 1

赵晓平 2

周子贤 2

JIA Feng

LI Shihao

SHEN Jianjun

Related Institution

南京信息工程大学信息与控制学院

Key Laboratory of Road Construction Technology and Equipment of Ministry of Education, Chang'an University

Key Laboratory of Education Ministry for Modern Design and Rotor-Bearing System, Xi'an Jiaotong University

Shaanxi Engineering Research Center of Nondestructive Testing and Structural Integrity Evaluation，Xi’an Jiaotong University

State Key Laboratory for Strength and Vibration of Mechanical Structures，Xi’an Jiaotong University

⁰