Study on the Interpretability of One-Dimensional Convolutional Neural Networks in Mechanical Fault Feature Extraction

WANG Fangzhen; ZHAGN Xiaoli; ZHAO Qiwu; WANG Baojian

doi:10.7652/xjtuxb202507003

您当前的位置：

首页 >

文章列表页 >

Study on the Interpretability of One-Dimensional Convolutional Neural Networks in Mechanical Fault Feature Extraction

Special Topic　Big Data Intelligent Fault Diagnosis and Prediction | 更新时间：2025-07-09

- Study on the Interpretability of One-Dimensional Convolutional Neural Networks in Mechanical Fault Feature Extraction
- Journal of Xi’an Jiaotong University Vol. 59, Issue 7, Pages: 24-35(2025)
- 作者机构：
  
  1.长安大学道路施工技术与装备教育部重点实验室， 710064，西安
  2.西安交通大学机械基础国家级实验教学示范中心， 710049，西安
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202507003
  CLC： TH133.33
- Received：08 January 2025，
  
  Online First：24 March 2025，
  
  Published：10 July 2025
- 稿件说明：
移动端阅览
WANG Fangzhen, ZHAGN Xiaoli, ZHAO Qiwu, et al. Study on the Interpretability of One-Dimensional Convolutional Neural Networks in Mechanical Fault Feature Extraction[J]. Journal of Xi’an Jiaotong University, 2025, 59(7): 24-35.
DOI：

WANG Fangzhen, ZHAGN Xiaoli, ZHAO Qiwu, et al. Study on the Interpretability of One-Dimensional Convolutional Neural Networks in Mechanical Fault Feature Extraction[J]. Journal of Xi’an Jiaotong University, 2025, 59(7): 24-35. DOI： 10.7652/xjtuxb202507003.

摘要

针对一维卷积神经网络在机械故障诊断中的内部决策和推断过程未知，导致结果可解释性与可信度不足的问题，从信号特征提取的视角建立信号分析与神经网络之间的相似性联系，通过提取神经网络卷积层权重，观察信号时域、频域特征随网络层的变化规律，从而揭示神经网络特征提取的本质，并采用实验测试数据和凯斯西储大学轴承公开数据进行验证。结果表明：卷积核可以等效为有限脉冲滤波器，最大池化层能够满足简单二分类任务中神经网络的非线性化要求，此时的卷积层无需添加激活函数；神经网络能够通过逐层提高频率分辨率，寻找到接近理论故障特征频率的频率成分，此行为与傅里叶变换存在相似性；当频谱范围最终分解到1~3倍故障特征频率时，能够更好地完成识别任务。该研究可为揭示卷积神经网络的“黑盒”机制与可解释性提供新的思路与方法。

Abstract

To address the limited interpretability and reliability caused by the unknown internal decision-making and inference processes of one-dimensional convolutional neural networks (CNNs) in mechanical fault diagnosis

a similarity connection between signal analysis and neural networks is established from the perspective of feature extraction. By extracting the weights of the convolutional layers in the neural network and observing the variations in time/frequency domain features as the network layers change

this study reveals the intrinsic feature extraction behavior of neural networks. Experimental test data and publicly available bearing data from Case Western Reserve University are used for validation. The results indicate that the convolutional kernel can be equivalent to a finite impulse filter

and the max pooling layer can meet the non-linear requirements of neural networks for simple binary classification tasks

therefore not requiring an activation function in the convolutional layer; the neural network is capable of incrementally increasing frequency resolution layer by layer to identify frequency components close to theoretical fault characteristic frequencies

exhibiting similarities to Fourier transforms. When the spectral range is ultimately decomposed to 1 to 3 times the fault characteristic frequency

the identification task is better accomplished. This study can provide new ideas and methods for revealing the “black box” mechanisms and interpretability of convolutional neural networks.

关键词

Keywords

references

卞文彬 , 邓艾东 , 刘东川 , 等 . 基于改进深度残差收缩网络的风电机组滚动轴承故障诊断方法 [J ] . 机械工程学报 , 2023 , 59 ( 12 ): 202 - 214 .

BIAN Wenbin , DENG Aidong , LIU Dongchuan , et al . Fault diagnosis method of wind turbine rolling bearing based on improved deep residual shrinkage network [J ] . Journal of Mechanical Engineering , 2023 , 59 ( 12 ): 202 - 214 .

ZHANG Xiaoli , WANG Fangzhen , ZHOU Yongqing , et al . The optimized local sparse parallel multi-channel deep convolutional neural network-LSTM and the application in bearing fault diagnosis under noise and variable working condition [J ] . Proceedings of the Institution of Mechanical Engineers: Part C　Journal of Mechanical Engineering Science , 2025 , 239 ( 1 ): 205 - 218 .

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 .

RUDIN C , CHEN Chaofan , CHEN Zhi , et al . Interpretable machine learning: fundamental principles and 10 grand challenges [EB/OL ] . ( 2021-07-10 ) [ 2024-10-12 ] . https://arxiv.org/abs/2103.11251 https://arxiv.org/abs/2103.11251 .

LIPTON Z C . The mythos of model interpretability [J ] . Communications of the ACM , 2018 , 61 ( 10 ): 36 - 43 .

林京 , 焦金阳 . 可解释机械智能诊断技术的研究进展与挑战 [J ] . 机械工程学报 , 2023 , 59 ( 20 ): 215 - 224 .

LIN Jing , JIAO Jinyang . Research progress and challenges of interpretable mechanical intelligent diagnosis [J ] . Journal of Mechanical Engineering , 2023 , 59 ( 20 ): 215 - 224 .

LI Sinan , LI Tianfu , SUN Chuang , et al . Multilayer grad-CAM: an effective tool towards explainable deep neural networks for intelligent fault diagnosis [J ] . Journal of Manufacturing Systems , 2023 , 69 : 20 - 30 .

王冉 , 石如玉 , 胡升涵 , 等 . 基于声成像与卷积神经网络的轴承故障诊断方法及其可解释性研究 [J ] . 振动与冲击 , 2022 , 41 ( 16 ): 224 - 231 .

WANG Ran , SHI Ruyu , HU Shenghan , et al . An acoustic fault diagnosis method of rolling bearings based on acoustic imaging and convolutional neural network [J ] . Journal of Vibration and Shock , 2022 , 41 ( 16 ): 224 - 231 .

YU Shihang , WANG Min , PANG Shanchen , et al . Intelligent fault diagnosis and visual interpretability of rotating machinery based on residual neural network [J ] . Measurement , 2022 , 196 : 111228 .

YANG Huixin , LI Xiang , ZHANG Wei . Interpretability of deep convolutional neural networks on rolling bearing fault diagnosis [J ] . Measurement Science and Technology , 2022 , 33 ( 5 ): 055005 .

廖才波 , 杨金鑫 , 邱志斌 , 等 . 一种基于夏普利值及油中溶解气体分析的可解释变压器故障诊断方法 [J ] . 电网技术 , 2024 , 48 ( 4 ): 1752 - 1761 .

LIAO Caibo , YANG Jinxin , QIU Zhibin , et al . Interpretable transformer fault diagnosis based on SHAP value and dissolved gas analysis of transformer oil [J ] . Power System Technology , 2024 , 48 ( 4 ): 1752 - 1761 .

JIA Sixiang , SUN Dingyi , NOMAN K , et al . Lifting wavelet-informed hierarchical domain adaptation network: an interpretable digital twin-driven gearbox fault diagnosis method [J ] . Reliability Engineering & System Safety , 2025 , 254 ( Part B ): 110660 .

GAO Lei , LIU Zhihao , GAO Qinhe , et al . Dual data fusion fault diagnosis of transmission system based on entropy weighted multi-representation DS evidence theory and GCN [J ] . Measurement , 2025 , 243 : 116308 .

BRITO L C , SUSTO G A , BRITO J N , et al . An explainable artificial intelligence approach for unsupervised fault detection and diagnosis in rotating machinery [J ] . Mechanical Systems and Signal Processing , 2022 , 163 : 108105 .

LI Yasong , ZHOU Zheng , SUN Chuang , et al . Variational attention-based interpretable transformer network for rotary machine fault diagnosis [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2024 , 35 ( 5 ): 6180 - 6193 .

LI Sinan , LI Tianfu , SUN Chuang , et al . WPConvNet: an interpretable wavelet packet kernel-constrained convolutional network for noise-robust fault diagnosis [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2024 , 35 ( 10 ): 14974 - 14988 .

ZHONG Jingshu , ZHENG Yu , RUAN Chengtao , et al . M-IPISincNet: an explainable multi-source physics-informed neural network based on improved SincNet for rolling bearings fault diagnosis [J ] . Information Fusion , 2025 , 115 : 102761 .

XU Z Q J , ZHANG Yaoyu , LUO Tao . Overview frequency principle/spectral bias in deep learning [J/OL ] . Communications on Applied Mathematics and Computation . ( 2024-09-04 ) [ 2024-12-20 ] . https://doi.org/10.1007/s42967-024-00398-7 https://doi.org/10.1007/s42967-024-00398-7 .

XU Zhiqin , ZHANG Yaoyu , XIAO Yangang . Training behavior of deep neural network in frequency domain [C ] // Neural Information Processing . Cham : Springer International Publishing , 2019 : 264 - 274 .

LIAO Jingxiao , DONG Hangcheng , SUN Zhiqi , et al . Attention-embedded quadratic network (qttention) for effective and interpretable bearing fault diagnosis [J ] . IEEE Transactions on Instrumentation and Measurement , 2023 , 72 : 1 - 13 .

MAGADÁN L , RUIZ-CÁRCEL C , GRANDA J C , et al . Explainable and interpretable bearing fault classification and diagnosis under limited data [J ] . Advanced Engineering Informatics , 2024 , 62 ( Part D ): 102909 .

CHEN Yikai , WANG Dong , HOU Bingchang , et al . Gaussian assumptions-free interpretable linear discriminant analysis for locating informative frequency bands for machine condition monitoring [J ] . Mechanical Systems and Signal Processing , 2023 , 199 : 110492 .

BORGHESANI P , HERWIG N , ANTONI J , et al . A Fourier-based explanation of 1D-CNNs for machine condition monitoring applications [J ] . Mechanical Systems and Signal Processing , 2023 , 205 : 110865 .

HOYER E , STORK R . The zoom FFT using complex modulation [C ] // ICASSP'77. IEEE International Conference on Acoustics, Speech, and Signal Processing . Piscataway, NJ, USA : IEEE , 1977 : 78 - 81 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

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

Dual-Wavelet Dictionary Based Feature Extraction Method for Compound Faults in Gearboxes

Application of Mayfly Algorithm Optimized Dual-Channel Network in Gearbox Fault Diagnosis

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

Diagnosis Method for Inter-Turn Short Circuit Faults in Doubly Fed Induction Generators Based on Spectral Amplitude Modulation of the Residual Current Harmonic Analysis

Related Author

LIN Jinshan

LI Yuan

FANG Qixuan

LIU Zhihao

LI Chunpeng

ZHANG Guanjun

WANG Xiangdong

LIU Qiang

Related Institution

西安交通大学电气工程学院，710049，西安School of Electrical Engineering, Xi'an Jiaotong University, Xi'an 710049, China

College of Hydraulic & Environmental Engineering, China Three Gorges University

China Yangtze Power Co., Ltd.

College of Mechanical and Power Engineering, China Three Gorges University

Hubei Key Laboratory of Hydroelectric Machinery Design and Maintenance, China Three Gorges University

⁰