WaveletKernelNet: An Interpretable Deep Neural Network for Industrial Intelligent Diagnosis
WaveletKernelNet: An Interpretable Deep Neural Network for Industrial Intelligent Diagnosis
复制标题
WaveletKernelNet：用于工业智能诊断的可解释深度神经网络
DOI：
10.1109/tsmc.2020.3048950
复制
发表时间：
2021-01-19
影响因子：
8.7
通讯作者：
Gao, Robert X.
中科院分区：
文献类型：
--
作者：
Li, Tianfu;Zhao, Zhibin;Gao, Robert X.
卷积神经网络（CNN）具有特征学习和非线性映射的能力，已在预测和健康管理（PHM）中证明了其有效性。然而，关于 CNN 架构物理意义的解释却很少被研究。在本文中，提出了一种新颖的小波驱动深度神经网络，称为 WaveletKernelNet (WKN)，其中设计了连续小波卷积 (CWConv) 层来取代标准 CNN 的第一个卷积层。这使得第一个 CWConv 层能够发现更有意义的内核。此外，在该 CWConv 层中，仅直接从原始数据中学习尺度参数和平移参数。这提供了一种非常有效的方法来获得定制的内核库，专门用于提取嵌入在振动信号中的与缺陷相关的冲击分量。此外，利用实验室环境的数据进行了三项实验研究，以验证所提出的机械故障诊断方法的有效性。实验结果表明WKNs的准确率比CNN高10%以上，这表明了设计的CWConv层的重要性。此外，通过理论分析和特征图可视化，发现WKN是可解释的，参数更少，并且能够在相同的训练周期内更快地收敛。
Convolutional neural network (CNN), with the ability of feature learning and nonlinear mapping, has demonstrated its effectiveness in prognostics and health management (PHM). However, an explanation on the physical meaning of a CNN architecture has rarely been studied. In this article, a novel wavelet-driven deep neural network, termed as WaveletKernelNet (WKN), is presented, where a continuous wavelet convolutional (CWConv) layer is designed to replace the first convolutional layer of the standard CNN. This enables the first CWConv layer to discover more meaningful kernels. Furthermore, only the scale parameter and translation parameter are directly learned from raw data at this CWConv layer. This provides a very effective way to obtain a customized kernel bank, specifically tuned for extracting defect-related impact component embedded in the vibration signal. In addition, three experimental studies using data from laboratory environment are carried out to verify the effectiveness of the proposed method for mechanical fault diagnosis. The experimental results show that the accuracy of the WKNs is higher than CNN by more than 10%, which indicate the importance of the designed CWConv layer. Besides, through theoretical analysis and feature map visualization, it is found that the WKNs are interpretable, have fewer parameters, and have the ability to converge faster within the same training epochs.