Model-driven deep unrolling: Towards interpretable deep learning against noise attacks for intelligent fault diagnosis
Intelligent fault diagnosis (IFD) has experienced tremendous progress owing to a great deal to deep learning (DL)-based methods over the decades. However, the “black box” nature of DL-based methods still seriously hinders wide applications in industry, especially in aero-engine IFD, and how to inter...
Saved in:
Published in | ISA transactions Vol. 129; pp. 644 - 662 |
---|---|
Main Authors | , , , , , , |
Format | Journal Article |
Language | English |
Published |
United States
Elsevier Ltd
01.10.2022
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Intelligent fault diagnosis (IFD) has experienced tremendous progress owing to a great deal to deep learning (DL)-based methods over the decades. However, the “black box” nature of DL-based methods still seriously hinders wide applications in industry, especially in aero-engine IFD, and how to interpret the learned features is still a challenging problem. Furthermore, IFD based on vibration signals is often affected by the heavy noise, leading to a big drop in accuracy. To address these two problems, we develop a model-driven deep unrolling method to achieve ante-hoc interpretability, whose core is to unroll a corresponding optimization algorithm of a predefined model into a neural network, which is naturally interpretable and robust to noise attacks. Motivated by the recent multi-layer sparse coding (ML-SC) model, we herein propose to solve a general sparse coding (GSC) problem across different layers and deduce the corresponding layered GSC (LGSC) algorithm. Based on the ideology of deep unrolling, the proposed algorithm is unfolded into LGSC-Net, whose relationship with the convolutional neural network (CNN) is also discussed in depth. The effectiveness of the proposed model is verified by an aero-engine bevel gear fault experiment and a helical gear fault experiment with three kinds of adversarial noise attacks. The interpretability is also discussed from the perspective of the core of model-driven deep unrolling and its inductive reconstruction property.
•A model-driven deep unrolling method is developed to design interpretable DL models.•GSC is solved gradually and its inducing optimization algorithm is unrolled into the LGSC-Net.•The interpretability is discussed from the perspective of the core of model-driven deep unrolling.•Experiments are performed to verify the diagnosis ability via adding adversarial noise attacks. |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 0019-0578 1879-2022 |
DOI: | 10.1016/j.isatra.2022.02.027 |