Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Scalable and reliable deep transfer learning for intelligent fault detection via multi-scale neural processes embedded with knowledge

Zhongzhi Li, Jingqi Tu, Jiacheng Zhu, Jianliang Ai, Yiqun Dong

TL;DR

A novel DTL-based method known as Neural Processes-based deep transfer learning with graph convolution network (GTNP) is proposed, which bridges the data distribution discrepancies of source domain and target domain in high-dimensional space.

Abstract

Deep transfer learning (DTL) is a fundamental method in the field of Intelligent Fault Detection (IFD). It aims to mitigate the degradation of method performance that arises from the discrepancies in data distribution between training set (source domain) and testing set (target domain). Considering the fact that fault data collection is challenging and certain faults are scarce, DTL-based methods face the limitation of available observable data, which reduces the detection performance of the methods in the target domain. Furthermore, DTL-based methods lack comprehensive uncertainty analysis that is essential for building reliable IFD systems. To address the aforementioned problems, this paper proposes a novel DTL-based method known as Neural Processes-based deep transfer learning with graph convolution network (GTNP). Feature-based transfer strategy of GTNP bridges the data distribution discrepancies of source domain and target domain in high-dimensional space. Both the joint modeling based on global and local latent variables and sparse sampling strategy reduce the demand of observable data in the target domain. The multi-scale uncertainty analysis is obtained by using the distribution characteristics of global and local latent variables. Global analysis of uncertainty enables GTNP to provide quantitative values that reflect the complexity of methods and the difficulty of tasks. Local analysis of uncertainty allows GTNP to model uncertainty (confidence of the fault detection result) at each sample affected by noise and bias. The validation of the proposed method is conducted across 3 IFD tasks, consistently showing the superior detection performance of GTNP compared to the other DTL-based methods.

Scalable and reliable deep transfer learning for intelligent fault detection via multi-scale neural processes embedded with knowledge

TL;DR

A novel DTL-based method known as Neural Processes-based deep transfer learning with graph convolution network (GTNP) is proposed, which bridges the data distribution discrepancies of source domain and target domain in high-dimensional space.

Abstract

Deep transfer learning (DTL) is a fundamental method in the field of Intelligent Fault Detection (IFD). It aims to mitigate the degradation of method performance that arises from the discrepancies in data distribution between training set (source domain) and testing set (target domain). Considering the fact that fault data collection is challenging and certain faults are scarce, DTL-based methods face the limitation of available observable data, which reduces the detection performance of the methods in the target domain. Furthermore, DTL-based methods lack comprehensive uncertainty analysis that is essential for building reliable IFD systems. To address the aforementioned problems, this paper proposes a novel DTL-based method known as Neural Processes-based deep transfer learning with graph convolution network (GTNP). Feature-based transfer strategy of GTNP bridges the data distribution discrepancies of source domain and target domain in high-dimensional space. Both the joint modeling based on global and local latent variables and sparse sampling strategy reduce the demand of observable data in the target domain. The multi-scale uncertainty analysis is obtained by using the distribution characteristics of global and local latent variables. Global analysis of uncertainty enables GTNP to provide quantitative values that reflect the complexity of methods and the difficulty of tasks. Local analysis of uncertainty allows GTNP to model uncertainty (confidence of the fault detection result) at each sample affected by noise and bias. The validation of the proposed method is conducted across 3 IFD tasks, consistently showing the superior detection performance of GTNP compared to the other DTL-based methods.
Paper Structure (24 sections, 23 equations, 17 figures, 7 tables)

This paper contains 24 sections, 23 equations, 17 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Problem statement
  3. Methodology
  4. The proposed GTNP framework
  5. Embedding
  6. GraphConstruction
  7. Distribution
  8. Loss Function
  9. Experiments and analysis
  10. Dataset and preprocessing
  11. Evaluation metrics
  12. Method configurations
  13. Case study of bearing fault detection with cross-working conditions
  14. Construct the representation of source domain knowledge
  15. Analysis of hyperparameter settings for GTNP training
  16. ...and 9 more sections

Figures (17)

  • Figure 1: Schematic diagram of applying the proposed GTNP to IFD tasks in complex environment.
  • Figure 2: The transfer principle of DTL for IFD tasks in this paper.
  • Figure 3: Illustration of the 2 IFD scenarios for GTNP. (a) Generalization Performance Improvement, (b) emerging Fault Detection.
  • Figure 4: The framework of GTNP proposed in this paper.
  • Figure 5: The calculation process of the dependency graphs G and A.
  • ...and 12 more figures