SWDAE: A New Degradation State Evaluation Method for Metro Wheels with Interpretable Health Indicator Construction based on Unsupervised Deep Learning

Machine learning has shown advantages in assessing wheel degradation in metro vehicles for fault prognostic and health management (PHM). However, practical implementation faces challenges due to disturbances in wheel vibration signals caused by factors like load, road conditions, and temperature, in...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on instrumentation and measurement Vol. 73; p. 1
Main Authors: Mao, Wentao, Wang, Yu, Feng, Ke, Kou, Linlin, Zhang, Yanna
Format: Journal Article
Language:English
Published: New York IEEE 01.01.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Algorithms
Deep autoencoder
Deep learning
Degradation
Degradation evaluation
Feature extraction
Frequencies
Health indicator
Interpretability
Machine learning
Maintenance engineering
Principal components analysis
Road conditions
Second-generation wavelet transform
Time-frequency analysis
Vibrations
Wavelet analysis
Wavelet transforms
Wheels
ISSN:0018-9456, 1557-9662
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Machine learning has shown advantages in assessing wheel degradation in metro vehicles for fault prognostic and health management (PHM). However, practical implementation faces challenges due to disturbances in wheel vibration signals caused by factors like load, road conditions, and temperature, introducing significant noise that masks evaluation tendencies. This paper develops a novel unsupervised approach to evaluate wheel degradation under strong noise disturbance. It utilizes a second-generation wavelet transform for time-frequency analysis of noisy signals and introduces a second-generation wavelet deep autoencoder network (SWDAE) to extract adaptive feature representations in different frequency bands. The training algorithm alternately optimizes the wavelet transform and deep autoencoder network.With frequency-saliency interpretability, health indicators (HIs) for the degradation process are constructed using principal component analysis on the obtained features in each frequency band, selecting the most representative frequency component based on the monotonicity of the HIs. State changes are automatically determined using a second-order derivative-based assessment method aligned with the first/second warning strategy. Comparative experiments using Beijing Subway wheel data demonstrate the monotonicity and physical significance of the constructed HIs, with warning locations accurately matching changes in wheel diameter recorded during repairs.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0018-9456
1557-9662
DOI:10.1109/TIM.2023.3348910