A domain adaptation model for early gear pitting fault diagnosis based on deep transfer learning network

Jialin Li, Xueyi Li, David He, Yongzhi Qu

Research output: Contribution to journalArticle

Abstract

In recent years, research on gear pitting fault diagnosis has been conducted. Most of the research has focused on feature extraction and feature selection process, and diagnostic models are only suitable for one working condition. To diagnose early gear pitting faults under multiple working conditions, this article proposes to develop a domain adaptation diagnostic model–based improved deep neural network and transfer learning with raw vibration signals. A particle swarm optimization algorithm and L2 regularization are used to optimize the improved deep neural network to improve the stability and accuracy of the diagnosis. When using the domain adaptation diagnostic model for fault diagnosis, it is necessary to discriminate whether the target domain (test data) is the same as the source domain (training data). If the target domain and the source domain are consistent, the trained improved deep neural network can be used directly for diagnosis. Otherwise, the transfer learning is combined with improved deep neural network to develop a deep transfer learning network to improve the domain adaptability of the diagnostic model. Vibration signals for seven gear types with early pitting faults under 25 working conditions collected from a gear test rig are used to validate the proposed method. It is confirmed by the validation results that the developed domain adaptation diagnostic model has a significant improvement in the adaptability of multiple working conditions.

Original languageEnglish (US)
Pages (from-to)168-182
Number of pages15
JournalProceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability
Volume234
Issue number1
DOIs
StatePublished - Feb 1 2020

Fingerprint

Pitting
Failure analysis
Gears
Feature extraction
Particle swarm optimization (PSO)
Deep neural networks

Keywords

  • Early gear pitting
  • improved deep neural network
  • multiple working conditions
  • transfer learning

Cite this

@article{d37a54f32d434f0fa1d3fb67b3f1ec96,
title = "A domain adaptation model for early gear pitting fault diagnosis based on deep transfer learning network",
abstract = "In recent years, research on gear pitting fault diagnosis has been conducted. Most of the research has focused on feature extraction and feature selection process, and diagnostic models are only suitable for one working condition. To diagnose early gear pitting faults under multiple working conditions, this article proposes to develop a domain adaptation diagnostic model–based improved deep neural network and transfer learning with raw vibration signals. A particle swarm optimization algorithm and L2 regularization are used to optimize the improved deep neural network to improve the stability and accuracy of the diagnosis. When using the domain adaptation diagnostic model for fault diagnosis, it is necessary to discriminate whether the target domain (test data) is the same as the source domain (training data). If the target domain and the source domain are consistent, the trained improved deep neural network can be used directly for diagnosis. Otherwise, the transfer learning is combined with improved deep neural network to develop a deep transfer learning network to improve the domain adaptability of the diagnostic model. Vibration signals for seven gear types with early pitting faults under 25 working conditions collected from a gear test rig are used to validate the proposed method. It is confirmed by the validation results that the developed domain adaptation diagnostic model has a significant improvement in the adaptability of multiple working conditions.",
keywords = "Early gear pitting, improved deep neural network, multiple working conditions, transfer learning",
author = "Jialin Li and Xueyi Li and David He and Yongzhi Qu",
year = "2020",
month = "2",
day = "1",
doi = "10.1177/1748006X19867776",
language = "English (US)",
volume = "234",
pages = "168--182",
journal = "Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability",
issn = "1748-006X",
publisher = "SAGE Publications Ltd",
number = "1",

}

TY - JOUR

T1 - A domain adaptation model for early gear pitting fault diagnosis based on deep transfer learning network

AU - Li, Jialin

AU - Li, Xueyi

AU - He, David

AU - Qu, Yongzhi

PY - 2020/2/1

Y1 - 2020/2/1

N2 - In recent years, research on gear pitting fault diagnosis has been conducted. Most of the research has focused on feature extraction and feature selection process, and diagnostic models are only suitable for one working condition. To diagnose early gear pitting faults under multiple working conditions, this article proposes to develop a domain adaptation diagnostic model–based improved deep neural network and transfer learning with raw vibration signals. A particle swarm optimization algorithm and L2 regularization are used to optimize the improved deep neural network to improve the stability and accuracy of the diagnosis. When using the domain adaptation diagnostic model for fault diagnosis, it is necessary to discriminate whether the target domain (test data) is the same as the source domain (training data). If the target domain and the source domain are consistent, the trained improved deep neural network can be used directly for diagnosis. Otherwise, the transfer learning is combined with improved deep neural network to develop a deep transfer learning network to improve the domain adaptability of the diagnostic model. Vibration signals for seven gear types with early pitting faults under 25 working conditions collected from a gear test rig are used to validate the proposed method. It is confirmed by the validation results that the developed domain adaptation diagnostic model has a significant improvement in the adaptability of multiple working conditions.

AB - In recent years, research on gear pitting fault diagnosis has been conducted. Most of the research has focused on feature extraction and feature selection process, and diagnostic models are only suitable for one working condition. To diagnose early gear pitting faults under multiple working conditions, this article proposes to develop a domain adaptation diagnostic model–based improved deep neural network and transfer learning with raw vibration signals. A particle swarm optimization algorithm and L2 regularization are used to optimize the improved deep neural network to improve the stability and accuracy of the diagnosis. When using the domain adaptation diagnostic model for fault diagnosis, it is necessary to discriminate whether the target domain (test data) is the same as the source domain (training data). If the target domain and the source domain are consistent, the trained improved deep neural network can be used directly for diagnosis. Otherwise, the transfer learning is combined with improved deep neural network to develop a deep transfer learning network to improve the domain adaptability of the diagnostic model. Vibration signals for seven gear types with early pitting faults under 25 working conditions collected from a gear test rig are used to validate the proposed method. It is confirmed by the validation results that the developed domain adaptation diagnostic model has a significant improvement in the adaptability of multiple working conditions.

KW - Early gear pitting

KW - improved deep neural network

KW - multiple working conditions

KW - transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85071490020&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071490020&partnerID=8YFLogxK

U2 - 10.1177/1748006X19867776

DO - 10.1177/1748006X19867776

M3 - Article

AN - SCOPUS:85071490020

VL - 234

SP - 168

EP - 182

JO - Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability

JF - Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability

SN - 1748-006X

IS - 1

ER -