RESEARCH ON ISOMAP AND DEEP BELIEF NETWORK-BASED MACHINE HEALTH STATE ASSESSMENT

CHEN JianYe; WANG ShaJing; WEI CunHai; WEI ZhiJiang; WU Lin

doi:10.16579/j.issn.1001.9669.2019.05.003

您当前的位置：

首页 >

文章列表页 >

RESEARCH ON ISOMAP AND DEEP BELIEF NETWORK-BASED MACHINE HEALTH STATE ASSESSMENT

更新时间：2022-09-22

- RESEARCH ON ISOMAP AND DEEP BELIEF NETWORK-BASED MACHINE HEALTH STATE ASSESSMENT
- Journal of Mechanical Strength Vol. 41, Issue 5, Pages: 1029-1034(2019)
- 作者机构：
  
  1. 兰州理工大学机电工程学院
  2. 兰州电源车辆研究所有限公司
  3. 甘肃省内燃机电站工程技术研究中心
- 作者简介：
- 基金信息：
- DOI：10.16579/j.issn.1001.9669.2019.05.003
  CLC：
扫描看全文
CHEN JianYe, WANG ShaJing, WEI CunHai, et al. RESEARCH ON ISOMAP AND DEEP BELIEF NETWORK-BASED MACHINE HEALTH STATE ASSESSMENT. [J]. 41(5):1029-1034(2019)
DOI：

CHEN JianYe, WANG ShaJing, WEI CunHai, et al. RESEARCH ON ISOMAP AND DEEP BELIEF NETWORK-BASED MACHINE HEALTH STATE ASSESSMENT. [J]. 41(5):1029-1034(2019) DOI： 10.16579/j.issn.1001.9669.2019.05.003.

摘要

运用等距映射(Isomap)算法将信号的特征数据从原始高维空间映射到低维空间,然后融合深度置信网络(DBN)来诊断齿轮的故障状态。进行了齿轮箱运行至故障的实验研究,经过一系列与未进行空间转换的原始特征、主成分分析(PCA)和拉普拉斯特征映射(LE)等两种空间转换方法及与隐马尔科夫模型(HMM)和BP神经网络等模型算法的对比分析,所提方法被证明在齿轮的故障诊断及状态评估中更加准确有效。

Abstract

This paper integrates Isomap that transforms data features from original high-dimensional space to projected low-dimensional space into DBN model for gear fault diagnosis. The run-to-failure experiment of gearbox was conducted for validation studies, through a series of comparison experiments with the original features without space transformation, PCA and LE, HMM and BPNN, the proposed method in this paper is proved more effective for gear fault diagnosis and state assessment.

关键词

特征提取等距映射特征转换深度置信网络齿轮故障诊断

Keywords

Feature extractionIsomapFeature transformationDBNGear fault diagnosis

references

Views

129

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

WEAK FAULT FEATURE EXTRACTION OF ROLLING BEARING BASED ON PARAMETER OPTIMIZED MOMEDA AND CEEMDAN

FAULT FEATURE EXTRACTION OF ROLLING ELEMENT BEARINGS BASED ON ADAPTIVE MCKD

THE REVIEW OF MECHANICAL FAULT DIAGNOSIS METHODS BASED ON CONVOLUTIONAL NEURAL NETWORK

DYNAMIC THRESHOLD CALCULATION METHOD FOR BEARING OF HELICOPTER SWASHPLATE BASED ON BAYESIAN INFERENCE

ULTRASONIC SIGNAL FEATURE EXTRACTION METHOD BASED ON GENERAL CROSS-VALIDATION THRESHOLDING IN SYNCHROSQUEEZING WAVELET DOMAIN

Related Author

No data

Related Institution

School of Photoelectric Information ＆ Computer Engineering,University of Shanghai for Science and Technology

State Key Laboratory of Traction Power,Southwest Jiaotong University

Campus of Shijiazhuang,Army Engineering University

Key Laboratory of Helicopter Rotor Dynamics,China Helicopter Researchand Development Institute

Key Laboratory of Image Processing and Pattern Recognition of Jiangxi Province,Nanchang Hangkong University

⁰