基于DCAE-CNN的自动倾斜器滚动轴承故障诊断

被引：34

作者：

万齐杨 ^{[1
]}

熊邦书 ^{[1
]}

李新民 ^{[2
]}

孙伟 ^{[2
]}

机构：

[1] 南昌航空大学图像处理与模式识别省重点实验室

[2] 中国直升机设计研究所直升机旋翼动力学重点实验室

来源：

振动与冲击 | 2020年 / 39卷 / 11期

关键词：

故障诊断; 小波时频图; 深度学习; 自动倾斜器;

D O I：

10.13465/j.cnki.jvs.2020.11.036

中图分类号：

V267 [航空器的维护与修理];

学科分类号：

082503 [航空宇航制造工程];

摘要：

针对直升机自动倾斜器滚动轴承工况复杂、噪声干扰大,造成故障诊断效果不佳的问题,提出一种基于深度卷积自编码器(Deep Convolutional AutoEncoder,DCAE)和卷积神经网络(Convolutional Neural Network,CNN)的轴承故障诊断方法。该方法首先采用小波变换方法构造不同状态下振动信号的时频图,然后使用DCAE对时频图进行图像去噪,最后利用CNN对去噪后的时频图进行故障分类。利用课题组和美国凯斯西储大学的滚动轴承故障数据开展诊断实验,并与CNN、堆叠降噪自编码器(Stacked Denoise AutoEncoder,SDAE)两种深度学习方法进行对比,结果表明,该方法在高噪声环境下具有更高的故障识别率。

引用

页码：273 / 279

页数：7

共 16 条

[1]

Multi-Layer domain adaptation method for rolling bearing fault diagnosis.[J].Xiang Li;Wei Zhang;Qian Ding;Jian-Qiao Sun.Signal Processing.2019,

[2]

An improved Bayesian network method for fault diagnosis [J].

Wang, Yalin ;

Yang, Haibing ;

Yuan, Xiaofeng ;

Cao, Yue .

IFAC PAPERSONLINE, 2018, 51 (21) :341-346

[3]

Laser stripe image denoising using convolutional autoencoder.[J].Zhuoqun Fang;Tong Jia;Qiusheng Chen;Ming Xu;Xi Yuan;Chengdong Wu.Results in Physics.2018,

[4]

Intelligent fault diagnosis of hot die forging press based on binary decision diagram and fault tree analysis.[J].Chunping Cao;Meng Li;Yu Li;Yu Sun.Procedia Manufacturing.2018,

[5]

Deep convolutional neural network model based chemical process fault diagnosis.[J].Hao Wu;Jinsong Zhao.Computers and Chemical Engineering.2018,

[6]

Rolling element bearing fault diagnosis using convolutional neural network and vibration image.[J].Duy-Tang Hoang;Hee-Jun Kang.Cognitive Systems Research.2018,

[7]

Performance evaluation of image denoising developed using convolutional denoising autoencoders in chest radiography.[J].Donghoon Lee;Sunghoon Choi;Hee-Joung Kim.Nuclear Inst. and Methods in Physics Research; A.2018,

[8]

A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load.[J].Wei Zhang;Chuanhao Li;Gaoliang Peng;Yuanhang Chen;Zhujun Zhang.Mechanical Systems and Signal Processing.2018,

[9]

A novel fault diagnosis scheme applying fuzzy clustering algorithms.[J].A. Rodríguez Ramos;O. Llanes-Santiago;J.M. Bernal de Lázaro;C. Cruz Corona;A.J. Silva Neto;J.L. Verdegay Galdeano.Applied Soft Computing.2017,

[10]

Hierarchical adaptive deep convolution neural network and its application to bearing fault diagnosis.[J].Xiaojie Guo;Liang Chen;Changqing Shen.Measurement.2016,

← 1 2 →