Research on fault diagnosis model of rotating machinery based on two-channel input LetNet-5 convolution neural network

FU Zhongguang; WANG Shiyun; GAO Yucai; ZHOU Xiangqi

doi:10.19666/j.rlfd.202210240

English Version Reader LoginJournal Login

您当前的位置：

首页 >

文章列表页 >

Research on fault diagnosis model of rotating machinery based on two-channel input LetNet-5 convolution neural network

更新时间：2025-09-28

- Research on fault diagnosis model of rotating machinery based on two-channel input LetNet-5 convolution neural network
- Thermal Power Generation Vol. 52, Issue 3, Pages: 81-87(2023)
- 作者机构：
  
  华北电力大学电站能量传递转化与系统教育部重点实验室
- 作者简介：
- 基金信息：
- DOI：10.19666/j.rlfd.202210240
  CLC： TH17;TP183
- Published：2023
- 稿件说明：
移动端阅览
FU Zhongguang, WANG Shiyun, GAO Yucai, et al. Research on fault diagnosis model of rotating machinery based on two-channel input LetNet-5 convolution neural network[J]. Thermal power generation, 2023, 52(3): 81-87.
DOI：

FU Zhongguang, WANG Shiyun, GAO Yucai, et al. Research on fault diagnosis model of rotating machinery based on two-channel input LetNet-5 convolution neural network[J]. Thermal power generation, 2023, 52(3): 81-87. DOI： 10.19666/j.rlfd.202210240.

摘要

针对旋转机械运行过程中伴随着诸多噪声，现有单通道网络在旋转机械故障诊断过程中抗噪性较差的问题，提出了一种加入并联机制的双通道输入Let Net-5卷积神经网络模型。模型合理性检验过程采用了凯斯西储大学轴承数据集，在此基础上，添加信噪比为–10 d B的高斯白噪声模拟真实噪声情形；采用短时傅里叶变换将电机风扇端和驱动端振动数据进行处理，获得的时频图像传递至双通道输入的Let Net-5卷积神经网络进行训练学习。研究结果表明：双通道输入Let Net-5卷积神经网络模型能够良好捕捉到强噪声环境下的故障特征；相比于多尺度特征融合残差模型、多模态耦合输入神经网络模型、传统的K近邻与决策树模型及单通道输入Let Net-5卷积神经网络模型，双通道输入Let Net-5卷积神经网络具有更高的效率和精度。

Abstract

Existing single-channel networks have poor noise immunity during fault diagnosis of rotating machinery due to the many noises associated with the operation of rotating machinery. To address this problem

a two-channel input LetNet-5 convolutional neural network model incorporating a parallel mechanism was proposed. Case Western Reserve University bearing dataset was used for the model plausibility check process

based on which Gaussian white noise with a signal-to-noise ratio of-10 dB was added to simulate the real noise situation. The shorttime Fourier transform was used to process the motor fan-side and drive-side vibration data

and the resulting timefrequency images were passed to a two-channel input LetNet-5 convolutional neural network for training and learning. The results show that

the dual-channel input LetNet-5 convolutional neural network model is able to capture the fault features in a strong noise environment well

it has higher efficiency and accuracy than the multiscale feature fusion residual model

the multimodal coupled input neural network model

the conventional K-nearest neighbour and decision tree model and the single-channel input LetNet-5 convolutional neural network model.

关键词

Keywords

references

Views

481

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Dynamic prediction of pollutants emission from circulating fluidized bed unit based on CNN-GRU-MHA

MULTISENSOR INFORMATION MERGING TECHNOLOGY AND IT'S APPLICATION IN VIBRITION FAULT DIAGNOSIS OF ROTARY MACHINES

NOISE CONTROL OF DIESEL GENERATOR SETS

CAUSE ANALYSIS OF FURNACE WALL VIBRATION FOR UTILITY BOILERS

ANALYSIS OF ABNORMAL SHAFT VIBRATION FOR BEARINGS NO 2 AND 3 ON HOMEMADE 200 MW STEAM TURBINE, AS WELL AS TREATMENT THEREOF

Related Author

WANG Yongquan

GAO Mingming

WANG Weihua

ZHANG Pengxin

CHENG Yongqiang

JIN Fenghua

LI Luping

WANG Shuxin

Related Institution

National Key Laboratory of New Energy Power System (North China Electric Power University)

School of Control and Computer Engineering (North China Electric Power University)

Changsha University of Science and Technology

Zhenjiang Shipbuilding Institute

AI问答

⁰