基于数据生成与迁移学习的轴承小样本故障诊断

吴定会; 方钦; 吴楚宜

doi:10.16578/j.issn.1004.2539.2020.11.023

您当前的位置：

首页 >

文章列表页 >

基于数据生成与迁移学习的轴承小样本故障诊断

试验分析 | 更新时间：2022-11-10

- 基于数据生成与迁移学习的轴承小样本故障诊断
- Bearing Small Sample Fault Diagnosis based on Data Generation and Transfer Learning
- 机械传动 2020年44卷第11期页码：139-144
- 作者机构：
  
  1.江南大学物联网应用技术教育部工程研究中心，江苏无锡 214122
- 作者简介：
  
  吴定会（1977— ），男，安徽合肥人，博士，教授，研究方向为风能发电。
  方钦（1996— ），男，山东青岛人，硕士研究生，研究方向为风电机轴承故障诊断。
- 基金信息：
- DOI：10.16578/j.issn.1004.2539.2020.11.023
  中图分类号：
扫描看全文
吴定会,方钦,吴楚宜.基于数据生成与迁移学习的轴承小样本故障诊断[J].机械传动,2020,44(11):139-144.

Wu Dinghui,Fang Qin,Wu Chuyi.Bearing Small Sample Fault Diagnosis based on Data Generation and Transfer Learning[J].Journal of Mechanical Transmission,2020,44(11):139-144.
吴定会,方钦,吴楚宜.基于数据生成与迁移学习的轴承小样本故障诊断[J].机械传动,2020,44(11):139-144. DOI： 10.16578/j.issn.1004.2539.2020.11.023.

Wu Dinghui,Fang Qin,Wu Chuyi.Bearing Small Sample Fault Diagnosis based on Data Generation and Transfer Learning[J].Journal of Mechanical Transmission,2020,44(11):139-144. DOI： 10.16578/j.issn.1004.2539.2020.11.023.

摘要

针对风电机轴承历史运行数据来源单一、数据量少，导致风电机轴承故障诊断性能受限问题，提出一种基于数据生成与迁移学习的轴承小样本故障诊断方法。首先，对于轴承数据集中存在类不平衡、数据稀缺的问题，提出一种基于门限机制的数据生成方法，采用与轴承驱动端同轴的桨叶端数据为模板产生足量的生成数据，结合真实数据作为源数据集；然后，根据数据的时序关联性和小样本的应用场景，提出一种基于一维卷积神经网络（One Dimensional Convolutional Neural Network，1DCNN）和双向门限单元（Bidirectional Gated Recurrent Unit，BiGRU）的迁移学习（Transfer Learning）方法，先用源数据集在训练网络上训练获得源模型，再用少量驱动端数据作为目标数据集对其进行微调（Fine-tuning）获得目标模型；最后，对目标模型全连接层的输出采用Softmax函数进行故障诊断。实验表明，提出的故障检测方法在目标集小样本数据的场景下平均精度达到99.67%，分类效果明显，泛化能力强。

Abstract

Aiming at the problem of limited fault diagnosis performance caused by a single source of historical operating data for wind turbine bearings and a small amount of data， a small sample fault diagnosis method of bearings based on data generation and transfer learning is proposed. First of all， for the problems of class imbalance and data scarcity in the wind turbine bearing dataset， a data generation method based on a gate mechanism is proposed. The blade-end data coaxial with the drive-end of the bearing is used as a template to generate a sufficient amount of generated data. The data is used as the source dataset. Then， according to the time series correlation of the data and the small sample application scenario， a transfer learning method based on one-dimensional convolutional neural network （1DCNN） and bidirectional gated recurrent unit （BiGRU） is proposed. First， the source dataset is trained on the training network to obtain the source model， and then a small amount of drive-end data is used as the target dataset to fine-tuning to obtain the target model. Finally， the Softmax function is used for fault diagnosis on the output of the fully-connected layer of the target model. Experiments show that the proposed fault detection method has an average accuracy of 99.67% in the scenario of small sample data of the target dataset， so the obvious classification effect and strong generalization ability could be seen.

关键词

风电机轴承小样本数据生成门限机制迁移学习微调故障诊断

Keywords

Wind turbine bearingSmall sampleData generationGate mechanismTransfer learningFine-tuningFault diagnosis

references

WEI Y，LI Y Q，XU M Q，et al.A review of early fault diagnosis approaches and their applications in rotating machinery［J］.Entropy，2019，21（4）：409-435.

KWON D H，KIM H J，KIM J，et al.A survey of deep learning-based network anomaly detection［J］.Cluster Computing，2019，22（1）：949-961.

GOODFELLOW I，POUGET-ABADIE J，MIRZA M，et al.Generative adversarial nets［J］.NIPS'14：Proceedings of the 27th Internationl Generative Adversarial Nets，2014（2）：2672-2680.

LI D，CHEN D C，GOH J，et al.Anomaly detection with generative adversarial networks for multivariate time series［J］.arXiv，2018：1809.04758.

ESTEBAN C，HYLAND S L，RATSCH G.Real-valued （Medical） time series generation with recurrent conditional GANs［J］.arXiv，2017：1706.02633.

PANOS S，TOBIAS H，ALEXANDRE M，et al.Enforcing constraints for interpolation and extrapolation in generative adversarial networks［J］.Journal of Computational Physics，2019，397：108844-108877.

NARUSE M，MATSUBARA T，CHAUVET N，et al.Generative adversarial network based on chaotic time series［J］.Scientific Reports，2019，9（1）：12963-12972.

GUO Z J，WAN Y M，YE H.A data imputation method for multivariate time series based on generative adversarial network［J］.Neurocomputing，2019，360：185-197.

LI X D，HU Y，LI M T，et al.Fault diagnostics between different type of components：A transfer learning approach［J］.Applied Soft Computing，2020，86：105950-105961.

WU Z H，JIANG H K，ZHAO K，et al.An adaptive deep transfer learning method for bearing fault diagnosis［J］.Measurement，2020，151：107227-107241.

WANG C Y，XIE Y H，ZHANG D.Deep learning for bearing fault diagnosis under different working loads and non-fault location point［J］.Journal of Low Frequency Noise，Vibration and Active Control，2019：1-13.DOI：10.1177/1461348419889511http://dx.doi.org/10.1177/1461348419889511.

CHE C C，WANG H W，FU Q，et al.Deep transfer learning for rolling bearing fault diagnosis under variable operating conditions［J］.Advances in Mechanical Engineering，2019，11（12）：1-11.

GOODFELLOW I，BING X，SHUO.Code and hyper parameters for the paper Generative Adversarial Networks［CP/OL］.［2020-01-20］.2014，https：//github.com/goodfeli/adversarialhttps://github.com/goodfeli/adversarial.

CHUNG J，GULCEHRE C，CHO K，et al.Empirical evaluation of gated recurrent neural networks on sequence modeling［J］.Eprint Arxiv，2014：1412.3555.

TRAN T U，HOANG H T T，HUYNH H X.Aspect extraction with bidirectional GRU and CRF［C］//2019 IEEE-RIVF International Conference on Computing and Communication Technologies，March 20-22，2019，Danang，Vietnam.New York：IEEE，2019：1-5.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于小波和深度小波自编码器的轴承故障诊断

基于时间子序列的轴承特征提取方法

一种基于改进VMD和UMAP的滚动轴承故障特征提取方法

基于深度迁移学习的齿轮故障诊断方法

基于DPD-1DCNN的行星齿轮箱故障诊断方法研究