样本空间基于多级高维特征表示的微小故障诊断

张彩霞, 王子涵, 文成林, 刘国文, 余伟

电子学报 ›› 2020, Vol. 48 ›› Issue (8) : 1647-1654.

PDF(801 KB)
PDF(801 KB)
电子学报 ›› 2020, Vol. 48 ›› Issue (8) : 1647-1654. DOI: 10.3969/j.issn.0372-2112.2020.08.026
学术论文

样本空间基于多级高维特征表示的微小故障诊断

  • 张彩霞1,3, 王子涵2, 文成林2, 刘国文1,3, 余伟1,3
作者信息 +

Sample Space Based on Multi-level High Dimensional Feature Representation Micro-fault Diagnosis

  • ZHANG Cai-xia1,3, WANG Zi-han2, WEN Cheng-lin2, LIU Guo-wen1,3, YU Wei1,3
Author information +
文章历史 +

摘要

传统主元分析(Principal Component Analysis,PCA)、相对主元分析等多元统计法基于阈值诊断故障,由于是原空间等价表示,并未增加任何信息量,使得微小故障难以诊断;且降维分成主元空间和残差空间,微小信息得不到充分表示.深度学习在模式识别方面有成功的应用,深度学习多层次网络对细节进行线性组合表示,但不具备可解释性,仅有训练结果无理论依据,机理分析困难.本文提出一种将主元分析思想与深度学习思想结合的故障诊断方法,在原PCA基础上先扩维再降维,使得原始空间中不能表达的信息充分表达,且具备可解释性.理论和仿真实验分析表明,本文方法能判断出传统PCA无法检测的微小故障,提高了故障检测的检出率,且具备可解释性.

Abstract

Traditional principal component analysis, relative principal component analysis and other multivariate statistical methods based on threshold to do the fault diagnosis. Since multivariate statistical method is an equivalent representation of the original space, it does not add any amount of information, making it difficult to diagnose minor faults. And the original space is reduced dimensionally into the principal component space and the residual space, so that the tiny information cannot be fully expressed. Deep learning has been successfully applied in pattern recognition. However, multilevel networks of deep learning represent linear combinations of details but do not have explanatory. Only with the result of training without theoretical basis, it is difficult to carry out mechanistic analysis. This paper presents a fault diagnosis method which combines PCA thought and deep learning thought. Based on the original PCA, this paper first expands and then reduces the dimension, making the inexplicit information in the original space fully expressed and interpreted. Theoretical and simulation experiments show that this method can judge the minor faults which cannot be detected by traditional PCA, improve the detection rate of fault detection and have interpretability.

关键词

多级高维 / 主元分析 / 投影标架 / 故障诊断

Key words

multilevel high-dimensional / principal component analysis / projection frame / fault diagnosis

引用本文

导出引用
张彩霞, 王子涵, 文成林, 刘国文, 余伟. 样本空间基于多级高维特征表示的微小故障诊断[J]. 电子学报, 2020, 48(8): 1647-1654. https://doi.org/10.3969/j.issn.0372-2112.2020.08.026
ZHANG Cai-xia, WANG Zi-han, WEN Cheng-lin, LIU Guo-wen, YU Wei. Sample Space Based on Multi-level High Dimensional Feature Representation Micro-fault Diagnosis[J]. Acta Electronica Sinica, 2020, 48(8): 1647-1654. https://doi.org/10.3969/j.issn.0372-2112.2020.08.026
中图分类号: TP273   

参考文献

[1] Ren L,Xu Z Y,Yan X Q.Single-sensor incipient fault detection[J].IEEE Sensors Journal,2011,11(9):2102-210.
[2] Yan R,Gao R X,Chen X.Wavelets for fault diagnosis of rotary machines:a review with applications[J].Signal Processing,2014,96(5):1-15.
[3] Wong P K,Yang Z,Chi M V,et al.Real-time fault diagnosis for gas turbine generator systems using extreme learning machine[J].Neurocomputing,2014,128(5):249-257.
[4] 周东华,李钢,李元.数据驱动的工业过程故障诊断技术[M].科学出版社,2011:23-27,58-59.
[5] 文成林,胡静,王天真,陈志国.相对主元分析及其在数据压缩和故障诊断中的应用研究[J].自动化学报,2008(09):1128-1139. WEN Cheng-Lin,HU Jing,Wang TIAN-Zhen,CHEN Zhi-Guo.Relative PCA with applications of data compression and fault diagnosis[J].Acta Automatica Sinica,2008,34(9):1128-1139.(in Chinese)
[6] 王天真,刘远,汤天浩,陈炎.基于相对主元分析的动态数据窗口故障检测方法[J].电工技术学报,2013,28(1):142-148. WANG Tian-zhen,LIU Yu-an,TANG Tian-ha,CHEN Yan.Dynamic data window fault detection method based on relative principal component analysis[J].Transactions of China Electrotechnical Society,2013,28(1):142-148.(in Chinese)
[7] 周福娜,文成林,陈志国,冷元宝.基于指定元分析的多级相对微小故障诊断方法[J].电子学报,2010,38(8):1874-1879. WANG Tian-zhen,LIU Yuan,TANG Tian-hao,CHEN Yan.Dynamic data window fault detection method based on relative principal component analysis[J].Acta Electronica Sinica,2010,38(8):1874-1879.(in Chinese)
[8] Harmouche J,Delpha C,Diallo D.Incipient fault detection and diagnosis based on Kullback-Leibler divergence using principal component analysis:Part I[J].Signal Processing,2014,94:278-28.
[9] Harmouche J,Delpha C,Diallo D.Incipient fault detection and diagnosis based on Kullback-Leibler divergence using principal component analysis:Part II[J].Signal Processing,2015,109:334-344
[10] 文成林,吕菲亚,包哲静,等.基于数据驱动的微小故障诊断方法综述[J].自动化学报,2016,42(9):1285-1299. WEN Cheng-Lin,LÜ Fei-Ya,BAO Zhe-Jing,LIU Mei-Qin.Areview of data driven-based incipient fault diagnosis[J].Acta Automatica Sinica,2016,42(9):1285-1299.(in Chinese)
[11] Wise B M,Gallagher N B.The process chemometrics approach to process monitoring and fault detection[J].Journal of Process Control,1996,6(6):329-348.
[12] Lv F,Wen C,Bao Z,et al.Fault diagnosis based on deep learning[A].American Control Conference(ACC)[C].Boston,MA,USA,IEEE,2016.6851-6856.
[13] Tamilselvan P,Wang P.Failure diagnosis using deep belief learning based health state classication[J].Reliability Engineering & System Safety,2013,115:124-135.
[14] Chang C H.Deep and shallow architecture of multilayer neural networks[J].IEEE Transactions on Neural Networks and Learning Systems,2015,26(10):2477-2486.
[15] Lu S,Liu H,Li C.Manifold Regularized stacked autoencoder for feature learning[A].The 2015 IEEE International Conference on Systems,Man,and Cybernetics[C].IEEE,2015.2950-2955.
[16] Lv F,Wen C,Liu M,Bao Z.Weighted time series fault diagnosis based on a stacked sparse autoencoder[J].Journal of Chemometrics,2017,31(4):e2912.
[17] R Raina,A Battle,H Lee,et al.Self-taught learning:transfer learning from unlabeled data[A].The 24th Int Conf on Machine Learning[C].New York:ACM Press,200.759-766.
[18] Baldi P,Hornik K.Neural networks and principal component analysis:Learning from examples without local minima[J].Neural Networks,1989,2(1):53-58.
[19] Japkowicz N,Hanson S J,Gluck M A.Nonlinear autoassociation is not equivalent to PCA[J].Neural Computation,2000,12(3):531-545.
[20] 胡静.基于多元统计分析的故障诊断与质量监测研究[D].浙江大学,2015.

基金

国家自然科学基金 (No.61803087); 广东省基础与应用基础研究基金粤港澳应用教学中心项目 (No.2019KTSCX192); 佛山市核心技术攻关项目 (No.1920001001367); 佛山市科技创新项目 (No.2016AG10011)
PDF(801 KB)

1124

Accesses

0

Citation

Detail

 
段落导航
相关文章

/