access icon openaccess Application of stack marginalised sparse denoising auto-encoder in fault diagnosis of rolling bearing

This is an open access article published by the IET under the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/)
Received 17/07/2018, Accepted 17/08/2018, Revised 17/08/2018, Published 28/08/2018

When a fracturing vehicle is working, it generally needs to bear high loads, media corrosion and erosion. For this special working environment, this study proposes a rolling bearing fault diagnosis method based on stack marginalised sparse denoising auto-encoder (SDAE). This method combines the sparse auto-encoder (SAE) and the denoising auto-encoder (DAE) and combines the characteristics of dimensionality reduction and robustness. The method adds marginalisation to optimise the SDAE. Finally, it uses a two-layer stacking method. The output results of the second marginalised SDAE are used as input to the softmax classifier for learning training and classification testing. This improved method (stack SDAE) improves the denoising ability, reduces the computational complexity, solves the problems of difficult parameter adjustment and slows training convergence. The experimental tests were carried out on the failure of pitting corrosion of the outer ring of the bearing, pitting failure of the inner ring, and cracking of the rolling element. The results show that the algorithm can effectively improve the accuracy of fault diagnosis of rolling bearings, and it has greatly improved than the algorithms of SAEs and DAE.

Inspec keywords: computational complexity; corrosion; fracture; rolling bearings; failure analysis; signal denoising; mechanical engineering computing; cracks; fault diagnosis

Other keywords: stack marginalised sparse denoising autoencoder application; pitting corrosion failure; computational complexity; denoising autoencoder; learning training; cracking; erosion; softmax classifier; fracturing vehicle; SDAE; SAE; two-layer stacking method; DAE; media corrosion; classification testing; sparse autoencoder; rolling bearing fault diagnosis method

Subjects: Digital signal processing; Signal processing and detection; Surface treatment and coating techniques; Mechanical components; Maintenance and reliability; Mechanical engineering applications of IT; Fracture mechanics and hardness (mechanical engineering); Computational complexity; Civil and mechanical engineering computing

References

    1. 1)
      • 7. Chen, R., Yang, X., Yang, L., et al: ‘Diagnosis of rolling bearing damage degree based on stacked sparse and noise auto-encoding deep neural network’, J. Vib. Shock, 2017, 36, (21), pp. 125–131. (in Chinese).
    2. 2)
      • 5. Yu, K., Jia, L., Chen, Y., et al: ‘Deep learning, yesterday, today and tomorrow’, J. Comput. Res. Dev., 2013, 50, (9), pp. 1799–1804(in Chinese).
    3. 3)
      • 4. Cao, Y.: ‘Intelligent recommendation system based on neural network and fuzzy logic’ (Chongqing University, Chongqing, 2006) (in Chinese).
    4. 4)
      • 15. Cheriyadat, A.M.: ‘Unsupervised feature learning for aerial scene classification’, IEEE Trans. Geosci. Remote Sens., 2013, 52, (1), pp. 439–451.
    5. 5)
      • 13. Ma, H., Ma, S., Xu, Y., et al: ‘Image denoising based on improved stacked sparse denoising auto-encoder’, Comput. Eng. Applic., 2018, (4), pp. 199–204(in Chinese).
    6. 6)
      • 10. Sun, W., Shao, S., Yan, R.: ‘Induction motor fault diagnosis based on sparse automatic coded deep neural network’, J. Mech. Eng., 2016, 52, (9), pp. 65–71. (in Chinese).
    7. 7)
      • 14. Yuan, K., Ling, Q., Yin, W.: ‘On the convergence of decentralized gradient descent’, Mathematics, 2016, 26, (3), pp. 1835–1854.
    8. 8)
      • 3. Ali, H.A., Sai Ho, L., Mohammed, H.J.: ‘Diagnosis system for Parkinson's disease using speech characteristics of patients and deep belief network’, CAAI Trans. Intell. Technol., 2017, 2, (4), pp. 246–253.
    9. 9)
      • 11. Mnih, V., Kavukcuoglu, K., Silver, D., et al: ‘Playing atari with deep reinforcement learning’, Comput. Sci., 2013, arXiv preprint arXiv: 1312.5602.
    10. 10)
      • 8. Rumelhart, D.E., Hinton, G.E., Williams, R.J.: ‘Learning representations by back-propagating errors’ in ‘Neurocomputing: foundations of research’ (MIT Press, 1988), pp. 533–536.
    11. 11)
      • 12. Yin, H., Jiao, X., Chai, Y., et al: ‘Scene classification based on single-layer SAE and SVM’, Expert Syst. Appl., 2015, 42, (7), pp. 3368–3380.
    12. 12)
      • 1. Patel, V.N., Tandon, N., Pandey, R.K.: ‘Vibrations generated by rolling element bearings having multiple local defects on races ⋆’, Procedia Technol., 2014, 14, pp. 312–319.
    13. 13)
      • 2. Humphrey, E.J., Bello, J.P., Lecun, Y.: ‘Feature learning and deep architectures: new directions for music informatics’, J. Intell. Inf. Syst., 2013, 41, (3), pp. 461–481.
    14. 14)
      • 16. Chen, M., Weinberger, K., Sha, F., et al: ‘Marginalized denoising auto-encoders for nonlinear representations’. Int. Conf. on Machine Learning, Beijing, China, June 2014, pp. 1476–1484.
    15. 15)
      • 9. Lecun, Y., Bottou, L., Orr, G.B., et al: ‘Efficient BackProp’ in ‘Neural networks: tricks of the trade’ (Springer, Berlin, Heidelberg, 1998), pp. 9–50.
    16. 16)
      • 6. Tang, F., Long, H.: ‘Sparse auto-coded deep neural network and its application in fault diagnosis of rolling bearings’, J. Mech. Eng. Tech., 2018, (3). (in Chinese).
http://iet.metastore.ingenta.com/content/journals/10.1049/joe.2018.8267
Loading

Related content

content/journals/10.1049/joe.2018.8267
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading