Novel paralleled extreme learning machine networks for fault diagnosis of wind turbine drivetrain

  • Xian-Bo Wang
  • Zhi-Xin YangEmail author
  • Pak Kin Wong
  • Chao Deng
Regular Research Paper


With the increasing installed power of the wind turbines, the necessity of condition monitoring for wind turbine drivetrain cannot be neglected any longer. A reliable and rapid response fault diagnosis is vital for the wind turbine drivetrain system. The existing manual inspection-based methods are difficult to accomplish the real-time compound-fault monitoring task. To solve this problem, this paper proposes a novel dual extreme learning machines (Dual-ELMs) based fault diagnostic framework for feature extraction and fault pattern recognition. At the stage of feature learning, this paper applies the local mean decomposition (LMD) method to extract the production functions from the raw vibration signals. Compared with the traditional empirical mode decomposition method, the LMD method has a stronger ability to restrain the mode mixing and endpoints effect. At the stage of compound-fault classification, unlike the other widely-used classifiers, the proposed Dual-ELM networks inherit the advantages of the original extreme learning machines (ELMs), that employs two basic ELM networks for the compound-fault classification, and it does not need iterative fine-tuning of parameters. Thus the learning speed is faster than the other combinations of classifiers. The experimental validity of the proposed algorithm was conducted on a test rig for vibration analysis, which demonstrated that the proposed Dual-ELMs based fault diagnostic method provides an effective measure for the observed machinery than the other available fault diagnostic methods in aspects of feature extraction and compound-fault recognition.


Fault diagnosis Vibration analysis Wind turbine drivetrain Local mean decomposition Multilayer extreme learning machines Wind energy 



This work was supported in part by the Science and Technology Development Fund of Macao SAR (FDCT) under MoST-FDCT Joint Grant 015/2015/AMJ and Grant FDCT/121/2016/A3, FDCT/194/2017/A3 in part by University of Macau under Grant MYRG2016-00160-FST and MYRG2018-00248-FST, and in part by the Ministry of Science and Technology of China under Grant 2016YFE0121700.


  1. 1.
    Wang X-B, Yang Z-X, Yan X-A (2018) Novel particle swarm optimization-based variational mode decomposition method for the fault diagnosis of complex rotating machinery. IEEE ASME Trans Mechatron 23:68–79CrossRefGoogle Scholar
  2. 2.
    Yang Y, Dong X, Peng Z, Zhang W, Meng G (2015) Vibration signal analysis using parameterized timefrequency method for features extraction of varying-speed rotary machinery. J Sound Vib 335:350–366CrossRefGoogle Scholar
  3. 3.
    Shao L, Liu L, Li X (2014) Feature learning for image classification via multiobjective genetic programming. IEEE Trans Neural Netw Learn Syst 25(7):1359–1371CrossRefGoogle Scholar
  4. 4.
    Dai X, Gao Z (2013) From model, signal to knowledge: a data-driven perspective of fault detection and diagnosis. IEEE Trans Ind Inf 9(4):2226–2238CrossRefGoogle Scholar
  5. 5.
    Yan Z, Miyamoto A, Jiang Z (2009) Frequency slice wavelet transform for transient vibration response analysis. Mech Syst Signal Process 23(5):1474–1489CrossRefGoogle Scholar
  6. 6.
    Kia SH, Henao H, Capolino G-A (2009) Diagnosis of broken-bar fault in induction machines using discrete wavelet transform without slip estimation. IEEE Trans Ind Appl 45(4):1395–1404CrossRefGoogle Scholar
  7. 7.
    Al-Badour F, Sunar M, Cheded L (2011) Vibration analysis of rotating machinery using time-frequency analysis and wavelet techniques. Mech Syst Signal Process 25(6):2083–2101CrossRefGoogle Scholar
  8. 8.
    Jauregui-Correa JC (2013) The effect of nonlinear traveling waves on rotating machinery. Mech Syst Signal Process 39(1):129–142CrossRefGoogle Scholar
  9. 9.
    Wang Z, Han Z, Gu F, Gu JX, Ning S (2015) A novel procedure for diagnosing multiple faults in rotating machinery. ISA Trans 55:208–218CrossRefGoogle Scholar
  10. 10.
    Smith JS (2005) The local mean decomposition and its application to eeg perception data. J R Soc Interface 2(5):443–454MathSciNetCrossRefGoogle Scholar
  11. 11.
    Cheng J, Yang Y, Yang Y (2012) A rotating machinery fault diagnosis method based on local mean decomposition. Digital Signal Process 22(2):356–366MathSciNetCrossRefGoogle Scholar
  12. 12.
    Zheng Z, Jiang W, Wang Z, Zhu Y, Yang K (2015) Gear fault diagnosis method based on local mean decomposition and generalized morphological fractal dimensions. Mech Mach Theory 91:151–167CrossRefGoogle Scholar
  13. 13.
    Li Y, Xu M, Haiyang Z, Wei Y, Huang W (2015) A new rotating machinery fault diagnosis method based on improved local mean decomposition. Digital Signal Process 46:201–214MathSciNetCrossRefGoogle Scholar
  14. 14.
    Ziad S, Hojjat A (2011) Probabilistic neural networks for diagnosis of alzheimer’s disease using conventional and wavelet coherence. J Neurosci Methods 197(1):165–70CrossRefGoogle Scholar
  15. 15.
    Tang J, Deng C, Huang G-B (2016) Extreme learning machine for multilayer perceptron. IEEE Trans Neural Netw Learn Syst 27(4):809–821MathSciNetCrossRefGoogle Scholar
  16. 16.
    Cheng X, Liu H, Xu X, Sun F (2017) Denoising deep extreme learning machine for sparse representation. Memet Comput 9(3):199–212CrossRefGoogle Scholar
  17. 17.
    Lu H, Du B, Liu J, Xia H, Yeap WK (2017) A kernel extreme learning machine algorithm based on improved particle swam optimization. Memet Comput 9(2):121–128CrossRefGoogle Scholar
  18. 18.
    Das SP, Padhy S (2016) Unsupervised extreme learning machine and support vector regression hybrid model for predicting energy commodity futures index. Memet Comput 3:1–14Google Scholar
  19. 19.
    Huang GB, Zhou H, Ding X, Zhang R (2012) Extreme learning machine for regression and multiclass classification. IEEE Trans Syst Man Cybern Part B Cybern 42(42):513–29CrossRefGoogle Scholar
  20. 20.
    Yang Z-X, Wang X-B, Zhong J-H (2016) Representational learning for fault diagnosis of wind turbine equipment: a multi-layered extreme learning machines approach. Energies 9(6):379CrossRefGoogle Scholar
  21. 21.
    Yang Y, Wu QJ, Wang Y (2016) Autoencoder with invertible functions for dimension reduction and image reconstruction. IEEE Trans Syst Man Cybern Syst PP(99):1–15Google Scholar
  22. 22.
    Yang Y, Wu QJ (2016) Multilayer extreme learning machine with subnetwork nodes for representation learning. IEEE Trans Cybern 46(11):2570–2583CrossRefGoogle Scholar
  23. 23.
    Huang G-B, Zhu Q-Y, Siew C-K (2006) Extreme learning machine: theory and applications. Neurocomputing 70(1):489–501CrossRefGoogle Scholar
  24. 24.
    Yang Y, Wu QJ (2016) Extreme learning machine with subnetwork hidden nodes for regression and classification. IEEE Trans Cybern 46(12):2885–2898CrossRefGoogle Scholar
  25. 25.
    Yang Y, Wu QJ, Wang Y, Zeeshan K, Lin X, Yuan X (2015) Data partition learning with multiple extreme learning machines. IEEE Trans Cybern 45(8):1463–1475CrossRefGoogle Scholar
  26. 26.
    Gong X, Qiao W (2013) Bearing fault diagnosis for direct-drive wind turbines via current-demodulated signals. IEEE Trans Ind Electron 60(8):3419–3428CrossRefGoogle Scholar
  27. 27.
    Huang NE (1998) Huang, n.e, et al.: The empirical mode decomposition and the hilbert spectrum for nonlinear and non-stationary time series analysis. proc. r. soc. lond. a 454, 903–995. Proc R Soc A Math Phys Eng Sci 454(1971):903–995MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Authors and Affiliations

  1. 1.State Key Laboratory of Internet of Things for Smart City, Department of Electromechanical Engineering, Faculty of Science and TechnologyUniversity of MacauMacauChina
  2. 2.School of Mechanical Science and EngineeringHuazhong University of Science and TechnologyWuhanChina

Personalised recommendations