Freight Car Roller Bearing Fault Detection Using Artificial Neural Networks and Support Vector Machines

  • Daniel MarainiEmail author
  • C. Nataraj
Conference paper
Part of the Mechanisms and Machine Science book series (Mechan. Machine Science, volume 23)


This paper deals with fault detection and diagnostics of freight rail tapered roller bearings using statistical features and auto-regressive m.odel parameters extracted from bearing vibration signals. Two classifiers are used to distinguish cup, cone, and roller defects, namely, artificial neural networks (ANNs) and support vector machines (SVMs). Time domain vibration data is collected from normal bearings, and defective bearings with combinations of cup, cone, and roller defects. Features are used to train both ANNs and SVMs, and performance is compared for both classifiers. Mutual information between time domain features and target classes is computed in order to rank features with the highest relevance. Excellent performance was observed using laboratory data and the trained ANNs and SVMs, with testing accuracy as high as 100 % in some cases.


Condition monitoring Artificial neural networks Support vector machines Mutual information Feature selection Bearing fault 


  1. 1.
    Ferreira JLA, Balthazar JC, Araujo APN (2003) An investigation of rail bearing reliability under real conditions of use. Eng Fail Anal 10(6):745–758CrossRefGoogle Scholar
  2. 2.
    Liu X, Saat MR, Barkan CP (2012) Analysis of causes of major train derailment and their effect on accident rates. Trans Res Rec: J Trans Res Board 2289(1):154–163CrossRefGoogle Scholar
  3. 3.
    Koester L, Turner et al (2012) Near-race ultrasonic inspection of tapered roller bearing components for non-metallic defects. In: ASME 2012 rail transportation division fall technical conference (pp 107–110). ASMEGoogle Scholar
  4. 4.
    Tarawneh CM, Cole KD et al (2008) Experiments and models for the thermal response of railroad tapered-roller bearings. Int J Heat Mass Transf 51(25):5794–5803CrossRefzbMATHGoogle Scholar
  5. 5.
    Tarawneh C, Wilson BM et al (2008) Dynamic bearing testing aimed at identifying the root cause of warm bearing temperature trending. In: Proceedings of the 2008 ASME RTD fall technical conference, RTDF2008-74036, Chicago, Illinois, 24–26 Sept 2008Google Scholar
  6. 6.
    Al-Dossary S, Hamzah RI, Mba D (2009) Observations of changes in acoustic emission waveform for varying seeded defect sizes in a rolling element bearing. Appl Acoust 70(1):58–81CrossRefGoogle Scholar
  7. 7.
    He Q, Wang J et al (2013) Wayside acoustic diagnosis of defective train bearings based on signal resampling and information enhancement. J Sound Vib 332(21):5635–5649CrossRefGoogle Scholar
  8. 8.
    Tarawneh CM, Kypuros JA et al (2009) Vibration signatures of temperature trended bearings in field and laboratory testing. In: ASMEGoogle Scholar
  9. 9.
    Tandon N, Choudhury A (1999) A review of vibration and acoustic measurement methods for the detection of defects in rolling element bearings. Tribol Int 32(8):469–480CrossRefGoogle Scholar
  10. 10.
    Randall RB, Antoni J (2011) Rolling element bearing diagnostics—a tutorial. Mech Syst Signal Process 25(2):485–520CrossRefGoogle Scholar
  11. 11.
    Kappaganthu K, Nataraj C, Samanta B (2009) Model based bearing fault detection using sample entropy features and support vector machines. (Preprint submitted to Elsevier)Google Scholar
  12. 12.
    Jardine AK, Lin D, Banjevic D (2006) A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mech Syst Signal Process 20(7):1483–1510CrossRefGoogle Scholar
  13. 13.
    Samanta B, Al-Balushi KR, Al-Araimi SA (2003) ANNs and SVMs with genetic algorithm for bearing fault detection. Eng Appl Artif Intell 16(7):657–665CrossRefGoogle Scholar
  14. 14.
    Sugumaran V, Ramachandran KI (2011) Effect of number of features on classification of roller bearing faults using SVM and PSVM. Expert Syst Appl 38(4):4088–4096CrossRefGoogle Scholar
  15. 15.
    Kappaganthu K, Nataraj C (2011) Feature selection for fault detection in rolling element bearings using mutual information. J Vib Acou 133(6):061001 Google Scholar
  16. 16.
    Peng H, Long F, Ding C (2005) Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy. IEEE Trans Pattern Anal Mach Intell 27(8):1226–1238CrossRefGoogle Scholar
  17. 17.
    Ljung L (1999) System identification. Wiley, New YorkGoogle Scholar
  18. 18.
    Bishop CM, Nasrabadi NM (2006) Pattern recognition and machine learning, vol 1. Springer, New York, p 740zbMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.Amsted Rail Company, IncChicagoUSA
  2. 2.Department of Mechanical EngineeringVillanova UniversityVillanovaUSA

Personalised recommendations