Your privacy, your choice

We use essential cookies to make sure the site can function. We also use optional cookies for advertising, personalisation of content, usage analysis, and social media.

By accepting optional cookies, you consent to the processing of your personal data - including transfers to third parties. Some third parties are outside of the European Economic Area, with varying standards of data protection.

See our privacy policy for more information on the use of your personal data.

for further information and to change your choices.

Skip to main content

Advertisement

SpringerLink
Log in

Acoustic Monitoring of Railway Defects Using Deep Learning with Audio to Spectrogram Conversion

  • Original Paper
  • Published:
Journal of Vibration Engineering & Technologies Aims and scope Submit manuscript

Abstract

Purpose

A railway vehicle traveling on a stable rail line has a fixed amplitude and frequency, and hence it has constant sound values. However, line defects on the rail line alter this constant sound. Related to this, each defect type varies the reference sound value in a different way. In this study, audio signals originating from the three most common structural defects in rail systems were investigated with a deep neural network using audio to spectrogram conversion.

Method

Short-term Fourier transform (STFT) was applied to sound signals collected from the rails and a convolutional neural network (CNN) model was developed to be able to acoustically detect the three most common types of defects (rail surface deformations, joint deformations, rail corrugations) in the railway superstructure at an early stage. The model learned five different classes, which are stable joint, stable rail surface, rail surface deformation, joint deformation, and rail corrugation with a train set (5000 observations).

Results

The model obtained 89.37% accuracy with the validation data set (1000 observations). Finally, we evaluated the model with the test data (1000 observations) set that showed 87.3% accuracy.

Conclusion

In rail transportation systems, high-frequency sound and vibration energy are formed by wheel–rail interaction. This study, which is easy to implement and shows the traceability of the changes occurring directly at the source of the noise, will provide a smart structural health monitoring way for rail systems in the future. For the next study, the data will be collected on the rail vehicle. A complete real-time system is a part of our ongoing research.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
¥17,985 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (Japan)

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

Data availability

The datasets generated during and/or analysed during the current study are not publicly available but are available from the corresponding author on reasonable request.

References

  1. Zarembski AM, Einbinder D, Attoh-Okine N (2016) Using multiple adaptive regression to address the impact of track geometry on development of rail defects. Constr Build Mater 127:546–555. https://doi.org/10.1016/j.conbuildmat.2016.10.012

    Article  Google Scholar 

  2. Ngamkhanong C, Kaewunruen S, Afonso Costa BJ (2018) State of the art review of railway track resilience monitoring. Infrastruct 3(1):3. https://doi.org/10.3390/infrastructures3010003

    Article  Google Scholar 

  3. Mair C, Fararooy S (1998) Practice and potential of computer vision for railways. IEE seminar condition monitoring for rail transport systems. IEEE, London, pp 10–13. https://doi.org/10.1049/ic:19980983

    Chapter  Google Scholar 

  4. Marais JJ, Mistry KC (2003) Rail integrity management by means of ultrasonic testing. Fatigue Fract Eng Mater Struct 26(10):931–938. https://doi.org/10.1046/j.1460-2695.2003.00668.x

    Article  Google Scholar 

  5. Fitzgerald SC (1995) Inspection for rail defects by magnetic induction. Non-destructive evaluation of aging railroads. SPIE, Oakland, pp 40–44. https://doi.org/10.1117/12.212672

    Chapter  Google Scholar 

  6. Rajamaki J, Vippola M, Nurmikolu A, Viitala T (2018) Limitations of eddy current inspection in railway rail evaluation. Proc of the Inst of Mech Eng Part F: J of Rail and Rapid Transit 232(1):121–129. https://doi.org/10.1177/0954409716657848

    Article  Google Scholar 

  7. Li Q, Zhong Z, Liang Z, Liang Y (2015) Rail inspection meets big data: Methods and trends. Proceedings of the 18th international conference on network-based information systems. Association for Computing Machinery, Taipei, pp 302–308. https://doi.org/10.1109/NBiS.2015.47

    Chapter  Google Scholar 

  8. Thompson D (2009) Railway noise and vibration. Elsevier, Oxford

    Google Scholar 

  9. Nielsen JCO, Pieringer A, Thompson DJ, Torstensson PT (2021) Wheel-rail impact loads, noise and vibration: a review of excitation mechanisms, prediction methods and mitigation measures. In: Degrande G et al (eds) noise and vibration mitigation for rail transportation systems. Notes on numerical fluid mechanics and multidisciplinary design. Springer, Cham, pp 3–40. https://doi.org/10.1007/978-3-030-70289-2_1

    Chapter  Google Scholar 

  10. Koenig F, Found PA, Kumar M (2019) Innovative airport 4.0 condition-based maintenance system for baggage handling DCV systems. Int J of Prod Perform Manag 68(3):561–577. https://doi.org/10.1108/IJPPM-04-2018-0136

    Article  Google Scholar 

  11. Kostrzewski M, Melnik R (2021) Condition monitoring of rail transport systems: a bibliometric performance analysis and systematic literature review. Sens 21(14):4710. https://doi.org/10.3390/s21144710

    Article  ADS  Google Scholar 

  12. An Z, Li S, Xin Y, Xu K, Ma H (2019) An intelligent fault diagnosis framework dealing with arbitrary length inputs under different working conditions. Meas Sci Technol 30:125107. https://doi.org/10.1088/1361-6501/ab26a2

    Article  CAS  ADS  Google Scholar 

  13. Strano S, Terzo M (2019) Review on model-based methods for on-board condition monitoring in railway vehicle dynamics. Adv in Mech Eng 11:1–10. https://doi.org/10.1177/1687814019826795

    Article  Google Scholar 

  14. Wang J, Wang Z, Gu F, Ma X, Fei J, Cao Y (2019) An investigation into the sensor placement of a marine engine lubrication system for condition monitoring. In: Ball A, Gelman L, Rao B (eds) Advances in Asset Management and Condition Monitoring. Springer International Publishing, Cham, pp 573–582. https://doi.org/10.1007/978-3-030-57745-2_48

    Chapter  Google Scholar 

  15. Liu J, Hu Y, Yang S (2021) A SVM-based framework for fault detection in high speed trains. Meas 172:108779. https://doi.org/10.1016/j.measurement.2020.108779

    Article  Google Scholar 

  16. Hajizadeh S, Nunez A, Tax DM (2016) Semi supervised rail defect detection from imbalanced image data. IFAC PapersOnLine 49(3):78–83. https://doi.org/10.1016/j.ifacol.2016.07.014

    Article  Google Scholar 

  17. Shafique R, Siddiqui HUR, Rustam F, Ullah S, Siddique MA, Lee E, Ashraf I, Dudley S (2021) A novel approach to railway track faults detection using acoustic analysis. Sens 21(18):6221. https://doi.org/10.3390/s21186221

    Article  ADS  Google Scholar 

  18. Santur Y, Karakose M, Akin E (2016) Random forest-based diagnosis approach for rail fault inspection in railways. Proceedings of the national conference on electrical, electronics and biomedical engineering. IEEE, Bursa, pp 745–750

    Google Scholar 

  19. Xia Y, Xie F, Jiang Z (2010) Broken railway fastener detection based on adaboost algorithm. Proceedings of the international conference on optoelectronics and image processing. IEEE, Haikou, pp 313–316. https://doi.org/10.1109/ICOIP.2010.303

    Chapter  Google Scholar 

  20. Famurewa SM, Zhang L, Asplund M (2017) Maintenance analytics for railway infrastructure decision support. J of Qual in Maint Eng 23(3):310–325. https://doi.org/10.1108/JQME-11-2016-0059

    Article  Google Scholar 

  21. Lasisi A, Attoh-Okine N (2018) Principal components analysis and track quality index: a machine learning approach. Transp Res Part C Emerg Technol 91:230–248. https://doi.org/10.1016/j.trc.2018.04.001

    Article  Google Scholar 

  22. Hu C, Liu X (2016) Modeling track geometry degradation using support vector machine technique. The American society of mechanical engineers (ASME) IEE joint rail conference. IEEE, Columbia, p V001T01A011. https://doi.org/10.1115/JRC2016-5739

    Chapter  Google Scholar 

  23. Krummenacher G, Ong CS, Koller S, Kobayashi S, Buhmann JM (2018) Wheel defect detection with machine learning. IEEE Trans on Intell Transp Syst 19(4):1176–1187. https://doi.org/10.1109/TITS.2017.2720721

    Article  Google Scholar 

  24. Gibert X, Patel VM, Chellappa R (2016) Deep multitask learning for railway track inspection. IEEE Trans on Intell Transp Syst 18(1):153–164. https://doi.org/10.1109/TITS.2016.2568758

    Article  Google Scholar 

  25. Li D, Xie Q, Gong X, Yu Z, Xu J, Sun Y, Wang J (2021) Automatic defect detection of metro tunnel surfaces using a vision-based inspection system. Adv Eng Inform 47:101206. https://doi.org/10.1016/j.aei.2020.101206

    Article  Google Scholar 

  26. Santur Y, Karakose M, Akin E (2017) A new rail inspection method based on deep learning using laser cameras. Proceedings of the international artificial intelligence and data processing symposium. IEEE, Malatya, pp 1–6. https://doi.org/10.1109/IDAP.2017.8090245

    Chapter  Google Scholar 

  27. Xu Q, Zhao Q, Yu G, Wang L, Shen T (2020) Rail defect detection method based on recurrent neural network. Proceedings of the 2020 39th Chinese control conference (CCC). IEEE, Shenyang, pp 6486–6490. https://doi.org/10.23919/CCC50068.2020.9188823

    Chapter  Google Scholar 

  28. De Bruin T, Verbert K, Babuska R (2016) Railway track circuit fault diagnosis using recurrent neural networks. IEEE Trans on Neural Netw and Learn Syst 28(3):523–533. https://doi.org/10.1109/TNNLS.2016.2551940

    Article  MathSciNet  Google Scholar 

  29. Yang C, Sun Y, Ladubec C, Liu Y (2021) Developing machine learning-based models for railway inspection. Appl Sci 11(1):13. https://doi.org/10.3390/app11010013

    Article  CAS  Google Scholar 

  30. Chen Y, Song B, Zeng Y, Du X, Guizani M (2021) Fault diagnosis based on deep learning for current carrying ring of catenary system in sustainable railway transportation. Appl Soft Comput 100:106907. https://doi.org/10.1016/j.asoc.2020.106907

    Article  Google Scholar 

  31. Mahfuz N, Dhali OA, Ahmed S, Nigar M (2017) Autonomous railway crack detector robot for bangladesh: SCANOBOT. Proceedings of the IEEE region 10 humanitarian technology conference. IEEE, Dhaka, Bangladesh, pp 524–527. https://doi.org/10.1109/R10-HTC.2017.8289014

    Chapter  Google Scholar 

  32. Wei X, Yang Z, Liu Y, Wei D, Jia L, Li Y (2019) Railway track fastener defect detection based on image processing and deep learning techniques: a comparative study. Eng Appl of Artif Intell 80:66–81. https://doi.org/10.1016/j.engappai.2019.01.008

    Article  Google Scholar 

  33. Kang G, Gao S, Yu L, Zhang D (2018) Deep architecture for high-speed railway insulator surface defect detection: denoising autoencoder with multitask learning. IEEE Trans on Instrum Meas 68(8):2679–2690. https://doi.org/10.1109/TIM.2018.2868490

    Article  ADS  Google Scholar 

  34. Amini A, Entezami M, Huang Z, Rowshandel H, Papaelias M (2016) Wayside detection of faults in railway axle bearings using time spectral kurtosis analysis on high-frequency acoustic emission signals. Adv in Mech Eng 8(11):1687814016676000. https://doi.org/10.1177/1687814016676000

    Article  Google Scholar 

  35. Sun Y, Liu Y, Yang C (2019) Railway joint detection using deep convolutional neural networks. IEEE 15th international conference on automation science and engineering (CASE). IEEE, Canada, pp 235–240. https://doi.org/10.1109/COASE.2019.8843245

    Chapter  Google Scholar 

  36. Hashmi MSA, Ibrahim M, Bajwa IS, Siddiqui HUR, Rustam F, Lee E, Ashraf I (2022) Railway track inspection using deep learning based on audio to spectrogram conversion: an on-the-fly approach. Sens 22(5):1983. https://doi.org/10.3390/s22051983

    Article  ADS  Google Scholar 

  37. Lee J, Choi H, Park D, Chung Y, Kim H-Y, Yoon S (2016) Fault detection and diagnosis of railway point machines by sound analysis. Sens 16(4):549. https://doi.org/10.3390/s16040549

    Article  ADS  Google Scholar 

  38. Sejdic E, Djurovic I, Jiang J (2009) Time-frequency feature representation using energy concentration: an overview of recent advances. Digit Signal Process 19(1):153–183. https://doi.org/10.1016/j.dsp.2007.12.004

    Article  Google Scholar 

  39. Wang Z (1984) Fast algorithms for the discrete W transform and for the discrete Fourier transform. IEEE Trans Acoust Speech Signal Process 32(4):803–816. https://doi.org/10.1109/TASSP.1984.1164399

    Article  MathSciNet  Google Scholar 

  40. Clayton P, Allery MBP (1982) Metallurgical aspects of surface damage problems in rails. Canadian Metall Q 21(1):31–46. https://doi.org/10.1179/cmq.1982.21.1.31

    Article  ADS  Google Scholar 

  41. Zefeng W, Xuesong J, Weihua Z (2005) Contact-impact stress analysis of rail joint region using the dynamic finite element method. Wear 258(7–8):1301–1309. https://doi.org/10.1016/j.wear.2004.03.040

    Article  CAS  Google Scholar 

  42. Grassie SL, Kalousek J (1993) Rail corrugation: characteristics, causes and treatments. Proc Inst Mech Eng Part F J Rail Rapid Transit 207(1):57–68. https://doi.org/10.1243/PIME_PROC_1993_207_227_02

    Article  Google Scholar 

  43. Heaton J, Goodfellow I, Bengio Y, Courville A (2018) Deep learning. Genet Program Evolvable Mach 19:305–307. https://doi.org/10.1007/s10710-017-9314-z

    Article  Google Scholar 

  44. Michie D, Spiegelhalter DJ, Taylor CC (1995) Machine learning, neural and statistical classification. Ellis Horwood, New York

    Google Scholar 

  45. Chollet F (2021) Deep learning with Python. Manning Publications, New York

    Google Scholar 

  46. Patentscope (2021) Accident prevention system and method for rail systems. WO/2021/194445. https://patentscope.wipo.int/search/en/detail.jsf?docId=WO2021194445

  47. German Patent and TM Office (2023) Unfallverhütungssystem für Schienensysteme. 212020000817. https://depatisnet.dpma.de/DepatisNet/depatisnet?action=pdf&docid=DE212020000817U1

Download references

Acknowledgements

The authors wish to thank the Turkish Republic State Railways (TCDD) for providing permission and cooperation. The authors thank Suleyman Demirel University for the international patenting (utility model) of this study with the reference number and name of “PCT/TR2020/051168-Accident Prevention System and Method for Rail Systems”. Finally, the authors would like to express their gratitude and appreciation to all the people who collaborated on the success of this project, including Elif Ceren YILDIRIM (Eindhoven University of Technology, Mathematics and Computer Science) and Murat Onur YILDIRIM (Eindhoven University of Technology, Mathematics and Computer Science).

Funding

The research is funded by Suleyman Demirel University (BAP FDK-2021-8315). Additionally, the author Uygun. E. was supported by The Scientific and Technological Research Council of Türkiye (TUBİTAK) 2211-C and by the Türkiye’s Council of Higher Education’s (YOK) 100/2000 doctoral scholarship programs.

Author information

Authors and Affiliations

  1. Department of Civil Engineering, Engineering Faculty, Suleyman Demirel University, Isparta, Turkey

    Emre Uygun & Serdal Terzi

Authors
  1. Emre Uygun
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Serdal Terzi
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding authors

Correspondence to Emre Uygun or Serdal Terzi.

Ethics declarations

Conflict of Interest

The authors have no financial or non-financial interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Uygun, E., Terzi, S. Acoustic Monitoring of Railway Defects Using Deep Learning with Audio to Spectrogram Conversion. J. Vib. Eng. Technol. 12, 2585–2594 (2024). https://doi.org/10.1007/s42417-023-01001-8

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s42417-023-01001-8

Keywords