Implementation of an AI-based predictive structural health monitoring strategy for bonded insulated rail joints using digital twins under varied bolt conditions

Abstract

Predictive maintenance is essential for the implementation of an innovative and efficient structural health monitoring strategy. Models capable of accurately interpreting new data automatically collected by suitably placed sensors to assess the state of the infrastructure represent a fundamental step, particularly for the railway sector, whose safe and continuous operation plays a strategic role in the well-being and development of nations. In this scenario, the benefits of a digital twin of a bonded insulated rail joint (IRJ) with the predictive capabilities of advanced classification algorithms based on artificial intelligence have been explored. The digital model provides an accurate mechanical response of the infrastructure as a pair of wheels passes over the joint. As bolt preload conditions vary, four structural health classes were identified for the joint. Two parameters, i.e. gap value and vertical displacement, which are strongly correlated with bolt preload, are used in different combinations to train and test five predictive classifiers. Their classification effectiveness was assessed using several performance indicators. Finally, we compared the IRJ condition predictions of two trained classifiers with the available data, confirming their high accuracy. The approach presented provides an interesting solution for future predictive tools in SHM especially in the case of complex systems such as railways where the vehicle–infrastructure interaction is complex and always time varying.

References

[1.] Sun Q, Wang X, Ma F, et al.. Synergetic effect and spatial-temporal evolution of railway transportation in sustainable development of trade: an empirical study based on the Belt and Road Sustainability, 2019, 11(6): 1721.

[2.] Lidén T. Railway infrastructure maintenance—a survey of planning problems and conducted research Transp Res Proc, 2015, 10: 574-583

[3.] Kaćunić DJ, Librić L, Car M. Application of unmanned aerial vehicles on transport infrastructure network Građevinar, 2016, 68(4): 287-300

[4.] Gbadamosi AQ, Oyedele LO, Delgado JMD, et al.. IoT for predictive assets monitoring and maintenance: an implementation strategy for the UK rail industry Autom Constr, 2021, 122. 103486

[5.] Davari N, Veloso B, Costa GA, et al.. A survey on data-driven predictive maintenance for the railway industry Sensors (Basel), 2021, 21(17): 5739.

[6.] Sedghi M, Kauppila O, Bergquist B, et al.. A taxonomy of railway track maintenance planning and scheduling: a review and research trends Reliab Eng Syst Saf, 2021, 215. 107827

[7.] La Placa A, Freddi F, Giuliani F. Bonded insulated rail joint monitoring using gap opening variation with fibre optic sensors: analytical validation and limits Transp Res Procedia, 2023, 74: 1007-1014.

[8.] Esveld C (2001) Modern railway track, 2nd edn. Delft university of Technology

[9.] Mandal NK, Dhanasekar M. Sub-modelling for the ratchetting failure of insulated rail joints Int J Mech Sci, 2013, 75: 110-122.

[10.] La Placa A, Freddi F, Giuliani F. Monitoring of insulated rail joints based on gap value measurement Urban Rail Transit, 2024, 10(1): 28-41.

[11.] Mundrey JS (2009) Railway track engineering. Tata McGraw-Hill Education

[12.] Kerr AD (2003) Fundamentals of railway track engineering. Simmons-Boardman Books, Inc.

[13.] Davis DD, Akhtar M, Kohake E et al (2005) Effects of heavy axle loads on bonded insulated joint performance. In: Proceedings of the AREMA 2005 annual conference, 25–28 Sept 2005, Chicago, Illinois

[14.] Kerr AD, Cox JE. Analysis and tests of boned insulted rail joints subjected to vertical wheel loads Int J Mech Sci, 1999, 41(10): 1253-1272.

[15.] Wen Z, Jin X, Zhang W. Contact-impact stress analysis of rail joint region using the dynamic finite element method Wear, 2005, 258(7–8): 1301-1309.

[16.] Kaewunruen S, Sresakoolchai J, Lin YH. Digital twins for managing railway maintenance and resilience Open Res Eur, 2021, 1: 91.

[17.] Flammini F. Digital twins as Run-time predictive models for the resilience of cyber-physical systems: a conceptual framework Phil Trans R Soc A, 2021, 379(2207): 20200369.

[18.] Dimitrova E, Tomov S (2021) Digital twins: an advanced technology for railways maintenance transformation. In: 2021 13th Electrical engineering faculty conference (BulEF), Varna, Bulgaria. IEEE, pp1–5

[19.] Sresakoolchai J, Kaewunruen S. Railway infrastructure maintenance efficiency improvement using deep reinforcement learning integrated with digital twin based on track geometry and component defects Sci Rep, 2023, 13(1): 2439.

[20.] Pillai N, Shih JY, Roberts C. Enabling data-driven predictive maintenance for S&C through digital twin models and condition monitoring systems The PWI Journal, 2021, 139(4): 14-19

[21.] Pillai N, Shih JY, Roberts C. Sensor selection and placement for track switch condition monitoring through validated structural digital twin models of train–track Interactions Eng Proc, 2021, 10(1): 49

[22.] Malekloo A, Ozer E, AlHamaydeh M, et al.. Machine learning and structural health monitoring overview with emerging technology and high-dimensional data source highlights Struct Health Monit, 2022, 21(4): 1906-1955.

[23.] Çınar ZM, Abdussalam Nuhu A, Zeeshan Q et al (2020) Machine learning in predictive maintenance towards sustainable smart manufacturing in industry 4.0. Sustainability 12(19):8211

[24.] Carvalho TP, Soares FAAMN, Vita R, et al.. A systematic literature review of machine learning methods applied to predictive maintenance Comput Ind Eng, 2019, 137. 106024

[25.] Mathew V, Toby T, Singh V et al (2017) Prediction of remaining useful lifetime (RUL) of turbofan engine using machine learning. In: 2017 IEEE international conference on circuits and systems (ICCS). Thiruvananthapuram, India. IEEE, pp 306–311

[26.] Allah Bukhsh Z, Saeed A, Stipanovic I, et al.. Predictive maintenance using tree-based classification techniques: a case of railway switches Transp Res Part C Emerg Technol, 2019, 101: 35-54.

[27.] Farahani MA, McCormick MR, Harik R, et al.. Time-series classification in smart manufacturing systems: an experimental evaluation of state-of-the-art machine learning algorithms Robot Comput Integr Manuf, 2025, 91. 102839

[28.] Li H, Parikh D, He Q, et al.. Improving rail network velocity: a machine learning approach to predictive maintenance Transp Res Part C Emerg Technol, 2014, 45: 17-26.

[29.] Böhm T. Remaining useful life prediction for railway switch engines using classification techniques Int J Progn Health Manag, 2017, 8(3): 2666

[30.] Tabaszewski M, Firlik B. Detection of the presence of rail corrugation using convolutional neural network Eng Trans, 2022, 70(4): 339-353

[31.] Malekjafarian A, Sarrabezolles CA, Khan MA, et al.. A machine-learning-based approach for railway track monitoring using acceleration measured on an in-service train Sensors, 2023, 23(17): 7568.

[32.] Farrar CR, Worden K Structural health monitoring: a machine learning perspective, 2012 John Wiley & Sons.

[33.] Dirnfeld R (2022) Digital twins in railways: state of the art, opportunities, and guidelines. Dissertation, Linnaeus University

[34.] Bhatti G, Mohan H, Raja Singh R. Towards the future of smart electric vehicles: digital twin technology Renew Sustain Energy Rev, 2021, 141. 110801

[35.] Torzoni M, Tezzele M, Mariani S, et al.. A digital twin framework for civil engineering structures Comput Meth Appl Mech Eng, 2024, 418. 116584

[36.] Rudskoy A, Ilin I, Prokhorov A. Digital twins in the intelligent transport systems Transp Res Procedia, 2021, 54: 927-935.

[37.] Bešinović N, De Donato L, Flammini F, et al.. Artificial intelligence in railway transport: taxonomy, regulations, and applications IEEE Trans Intell Transp Syst, 2022, 23(9): 14011-14024.

[38.] Karlsson A, Kalevirta S (2021) The digital revolution reaches rolling stock component services. Global Railway Review (3). https://www.globalrailwayreview.com/article/126590/digital-rolling-stock-component/ Accessed 8 Feb 2024.

[39.] Smith M (2009) ABAQUS/standard user’s manual, version 6.9

[40.] Rete Ferroviaria Italiana (1967) Disegno tecnico FS8239 12, Materiale metallico occorrente per la confezione di giunzioni isolanti incollate armamento tipo 50 UNI. 1967

[41.] Rete Ferroviaria Italiana (2003) Disegno tecnico FS8335 10, Materiale isolante per giunzioni isolanti incollate armamento tipo 50 UNI

[42.] Mandal NK, Spiryagin M, Wu Q, et al.. FEA of mechanical behaviour of insulated rail joints due to vertical cyclic wheel loadings Eng Fail Anal, 2022, 133. 105966

[43.] Igwemezie JO, Nguyen AT. Anatomy of joint bar failures II Railway Track and Structures, 2009, 105(10): 43-48

[44.] Himebaugh AK, Plaut RH, Dillard DA. Finite element analysis of bonded insulated rail joints Int J Adhes Adhes, 2008, 28(3): 142-150.

[45.] Gallou M, Frost M, El-Hamalawi A, et al.. Assessing the deflection behaviour of mechanical and insulated rail joints using finite element analysis Proc Inst Mech Eng Part F J Rail Rapid Transit, 2018, 232(9): 2290-2308.

[46.] Mandal NK, Bracciali A, Megna G. Evaluation of railhead material damage of insulated rail joints due to cyclic wheel loadings Wear, 2023, 528. 204966

[47.] Kabo E, Nielsen JCO, Ekberg A. Prediction of dynamic train–track interaction and subsequent material deterioration in the presence of insulated rail joints Veh Syst Dyn, 2006, 44(sup1): 718-729.

[48.] Harmon M, Santa JF, Jaramillo JA, et al.. Evaluation of the coefficient of friction of rail in the field and laboratory using several devices Tribol Mater Surf Interfaces, 2020, 14(2): 119-129.

[49.] Mandal NK, Brendan P. An engineering analysis of insulated rail joints: a general perspective Int J Eng Sci Technol, 2010, 2(8): 3964-3988

[50.] Bono G, Focacci C, Lanni S (1997) La sovrastruttura ferroviaria. Collegio ingegneri ferroviari italiani, Roma

[51.] Zhu K, Edwards JR, Qian Y, et al.. Finite element analysis of the effects of bolt condition on bolted rail joint stresses Transp Res Rec J Transp Res Board, 2016, 2545(1): 36-45.

[52.] Zhu K, Qian Y, Edwards JR, et al.. Finite element analysis of rail-end bolt hole and fillet stress on bolted rail joints Transp Res Rec J Transp Res Board, 2017, 2607(1): 33-42.

[53.] Regione Emilia-Romagna (2021) Rapporto annuale di monitoraggio della mobilita e del trasporto in Emilia-Romagna 2021, Mobilita

[54.] Gardie E, Dubale H, Tefera E, et al.. Numerical analysis of rail joint in a vertical applied load and determining the possible location of joints Forces Mech, 2022, 6. 100064

[55.] Wondimu A, Alemu N, Sivaprakasam P. Study on effect of dynamic load on rail joint bar by finite element method (FEM) Adv Mater Process Technol, 2022, 8(1): 1121-1134

[56.] Brahimi M, Medjaher K, Leouatni M et al (2016) Development of a prognostics and health management system for the railway infrastructure—Review and methodology. In: 2016 Prognostics and System Health Management Conference (PHM-Chengdu), Chengdu, China. IEEE, pp 1–8

[57.] Sun YQ, Spiryagin M, Wu Q, et al.. Feasibility in assessing the dipped rail joint defects through dynamic response of heavy haul locomotive J Mod Transp, 2018, 26(2): 96-106.

[58.] Stephen JT, Hardwick C, Beaty P, et al.. Ultrasonic monitoring of insulated block joints Proc Inst Mech Eng Part F J Rail Rapid Transit, 2019, 233(3): 251-261.

[59.] Ramezan CA, Warner TA, Maxwell AE, et al.. Effects of training set size on supervised machine-learning land-cover classification of large-area high-resolution remotely sensed data Remote Sens, 2021, 13(3): 368.

[60.] Xie J, Huang J, Zeng C, et al.. Systematic literature review on data-driven models for predictive maintenance of railway track: implications in geotechnical engineering Geosciences, 2020, 10(11): 425.

[61.] Ciaburro G MATLAB for machine learning: unlock the power of deep learning for swift and enhanced results, 2024 Birmingham Packt Publishing Ltd.

[62.] MathWorks (2023) Machine Learning with MATLAB. https://www.mathworks.com/campaigns/offers/machine-learning-with-matlab.html. Accessed 20 June 2023

[63.] MathWorks, (2023) Mastering Machine Learning. A step-by-step guide with MATLAB. https://www.mathworks.com/campaigns/offers/mastering-machine-learning-with-matlab.html. Accessed 20 June 2023

[64.] Jamian S, Gunawan TS, Kartiwi M et al (2022) Human activity and posture classification using smartphone sensors and Matlab mobile. In: 2022 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), Ottawa, ON, Canada. IEEE, pp 1–6

[65.] Chen C, Seo H, Zhao Y. A novel pavement transverse cracks detection model using WT-CNN and STFT-CNN for smartphone data analysis Int J Pavement Eng, 2022, 23(12): 4372-4384.

[66.] MOST (2024) Centro Nazionale per la Mobilità Sostenibile. Spoke 4 - Research mission statement. https://www.centronazionalemost.it/spoke4.html. Accessed 5 August 2024

[67.] European Union (2024) NextGenerationEU. https://next-generation-eu.europa.eu/index_en. Accessed 5 Aug 2024

Funding

National Recovery and Resilience Plan (NRRP), Mission 4 Component 2 Investment 1.4—Call for tender No. 3138 of 16/12/2021 of Italian Ministry of University and Research funded by the European Union—NextGenerationEU.(Project code CN00000023, Concession Decree No. 1033 of 17/06/2022 adopted by the Italian Ministry of University and Research, CUP D93C22000400001, “Sustainable Mobility Center” (CNMS). Spoke 4—Rail Transportation.)