The wheel flat detection in trains using Artificial Intelligence (AI) has emerged as a critical advancement in railway maintenance and safety practices. AI systems can effectively identify geometric deformation in wheel rotation patterns, indicative of potential wheel flat damage, resorting to wayside monitoring systems and machine learning algorithms. This study aims to propose an unsupervised learning algorithm to identify and localize railway wheel flats, which considers three stages: (i) wheel flat detection to distinguish a healthy wheel from a damaged one using outlier analysis, achieving 100 percent accuracy; (ii) localizing the damage to pinpoint the location of the defective wheel through the Hidden Markov Model (HMM); (iii) classification of wheel damage based on its severity using k-means clustering technique. The unsupervised learning algorithm is validated with artificial data attained from a virtual wayside monitoring system related to freight train passages with healthy wheels and defective wheels with single and multiple defects. The proposed methodology demonstrated efficiency and robustness for wheel flat detection, localization, and damage severity classification regardless of the number of defective wheels and their position.
MoslehAMeixedoARibeiroD, et al.Automatic clustering-based approach for train wheels condition monitoring. Int J Real Ther2023; 11(5): 639–664.
2.
MoslehAMeixedoARibeiroD, et al.Early wheel flat detection: an automatic data-driven wavelet-based approach for railways. Veh Syst Dyn2023; 61(6): 1644–1673.
3.
MoslehAMontenegroPAlves CostaP, et al.An approach for wheel flat detection of railway train wheels using envelope spectrum analysis. Structure and Infrastructure Engineering2021; 17(12): 1710–1729.
4.
European Commission. Directorate General for Mobility and Transport. White paper on transport: roadmap to a single European transport area: towards a competitive and resource-efficient transport system. Luxembourg City, Luxembourg: Publications Office of the European Union, 2011.
5.
ValeC. Wheel flats in the dynamic behavior of ballasted and slab railway tracks. Appl Sci2021; 11(15): 7127.
6.
Rail Industry Safety and Standards Board. Code of practice for wheel defects. Spring Hill, QLD: Rail Industry Safety and Standards Board, 2020.
7.
BarkeDWChiuWK. A review of the effects of out-of-round wheels on track and vehicle components. Proc Inst Mech Eng F J Rail Rapid Transit2005; 219(3): 151–175.
8.
JohanssonANielsenJC. Out-of-round railway wheels—wheel-rail contact forces and track response derived from field tests and numerical simulations. Proc Inst Mech Eng F J Rail Rapid Transit2003; 217(2): 135–146.
9.
MohammadiMMoslehAValeC, et al.An unsupervised learning approach for wayside train wheel flat detection. Sensors2023; 23(4): 1910.
10.
UIC. Code of practice for the loading and securing of goods on railway wagons. Paris, France: UIC, 2022.
11.
AlemiACormanFLodewijksG. Condition monitoring approaches for the detection of railway wheel defects. Proc Inst Mech Eng F J Rail Rapid Transit2017; 231(8): 961–981.
FilogranoMLCorrederaPRodriguez-PlazaM, et al.Wheel flat detection in high-speed railway systems using fiber Bragg gratings. IEEE Sensor J2013; 13(12): 4808–4816.
14.
MishraSSharanPSaaraK. Real time implementation of fiber Bragg grating sensor in monitoring flat wheel detection for railways. Eng Fail Anal2022; 138: 106376.
15.
AktasMGunelEHYılmazerP, et al.Detection of wheel flatten defect on the moving train with acoustic emission sensor. In: 2018 IEEE 29th annual international symposium on personal, indoor and mobile radio communications (PIMRC), Bologna, Italy, 9–12 September 2018. IEEE.
16.
AminiAEntezamiMHuangZ, et al.Wayside detection of faults in railway axle bearings using time spectral kurtosis analysis on high-frequency acoustic emission signals. Adv Mech Eng2016; 8(11): 1687814016676000.
17.
MoslehAMontenegroPACostaPA, et al.Railway vehicle wheel flat detection with multiple records using spectral kurtosis analysis. Appl Sci2021; 11(9): 4002.
18.
SunZYeX-WLuJ. Estimating stay cable vibration under typhoon with an explainable ensemble learning model. Struc and Infrastructure Eng2023; 20: 1912–1924.
19.
GonçalvesVMoslehAValeC, et al.Wheel out-of-roundness detection using an envelope spectrum analysis. Sensors2023; 23(4): 2138.
20.
ChenSWangKChangC, et al.A two-level adaptive chirp mode decomposition method for the railway wheel flat detection under variable-speed conditions. J Sound Vib2021; 498: 115963.
21.
NowakowskiTKomorskiPSzymańskiGM, et al.Wheel-flat detection on trams using envelope analysis with Hilbert transform. Lat Am J Solid Struct2019; 16: 78255010.
22.
WangRCrosbeeDBevenA, et al.Vibration-based detection of wheel flat on a high-speed train. In: Advances in asset management and condition monitoring: comadem 2019. Berlin, Germany: Springer, 2020, pp. 159–169.
23.
AzimiMEslamlouADPekcanG. Data-driven structural health monitoring and damage detection through deep learning: state-of-the-art review. Sensors2020; 20(10): 2778.
24.
DernbachGLykartsisASieversL,et al. Acoustic identification of flat spots on wheels using different machine learning techniques. Fortschritte der Akustik - DAGA. 2020: 46. Deutsche Jahrestagung für Akustik. Berlin, Germany: Deutsche Gesellschaft für Akustik, 2020, pp. 367–370.
25.
WanTHTsangCWHuiK, et al.Anomaly detection of train wheels utilizing short-time Fourier transform and unsupervised learning algorithms. Eng Appl Artif Intell2023; 122: 106037.
26.
ShaikhKHussainIChowdhryBS. Wheel defect detection using a hybrid deep learning approach. Sensors2023; 23(14): 6248.
27.
NiY-QZhangQ-H. A Bayesian machine learning approach for online detection of railway wheel defects using track-side monitoring. Struct Health Monit2021; 20(4): 1536–1550.
28.
GuedesASilvaRRibeiroD, et al.Detection of wheel polygonization based on wayside monitoring and artificial intelligence. Sensors2023; 23(4): 2188.
29.
MoslehAMeixedoARibeiroD, et al.Machine learning approach for wheel flat detection of railway train wheels. Transport Res Procedia2023; 72: 4199–4206.
30.
MoslehAMohammadiMValeC, et al.Smart detection of wheel defects using artificial intelligence and wayside monitoring system. International J of Railway Res2023; 10(2): 9–18.
31.
MoslehAMeixedoARibeiroD, et al.Intelligent clustering-based approach for railway wheel flat detection. In: Proceedings of the fifth international conference on railway technology: research, development and maintenance, Montpellier, France, 22–25 August 2022. Civil-Comp Press.
32.
MontenegroPACalçadaR. Wheel–rail contact model for railway vehicle–structure interaction applications: development and validation. Rail Eng Science2023; 31(3): 181–206.
33.
MATLAB®. Version R2022a. Natick, MA: The MathWorks Inc, 2022.
34.
HertzH. Ueber die Berührung fester elastischer Körper. J Für Die Reine Und Angew Math1882; 92: 156–171.
35.
KalkerJJ. Book of tables for the herzian creep-force law. Delft, The Netherlands: Delft University of Technology, Faculty of Technical Mathematics and Informatics, 1996.
MeixedoASantosJRibeiroD, et al.Online unsupervised detection of structural changes using train–induced dynamic responses. Mech Syst Signal Process2022; 165: 108268.
38.
KerschenGFeenyBFGolinvalJC. On the exploitation of chaos to build reduced-order models. Comput Methods Appl Mech Eng2003; 192(13): 1785–1795.
39.
HanSFeenyBF. Enhanced proper orthogonal decomposition for the modal analysis of homogeneousstructures. J Vib Control2002; 8: 19–40.
40.
LenaertsVKerschenGGolinvalJC. Identification of a continuous structure with a geometrical non-linearity. Part II: proper orthogonal decomposition. J Sound Vib2003; 262(4): 907–919.
41.
De BoePGolinvalJC. Principal component analysis of a piezosensor array for damage localization. Struct Health Monit2003; 2(2): 137–144.
42.
Sousa ToméEPimentelMFigueirasJ. Damage detection under environmental and operational effects using cointegration analysis – application to experimental data from a cable-stayed bridge. Mech Syst Signal Process2020; 135: 106386.
43.
BullLAWordenKFuentesR, et al.Outlier ensembles: a robust method for damage detection and unsupervised feature extraction from high-dimensional data. J Sound Vib2019; 453: 126–150.
44.
FigueiredoEParkGFarrarCR, et al.Machine learning algorithms for damage detection under operational and environmental variability. Struct Health Monit2010; 10(6): 559–572.
45.
RabinerLR. A tutorial on hidden Markov models and selected applications in speech recognition. Proc IEEE1989; 77(2): 257–286.
46.
BilmesJA. A gentle tutorial of the EM algorithm and its application to parameter estimation for Gaussian mixture and hidden Markov models. Intern Comp Sci Insti1998; 4(510): 126.
47.
StampM. A revealing introduction to hidden Markov models. San Jose, CA: Department of Computer Science San Jose State University, 2004, pp. 26–56.
48.
SkarlatosDKarakasisKTrochidisA. Railway wheel fault diagnosis using a fuzzy-logic method. Appl Acoust2004; 65(10): 951–966.
49.
MøllerMF. A scaled conjugate gradient algorithm for fast supervised learning. Neural Network1993; 6(4): 525–533.
50.
ZhaiWMWangQCLuZW, et al.Dynamic effects of vehicles on tracks in the case of raising train speeds. Proc Inst Mech Eng F J Rail Rapid Transit2001; 215(2): 125–135.
51.
ValeCCalçadaR. Dynamic response of a coupled vehicle-track system to real longitudinal rail profiles. In: Proceedings of the tenth international conference on computational structures technology (CST 2010), Valencia, Spain, 14–17 September 2010.
52.
ValeCCalçadaR. A dynamic vehicle-track interaction model for predicting the track degradation process. J Infrastruct Syst2014; 20(3): 04014016.
53.
European Committee for Standardization. Railway applications - track - Track geometry quality - part 2: measuring systems - track recording vehicles (EN 13848-2). Brussels, Belgium: European Committee for Standardization, 2006.
54.
MoslehACostaPACalçadaR. A new strategy to estimate static loads for the dynamic weighing in motion of railway vehicles. Proc Inst Mech Eng F J Rail Rapid Transit2019; 234(2): 183–200.
55.
MoslehAMeixedoARibeiroD, et al.Feature extraction and normalization for wheel flat detection on railway vehicles. In: International conference on structural health monitoring of intelligent infrastructure: transferring research into practice, SHMII, Porto, Portugal, 30 June 2021–2 July 2021.
56.
LourençoAFerrazCRibeiroD, et al.Adaptive time series representation for out-of-round railway wheels fault diagnosis in wayside monitoring. Eng Fail Anal2023; 152: 107433.
