Drive-by damage detection methodology for high-speed railway bridges using sparse autoencoders

Abstract

High-speed railway bridges are essential components of any railway transportation system that should keep adequate levels of serviceability and safety. In this context, drive-by methodologies have emerged as a feasible and cost-effective monitoring solution for detecting damage on railway bridges while minimizing train operation interruptions. Moreover, integrating advanced sensor technologies and machine learning algorithms has significantly enhanced structural health monitoring (SHM) for bridges. Despite being increasingly used in traditional SHM applications, studies using autoencoders within drive-by methodologies are rare, especially in the railway field. This study presents a novel approach for drive-by damage detection in HSR bridges. The methodology relies on acceleration records collected from multiple bridge crossings by an operational train equipped with onboard sensors. Log-Mel spectrogram features derived from the acceleration records are used together with sparse autoencoders for computing statistical distribution-based damage indexes. Numerical simulations were performed on a 3D vehicle–track–bridge interaction system model implemented in Matlab to evaluate the robustness and effectiveness of the proposed approach, considering several damage scenarios, vehicle speeds, and environmental and operational variations, such as multiple track irregularities and varying measurement noise. The results show that the proposed approach can successfully detect damages, as well as characterize their severity, especially for very early-stage damages. This demonstrates the high potential of applying Mel-frequency damage-sensitive features associated with machine learning algorithms in the drive-by condition assessment of high-speed railway bridges.

References

[1.] Givoni M. Environmental benefits from mode substitution: Comparison of the environmental impact from aircraft and high-speed train operations. Int J Sustain Transp 2007, 1 4 209-230

[2.] Adler N, Pels E, Nash C. High-speed rail and air transport competition: Game engineering as tool for cost-benefit analysis. Transportation Research Part B: Methodological 2010, 44 7 812-833

[3.] Kang C, Schneider S, Wenner M . Development of design and construction of high-speed railway bridges in Germany. Eng Struct 2018, 163 184-196

[4.] Malekjafarian A, McGetrick PJ, Obrien EJ. A review of indirect bridge monitoring using passing vehicles. Shock Vib 2015, 2015 286139

[5.] Malekjafarian A, Corbally R, Gong W. A review of mobile sensing of bridges using moving vehicles: progress to date, challenges and future trends. Structures 2022, 44 1466-1489

[6.] Locke W, Sybrandt J, Redmond L . Using drive-by health monitoring to detect bridge damage considering environmental and operational effects. J Sound Vib 2020, 468 115088

[7.] Yang Y-B, Lin CW, Yau JD. Extracting bridge frequencies from the dynamic response of a passing vehicle. J Sound Vib 2004, 272 3–5 471-493

[8.] González A, Obrien EJ, McGetrick PJ . Identification of damping in a bridge using a moving instrumented vehicle. J Sound Vib 2012, 331 18 4115-4131

[9.] Yang YB, Li YC, Chang KC. Constructing the mode shapes of a bridge from a passing vehicle: a theoretical study. Smart Struct Syst 2014, 13 5 797-819

[10.] Malekjafarian A, Obrien EJ. Identification of bridge mode shapes using short time frequency domain decomposition of the responses measured in a passing vehicle. Eng Struct 2014, 81 386-397

[11.] Tan C, Uddin N, Obrien EJ . Extraction of bridge modal parameters using passing vehicle response. J Bridg Eng 2019, 24 9 04019087

[12.] Zhang Y, Wang L, Xiang Z. Damage detection by mode shape squares extracted from a passing vehicle. J Sound Vib 2012, 331 2 291-307

[13.] E. OBrien, A. Malekjafarian,. A mode shape-based damage detection approach using laser measurement from a vehicle crossing a simply supported bridge. Struct Control Health Monit 2016, 23 1273-1286

[14.] Corbally R, Malekjafarian A. Examining changes in bridge frequency due to damage using the contact-point response of a passing vehicle. J Struct Integ Maint 2021, 6 3 148-158

[15.] Meixedo A, Santos J, Ribeiro D . Damage detection in railway bridges using traffic-induced dynamic responses. Eng Struct 2021, 238

[16.] Souza EF, Bragança C, Meixedo A . Drive-by methodologies applied to railway infrastructure subsystems: A literature review—part I: bridges and viaducts. Appl Sci 2023, 13 12 6940

[17.] Bragança C, Souza EF, Ribeiro D . Drive-by methodologies applied to railway infrastructure subsystems: A literature review—part II: track and vehicle. Appl Sci 2023, 13 12 6982

[18.] Bowe, C. Quirke, P. Cantero D. (2015) Drive-by structural health monitoring of railway bridges using train-mounted accelerometers, In: Proceedings of the 5th ECCOMAS Thematic Conference on Computational Methods in Structural Dynamics and Earthquake Engineering, Crete Island pp. 1652–1663

[19.] Yang YB, Zhang B, Qian Y . Further revelation on damage detection by IAS computed from the contact-point response of a moving vehicle. Int J Struct Stab Dyn 2018, 18 11 1850137

[20.] Fitzgerald PC, Malekjafarian A, Cantero D . Drive-by scour monitoring of railway bridges using a wavelet-based approach. Eng Struct 2019, 191 1-11

[21.] Bernardini L, Carnevale M, Collina A. Damage identification in warren truss bridges by two different time–frequency algorithms. Appl Sci 2021, 11 22 10605

[22.] Quirke P, Bowe C, Obrien EJ . Railway bridge damage detection using vehicle-based inertial measurements and apparent profile. Eng Struct 2017, 153 421-442

[23.] Ren Y, Obrien EJ, Cantero D. Railway bridge condition monitoring using numerically calculated responses from batches of trains. Appl Sci 2022, 12 10 4972

[24.] Wang C, Zhan J, Wang Y. A drive-by methodology for rapid inspection of hsr bridge substructures using dynamic responses of passing marshaling trains. Int J Struct Stab Dyn 2024, 24 6 2450068

[25.] Corbally R, Malekjafarian A. Experimental verification of a data-driven algorithm for drive-by bridge condition monitoring. Struct Infrastruct Eng 2024, 20 7/8 1174-1196

[26.] Lan Y, Li Z, Lin W. Physics-guided diagnosis framework for bridge health monitoring using raw vehicle accelerations. Mech Syst Signal Process 2024, 206 110899

[27.] OBrien EJ, McCrum DP, Wang S. Monitoring bearing damage in bridges using accelerations from a fleet of vehicles, without prior bridge or vehicle information. Eng Struct 2024, 302 117414

[28.] Erduran E, Gonen S. Contact point accelerations, instantaneous curvature, and physics-based damage detection and location using vehicle-mounted sensors. Eng Struct 2024, 304 117608

[29.] Talebi-Kalaleh M, Mei Q. Damage detection in bridge structures through compressed sensing of crowdsourced smartphone data. Struct Control Health Monit 2024, 2024 5436675

[30.] Lei Y, Jin Z, Qi C. Drive-by bridge damage detection based on wavelet analysis of residual contact response of a moving vehicle. Acta Mech 2024, 235 1437-1452

[31.] Corbally R, Malekjafarian A. A deep-learning framework for classifying the type, location, and severity of bridge damage using drive-by measurements. Comput-Aided Civil and Infrastruct Eng 2024, 39 6 852-871

[32.] Alamdari MM. An evolutionary vehicle scanning method for bridges based on time series segmentation and change point detection. Mech Syst Signal Process 2024, 210 111173

[33.] Oppenheim AV, Schafer RW. From frequency to quefrency: A history of the cepstrum. IEEE Signal Process Mag 2004, 21 5 95-106

[34.] Randall RB. (2009) Cepstral methods of operational modal analysis. In: BoIler C, Chang FK, Fujino Y eds, Encyclopedia of structural health monitoring, John Wiley & Sons, Ltd.

[35.] Randall RB. A history of cepstrum analysis and its application to mechanical problems. Mech Syst Signal Process 2017, 97 3-19

[36.] Joseph GV, Pakrashi V. Spiking neural networks for structural health monitoring. Sensors 2022, 22 23 9245

[37.] Li L, Morgantini M, Betti R. Structural damage assessment through a new generalized autoencoder with features in the quefrency domain. Mech Syst Signal Process 2023, 184 109713

[38.] Naranjo-Alcazar J, Perez-Castanos S, Zuccarello P . Open set audio classification using autoencoders trained on few data. Sensors 2020, 20 13 3741

[39.] Faridh MH, Zulpratita US. HiVAD: a voice activity detection application based on deep learning, elkomika: jurnal teknik energi elektrik. Teknik Telekomunikasi, Teknik Elektronika 2021, 9 856

[40.] Parvathy SR, Jayan DP, Pathrose N et al (2021) Convolutional autoencoder based deep learning model for identification of red palm weevil signals. In: 2021 Asia-Pacific Signal and Inf Process Assoc Annual Summit and Conf, Tokyo, pp 1987–1992

[41.] Natsiou A, Longo L, Leary SO’ (2022) An investigation of the reconstruction capacity of stacked convolutional autoencoders for log-mel-spectrograms. In: Proceedings of the 16th Int Conf Signal-Image Technol Internet-Based Systems, Dijon, pp. 155–162

[42.] Mei Q, Gül M, Boay M. Indirect health monitoring of bridges using Mel-frequency cepstral coefficients and principal component analysis. Mech Syst Signal Process 2019, 119 523-546

[43.] Bochud N, Gomez AM, Rus G et al (2011) Robust parametrization for non-destructive evaluation of composites using ultrasonic signals. In: IEEE International Conf Acoustics, Speech and Signal Process, Prague, pp 1789–1792

[44.] Zhang G, Harichandran RS, Ramuhalli P. Application of noise cancelling and damage detection algorithms in NDE of concrete bridge decks using impact signals. J Nondestr Eval 2011, 30 259-272

[45.] Dackermann U, Smith WA, Randall RB. Damage identification based on response-only measurements using cepstrum analysis and artificial neural networks. Struct Health Monit 2014, 13 4 430-444

[46.] Balsamo L, Betti R, Beigi H. A structural health monitoring strategy using cepstral features. J Sound Vib 2014, 333 19 4526-4542

[47.] Lee J, Choi H, Park D . Fault detection and diagnosis of railway point machines by sound analysis. Sensors (Switzerland) 2016, 16 4 549

[48.] Abdul ZK, Al-Talabani AK. Mel frequency cepstral coefficient and its applications: a review. IEEE Access 2022, 10 122136-122158

[49.] Mei Q, Gül M. A crowdsourcing-based methodology using smartphones for bridge health monitoring. Struct Health Monit 2019, 18 5/6 1602-1619

[50.] de Souza EF, Bittencourt TN, Ribeiro D . Feasibility of applying Mel-frequency cepstral coefficients in a drive-by damage detection methodology for high-speed railway bridges. Sustainability 2022, 14 20 13290

[51.] Eltouny K, Gomaa M, Liang X. Unsupervised learning methods for data-driven vibration-based structural health monitoring: a review. Sensors 2023, 23 6 3290

[52.] Giglioni V, Venanzi I, Poggioni V . Autoencoders for unsupervised real-time bridge health assessment. Comput Aid Civil Infrastruct Eng 2023, 38 8 959-974

[53.] Liu J, Chen S, Berges M . Diagnosis algorithms for indirect structural health monitoring of a bridge model via dimensionality reduction. Mech Syst Signal Process 2020, 136 106454

[54.] Li Z, Lin W, Zhang Y. Real-time drive-by bridge damage detection using deep auto-encoder. Structures 2023, 47 1167-1181

[55.] Calderon Hurtado A, Kaur K, Makki Alamdari M . Unsupervised learning-based framework for indirect structural health monitoring using adversarial autoencoder. J Sound Vib 2023, 550 117598

[56.] Zhang J, Jiang Q, Chang F (2018) Fault diagnosis method based on MFCC fusion and SVM. In: 2018 IEEE International Conf Inf Automation, Wuyishan, pp 1617–1622

[57.] Abdul ZK, Al-Talabani AK, Ramadan DO. A hybrid temporal feature for gear fault diagnosis using the long short term memory. IEEE Sens J 2020, 20 23 14444-14452

[58.] Akpudo UE, Hur JW. A cost-efficient MFCC-based fault detection and isolation technology for electromagnetic pumps. Electronics 2021, 10 4 439

[59.] Lu L, Tao W, Liu G et al (2022) Effectiveness analysis of the Mel spectrum features of AE signals in the detection of partial discharge faults. In: 2022 International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, Harbin, pp 1–5

[60.] Li G, Wu J, Deng C, Chen Z. Parallel multi-fusion convolutional neural networks based fault diagnosis of rotating machinery under noisy environments. ISA Trans 2022, 128 545-555

[61.] Liu Y, Guan J, Zhu Q et al (2022) Anomalous sound detection using spectral-temporal information fusion, In: IEEE Int Conf Acoustics, Speech and Signal Process, Singapore, pp 816–820

[62.] Mobtahej P, Zhang X, Hamidi M . (2022) An LSTM-autoencoder architecture for anomaly detection applied on compressors audio data. Comput Math Methods 2022, 2 3622426

[63.] Xie S, Liu R, Du L . Anomaly detection in rolling bearings based on the Mel-frequency cepstrum coefficient and masked autoencoder for distribution estimation. Struct Control Health Monit 2022, 29 11 e3096

[64.] de Paula Monteiro R, Lozada MC, Mendieta DRC . A hybrid prototype selection-based deep learning approach for anomaly detection in industrial machines. Expert Syst Appl 2022, 204 117528

[65.] Hua F, Li L (2023) Sound anomaly detection of industrial products based on MFCC fusion short-time energy feature extraction. In: 2022 IEEE Conference on Telecommunications, Optics and Computer Science, Dalian, pp 861–864

[66.] Ribeiro A, Matos LM, Pereira PJ et al (2021) Deep dense and convolutional autoencoders for unsupervised anomaly detection in machine condition sounds. In: IFIP International Conference on Artificial Intelligence Applications and Innovations, Hersonissos, pp 337–348

[67.] Abbasi S, Famouri M, Shafiee MJ . Outliernets: Highly compact deep autoencoder network architectures for on-device acoustic anomaly detection. Sensors 2021, 21 14 4805

[68.] Bayram B, Duman TB, Ince G. Real time detection of acoustic anomalies in industrial processes using sequential autoencoders. Expert Syst 2021, 38 1 e12564

[69.] Li Z, Lin W, Zhang Y. Drive-by bridge damage detection using Mel-frequency cepstral coefficients and support vector machine. Struct Health Monit 2023, 22 5 3302-3319

[70.] Sarwar MZ, Cantero D. Deep autoencoder architecture for bridge damage assessment using responses from several vehicles. Eng Struct 2021, 246 113064

[71.] Lourenço A, Ferraz C, Ribeiro D . Adaptive time series representation for out-of-round railway wheels fault diagnosis in wayside monitoring. Eng Fail Anal 2023, 152 107433

[72.] Pathirage CSN, Li J, Li L . Development and application of a deep learning–based sparse autoencoder framework for structural damage identification. Struct Health Monit 2019, 18 1 103-122

[73.] Finotti RP, F. de S. Barbosa, A.A. Cury,. Numerical and experimental evaluation of structural changes using sparse auto-encoders and svm applied to dynamic responses. Appl Sci 2021, 11 24 11965

[74.] Finotti RP, Gentile C, Barbosa F. Structural novelty detection based on sparse autoencoders and control charts. Struct Eng Mech 2022, 81 5 647-664

[75.] Huang J, Yang S, Li J . Prediction model of sparse autoencoder-based bidirectional LSTM for wastewater flow rate. J Supercomput 2023, 79 4412-4435

[76.] Shi R, Ji J, Zhang C. Boosting sparsity-induced autoencoder: A novel sparse feature ensemble learning for image classification. Int J Adv Rob Syst 2019, 16 3 1-9

[77.] Węglarczyk S. Kernel density estimation and its application. ITM Web of Conf 2018, 23 2 00037

[78.] Zhai W, Xia H, Cai C . High-speed train–track–bridge dynamic interactions—Part I: theoretical model and numerical simulation. Int J Rail Transp 2013, 1 1–2 3-24

[79.] Zhai W. Vehicle-Track coupled dynamics: theory and applications 2020 Singapore Springer

[80.] Goicolea JM (2014) Simplified mechanical description of AVE S-103-ICE3 Velaro E high-speed train. https://oa.upm.es/43946/1/aves103_ice3.pdf, Accessed March 21, 2023