The growing volume of available infrastructural monitoring data enables the development of powerful data-driven approaches to estimate infrastructure health conditions using direct measurements. This paper proposes a deep learning methodology to estimate infrastructure physical parameters, such as railway track stiffness, using drive-by vibration response signals. The proposed method employs a long short-term memory (LSTM) feature extractor accounting for temporal dependencies in the feature extraction phase, and bidirectional long short-term memory (BiLSTM) networks to leverage bidirectional temporal dependencies in both the forward and backward paths of the drive-by vibration response in the condition estimation phase. In addition, a framing approach is employed to enhance the resolution of the monitoring task to the beam level by segmenting the vibration signal into frames equal to the distance between individual beams, centering the frames over the beam nodes. The proposed LSTM-BiLSTM model offers a versatile tool for various bridge and railway infrastructure conditions monitoring using direct drive-by vibration response measurements. The results demonstrate the potential of incorporating temporal analysis in the feature extraction phase and emphasize the pivotal role of bidirectional temporal information in infrastructure health condition estimation. The proposed methodology can accurately and automatically estimate railway track stiffness and identify local stiffness reductions in the presence of noise using drive-by measurements. An illustrative case study of vehicle–track interaction simulation is used to demonstrate the performance of the proposed model, achieving maximum mean absolute percentage errors of 1.7% and 0.7% in estimating railpad and ballast stiffness, respectively.

Axle box acceleration (ABA) measurements can be used for continuously monitoring rail infrastructure and detecting rail surface defects such as squats. However, accurately detecting squats is challenging due to their short-duration responses and low occurrence in ABA signals, particularly for light squats that exhibit subtle ABA responses. To address this challenge, we propose using a spiking neural network (SNN) with time-varying weights to enhance the detection performance of rail squats based on ABA measurements. Our approach employs a simple SNN architecture without hidden layers, trained using a method that combines genetic algorithms, k-fold cross-validation, and multi-start gradient-based approach to optimise hyperparameters and weights. The proposed methodology demonstrates competitive accuracy compared to other state-of-the-art SNN-based methods on UCI benchmarks for both binary and multi-class nonlinear problems. Part of its advantages include higher efficiency with a simpler architecture and training approach that reduces computational times while achieving effective spatiotemporal pattern detection. As shown by real-field measurements from Dutch and Swedish railways in anomaly detection, it effectively captures subtle changes in light squat defect responses in ABA signals and achieves a detection performance of 100% for severe squat defects and over 93% for light squat defects. Furthermore, we show that the spike responses, postsynaptic potentials, and membrane potentials can be used as a new way to explain and analyse the ABA signals. The proposed method using time-varying weights highlights a correspondence with the physical problem and offers an ability to capture sudden and subtle changes in the responses, which is crucial, particularly for detecting defects in their early stages.

Nowadays, rolling stock can be equipped with high-frequency vibration sensors to continuously monitor rail infrastructures and detect defects. These moving sensors measure at high speeds and sampling frequencies, generating a massive amount of data that covers each track position with very short signal durations. These data contain a variety of dynamic and transient responses that vary significantly along the track and are affected by noise. This leads to a large amount of unlabeled and noisy data, complicating the extraction of dynamic responses for effective anomaly detection. To address these challenges, this paper proposes an unsupervised representation learning methodology to automatically capture and extract characteristic features of dynamic responses that reflect the conditions of rail infrastructures. The unsupervised nature allows exploratory analysis of high-frequency vibration signals when prior knowledge or reference information about infrastructure conditions is unavailable or very limited. A collaborative optimization process that synchronizes empirical mode decomposition (EMD) with a convolutional autoencoder (CAE) is presented. The EMD level is tuned to remove noise while preserving effective vibration responses. The CAE is trained using demodulated signals that are considered normal to generate representations that ensure reconstruction quality and differentiate between normal and abnormal conditions. Furthermore, a Gaussian mixture model is used to showcase the effectiveness of the learned representations for rail infrastructures. Applied to validated axle box acceleration data for rail defect detection and train-borne laser Doppler vibrometer data for rail fastener monitoring, our method outperforms other variants of autoencoder-based models and the wavelet-based CAE in accurately identifying the conditions. It achieves an average improvement of 16% with the axle box acceleration data and 21% with the laser Doppler vibrometer data.

Transition zones in railway tracks often degrade faster than other locations, yet traditional health assessments rely on infrequent track geometry measurements, limiting early detection of dynamic changes. This research presents an approach for more frequent evaluation of transition zone health by integrating data sources from multiple monitoring technologies: track geometry, interferometric synthetic aperture radar (InSAR), and axle box acceleration (ABA). Missing InSAR data are addressed through spatio-temporal interpolation, and track longitudinal levels are predicted using a hybrid neural model that includes a hybrid convolutional neural network (CNN) with gated recurrent units (GRU) network and a hybrid CNN with a long short-term memory (LSTM) network. The models fuse historical and interpolated data from InSAR and ABA, enabling higher-frequency insights. A novel key performance index (KPI) based on predicted longitudinal levels is proposed to quantify track condition. The framework is validated on a transition zone at a railway bridge between Dordrecht and Lage Zwaluwe in the Netherlands. Results show that the hybrid model outperforms standalone methods and offers a good balance between accuracy and computational efficiency. The proposed approach enables earlier detection of irregularities, supporting prescriptive maintenance decisions.

This paper proposes an onboard measurement technology that combines a Laser Doppler Vibrometer (LDV) and an Axle Box Accelerometer (ABA) to approximate the dynamic load-response relationship of railway tracks. Unlike existing track-side and onboard technologies, this paper eliminates the need for load measurement, estimation, or control, enabling continuous measurements under operational conditions. The LDV mounted on the moving vehicle captures the track vibration response contactlessly, while the wheel vibration measured by ABA is used to directly represent the dynamic vehicle load. The LDV and ABA signals are combined to approximate the load-response relationship in the frequency domain. Experimental validation on a vehicle-track test rig demonstrates the effectiveness of the developed system at different speeds. Further comparisons with the hammer test result confirm its ability to capture the local dynamic properties of consecutive track segments along a railway track and also its superiority in measurement efficiency. This paper offers a promising solution for monitoring railway tracks on a large scale and allowing prescriptive maintenance of rail infrastructures.

Regional non-electrified railway networks require replacement of diesel traction to meet increasingly stringent emission reduction targets. Since full electrification of these networks is often not economically viable due to their low utilization, battery-electric multiple units (BEMUs) are recognized as a potentially suitable long-term solution, offering zero-emission train operation while requiring only partial tracks electrification. One of the main challenges when introducing BEMUs is determining an optimal electrification layout, i.e. the location and the length of electrified track sections while taking into account the vehicles’ and infrastructure technical characteristics and constraints alongside the requirements related to maintaining current timetable and quality of service. This paper formulates this as an intermittent partial electrification network design problem and develops an optimization framework that integrates high-fidelity BEMU simulation model in deriving a cost-optimized network electrification configuration. The proposed method is demonstrated using the existing non-electrified regional railway network in the Netherlands with the rolling stock and transport services of Arriva as a case. The obtained solution provides about 30% lower capital costs compared to the conventional continuous partial electrification approach, and about 3.5 times cut in these costs compared to the fully electrified network. Additionally, further costs reduction is observed by increasing the maximum current absorption limits at standstill and by introducing flexibility in terms of operational margins.

This paper presents a methodology for detecting and monitoring short pitch corrugation (SPC) under varying measurement conditions using vertical and longitudinal axle box acceleration (ABA) measurements. The main objective of the detection algorithm is to determine the likelihood and approximate severity of SPC presence, providing insights for maintenance planning. The methodology combines a validated three-dimensional finite element (3D-FE) model of the ABA responses at SPC and signal processing techniques to extract meaningful data from the real-world on-board measurements. First, a 3D-FE vehicle-track model is validated and used to quantify the physical relationships in the time–frequency responses of ABA at SPC under different levels of corrugation severity and measurement speeds. Then, a measurement train is instrumented with multiple accelerometers to capture field data on ABA at SPC, which is validated with field inspections and Railprof measurements. Finally, the ABA responses are analyzed based on the number of signals detecting SPC and an assessment of severity based on impact energy due to SPC. The methodology is demonstrated by analyzing the track between Assen and Groningen on the Dutch rail network. Results show that the methodology accurately detects registered SPC locations. Further, a whole track analysis is conducted, from which the methodology proposes new locations and severities of SPC, providing crucial information for rail maintenance planning.

Modern smartphones, widely available and equipped with multiple sensors, offer the potential for railway infrastructure condition monitoring through mobile crowdsensing without disrupting operational railway services. This paper investigates key factors influencing the use of smartphone accelerometers for railway track quality assessment and highlights the associated challenges. Differences in accelerometer characteristics across three budget smartphones, including variations in sampling frequencies and dynamic sensitivity, are analyzed. A case study conducted on in-service passenger trains examines the effects of operational conditions, specifically vehicle speed and smartphone placement within the car body, on the frequency content and magnitude of recorded signals. Additionally, this study compares smartphone measurement results with a conventional track quality index derived from track geometry measured by specialized vehicles. Promising correlations are observed between the standard deviation of the smartphone vertical acceleration signals and the longitudinal level-based track quality index, demonstrating the potential of smartphones to assess track quality and detect local anomalies. However, variations caused by vehicle speed and smartphone placement pose challenges for standardizing track quality assessments. These findings highlight the potential of mobile crowdsensing for railway infrastructure monitoring while emphasizing the need for strategies to address variability in operational conditions and device characteristics.

This work presents the results of a measurement campaign to demonstrate the effectiveness of the axle box acceleration (ABA) technology for detecting rail defects. The measurements were conducted along the Iron Ore line between Sweden and Norway for the IN2TRACK3 project. This line is mostly single-track with passenger-freight mixed traffic and heavy axle load. Historical data and track information data were not considered in this study. By analyzing data acquired from the accelerometers in vertical and longitudinal directions, rail defects were detected in near real-time using big-data analytics. For our validated sections, 100% of rail defects (including squats) were detected using time-frequency analysis and an outlier detection approach. The methodology also allows for identifying priority locations, e.g., defective welds, joints, transition zones, etc., and its use for prescriptive maintenance recommendations is being explored in the framework of the IAM4RAIL project.

Railway transition zones connecting conventional embankments and rigid struc-tures, such as bridges and tunnels, usually degrade much faster than other railway sections. Efficient health condition monitoring of transition zones is important for preventative track maintenance. In this paper, a methodology for monitoring rail-way transition zones using acceleration measurements on multiple axle boxes (multi-ABA) of a passing train is presented. To showcase its capability, the measurements in the Netherlands, Sweden, and Norway are analyzed and dis-cussed. It is found that different bridges and transition zones exhibit unique char-acteristics including dominant wavelengths and energy distribution. Based on these unique characteristics, the geometry and support conditions at different lo-cations of a transition zone can be evaluated. Higher train speed makes the char-acteristics more pronounced. The results demonstrate that the multi-ABA meas-urement has the potential to evaluate and thus monitor the health conditions of various transition zones.

A train-borne laser Doppler vibrometer (LDV) directly measures the dynamic response of railway track components from a moving train, which has the potential to complement existing train-borne technologies for railway track monitoring. This paper proposes a holistic methodology to characterize train-borne LDV measurements by combining computer-aided approaches and real-life measurements. The focus is on the speed-dependent characteristics because the train speed affects the intensity of railway sleeper vibrations and the intensity of speckle noise, which further affects the quality and usability of the measured signals. First, numerical models are established and validated to simulate sleeper vibrations and speckle noise separately. Then, a vibration–noise separation method is proposed to effectively extract speckle noise and structural vibrations from LDV signals measured at different speeds. The parameters of the separation method are tuned using simulation signals. The method is then validated using laboratory measurements in a vehicle-track test rig and applied to field measurements on a railway track in Rotterdam, the Netherlands. Further, the speed-dependent characteristics of train-borne LDV measurement are determined by analyzing the competition between sleeper vibrations and speckle noise at different speeds. Simulation and measurement results show that an optimal speed range yields the highest signal-to-noise ratio, which varies for different track structures, measurement configurations, and operational conditions. The findings demonstrate the potential of train-borne LDV for large-scale rail infrastructure monitoring.

This paper presents a new approach to simulate forward and inverse problems of moving loads using physics-informed machine learning (PIML). Physics-informed neural networks (PINNs) utilize the underlying physics of moving load problems and aim to predict the deflection of beams and the magnitude of the loads. The mathematical representation of the moving load considered involves a Dirac delta function, to capture the effect of the load moving across the structure. Approximating the Dirac delta function with PINNs is challenging because of its instantaneous change of output at a single point, causing difficulty in the convergence of the loss function. We propose to approximate the Dirac delta function with a Gaussian function. The incorporated Gaussian function physical equations are used in the physics-informed neural architecture to simulate beam deflections and to predict the magnitude of the load. Numerical results show that PIML is an effective method for simulating the forward and inverse problems for the considered model of a moving load.

This study presents a measuring framework for railway transition zones using a case study on the Swedish line between Boden and Murjek. The final goal is to better understand the vertical dynamics of transition zones using hammer tests, falling weight measurements, and axle box acceleration (ABA) measurements. Frequency response functions (FRFs) from hammer tests indicate two track resonances, for which the FRF magnitudes on the plain track are at least 30% lower than those at the abutment. The falling weight measurements indicate that the track on the bridge has a much higher deflection than the track on the embankment. Two features from ABA signals, the dominant spatial frequency and the scale average wavelet power, show variation along the transition zone. These variations indicate differences in track conditions per location. Finally, the ABA features in the range of 1.05–2.86 m⁻¹ are found to be related to the track resonance in the range of 30–60 Hz. The findings in this paper provide additional support for physically interpreting train-borne measurements for monitoring transition zones.

A transfer function (TF) is an effective representation of the load-response relationship of railway track structures. To fill the gap in measuring track structure TFs over a wide frequency range from a moving vehicle, we develop a TF measurement system and the associated TF estimation methodology. Accelerometers are utilized to estimate the dynamic vehicle load to track structures, and a laser Doppler vibrometer (LDV) is used to scan track structures and measure their vibration response. First, operational modal analysis is applied to vehicle impact response over joints to identify its modal parameters, which support the estimation of dynamic wheel-rail forces from vehicle vibrations. This combination eliminates the need to pre-define the vehicle stiffness, vehicle damping, and vehicle body mass and enables the vehicle parameters to be updated under operational conditions. Meanwhile, a signal processing method is applied to LDV signals to reduce speckle noise and compensate for the effect of vehicle vibration. Then, a continuous track structure is segmented into distributed sections, and a TF is estimated for each track section using the estimated wheel-rail force as input and the extracted track vibration as output. We validate the methodology in a vehicle-track test rig on different track sections (with or without joints) and at different speeds (from 8 km/h to 16 km/h). The results are further compared with trackside measurements and hammer tests. We demonstrate that the track vibrations extracted from the LDV signals are consistent with those measured by trackside accelerometers. The shapes and resonance frequencies of the estimated TFs are in good agreement with those measured from the hammer tests in the frequency range of 200–800 Hz. The developed system captures differences in the TFs between different track sections, suggesting its potential to be used for structural health monitoring of railway tracks.

This article develops and tests a self-contained railway track monitoring system that fits in existing vehicles without the need for speed and load control. Combining a train-borne laser Doppler vibrometer and axle box accelerometers enables synchronized measurements of train-track response under operational conditions. Utilizing a GPS antenna and video camera, we propose the multisignal processing method to obtain train-track vibrations with train position and speed. Then, we fuse the multiple signals to extract an impact index and a resonance index and further propose an interpretable anomaly detection strategy. We test the system on an operational line at 20-60 km/h under different working conditions and verify the detection results using information from conventional technologies. The impact index peaks near joints and welds, and the resonance index yields a good correlation with the measured track geometry. The developed solution achieves the detection, localization, and quantification of surface and support anomalies in railway tracks.

Inefficient management of rail surface defects can increase maintenance costs, safety hazards, service disruptions, and catastrophic failures like rail breaks. To achieve adequate management, having effective technology capable of timely detecting and frequently monitoring rail defects is of utmost importance. The aim is early detection of defects to maintain safety levels and prevent the re-appearance due to residual damages.

Various measurement technologies, such as visual inspections, geometry profile measurements, and other measurement techniques, have been used for the detection of rail defects. While these methods provide insights, they often lack the capability for early-stage defect detection. Thus, most of these technologies are suitable for reactive maintenance since they detect defects when they reach a certain severity level. Axle box acceleration (ABA) technology provides a solution capable of frequent monitoring, mounted on trains in operation without dedicated measurement vehicles (see figure 71-1). Its basic principle is to use a train as a moving load that excites the infrastructure and to detect defects by evaluating the time-frequency characteristics of the dynamic response measured by accelerometers installed on axle boxes of the train. ABA systems have shown promise in detecting defects in the early stages. However, its widespread application and need for robustness require further validation and development. This work presents the results of detecting and monitoring rail surface defects using ABA technology.

This paper presents a method for estimating Well-to-Wheel (WTW) energy use and greenhouse gas (GHG) emissions attributed to the advanced railway propulsion systems implemented in conjunction with different energy carriers and their production pathways. The analysis encompasses diesel-electric multiple unit vehicles converted to their hybrid-electric, plug-in hybrid-electric, fuel cell hybrid-electric or battery-electric counterparts, combined with biodiesel or hydrotreated vegetable oil (HVO) as the first and second generation biofuels, liquefied natural gas (LNG), hydrogen and/or electricity. The method is demonstrated using non-electrified regional railway network with heterogeneous vehicle fleet in the Netherlands as a case. Battery-electric system utilizing green electricity is identified as the only configuration leading to emission-free transport while offering the highest energy use reduction by 65–71% compared to the current diesel-powered hybrid-electric system. When using grey electricity based on the EU2030 production mix, these savings are reduced to about 27–39% in WTW energy use and around 68–73% in WTW GHG emissions. Significant reductions in overall energy use and emissions are obtained for the plug-in hybrid-electric concept when combining diesel, LNG, or waste cooking oil-based HVO with electricity. The remaining configurations that reduce energy use and GHG emissions are hybrid-electric systems running on LNG or HVO from waste cooking oil. The latter led to approximately 88% lower WTW emissions than the baseline for each vehicle type. When produced from natural gas or EU2030-mix-based electrolysis, hydrogen negatively affected both aspects, irrespective of the prime mover technology. However, when produced via green electricity, it offers a GHG reduction of approximately 90% for hybrid-electric and fuel cell hybrid-electric configurations, with a further reduction of up to 92–93% if combined with green electricity in plug-in hybrid-electric systems. The results indicate that HVO from waste cooking oil could be an effective and instantly implementable transition solution towards carbon–neutral regional trains, allowing for a smooth transition and development of supporting infrastructure required for more energy-efficient and environment-friendly technologies.

This paper proposes a novel framework for simulating the dynamics of beams on elastic foundations. Specifically, partial differential equations modeling Euler–Bernoulli and Timoshenko beams on the Winkler foundation are simulated using a causal physics-informed neural network (PINN) coupled with transfer learning. Conventional PINNs encounter challenges in handling large space–time domains, even for problems with closed-form analytical solutions. A causality-respecting PINN loss function is employed to overcome this limitation, effectively capturing the underlying physics. However, it is observed that the causality-respecting PINN lacks generalizability. We propose using solutions to similar problems instead of training from scratch by employing transfer learning while adhering to causality to accelerate convergence and ensure accurate results across diverse scenarios. The primary contribution of this paper lies in introducing a causality-respecting PINN loss function in the context of structural engineering and coupling it with transfer learning to enhance the generalizability of PINNs in simulating the dynamics of beams on elastic foundations. Numerical experiments on the Euler–Bernoulli beam highlight the efficacy of the proposed approach for various initial conditions, including those with noise in the initial data. Furthermore, the potential of the proposed method is demonstrated for the Timoshenko beam in an extended spatial and temporal domain. Several comparisons suggest that the proposed method accurately captures the inherent dynamics, outperforming the state-of-the-art physics-informed methods under standard L²-norm metric and accelerating convergence.

The growing volume of available infrastructural monitoring data enables the development of powerful data driven approaches to estimate infrastructure health conditions using direct measurements. This paper proposes a deep learning methodology to estimate infrastructure physical parameters, such as railway track stiffness, using vehicle-based vibration response signals. The proposed method employs a Long Short-term Memory (LSTM) feature extractor accounting for temporal dependencies in the feature extraction phase, and a bidirectional Long Short-term Memory (BiLSTM) networks to leverage bidirectional temporal dependencies in both the forward and backward paths of the drive-by vibration response in condition estimation phase. Additionally, a framing approach is employed to enhance the resolution of the monitoring task to the beam level by segmenting the vibration signal into frames equal to the distance between individual beams, centering the frames over the beam nodes. The proposed LSTM-BiLSTM model offers a versatile tool for various bridge and railway infrastructure conditions monitoring using direct drive-by vibration response measurements. The results demonstrate the potential of incorporating temporal analysis in the feature extraction phase and emphasize the pivotal role of bidirectional temporal information in infrastructure health condition estimation. The proposed methodology can accurately and automatically estimate railway track stiffness and identify local stiffness reductions in the presence of noise using drive-by measurements. An illustrative case study of vehicle-track interaction simulation is used to demonstrate the performance of the proposed model, achieving a maximum mean absolute percentage error of 1.7% and 0.7% in estimating railpad and ballast stiffness, respectively.

Computer-aided simulations are routinely used to predict a prototype's performance. High-fidelity physics-based simulators might be computationally expensive for design and optimization, spurring the development of cheap deep-learning surrogates. The resulting surrogates often struggle to generalize and predict novel scenarios beyond their training domain. We propose a two-stage methodology addressing the challenge of generalization. It employs physics-based simulators, supplemented with ordinary differential equations integrated into the recurrent architecture, to learn the intrinsic dynamics. The proposed approach captures the inherent causality and generalizes the dynamics irrespective of a data source. The presented numerical experiments encompass five fundamental structural engineering scenarios, including beams on Winkler foundations based on Euler-Bernoulli and Timoshenko theories, beams under moving loads, and catenary-pantograph interactions in railways. The proposed methodology outperforms conventional recurrent methods and remains invariant to data sources, showcasing its efficacy. Numerical experiments highlight its prospects for design optimization, predictive maintenance, and enhancing safety measures.