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.

The conventional vertical track quality index (TQI) based on the standard deviation of longitudinal levels yields standardized railway track condition assessment. Nevertheless, its capability to identify problems is limited, particularly in the ballast and substructure layers when abrupt changes affect train-track interaction. Previous research shows that dynamic responses from moving trains via axle box acceleration (ABA) measurements can quantify abrupt changes in the vertical dynamic responses. Thus, this paper proposes a framework to design an enhanced vertical TQI, called EnVTQI, by integrating track longitudinal levels and dynamic responses from ABA measurements. First, measured ABA signals are processed to mitigate the influence of variation in measurement speed. Then, substructure and ballast-related features are extracted, including scale average wavelet power (SAWP) in the ranges 0.04 m^-1 to 0.33 m^-1 (substructure) and 1.25 m^-1 to 2.50 m^-1 (ballast). This enables identifying track conditions at different track layers. Finally, EnVTQI is determined by weight averaging between the conventional vertical TQI and the ABA features from moving trains. The performance of EnVTQI is evaluated based on 48 segments of a 200-m track on a Dutch railway line. The results indicate that EnVTQI helps to distinguish track segments that cause poor train-track interaction, which the conventional TQI does not indicate. EnVTQI can supplement the conventional TQI, improving the effectiveness of track maintenance decision-making.

The railway industry has the potential to make a strong contribution to the achievement of various sustainable development goals, by an expansion of its role in the transportation system of different countries. To realize this, complex technological and societal challenges are to be addressed, along with the development of suitable state-of-the-art methodologies fully tailored to the particular needs of the wide variety of railway infrastructure types and conditions. Artificial intelligence (AI) methods have been increasingly and successfully applied to solve practical problems in the railway infrastructure domain for over two decades. This paper proposes a review of the development of AI methods in railway infrastructure. First, we present a survey limited to selected journal papers published between 2010 and 2022. Bibliographical statistics are obtained, showing the increasing number of contributions in this field. Then, we select key AI methodologies and discuss their applications in the railway infrastructure. Next, AI methods for key railway components are analyzed. Finally, current challenges and future opportunities are discussed.