Axle Box Research Articles

Damage Identification for Railway Tracks Using Onboard Monitoring Systems in In-Service Vehicles and Data Science

Nowadays, railway track monitoring strategies are based on the use of railway inspection vehicles and wayside dynamic monitoring systems. The latter sometimes requires traffic disruption, as well as higher time and cost-consumption activities, and the use of dedicated inspection vehicles is less economical and efficient as the use of in-service vehicles. Furthermore, the use of non-automated algorithms faces challenges when it comes to early damage detection in railway infrastructure, considering operational, environmental, and big data aspects, and may lead to false alarms. To overcome these challenges, the application of artificial intelligence (AI) algorithms for early detection of track defects using accelerations, measured by dynamic monitoring systems in in-service railway vehicles is attracting the attention of railway managers. In this paper, an AI-based methodology based on axle box acceleration signals is applied for the early detection of distributed damage to track in terms of the longitudinal level and lateral alignment. The methodology relies on feature extraction using an autoregressive model, data normalization using principal component analysis, data fusion and feature discrimination using Mahalanobis distance and outlier analysis, considering eight onboard accelerometers. For the numerical simulations, 75 undamaged and 45 damaged track scenarios are considered. The alert limit state defined in the European Standard for assessing track geometry quality is also assumed as a threshold. It was found that the detection accuracy of the AI-based methodology for different sensor layouts and types of damage is greater than 94%, which is acceptable.

Analysis of Acoustic Flow Field Characteristics of Key Components of High-Speed Train Bogies

In this paper, we aim to address the challenge of airflow interference during fault detection in high-speed train bogies by introducing a flow field and investigating the characteristics of the sound field distribution of critical components under its influence. This approach overcomes the limitation observed in previous studies that ignored the impact of the flow field. Furthermore, we evaluate and compare various layouts for inter-track acoustic sensor arrays. The study results demonstrate that the axle box and gearbox are primarily affected by frequency bands above 200 Hz, while the motor is mainly influenced by frequencies above 150 Hz. Moreover, the motor is most significantly impacted by vehicle speed, followed by the gear box and finally the shaft box. When altering the vibration frequency of the sound source, there is a decreasing trend in minimum sound pressure level with increasing frequency. Additionally, the gradient of the sound pressure level change at three locations (motor, gear box, and axle box) exceeds that of both vehicle speed variation and vibration intensity of the sound source. By adjusting the vibration intensity of the sound source, similar gradients are observed for both maximum and minimum values of sound pressure level at these three locations. Therefore, vibration intensity has a consistent effect on changes in sound pressure level at these three places but only influences its magnitude. The aforementioned methods can be employed for the analysis of wind noise and structural noise, thereby offering novel insights for future advancements in noise reduction processing.

Vold–Kalman filter order tracking of axle box accelerations for track stiffness assessment

Intelligent data-driven monitoring procedures hold enormous potential for ensuring safe operation and optimal management of railway infrastructure in the face of increasing demands on cost and operations efficiency. Numerous studies have highlighted track stiffness as a main parameter influencing the evolution of degradation that drives maintenance processes. As such, the measurement of track stiffness is fundamental for characterizing the performance of the track in terms of deterioration rate and noise emission. In this work, we propose a rail stiffness assessment scheme relying on low-cost, on board monitoring sensors, namely axle-box accelerometers, that are mounted on in-service trains and enable frequent, real-time monitoring of the railway infrastructure network. A Vold–Kalman filter is proposed for decomposing the signal into periodic wheel and track related excitation–response pair functions. We demonstrate that these components are in turn correlated to operational conditions, such as wheel out-of-roundness and the rail type. We further illustrate the relationship between the track stiffness, the measured wheel-rail forces and the sleeper passage amplitude, which can ultimately serve as an indicator for predictive track maintenance and prediction of track durability.

Multi-Objective Sparsity Maximum Mode De-Composition: A New Method for Rotating Machine Fault Diagnosis on High-Speed Train Axle Box

Signal decomposition techniques generally have two limitations. First, for adaptive correlated kurtogram (ACK) and variational mode decomposition (VMD), etc., the definition of the decomposition target has no direct relationship with the features of fault impacts, i.e., the decomposition is not guided by the health indices used to characterize bearing faults, resulting in fault impacts cannot be completely separated from vibration data. Second, it is hard to accurately determine the decomposition stop parameter (the number of final output modes), resulting in frequent over- or under-decomposition. All these are not conducive to bearing fault detection in engineering practice. In this paper, a new decomposition method, multi-objective sparse maximum mode decomposition (MOSMMD), is proposed. Firstly, by analyzing the Fourier spectrum of vibration data, MOSMMD constructs multiple finite impulse response filters covering different frequency bands to obtain initial candidate modes. Then, aiming at maximizing <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L2</i> / <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L1</i> norm of envelope spectrum (ES), the upper and lower cutoff frequencies of each filter are adjusted adaptively through iterative optimization, and the frequency bands covered by each mode are adjusted accordingly. Finally, by using that fault feature harmonics have periodic natures in ES, the redundant modes are eliminated and only the modes with obvious fault features are output. Compared with VMD and ACK, the decomposition target of MO-SMMD is directly related to the fault features, i.e., cyclostationarity of fault impacts, which makes the extraction of fault impacts more accurate. Moreover, the number of final output modes of MOSMMD is determined adaptively and does not need to be set in advance as an input parameter. Also, MOSMMD can decompose fault modes from vibration data without prior fault information, even when processing concurrent fault signals. Simulation and experiments demonstrate the superiority of the proposed MOSMMD.

A study of the axle box ASD spectrum of high-speed rail vehicles based on field test measurements in China

The axle box ASD (Acceleration Spectral Density) spectrum serves as the main resource of high frequency vibration for rail vehicles, which is usually taken as a critical load spectrum for the evaluation of vibration fatigue of rail vehicle sub-assemblies. However, the continuous in-service failures occurring in a plenty of rail vehicles invariably suggested a low reliability for the ASD defined in IEC 61373 standard for some scenarios. This paper thus investigated on the characteristics of axle box ASD spectrum through measurements acquired from a number of high speed rail vehicles in China, so as to develop a more realistic vibration spectrum for the considered operation condition. Firstly, a brief review for typical fatigue failures of rail vehicles was conducted to understand fatigue failure modes caused by the high frequency vibration of wheel/rail system. Subsequently, the definition of axle box ASD spectra in IEC 61373 was analyzed and further compared with the ASD spectra of axle box obtained from measurements to demonstrate the reliability of standard vibration level. The results suggest that the ASD spectrum given in IEC 61373 underestimates the vibration level in high frequency ranges, especially the frequency band above 500 Hz. Finally, a modified ASD spectrum profile of axle box considering the characteristic frequency bands of wheel/rail system was proposed to better representation of vibration characteristics of axle box, especially for those arising from the wheel/track vibration modes. This is expected to better characterize the vibration characteristics of axle box for rail vehicles, and improve the estimation accuracy of fatigue lifetime for the components subjected to the axle box vibration.

Connection Between Track Geometry Quality and Dynamic Vehicle Response at Various Speeds

Track geometry measurements are widely used for describing track quality. However, derailments and track deterioration are caused by forces arising in vehicle-track system. This research focuses on two types of vehicle response. Firstly, the influence of the longitudinal level irregularities on the vertical wheel-rail forces was examined. Secondly, the correlation between the lateral axle box acceleration and the cross level irregularities was investigated. Track geometry and vehicle response data were acquired simultaneously by a track recording car, formed from a passenger car, at various speeds up to 130 km/h. Vehicle-track forces were calculated based on accelerometers mounted on the car body, bogies and axle boxes, considering mass and moment of inertia. Non-linear regressions resulted in vertical vehicle-track force estimation functions. It was proven that the use of second spatial derivatives of the longitudinal level gave a better estimation than the use of reference TQIs according to European Standard EN 13848-6. A linear relationship was found between the speed and standard deviation of vertical vehicle-track forces. On straight sections with constant speed, correlation coefficients of around 0.8 were found between second spatial derivatives of cross level and lateral axle box acceleration.

Анализ современных подшипниковых узлов, применяемых для высокоскоростного состава, и перспективы их развития

Purpose: To identify the main directions for increasing the service life and reliability of bearing assemblies for high-speed rolling stock. Methods: An analysis of modern bearing designs and new materials used in their manufacturing has been conducted. Results: The analysis of the conducted research has shown that all the examined designs of bearing assemblies produced by NSK company mainly meet the current demands of railways operating at speeds up to 320 km/h, and can ensure the safety of train movement under all existing operating conditions. Recommendations for equipping the axle box unit s of rolling stock with bearings that perform most successfully under special operating conditions have been developed. The issues of monitoring the faults of axle box units operating on high-speed trains, such as TGV, ICE, and Talgo, and the systems of their onboard diagnostics, have been addressed. An assessment of the domestic control system is provided, and its advantages over foreign counterparts are demonstrated. A solution for diagnosing trains under the ‘black box’ regime has been proposed. Practical significance: The application of modern bearing designs, new materials, and diagnostic systems allows for an increase in the reliability and performance of railway rolling stock.

A small sample diagnosis method driven by simulation and test data: applied to axle box bearings of high-speed train

Abstract Fault diagnosis of axle box bearings is an important technology to improve the service safety and economy of high-speed trains. However, it is difficult to obtain sufficient fault samples in actual train operation, which limits the application of deep learning methods in the field of high-speed train fault diagnosis. Therefore, in this paper, a simulation and test data fusion-driven fault diagnosis method for axle box bearings with few samples is proposed. In this method, a bearing dynamics model is constructed to obtain simulation data for fault bearings, and the accuracy of the constructed dynamics model is verified by experimental data. Data fusion methods are proposed which use massive simulated fault samples and only normal test samples to construct new samples for the pre-training of deep learning diagnostic models. The parameter transfer strategy is used to fine-tune the diagnostic model with a small amount of test fault samples. A real train experiment is carried out for validating the proposed methods which use the real fault bearings that fail during operation. The results show that the proposed method via the fusion of simulation and test data has higher generalization ability and diagnostic accuracy in small sample situations.

Multiaxial Fatigue Analysis of Connecting Bolt at High-Speed Train Axle Box under Structural Subharmonic Resonance.

Based on the dynamic characteristics of the axle box front cover of high-speed trains in the subharmonic resonance state, the nonlinear single-degree-of-freedom (SDOF) model was proved to be reasonable, and reasons for the ineffectiveness of the common prevention methods for preventing bolt failure were analyzed firstly. Then, dynamic stress of the bolt was simulated by innovatively adopting the linear method based on frequency response analysis. The stress simulation method was verified to be practical under the subharmonic resonance state by analyzing and comparing the experimental and numerical results of the bolted front cover. It was proved that the linear method was accurate enough to simulate the dynamic stress of bolts, which is of great engineering significance. In addition to the transverse resonance stress of bolts caused by drastic vertical vibration of the front cover, the tensile resonance stress at the root of the first engaged thread was too large to be neglected on account of the first-order bending modes of bolts. Next, equivalent stress amplitude of the multiaxial stresses was obtained by means of the octahedral shear stress criterion. Finally, fatigue life of bolts was predicted in terms of S-N curve suitable for bolt fatigue life analysis. It argued that the bolts were prone to multiaxial fatigue failure when the front cover was in subharmonic resonance for more than 26.8 h, and the fatigue life of bolts could be greatly improved when the wheel polygonization was eliminated by shortening the wheel reprofiling interval.

Investigation of the Effect of Background Radiations on Infrared Sensor Measurement Data in Hot Axle Box Detection Systems

Kızılötesi sıcaklık ölçüm teknolojisi, demiryolu taşıtlarının güvenliği ve verimliliği için kritik öneme sahiptir. Ancak çevresel koşullar, özellikle yansıyan radyasyona bağlı arka plan ışımaları (back-ground radiation), doğru ölçümleri engelleyebilir veya yanıltıcı sonuçlara yol açabilir. Bu çalışma, arka plan ışımalarının sıcak aks kutusu kızılötesi sıcaklık sensör ölçüm verilerine olan etkisini incelemekte ve bu etkiyi elimine edecek analitik bir yöntem önermektedir. Bu sayede daha doğru ve güvenilir sonuçlar elde edilmesine katkı sunmaktadır. Bu amaçla, aks kutusunun gerçek verilerine dayalı bir simülasyon modeli kurulmuş ve farklı ortam sıcaklıklarının sensör ölçüm sonuçlarına etkisi incelenmiştir. Simülasyon sonuçlarından elde edilen veriler, ortam ile ölçüm yüzeyi arasındaki sıcaklık farkı azaldıkça arka plan ışıma etkisinin daha yüksek olduğunu göstermiştir. Örneğin 50°C bir hedef yüzey sıcaklığı ve 45°C ortam sıcaklığı altında 6,8°C bir sıcaklık sapması oluşmuştur. Ayrıca kızılötesi sensörün sıcaklığı doğru algılaması için yüzeyin yayıcılık katsayısının düzgün belirlenmesinin önemli olduğu görülmüştür.

Evaluating railway track stiffness using axle box accelerations: A digital twin approach

While various train-borne techniques have been developed for measuring railway track stiffness, differentiating stiffness at different track layers remains a challenge. This study proposes a digital twin framework for the vehicle–track interaction system, which enables track stiffness evaluations based on axle box accelerations (ABA). The digital twin consists of a physics-based model, a model library and data-driven models. Compared to existing techniques, the proposed method simultaneously evaluates the stiffness of the railpad, sleeper and ballast layers at a sleeper spacing resolution, while being robust to varying track conditions, such as track irregularities and vehicle speeds. This is accomplished by employing a localized frequency-domain ABA feature capable of distinguishing between the characteristics of different track layers. Furthermore, track stiffness is evaluated in near real-time. This is achieved using a model library derived from physics-based simulations of a range of track conditions. Two data-driven models that can quickly select or interpolate model instances contained in the library are developed. During operation, the data-driven models use the measured ABA features as input and then infer the stiffness for the different track layers. The proposed method is applied to evaluate the track stiffness of a downscale test rig in a case study. The track stiffness evaluated by the proposed method is compared with that obtained through hammer tests and with the observations of the track component conditions. These comparisons show that the proposed method can capture the stiffness variations due to periodically fastened clamps and substructure misalignments at different speeds. In addition, the proposed method is demonstrated to be superior to the commonly used hammer test method for evaluating track stiffness under loaded conditions.

Identification of high-speed railway turnout generated vibrations based on time–frequency slice and mutual correlation of vibration responses from axle box

Railway turnouts are one of the weaknesses of the high-speed railway; their health condition inspection is, therefore, crucial for the safety of high-speed trains. Vibrations collected from train axle boxes contain rich information of health conditions of railway turnouts. Identification of high-speed railway turnout generated vibrations is necessary for the subsequent assessment of the overall health status of the turnout based on the vibration signals generated by each element of the turnout (including welded joints, insulation joints, switch area, frog area, etc.). However, the turnout location in collected vibration data may vary from the actual one recorded. Hence, in order to detect the turnout location in the collected vibration data and identify the difference between the turnout location in the vibration and the actual turnout location recorded, this paper proposes a time–frequency slice and numerical model-driven mutual correlation method. Ensemble empirical mode decomposition (EEMD) is first applied to decompose the vibration signals collected from the train axle box. A time–frequency slice strategy is developed based on time-reassigned multi-synchrosqueezing transform (TMSST) to extract the impacts from joints and frog of the turnout. A vehicle turnout coupling dynamics model is then established to drive the mutual correlation to perform the turnout-induced impact extraction and the position difference identification. The proposed method is validated by on-site experimental data.

Multilevel Feature Extraction Method for Adaptive Fault Diagnosis of Railway Axle Box Bearing

Railway axle box bearing fault signal contains high Q-factor resonance and low Q-factor transient shock components with periodic transient shock features that can characterize bearing faults. However, extracting fault features is usually difficult due to noise, transmission paths, and high-amplitude accidental shocks. Therefore, to address the abovementioned problems, the multilevel feature extraction method for adaptive fault diagnosis of railway axle box bearing was proposed. This paper used the maximum second-order cyclostationary blind convolution (CYCBD) to weaken the influence of disturbances, enhancing weak transient shock components. Resonant sparse decomposition(RSSD) is used for the selection of quality factor Q due to its excellent fault feature separation property. Considering the strict fault frequency component periodicity in the envelope spectrum, the envelope spectrum multipoint kurtosis (ESMK) is proposed as a metric. The Q factor is optimized using the gray wolf optimization algorithm (GWO) to obtain an adaptive sparse decomposition method (GWO-RSSD) for extracting bearing transient fault shocks to eliminate the signal effects of high-amplitude disturbance shocks and background noise. The simulation results and measured signal analysis of railway axle box bearing show that ESMK can effectively measure periodic transient shocks under strong shock interference compared with the MED-RSSD method. Thus, GWO-RSSD can adaptively separate the optimal low-Q resonance components, verifying the methods’ effectiveness and superiority.

Research on Rolling Contact Fatigue Failure of the Bearing Used in High-Speed Electric Multiple Units’ Axle Box Based on a Damage-Coupled Elastic–Plastic Constitutive Model

The axle box bearing is a crucial component of high-speed electric multiple units (EMU) and is exposed to harsh working conditions, making it susceptible to subsurface-induced rolling contact fatigue (RCF) under long-term alternating stress. The objective of this paper is to develop a damage-coupled elastic–plastic constitutive model that can accurately predict the RCF life of EMU axle box bearings made from AISI 52100 bearing steel. The total damage is divided into elastic damage related to the shear stress range and plastic damage associated with plastic deformation. Material parameters are determined based on experimental data from the literature, and validation is conducted to ensure the validity of the model. Finally, the RCF behavior of the EMU axle box bearing, including crack initiation, crack propagation, and spalling, is simulated, and reasonable results are obtained. This study provides valuable insights into the RCF behavior of EMU axle box bearings and contributes to the accurate prediction of the fatigue life.

On the use of transmissibility for the detection of damaged springs in the primary suspension of a locomotive

The condition monitoring of the suspensions of railway vehicles is of utmost importance, allowing the reduction of the maintenance actions and the increase in the operational safety. However, the available methods often require a simplification of the vehicle through linearised models, a high number of sensors, or the use of complex algorithms that disregard the mechanical phenomena that explain the vehicle dynamics. This work suggests the Localized Transmissibility Damage Indicator (LTDI), based on the existing Transmissibility Damage Indicator (TDI), to detect damage in the springs of a locomotive, using pairs of sensors placed in the bogie frame and the axle boxes. For that purpose, multibody simulations are used to simulate the dynamic behaviour of the vehicle in tangent tracks under nominal and damaged conditions. The results from multibody simulations allow the calculation of the LTDI values for different levels of damage and various operation conditions, as well as the study of the effect of the variability inherent to the railway operation. The results show that the LTDI is significantly sensitive to damage. However, depending on the use of the lateral or vertical response, the LTDI is more suitable to detect the stiffness increase or decrease, or even to locate the damage. In conclusion, the LTDI is a promising method for the detection of damage on suspension elements of railway vehicles.

High-speed railway wheel polygon detection framework using improved frequency domain integration

Timely monitoring of wheel polygon is of great importance for the formulation of railway wheel maintenance strategies. In this study, a novel data-driven method for onboard and quantitative detection of wheel polygon is presented. First, the axle box acceleration (ABA) signal preprocessing method and stationarity test are introduced to select the relatively stationary signal from the measured data of ABA. Next, an iterative algorithm is developed to accurately extract the quasi-stationary ABA signals, representing each wheel rotation period. Then, an improved frequency domain integration method is developed to quantitatively capture the orders and roughness levels of the wheel polygon. Finally, the effectiveness and superiority of the proposed method is verified using the field-measured data of ABA and the wheel polygon in one cycle of wheel re-profiling. The results show that the proposed method can quantitatively capture the dominant characteristics of single- and multi-orders wheel polygons at different operating mileages with minimum and maximum absolute errors of 0.04 dB re 1 µm and −2.33 dB re 1 µm, respectively. The comparative analysis demonstrates that the proposed method outperforms the traditional time and frequency domain integration algorithms in the detailed characterisation of wheel polygon roughness levels.

A modelling methodology of the axle box bearing-vehicle coupled system dynamics

The axle box bearing serves as a critical rotational component for high-speed trains. Thus, it is of great significance to obtain accurate loading conditions for axle box bearings for its reliability design. In this study, a bearing-vehicle coupled system dynamic model, consisting of an equivalent bearing model characterized by the time-varying stiffness in the bearing, a nonlinear high-speed train dynamic model, and a slab track model, is developed. The validity of the proposed coupled dynamic model is demonstrated by comparing the simulated results with those obtained in a field test and the literature, respectively. Subsequently, the influence of long- and short-wavelength irregularities on the dynamic responses of bearings is investigated using the proposed coupled model in terms of dynamic loads of the axle box bearing. The results obtained by the proposed model are further compared with those calculated using a traditional vehicle dynamic model that represents the bearing as a revolute joint, so as to illustrate the advantages of the proposed model. The results suggest that the loading responses of the bearing calculated by the two models exhibit differences, which are simultaneously related to the structural damping of the bearing and the mode frequency change of the system.

Avoid Bogie Bearing Failure of IGBT Inverter Fed EMUs and Locomotives

Three current paths are proposed, and theoretical analysis and laboratory tests are carried out to investigate the root causes of bearing failure in IGBT inverter-fed locomotives and EMUs. The three types of current paths that run through the drive unit bearings and axle box bearings used on EMUs and electric locomotives are classified as the primary side current path, the main traction system current path, and the current path between the vehicles of the EMU or electric locomotive and the vehicles it hauls. The research found that the EDM current path in the main traction system caused by common mode voltage is distinguished as the main cause resulting in the failure of the bogie motor bearings or the bearings of the load connected to the motor shaft. The cause of common mode voltage is analyzed, and the thresholds of current density and voltage without causing bearing damage are analyzed and presented. The lab tests carried out on the bearings on the main traction system’s current path verified that the current path does exist. The proof to identify electric erosion, such as craters and washboards, and corresponding measures to prevent the failure of bogie bearings are proposed. Further research about the other two current paths is urgent and necessary.

Performance Degradation Assessment of Railway Axle Box Bearing Based on Combination of Denoising Features and Time Series Information.

In the existing rolling bearing performance degradation assessment methods, the input signal is usually mixed with a large amount of noise and is easily disturbed by the transfer path. The time information is usually ignored when the model processes the input signal, which affects the effect of bearing performance degradation assessment. To solve the above problems, an end-to-end performance degradation assessment model of railway axle box bearing based on a deep residual shrinkage network and a deep long short-term memory network (DRSN-LSTM) is proposed. The proposed model uses DRSN to extract local abstract features from the signal and denoises the signal to obtain the denoised feature vector, then uses deep LSTM to extract the time-series information of the signal. The healthy time-series signal of the rolling bearing is input into the DRSN-LSTM reconstruction model for training. Time-domain, frequency-domain, and time-frequency-domain features are extracted from the signal both before and after reconstruction to form a multi-domain features vector. The mean square error of the two feature vectors is used as the degradation indicator to implement the performance degradation assessment. Artificially induced defects and rolling bearings life accelerated fatigue test data verify that the proposed model is more sensitive to early failures than mathematical models, shallow networks or other deep learning models. The result is similar to the development trend of bearing failures.

Research on vibration fatigue and failure mechanism of the wire bracket for high-speed train

A wire bracket is a cantilever structure connected to the axle box of a high-speed train. Its force is complex and it is prone to vibration fatigue failure. Combined with fracture analysis and on-track tests, the local cracking of the wire bracket is studied and the mechanism and cause of structural failure are revealed. The results show that a fatigue crack in the wire bracket initiates from the incomplete penetration defect of the bolt hole weld of the cantilever side mounting seat and propagates to the upper cover plate and vertical plate. The joint action of wheel polygon and rail corrugation results in higher vibration energy of the wire bracket in 400–800 Hz. The welding defects reduce the fatigue strength of the weld, and the high-frequency and high-amplitude vibration caused by wheel-rail excitation makes the stress concentration near the weld, which brings the possibility of fatigue crack initiation and propagation of the wire bracket during the on-track operation.