Wheel Flats Research Articles

M-CLUSTER: multistage clustering for unsupervised train wheel condition monitoring

Wheel out-of-roundness (OOR) is a common wheel defect that raises maintenance costs and increases the risk of failure or damage to track components. This paper proposes a novel multistage clustering framework (M-CLUSTER) for unsupervised condition monitoring of train wheels. The framework initially extracts time-domain features from raw acceleration responses collected by rail-mounted sensors. Sensitive features are then selected through an unsupervised feature selection algorithm called local learning-based clustering (LLC). Next, a detector model is trained using density-based spatial clustering of applications with noise (DBSCAN), a data clustering method effective for clusters with similar density. Since this algorithm does not originally involve separate training and testing phases, a new two-step mechanism is introduced: (1) training on a healthy dataset and (2) testing on an unlabelled dataset. Finally, the severity of train wheel defects is classified by K-means, with cluster validity indices (CVI) automatically determining the number of severity clusters (classes). The framework’s efficiency is demonstrated through the detection of defective wheels using the Alfa Pendular passenger model. Results indicate that M-CLUSTER accurately identifies train wheel flats and polygonal wear without labelled data, achieving 98% accuracy by selecting 10 features from the set.

Effects of Wheel Flats on Vibration and Stress Response of the Wheel-Rail System in High-Speed Trains

With the advancement of high-speed and lightweight vehicles, the occurrence of wheel tread wear, including wheel flats, has escalated. The resulting impact loads on axle-end components and the dynamic response of the axle cannot be overlooked. On-track tests were conducted on wheel flats to obtain wheel-rail forces, axle box vibration accelerations, and axle stresses during service. Time–frequency analysis was employed to uncover patterns of variation with speed. Additionally, a dynamic model of the high-speed train–track system was developed to replicate test conditions, investigating the characteristics of wheel–rail loads and structural elastic vibrations under high-frequency excitation for various flat sizes and higher operating speeds. The findings reveal that under wheel flats, the time-domain response of the wheel–rail system exhibits periodic impacts, while the frequency-domain shows high-frequency energy comprising multiple harmonics of the wheel rotation frequency. Wheel–rail forces, axle box vibration accelerations, and axle stresses exhibit inflection points with increasing speed, particularly within the 120–160 km/h range. In the high-frequency domain, axle stresses display characteristic frequencies that remain unchanged with speed, attributed to the coupled vertical bending vibration modes of the wheel–rail system. These research results offer foundational support for predicting fatigue damage to axle-end components, assessing axle stress, and establishing tolerances for tread defect limits.

Blockchain-enabled Data Mining for Pre-Condition Monitoring in Railway Infrastructure

Early detection of abnormalities in railway systems is crucial for risk mitigation and cost-effective maintenance. Damaged wheels in freight trains cause a significant threat to transport infrastructure integrity, particularly when wheel flats occur. Monitoring wheel conditions relies on sensing mechanisms, requiring robust data integrity checks. This paper proposes a data-driven approach utilising blockchain for pre-condition monitoring to enhance data reliability. To achieve this goal, a data mining procedure is developed to establish a local blockchain. The blockchain continuously expands with new blocks, with each block linked to the previous one via a hash function. Only validated data are recorded in order to enhance the reliability of train measurements.

The Impact of Wheel Flat on Traction Drive System of Electric Locomotives

Wheel flat can seriously affect the wheel–rail contact performance and cause impact vibrations, which significantly threaten the safe operation of railway vehicles. It is crucial to accurately assess the effects of wheel flats on the wheel–rail contact performance and the vibration impact on the traction drive system. However, studies related to wheel flats have focused on the mechanical part and have yet to fully consider the electrical and mechanical parts of the entire traction drive system. Therefore, this paper first builds an electromechanical coupling model based on the electric traction drive system and the locomotive-track coupling dynamics model. Then, based on this model, the impact of the wheel flat on the electrical and mechanical parts of the traction drive system under different flat lengths, different flat depths, and different vehicle speeds was analyzed. The results indicate that with the increasing depth of the flat spot, the wheel–rail dynamic response, motor rotor speed, and motor torque exhibit more significant fluctuations. Additionally, as the locomotive speed increases, the impact of the flat on the wheel–rail contact performance intensifies. Wheel flats can excite the 1st bending mode of the wheelset, resulting in vibration shocks at 90 Hz.

Early identification of out-of-roundness damage wheels in railway freight vehicles using a wayside system and a stacked sparse autoencoder

Early identification of wheel defects can prevent serious damage to railways, considerably lowering maintenance costs for both railway administrations and rolling stock operators. Within this context, an unsupervised methodology based on artificial intelligence techniques is presented, which allows the detection and classification of out-of-roundness damage wheels, such as wheel flats and polygonal wheels, based on dynamic responses induced on the track by crossing freight railway vehicles. The methodology involves the following steps: (i) data collection and pre-processing, (ii) feature extraction (iii) data fusion and (iv) feature discrimination. In the first phase, an FFT algorithm is applied to the acceleration track responses. Then, features are extracted after training a Stacked Sparse Autoencoder, in which the main features of the responses are obtained after a compression stage using an encoder network. This lower dimensional layer forces the model to learn a compression of the input data. Then, these extracted features are merged using the Mahalanobis distance, which enhances the sensitivity to the damage recognition. Posteriorly, an outlier analysis is performed to distinguish a healthy wheel from a defective one and a cluster analysis to discriminate the two types of out-of-roundness (OOR) damage and classify the severity of each type of damage.

Wheel flat detection by using the angular domain synchronous averaging method and axle box acceleration: Simulation and experiment

Wheel flats (WFs) can generate periodic impact forces and aggravate the wheel-rail interaction. Therefore, the early detection of WFs is significant for railway operators to lower the maintenance cost. In this work, the angular domain synchronous averaging (ADSA) method is employed to detect WFs on railway vehicles since it can handle the non-stationary vibration process of rotating machines by processing the vibration data in the angular domain. Furthermore, the short-time irregular impulses, such as those caused by crossings and rail joints, can be attenuated by using synchronous averaging technology. The periodic impact caused by WFs can be easily observed in the polar coordinate using the ADSA method, which is verified by one multibody system (MBS) simulation and one field experiment. Besides, the sensitivity of this method to various flat dimensions and multiple flats is also studied in the MBS simulation.

Dynamic analysis of axle box bearings on the high-speed train caused by wheel-rail excitation

To explore the impact of wheel-rail excitation on the dynamic performance of axle box bearings, a dynamic model of the high-speed train including axle box bearings is developed. Subsequently, the dynamic response characteristics of the axle box bearing are examined. The investigation focuses on the acceleration characteristics of bearing vibration under excitation of track irregularities and wheel flats. In addition, experiments on both normal and faulty bearings are conducted separately, and the correctness of the model and some conclusions are verified. According to the research, track irregularity is unfavorable for bearing fault detection based on resonance demodulation. Under the same speed conditions, the acceleration peak of bearing is inversely proportional to the length of the wheel flat and directly proportional to its depth. The paper will contribute to a deeper understanding of the dynamic performance of axle box bearings.

Wheel-Rail Stick-Slip Characteristics under Coupling Damages of Rail Corrugation and Wheel Flat in Metro

To investigate wheel-rail contact stick-slip characteristics (SSCs) under coupling damages of rail corrugation and wheel flat and explain the co-action mechanism of the two damages, a vehicle-track system model was built, and the SSCs were analyzed by employing the index of the overall dispersion rate of adhesion coefficient (ODRAC). The results show that both inner and outer wheel-rail systems (WRSs) on the curve have a propensity for stick-slip motion. Since the values of SSCs of the inner WRS are 1.204 to 2.963 times higher than those of the outer one, the inner WRS is easier to stick-slip vibration (SSV). When the rail corrugation is fixed, the overall dispersion rates of adhesion coefficients (ODRACs) progressively decrease with the increase of flat depth and length, and the decrease range is 2.766–41.775%. When the wheel flat is fixed, the ODRACs also progressively decrease as the corrugation wavelength increases, and the decrease ranges from 7.129% to 42.474%; the ODRACs incrementally increase together with the increase of the corrugation wave depth, and the increase ranges from 10.030% to 30.380%. The presence of flat will attenuate the level of SSV of the original system with rail corrugation, whilst the presence of corrugation will intensify the level of SSV of the original system with wheel flat.

A Methodology of Condition Monitoring System Utilizing Supervised and Semi-Supervised Learning in Railway.

In this paper, research was conducted on anomaly detection of wheel flats. In the railway sector, conducting tests with actual railway vehicles is challenging due to safety concerns for passengers and maintenance issues as it is a public industry. Therefore, dynamics software was utilized. Next, STFT (short-time Fourier transform) was performed to create spectrogram images. In the case of railway vehicles, control, monitoring, and communication are performed through TCMS, but complex analysis and data processing are difficult because there are no devices such as GPUs. Furthermore, there are memory limitations. Therefore, in this paper, the relatively lightweight models LeNet-5, ResNet-20, and MobileNet-V3 were selected for deep learning experiments. At this time, the LeNet-5 and MobileNet-V3 models were modified from the basic architecture. Since railway vehicles are given preventive maintenance, it is difficult to obtain fault data. Therefore, semi-supervised learning was also performed. At this time, the Deep One Class Classification paper was referenced. The evaluation results indicated that the modified LeNet-5 and MobileNet-V3 models achieved approximately 97% and 96% accuracy, respectively. At this point, the LeNet-5 model showed a training time of 12 min faster than the MobileNet-V3 model. In addition, the semi-supervised learning results showed a significant outcome of approximately 94% accuracy when considering the railway maintenance environment. In conclusion, considering the railway vehicle maintenance environment and device specifications, it was inferred that the relatively simple and lightweight LeNet-5 model can be effectively utilized while using small images.

Operational Wheel Flat Detector in Railway Vehicles

Maintenance of railway systems is shifting from being based on scheduled interventions to a continuous regime based on the actual status of assets. This change is supported mainly on three pillars: the development of new sensors and signal processing techniques, the capability to store and analyze all the information gathered by this huge amount of new sensors, and the capability of modifying dynamically the maintenance plans. This paper presents a new wayside system for detecting flats whose development has been based on combining physical models with Machine Learning Techniques. Physical models are used to understand the phenomena, define the key indicators to characterize the phenomena and generate synthetic data to train Machine Learning algorithms. Subsequently, regression models are generated to relate the key parameters with the flat severity. The last part of the paper is focused on validating the proposed methodology in a real environment.

Fatigue life assessment of high-speed train’s bogie frame due to dynamic loads under the influence of wheel flat

A bogie frame is the most important railway vehicle’s component that supports almost the weight of the system, and it is subjected to various excitations and loads that can result in the failure of the structure. The dynamic forces influenced by various excitations, both internal and external, can speed up the system's fatigue failure. This paper aims to analyse the impact of wheel flats on the life of a high-speed bogie frame. To accomplish this study, a Chinese railway high-speed (CRH2) bogie frame, has been chosen for the analysis. The required input data for CRH2 have been collected from literature; the wheel flat defect models under the wheel flat lengths of 20 mm, 40 mm, and 60 mm have been developed with a MATLAB tool, and the results have been inputted into a multibody model with the SIMPACK tool. The dynamic response loads for each case have been exported and applied to the finite element model in ANSYS, where stresses and strain have been used to estimate the fatigue life. The results from the study show that as the wheel flat length increases, it increases the dynamic forces and stresses that decrease the life of the bogie frame; thus, the influence of wheel flat should be considered in the life prediction of a bogie frame.

Wheel flat detection on railway vehicles using the angular domain synchronous averaging method: an experimental study

Recently, composite brake blocks have been widely adopted in European railway systems to replace cast iron brake blocks in order to reduce braking noise. However, the poor heat dissipation of composite materials has caused more wheel tread damage, including wheel flats (WFs). The WF, as an undesirable tread defect in railway vehicles, can cause severe wheel/rail impact and increase the damage probability of wheelset components and track structures. Early detection of WFs is vital to ensure operational safety and reduced maintenance costs. In this work, to help understand the dynamic impact caused by WFs, a single WF with a length of 20 mm is artifically produced on a wheel of a tank wagon, and field experiments are performed. The influence of the WF on the axle box accelerations (ABAs), vertical accelerations of the bogie frame and vertical accelerations of the car body is discussed in the time and time-frequency domains. On the other hand, the angular domain synchronous averaging (ADSA) method is used to detect WFs by averaging the ABA signals in the angular domain over a fixed time interval, and it requires ABA data and the axle rotational speed as inputs. This method can greatly reduce the amount of measured data by averaging it in the angle domain and can eliminate the short-time influence of interference signals caused by switches and crossings, rail joints, and so on. The results show that the ADSA of a defective wheel with the WF has a clear peak with an amplitude of 21–104 m/s2 while that of a healthy wheel fluctuates around zero in the polar coordinate. Also, the waveform of defective signals can be well presented in polar coordinates. The effectiveness of this method is compared with the envelope spectrum technology in three scenarios and its application to three journeys is presented.

Recent Advances in Wayside Railway Wheel Flat Detection Techniques: A Review.

Wheel flats are amongst the most common local surface defect in railway wheels, which can result in repetitive high wheel-rail contact forces and thus lead to rapid deterioration and possible failure of wheels and rails if not detected at an early stage. The timely and accurate detection of wheel flats is of great significance to ensure the safety of train operation and reduce maintenance costs. In recent years, with the increase of train speed and load capacity, wheel flat detection is facing greater challenges. This paper focuses on the review of wheel flat detection techniques and flat signal processing methods based on wayside deployment in recent years. Commonly used wheel flat detection methods, including sound-based methods, image-based methods, and stress-based methods are introduced and summarized. The advantages and disadvantages of these methods are discussed and concluded. In addition, the flat signal processing methods corresponding to different wheel flat detection techniques are also summarized and discussed. According to the review, we believe that the development direction of the wheel flat detection system is gradually moving towards device simplification, multi-sensor fusion, high algorithm accuracy, and operational intelligence. With continuous development of machine learning algorithms and constant perfection of railway databases, wheel flat detection based on machine learning algorithms will be the development trend in the future.

Out-of-round railway wheels and polygonisation

Out-of-roundness in railway wheels, in particular polygonal wear resulting in regular, multi-lobed, out-of-round wheels, has become a significant problem in recent years. It is of concern to railway operators due to the increased noise and vibration it can cause. This polygonisation is caused by cyclic wear but the exact mechanism leading to this type of wear is not fully understood. It appears to be the result of dynamic linking between a resonance or other periodic excitation in the coupled vehicle–track system and the existing wear at the wheel. This paper reviews the developing body of research being carried out in many countries by research groups, manufacturers and operators. Some examples of polygonisation on different types of railway vehicles are reported including feight trains, urban transit trains and high-speed trains. The main theories for the formation mechanisms are presented and the current measurement methods, computer simulation techniques and the effects and potential mitigation methods are reviewed. In parallel, the mechanisms and consequences of discrete wheel tread irregularities, such as wheel flats and material fall-out due to rolling contact fatigue cracking, are addressed.

Friction and vibration behaviours of high-speed train brake system excited by wheel flats

ABSTRACT To study the interface friction and vibration behaviours of brake systems with wheel flat, a rigid-flexible vehicle dynamics model and a finite element model of brake system are developed. These two models are validated using experimental test data. Additionally, an analysis formula of brake system excited by wheel flat is proposed to reveal the tribological behaviours. The results indicate that the wheel flat directly induces the elastic vibration of brake disc and further affects the disc contact, resulting in a significant increase in the contact force, contact stress, friction force and vibration acceleration between the brake interfaces. Furthermore, the proposed model can be used to investigate the related tribological behaviour of brake systems under other complex wheel-rail excitations.

Wheel Out-of-Roundness Detection Using an Envelope Spectrum Analysis.

This paper aims to detect railway vehicle wheel flats and polygonized wheels using an envelope spectrum analysis. First, a brief explanation of railway vehicle wheel problems is presented, focusing particularly on wheel flats and polygonal wheels. Then, three types of wheel flat profiles and three periodic out-of-roundness (OOR) harmonic order ranges for the polygonal wheels are evaluated in the simulations, along with analyses implemented using only healthy wheels for comparison. Moreover, the simulation implements track irregularity profiles modelled based on the US Federal Railroad Administration (FRA). From the numerical calculations, the dynamic responses of several strain gauges (SGs) and accelerometer sensors located on the rail between sleepers are evaluated. Regarding defective wheels, only the right wheel of the first wheelset is considered as a defective wheel, but the detection methodology works for various damaged wheels located in any position. The results from the application of the methodology show that the envelope spectrum analysis successfully distinguishes a healthy wheel from a defective one.

An Unsupervised Learning Approach for Wayside Train Wheel Flat Detection.

One of the most common types of wheel damage is flats which can cause high maintenance costs and enhance the probability of failure and damage to the track components. This study aims to compare the performance of four feature extraction methods, namely, auto-regressive (AR), auto-regressive exogenous (ARX), principal component analysis (PCA), and continuous wavelet transform (CWT) capable of automatically distinguishing a defective wheel from a healthy one. The rail acceleration for the passage of freight vehicles is used as a reference measurement to perform this study which comprises four steps: (i) feature extraction from acquired responses using the specific feature extraction methods; (ii) feature normalization based on a latent variable method; (iii) data fusion to enhance the sensitivity to recognize defective wheels; and (iv) damage detection by performing an outlier analysis. The results of this research show that AR and ARX extraction methods are more efficient techniques than CWT and PCA for wheel flat damage detection. Furthermore, in almost every feature, a single sensor on the rail is sufficient to identify a defective wheel. Additionally, AR and ARX methods demonstrated the potential to distinguish a defective wheel on the left and right sides. Lastly, the ARX method demonstrated robustness to detect the wheel flat with accelerometers placed only in the sleepers.

Towards Wayside Wheel Flat Detection and Classification Based on Psychoacoustic Quantities

Somewhat similar to the acoustic problems that were uncovered with the replacement of combustion engines with electric motors in cars, the retrofitting of all DB Cargo freight wagons with low-noise brakes caused other disturbing noise emissions to become more prominent. Among these, the periodic beating sounds originating from flattened wheel treads attract most attention. As there is no sufficiently validated solution for wayside wheel flat detection which meets the requirements of precise detection, focus on acoustic relevance and economic feasibility, the German Federal Environmental Agency has initiated a research project to determine an acoustic maintenance criterion for flat spots. In this contribution, we present the results of psychoacoustic analyses performed on a dataset comprising audio recordings of 200 cargo train passings. Regions with perceptible flat spots are labelled by trained ears. Common acoustic and psychoacoustic quantities are calculated for signal windows of classes "flat spot" and "no flat spot" and statistical analyses are performed to find quantities which qualify best for a separation of both classes. The effects of window size and sample rate are examined. Based on the results of listening tests, an acoustic classification criterion for the annoyance of wheel flats is proposed.

Auralisation of combined mitigation measures in railway pass-by noise

To reduce noise exposure along railway lines various combinations of noise mitigation measures can be considered. However, predicting and assessing their effects is non-trivial and the potential need for multiple measures is difficult to communicate to stakeholders. Auralisation is a promising tool that can help to support communication and decision-making, and enable psychoacoustic evaluations. This paper presents developments of a physics-based auralisation model for train pass-bys that allows various mitigation measures to be included. The work is conducted within the European research project SILVARSTAR. The proposed model includes contribution from rolling noise, impact noise, traction, auxiliary systems, and aerodynamic noise. It is physically based and allows a direct assessment of pass-by parameters such as speed, roughness, wheel flats and track design. Based on the TWINS model, five structural transfer paths for rolling noise are considered to integrate mitigation measures such as wheel and rail dampers. Shielding by noise barriers is simulated with analytical models. Reflection at different ground types is considered and can account for track embankments. The results can be coupled to an immersive Virtual Reality environment, by first panning the synthesised sounds to a small virtual speaker array and subsequently dynamic binaural rendering for headphones.

Experimental detection of train wheel defects using wayside vibration signal processing

While many studies have been used onboard monitoring methods to detect train wheel defects, this paper aims to extract damage features from vibrational signals obtained by the wayside monitoring system. To provide an appropriate tool for the decomposition of experimental nonstationary vibration signals containing impacts, an analytical amplitude-based signal decomposition method is provided. A two-axle motor car with a flat wheel defect is studied at different operating conditions. Vibrational signals arising from wheel-rail interactions are gathered by the Wayside Data Collection Setup (WDCS) with five piezoelectric accelerometers. The tests are performed at the constant speeds of 10, 20, 30, and 40 km/h. The acquired signals are decomposed by the Multi-channel Singular Spectrum Analysis (MSSA) into the contributing components. Based on the results, increases in the amplitudes of the second reconstructed components (RCs) of the vibrational signals sensed by all the accelerometers can be observed at a specific time instance as a repeating pattern. Similar patterns are obtained at different motor car positions relative to the measurement area, as well as different motor car speeds. To verify the consistency of the second RCs, the crest factors are obtained for all the angular positions. The results show sudden increases in the crest factor at a fixed angular position that indicate the existence of a defect at that specific angular position of the wheel.