Squat Defects Research Articles

Corrugation and Squat Classification and Detection with VGG16 and YOLOv5 Neural Network Models

Railway track defects in Malaysia pose significant risks of train derailments and accidents, underscoring the urgency for early and accurate defect detection and classification. This study presents a novel approach utilizing deep learning models, VGG16 and YOLOv5, for detecting and classifying railway track defects, explicitly focusing on corrugation and squat defects. The research's uniqueness lies in its application of these specific models and the composition of a dataset collected from extensive field measurements and inspections across various railway tracks within the Track Network Maintenance Ampang Line in Malaysia. The results demonstrate that these models achieve high precision in defect classification and detection of defects by more than 80%. The proposed methodology provides the railway industry with a powerful tool to streamline maintenance planning and prioritize defect remediation efficiently. Early defect detection can prevent potential accidents and improve safety and operational efficiency. Future studies can expand on these findings by exploring the extension of the proposed techniques to address other types of rail defects. Incorporating a diverse range of scenarios and operating conditions in the dataset could further enhance the models' performance and generalization. Real-time deployment and integration with existing maintenance systems are crucial for practical adoption. This research has strengths but acknowledges limitations. Additional evaluation metrics and a diverse dataset are essential for model performance. Leveraging deep learning models offers a reliable solution for railway maintenance, enhancing safety and efficiency. Addressing these limitations will drive proactive defect management, ensuring safe and reliable railway networks.

Expanded statistical analysis of squats on the Great Britain (GB) mainline network

Squat defects are one of the most common rail surface defects. Significant research effort has gone into understand squat defects over the last 10 years which has brought about important developments in the understanding of their initiation mechanism; however, further work is still required to fully understand squat and the best methods to control them.This study considers records of squat defects over a period 9 years, considering 2600 km of track across 8 different routes on the GB mainline network. The analysis separately reviews squats on: plainline, crossings, joints and welds. Results include an overview of the main factors influencing the development of each type of squats, practical methods to immediately reduce and manage squat defects and recommends focus areas for further research to understand squat defects.Results suggest that squats on plainline, crossings, joints and welds, all correlate with different influencing factors; headcheck defects appear to significantly influence the probability of squats and how other factors influence squat development.There is a strong connection between total head wear rate (combined material removal due to traffic and grinding) and squats; 90 % of all squats appear on rail with a headwear rate of <0.2 mm/year. Overall larger section rail (60 kg/m vs 56 kg/m) and harder material (260 Brinell vs 220 Brinell) is significantly less susceptible to squat damage. Track curvature has an influence of squat development, especially in rail with no headcheck cracking, where the tightest curves are significantly more likely to sustain squat damage. The probability of squat at vertical discontinuities, i.e. joints and crossings are significantly more likely as train speed increases. Whilst squats on joints are 1000 time more likely than squats on welds.

Microstructural investigation on a rail fracture failure associated with squat defects

Squats, observed in numerous railway networks worldwide, are a significant concern for rail track maintenance. This work presents a specific case study of a rail fracture failure associated with squat defects. A detailed microstructural analysis was conducted to investigate the causes of squat formation, subsequent crack propagation, transverse defects, and fracture. According to the findings, the initiation of squats can be attributed to the presence of White Etching Layers on the rail surface and rolling contact fatigue. After the initiation of Squats, the cracks extended downwards into the bulk rail at a shallow angle until reaching a sizeable martensitic island located in the subsurface of the rail along the path of crack propagation. The brittleness of the undesired martensitic structure triggered rapid crack development, facilitated transverse cracking, and ultimately resulted in rail fracture failure. Transmission electron microscopy examination confirmed that the abnormal defects are plate martensite with a twin substructure and high carbon content, which could be generated as a result of improper heat treatment during the manufacturing of rails.

Squat Detection of Railway Switches and Crossings Using Wavelets and Isolation Forest.

Railway switches and crossings (S&Cs) are critical, high-value assets in railway networks. A single failure of such an asset could result in severe network disturbance and considerable economical losses. Squats are common rail surface defects of S&Cs and need to be detected and estimated at an early stage to minimise maintenance costs and increase the reliability of S&Cs. For practicality, installation of wired or wireless sensors along the S&C may not be reliable due to the risk of damages of power and signal cables or sensors. To cope with these issues, this study presents a method for collecting and processing vibration data from an accelerometer installed at the point machine to extract features related to the squat defects of the S&C. An unsupervised anomaly-detection method using the isolation forest algorithm is applied to generate anomaly scores from the features. Important features are ranked and selected. This paper describes the procedure of parameter tuning and presents the achieved anomaly scores. The results show that the proposed method is effective and that the generated anomaly scores indicate the health status of an S&C regarding squat defects.

An in-depth assessment of convolutional neural networks for rail surface defect detection

The consistent monitoring of rails is based on correctly identifying defects to support corrective measures. Recently, convolutional neural networks (CNN), a deep learning method, have been providing outstanding results for the automatic detection of defects. However, several aspects of CNN-based approaches such as network architecture, transfer learning and processing time remains not fully understood. In this work, we performed an in-depth assessment of ten widely used CNN models with the objective of finding the one with the best performance in identifying defects in rail surface images. The classification results are promising, reaching an average accuracy of 83.7% on detection of mild defects and squat. The Inceptionv3 network provided the best results by correctly identifying 92% of images with severe squat defects.

A laser scanner based approach for identifying rail surface squat defects

The defect identification process within the UK rail industry has seen significant improvements over the past decade with the introduction of new measurement systems and defect detection systems. Although significant work has been on the defect identification little work has been done on the process after the defect has been detected. This repair process is still extremely manual. Due to the current process being manual the repair operation has very little traceability and transparency. This paper has therefore presented the need for not only a defect detection system but a defect repair system for the UK railway industry. Further to this, this paper has acknowledged that the rise of defects occurring on the UK railway lines requires a solution that can fully repair a defect with little to no user intervention in a timely manner. To address this, this paper has taken the extremely manual process of rail repair and has laid out the possibilities to automate this process. By doing this a work flow diagram has been generated to show how the system could be used to repair surface defects with a specific focus being made on squat defects. To achieve this a defect detection and measurement system has been explored, as this will make up the first stage of the automated repair system. The literature on various defect detection algorithms was reviewed and two variations of existing defect detection algorithms were created, i.e. the Covariance method and the Normal Intersection method. These algorithms have been tested against 100 simulated squat defects and have been verified using 4 experimentally generated defects. Both algorithms have been proven to not only identify the approximate size of the defect but also its location. This successful defect identification will be integrated into an automated rail repair system.

Metallurgical and tribological aspects for squat formation in the aluminothermic weld HAZ edges of rails welded using aluminothermy

PurposeThe purpose of this paper is to present the results of a metallurgical, mechanical and tribological characterization of the weld and heat-affected zone (HAZ) of aluminothermic welding of premium rails used in heavy haul, looking into the origins of the squat defects associated with rail wear.Design/methodology/approachA full factorial design of experiment was carried out for 24 welds of premium and super premium rails. The factors studied were chemical composition, welding gap and preheating time. The welds were inspected visually and by ultrasound to detect superficial and internal defects and characterized by macrographic analysis, hardness profile, tensile tests and microstructural characterization in scanning electronic microscopy. Pin-on-disk test were carried out to compare the tribological behavior of the different regions of the weld rail.FindingsSquat formation was shown to be associated with spheroidized pearlite regions formed on the HAZ of the welds, presenting near half the hardness of the weld metal. Thermal analysis showed that spheroidized pearlite is a result of partial austenitization at these positions. Tribological tests showed that low hardness regions presented smaller wear resistance than both the weld metal and the parent rail. Tensile test of the whole region resulted in brittle fracture along the weld metal.Originality/valueThe results showed that it is essential to reduce the dimensions of the HAZ and the width of the hardness drop area to mitigate squat formation in the HAZ edges.Peer reviewThe peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-01-2020-0020/

Understanding and treatment of squat defects in a railway network

A comprehensive investigation of the Sydney Trains railway network has been carried out to control and minimize rail degradation, especially squat defects. Squat occurrence was analysed with respect to the parameters of traffic operation and track geometry. Detailed metallurgical examination of the ex-service rails revealed a possible connection between brittle White Etching Layers (WELs) and the occurrence of squats in the running band of the rail. Initiation of squats was found to be associated with the progression of surface cracks in the WELs. Furthermore, two distinguishable types of WELs were found, based on different track operational conditions. These are described as Thermally Produced WEL (TP-WEL) which was found in heavy braking regions, and Severe Deformation WEL (SD-WEL) which was found in track regions under steady traffic speeds. In addition to the explanations based on thermal phase transformation and severe plastic deformation, the arcing phenomenon associated with electrical leakage at the wheel/rail interface was considered an alternative formation mechanism of WELs observed on rails. A revised rail grinding strategy was proposed by considering the formation of WELs and addressing it through more regular, relatively shallow grinding to effectively control squat defects.

The formation of martensite during the propagation of fatigue cracks in pearlitic rail steel

Squat is a kind of railhead defects related to rolling contact fatigue (RCF), and regarded as a potential safety hazard in railway networks. The mechanisms of squat initiation and propagation remain unclearly understood. In the current study, several branches of subsurface cracks inside squat defects had been intensively investigated for understanding the development of RCF cracks. From metallographic observations, a few white etching particles were clearly found either located within crack gaps or attached with crack faces. Based on nanoindentation measurements, the hardness of white particles was higher than that in matrix, presenting a possible mechanical property of martensite. A large density of nano-twins in the substructure of white particles can be identified in the transmission electron microscopy bright-field imaging. Their corresponding selected area electron diffraction patterns proved white particles as Fe-C martensite with body-centered tetragonal crystallographic structure. Besides the martensite, a random distribution of oxidized sulphide inclusions (MnS) and silicate composites were found included in white etching particles as well by the energy-dispersive x-ray spectroscopy analyses. The discovery of martensitic transformation and inclusions in the crack of squat defects reveals an accelerated failure mechanism which would be detrimental to operation safety. It is necessary to pay more attentions so that the risk of fatigue failures can be minimized.

Squat formation and the occurrence of two distinct classes of white etching layer on the surface of rail steel

Squats are recognized as an important rail track issue related to rolling contact fatigue, and represent a serious problem affecting significant parts of railway networks across the world. The mechanisms of squat initiation and propagation are still a matter of controversy. Research efforts have included understanding of possible relationships between the narrow and brittle white etching layers (WELs) observed on rail surfaces and squat formation in rails. In the present investigation, structural and microstructural features of ex-service damaged rails containing squats and WELs were investigated in order to determine the formation mechanism of white etching layers (WELs) on the rail surface. Synchrotron XRD analysis of rail surfaces, combined with microstructural and microhardness investigations, revealed two distinguishable types of WELs; one characterised by the presence of fine martensite and the other characterised by the presence both of retained austenite and martensite. WELs containing martensite were found to be associated with a deformed subsurface pearlite structure and it was concluded that this type of WEL is induced by severe plastic deformation at wheel-rail interface. In contrast, the WELs containing both of significant amounts of retained austenite and martensite, were associated with the presence of undeformed pearlite in the underlying structure and it was concluded that this type of WEL is either thermally or thermomechanically induced. Synchrotron XRD results obtained from regions of squat defects themselves are consistent to that of WELs containing austenite and martensite in regions near the squats, which indicates a relationship between WELs and squat initiations. Occurrences of cracks extending from WELs down into the rail in squat regions were confirmed by microstructural observations, which is consistent with the premise that cracks in squats originate from an extension of cracks in WELs.

Reply to Giannakos, K. Comment on: Toughness of Railroad Concrete Crossties with Holes and Web Openings. Infrastructures 2017, 2, 3

This paper responds to the discussion [1] by Dr. K. Giannakos over our technical note [2].[...]

Rolling contact fatigue: Spalling versus transverse fracture of rails

Rolling contact fatigue (RCF) defects in the running band of the rail may develop, as a function of born tonnage, either superficially and spall off, or penetrate into the subsurface. In practice, the first type is found to occur notably (but not exclusively) on heat-treated pearlitic rails. Both possibilities involve an essentially different operational risk with respect to transverse rail fracture and require therefore different inspection and maintenance regimes. This study presents a validated hypothesis that explains both similarities and differences of spalling defects and classical squat defects that develop also in depth. It is shown that their microstructural/-mechanical initiation mechanism is different and not necessarily governed by the local tangential stress history in the case of spalling. A model is set up and validated for subsurface crack propagation directivity, distinguishing a spalling and a transverse fracture domain for development of running band defects for both straight track and high and low legs of curves. This model allows for understanding and recognition of the nature of running band defects and for adjustment of control actions.

Integrated analysis of dynamic vehicle–track interaction and plasticity induced damage in the presence of squat defects

Despite significant efforts, the mechanisms behind the formation of squats – a form of rolling contact fatigue (RCF) damage – are not fully understood. This study employs numerical simulations to investigate the propensity of squat initiation in the vicinity of small, isolated rail surface irregularities. Time-domain dynamic vehicle–track interaction analysis is used to obtain wheel–rail contact stress distributions, which are mapped onto a continuum finite element model that accounts for plastic deformation of the rail material. The evaluated stress and strain fields are quantified using two RCF impact measures: accumulated effective strain and the Jiang–Sehitoglu multiaxial low cycle fatigue parameter. It is shown that the RCF impact increases with increasing size of the surface irregularity and that clustering of irregularities might strongly promote RCF. The friction coefficient is identified as a very influential parameter and also the effect of variations in friction along the rail is evaluated.

LASER SCATTEROMETRY FOR DETECTION OF SQUAT DEFECTS IN RAILWAY RAILS

The paper concerns a defectoscopy of squats occurring on the surface of the rail head. Disadvantages of traditional methods being used in detection of such defects were also presented. Authors proposed a new method based on simplified laser scatterometry to detect these defects. Physical model of the laser beam scattering by edges of such defects and simulation results were given. An experimental set-up for practical testing of this method was designed and constructed. This system enabled measurements of squat defects occurring in the section of the rail extracted from the track. The analysis of obtained results was performed. Authors also indicated directions of further research and development.

Dynamic Wheel–Rail Interaction Over Rail Squat Defects

Dynamic interaction between wheel and rail generates vibration and acoustic radiations transmitted through other media such as structural track systems, air, and rolling stock. Noteworthy, both wheel and rail irregularities often induce additional impact vibration and noise along the rail corridor. Rail squats and studs are typically classified as the development of any cracks that have grown longitudinally and laterally through the subsurface and become a laminated crack resulting in a depression of the rail surface. The rail defects are commonly referred to as ‘squats’ when they are initiated from rolling contact fatigue cracks, and as ‘studs’ when they are associated with a white etching layer due to wheel slips or excessive tractive effort. Interestingly, recent research shows that rail squats can be found almost everywhere, and every type of track structures, gradients and geometries in either light passenger or heavy freight traffics and for low, medium or high speed trains all over the world. A more pronounced issue appears when the crack grows and finally flakes off the rail. This paper highlights the monitoring of wheel–rail interaction over rail squat defects. Using the axle box acceleration data obtained from the calibrated track recording vehicle, the frequency spectra can be analysed. The results demonstrate an apparent influence of squats on the wheel–rail interaction over the defective sections and increased acoustic radiation at the high frequency range. The practical understanding will also enhance the development of acoustic- and vibration-based monitoring systems of rail surface defects.

Early stages of rail squat formation and the role of a white etching layer

A squat is a rolling contact fatigue crack below the surface of a rail that leads to a depression in the surface. Squats were thought to initiate by ratcheting (cyclic plastic shear deformation of the surface layer). However, the rail is often found to have a martensitic hard layer which refuses to etch (a white etching layer) on the surface, indicating a thermal crack initiation process. Grassie et al. [1] has suggested these cracks should be given a different name “studs” to distinguish their thermal origin. This paper describes detailed observations of cracks at a site where no white etching layer was evident. Hence these cracks can confidently be labelled as squats. Three stages of squat formation have been identified. Firstly, there is an early stage of crack initiation due to a white etching layer or ratcheting, and then growth of a subsurface crack leading to formation of a squat, caused by entrapped water inside the crack, and finally spalling off of a piece of rail surface. Rails with squat defects in different stages were removed from the site, investigated and inspected visually, ultrasonically, optically and by using electron microscopy to find the main cause of crack initiation and squat formation Microscopic observations revealed primary, secondary and tertiary cracks in the transverse section of rails, and differences in crack surface appearance. Primary cracks initiated at an angle of 10–15° with respect to the rail surface and then turned down at an angle of 50° with respect to the rail surface. Secondary cracks initiated at an angle of 160° with respect to the rail surface and further bifurcated in tertiary cracks. Microhardness of the damaged rail surface was higher than that of the adjacent rail matrix but was not comparable with that of a white etching layer or martensitic structure.

Study on track dynamic forces due to rail short-wavelength dip defects using rail vehicle-track dynamics simulations

A rail vehicle-track interaction dynamics model has been applied to determine the track vertical dynamic forces due to rail shortwavelength dip defects such as squat, dip joints and welds, etc., which are required in both rail vehicle acceptance procedure and track maintenance. The model is validated using the field measurement data of rail squat defects and accelerations on a vehicle axlebox. The simulated track dynamic forces — the P2 forces due to rail dipped joints have been compared with those calculated by using a well-known formula. The results are compared and the formula’s limitations have been discussed. The dependence of the track vertical dynamic forces on the rail dip defect size and vehicle speed has also been investigated.

Lock-in thermographic inspection of squats on rail steel head

The development of squat defects has become a major concern in numerous railway systems throughout the world. Infrared thermography is a relatively new non-destructive inspection technique used for a wide range of applications. However, it has not been used for rail squat detection. Lock-in thermography is a non-destructive inspection technique that utilizes an infrared camera to detect the thermal waves. A thermal image is produced, which displays the local thermal wave variation in phase or amplitude. In inhomogeneous materials, the amplitude and phase of the thermal wave carries information related to both the local thermal properties and the nature of the structure being inspected. By examining the infrared thermal signature of squat damage on the head of steel rails, it was possible to generate a relationship matching squat depth to thermal image phase angle, using appropriate experimental/numerical calibration. The results showed that with the additional data sets obtained from further experimental tests, the clarity of this relationship will be greatly improved to a level whereby infrared thermal contours can be directly translated into the precise subsurface behaviour of a squat.

Modelling of the lock-in thermography process through finite element method for estimating the rail squat defects

Squats are a major problem on the world railways. The non-destructive evaluation technique is becoming increasingly attractive in the detection of near surface defects on track. Non-destructive thermal evaluation is one such method of inspection technique that can be used for the detection of near surface defects. Its sub-group of lock-in thermography is under analysis. Lock-in thermography utilizes an infrared camera to detect the thermal waves and then produces a thermal image, which displays the local thermal wave variation in phase or amplitude. There are few studies into the actual experimental representation of complex subsurface defects when concerning lock-in thermography processes. While this may be less of a concern given the purpose of numerical defect characterization to reduce the need for extensive experimental pre-tests, the necessity for (artificial) representations of a defect will inevitably be required for validation. The research outlined in this paper examines the use of 3D finite element modelling (FEM) as a potential flexible tool in simulating the lock-in thermography process for detecting squats in track. In addition, lock-in analysis proved that the correct frequency range had to be selected for the material to detect the defect. As maximum positive and negative phase angles were located at “optimum” frequencies, at certain frequencies lead to minimal phase angle difference to which the defects were not detectable (blind frequency) by using the incorrect testing. The 3D finite element method has advantage for determining the “optimum” thermal excitation frequencies compare with experimental investigation. The experimental results show that 3D FEM models can be used to defect the location and the depth of squats in the railway.

Tribological characterization of rail squat defects

Rail squats remain a major rolling contact fatigue problem for mass transit authorities. Several studies have been conducted to supplement the standard characterization of rail squats. In particular an extensive tribological and metallurgical investigation was performed on samples of incipient squats from various areas. Surfaces and cross-sections of rail samples were observed by optical microscopy and scanning electron microscopy. These observations highlighted that the running band is composed of longitudinal contact strips with various surface and subsurface morphologies. This microstructure seems to be essential for the onset of squats. This new information makes it possible to account for the entire damage mechanism of a squat defect from initiation to propagation.