Rail Neutral Temperature Research Articles

Data-driven prediction of rail neutral temperature for continuously welded rails using impulse-based vibration frequencies

Continuously welded rails (CWR) are prone to the development of high thermal-induced load along the axial direction. Excessive levels of load lead to risk of rail buckling and potential for derailment. Knowledge of the in situ rail axial load in CWRs is therefore important to ensure safe rail management. Field-deployable, nondestructive evaluation techniques for measuring the rail load, or a widely adopted alternative called rail neutral temperature (RNT), are desired. This study uses a data-driven approach to investigate if rail dynamic response data, collected in a non-destructive fashion, can be used to predict RNT. The study is based on a data set comprising rail equivalent strain, temperature and vibration resonance frequencies that was collected from a revenue-service rail over a period of nearly two years. All excited vibration resonance peaks are identified from other peaks caused by noise using spectral amplitude variance. Among these resonance peaks, potentially useful resonances are identified with respect to stacked spectra collected across a testing day using an assumed temperature-frequency relation. A subset of the identified useful resonances is then identified based on their consistent appearance across both testing locations and all testing days, strong correlation to effective strain, and strong correlation to each other. Three particular vibration resonances (or vibration modes -- these terms will be used interchangeably throughout this paper unless specified otherwise. The term mode does not necessarily indicate mode shapes or mode families.) emerge from this process as best candidates. A classic feature selection technique, Lasso linear regression, is then employed to identify critical power combinations of the three resonant mode frequencies. Two power combinations exhibit unique correlation to the measured equivalent axial strain at both test locations across all testing days, and thus show particular ability to predict RNT. The RNT is predicted at one test location using different models based on the power combination data from the other location, and vice versa, where the predictions satisfy standard RNT measurement accuracy expectations.

Quantifying the effect of fixed structures on rail stress management: A numerical investigation

From 2012 to 2021, FRA accident data indicate that there have been 132 derailments caused by track buckles. Out of the 132 total derailments, 59 (or 45 %) occurred within 152 m (500 ft.) of a fixed structure. Therefore, this paper presents, validates, and applies a 3D finite element-based model to study the effect of the fixed structure on the rail gap size given a rail break. Specifically, the model was used to quantify the effect of the: 1) temperature differential between the rail neutral temperature (RNT) and the actual rail temperature (dT), 2) distance of the cut to the fixed structure, and 3) fixed structure length and anchoring patterns (ETA and EOTA). Model results indicated that at low dTs (e.g., <13.9ºC (25ºF)) the effect of the fixed structure can be considered if the cut occurred at the edge of the fixed structure with EOTA anchoring pattern and the fixed structure length of at least 18.3 m (60 ft.). At a dT of 22.2ºC (40ºF), the fixed structure should be accounted for when there is an edge cut. At extreme dTs (e.g., > 44.4ºC (80°F)) the fixed structure should be considered if the cut occurs within 152.3 m (500 ft.). Therefore, it is recommended that railroads refrain from strictly following open track guidelines for CWR (continuous welded rail) management and rather, consider the influence of fixed structures in relevant situations.

Predicting Axial Stress State In Continuously Welded Rail Using Impulse-Generated Vibration Measurements

Continuously welded rails are connected without stress relief joints and, thus, thermally induced rail movement is constrained, which can result in the development of excessive axial stress and risk of rail failure. Nondestructive testing (NDT) methods that estimate in-place rail stress state or rail neutral temperature are desired. Some methods have been developed, but none satisfy the requirements for ideal monitoring in practice. We propose an NDT technique based on impulse-generated vibration, seeking high-frequency rail vibration resonances whose frequency maintains a consistent correlation with rail axial stress/strain across different temperatures, stress states, and rail support conditions. Rail temperature, axial strain, and vibration data were collected from an active Class 1 commercial rail line over a period of nearly two years. The frequencies of four consistent and clear resonance modes of the rail were monitored. One of the identified modes demonstrates a unique linear relation with axial strain across a range of temperatures and stress states at each of the two measurement locations. The developed linear relations were used to predict in-place strain and rail neutral temperature with acceptable accuracy across all the measurement data, although each test location exhibits a unique relation.

Nondestructive Estimation of Neutral Temperature in Rails: A Comparative Study of Machine Learning Strategies

This paper presents the latest findings of a nondestructive evaluation technique currently under development at the University of Pittsburgh to determine the rail neutral temperature (RNT) in continuous welded rails. The technique is based on the extraction of relevant features from rail vibrations and the use of machine learning (ML) to associate these features to the longitudinal stress of the rail of interest. The features contain the spectral information of the vibrations and are pooled together by frequency domain decomposition for input to ML algorithms. Minimum redundancy–maximum relevance and neighboring component analysis are used to identify relevant features to reduce the size of the input vector. In addition, seven algorithms were considered to identify the most accurate model for neutral temperature with respect to the ground truth RNT measured with a strain-gage rosette. The data used in this study were collected from a 5° curved rail on concrete ties. The vibrations were triggered with a hammer and recorded with a few wireless and wired accelerometers attached on the railhead. The results showed that the Gaussian process regressor performs best, and as few as 20 frequencies can be used to predict the RNT with sufficient accuracy.

Birefringence Technique for Evaluating Thermal Stresses in Railroad Rails

This paper discusses the development of an in situ noncontact electromagnetic acoustic transducer (EMAT) nondestructive evaluation technology to determine rail neutral temperature and estimate rail stress in continuous welded rail (CWR). Stresses develop in CWR due to a lack of expansion joints to accommodate thermal expansion and contraction of the rail when ambient temperatures vary over time. The novelty of the work presented is the usage of ultrasonic birefringence properties using EMATs to estimate thermally induced stresses in rails. EMATs produce polarized shear waves propagating through the rail web in the pulse-echo mode. Experimental tests were performed on machined 136RE and 141RE rail material with applied compressive and tensile stresses to explore the stress-birefringence behavior. Two additional sets of experimental tests were conducted on full-size rail sections with in situ surface conditions to study variations in the in situ birefringence and the acoustic stress constant in different rail materials including 115RE rail, 119RE rail, two different 136RE rails, and 141RE rail. The results show a highly linear relationship between the stresses applied and the measured acoustic birefringence.

Effect of Critical Factors Influencing Longitudinal Track Resistance Leveraging Field Experimentation

As extreme temperatures increase, there is an increased need to ensure that continuous welded rail (CWR) stresses are more accurately set and better maintained. Therefore, to improve the management of CWR rail stresses and to increase safety through the reduction of buckled track derailments this paper documents the results from a field experimentation program of controlled single rail breaks (SRBs). SRBs were conducted at multiple field site locations to quantify the longitudinal resistance on both timber and concrete sleeper track. Accurate longitudinal resistance values are critical, because if the wrong value is selected, the rail neutral temperature could easily be set to a low value, increasing the probability of a track buckle or rail pull-apart, further compounding the impact of extreme temperatures. The results presented quantify the effect of sleeper and fastener type as well as ballast consolidation, and compare them with relevant results from the literature. From this, it was found that well-maintained concrete sleeper track exhibited a longitudinal resistance greater than the currently recommended value. Further, well-maintained timber sleeper track with anchors on every other sleeper exhibited a longitudinal resistance 44% lower than concrete sleeper track and 30% lower than timber track with anchors installed on every sleeper. Additionally, disturbing the ballast reduced the longitudinal resistance by 15%. Finally, the results indicated that slip was primarily occurring at the rail–sleeper interface for unanchored timber sleepers and at the sleeper–ballast interface for anchored timber sleepers and concrete sleepers.

Machine Learning for the Nondestructive Prediction of Neutral Temperature in Continuous Welded Rails

ABSTRACT This paper describes the application of machine learning (ML) in the framework of a data-driven nondestructive evaluation (NDE) method to estimate the rail neutral temperature (RNT) of continuous welded rails (CWR). The method consists of triggering vibration of the rail of interest and extracting the power spectral densities (PSDs) of the accelerations associated with the lowest modes of vibration. The PSDs then become the input of an ML algorithm trained to associate the PSD to longitudinal stress and then RNT. In the study presented in this article, the proposed NDE method was tested on a tangent track on wood cross-ties. Vibrations were induced with a hammer and detected with several wireless and wired accelerometers. The PSDs across the 0–700 Hz range were extracted from the time-series. These densities in both the lateral and vertical directions constituted part of the input of an artificial neural network trained and tested with experimental data. The predicted neutral temperatures showed very good agreement with the RNT estimated by an independent party and based on conventional strain-gage rosettes.

Nondestructive rail neutral temperature estimation based on low-frequency vibrations and machine learning

This paper describes the latest advancements of a non-invasive inspection/monitoring technique to estimate the rail neutral temperature (RNT), which is the temperature at which the net longitudinal force in continuous welded rails is zero. The technique is based on the measurement of the lowest modes of vibration of the rail of interest, and the application of machine learning (ML) to predict the RNT from the spectral densities of the vibrations’ frequencies. In the study presented in this article, the proposed technique was tested twice in the field. Vibrations were induced with an instrumented hammer on a 5° curved rail on concrete ties, and detected with a few wireless and wired accelerometers. The time-series were post-processed to extract the power spectral densities across the 0–700 Hz range. These densities constituted the input vector of a ML algorithm trained with experimental data and then tested with different experimental data. The RNT predicted by the algorithm were compared and showed very good agreement with the temperature estimated by an independent party and based on conventional strain-gage rosettes.

Review of Critical Factors Influencing Longitudinal Track Resistance

Longitudinal track resistance is one of the most critical parameters required to accurately analyze longitudinal load propagation and refine rail neutral temperature (i.e., stress-free temperature) maintenance practices. However, longitudinal track resistance has not been consistently defined and has been quantified using multiple methods that should not be conflated. This paper documents common definitions of longitudinal track resistance and the two methods regularly used for its quantification: track panel pull test (TPPT) and single rail break (SRB). Further, this paper documents the differences in mechanics between these methods, summarizes and discusses the critical factors influencing the longitudinal track resistance values found in the literature, and adds novel TPPT longitudinal track resistance values for timber sleeper track to address the current scarcity of data found in the literature. In summary, TPPT values provide insight into the mechanics of load propagation and pre-buckle analysis while the SRB values aid in the maintenance and restoration of rail neutral temperature after a rail break or destress. Additionally, the TPPT longitudinal track resistance values reported in the literature were independent of panel length, were influenced most by the presence of a vertical load, and were reduced by 25% when the ballast was disturbed. Finally, novel TPPT results indicated the longitudinal resistance of timber sleeper track was 19% lower than concrete sleeper track and that unfastened sleepers still transferred longitudinal load when the cribs were full and ballast was compacted.

Stress Evaluation in Rails Based on Vibration Data and Artificial Intelligence

An inspection technique for the estimation of longitudinal stress and rail neutral temperature (RNT) in continuous welded rails (CWR) is presented. The technique is based on the use of finite element analysis (FEA), vibration measurements, and machine learning (ML). The FEA is used to establish the dependency of the vibration characteristics (mode shapes and frequencies) of any given CWR to the conditions of the track and the longitudinal stress of the rail. The dependency is then used to train an ML algorithm that is tested using experimental data obtained with two accelerometers bonded to the rail of interest, during a field test conducted in Colorado, U.S. A commercial finite element (FE) software was used to model the rail track as a short segment repeated indefinitely and under varying boundary conditions and stress. Three ML models were developed using hyperparameter search optimization techniques and k-fold cross-validation to infer the stress or RNT from the frequencies of vibration extracted from the time waveforms recorded by the accelerometers. The results demonstrated that the success of the technique is highly dependent on the accuracy of the FE model. The results also demonstrated that ML was mostly able to learn from the experimental data and successfully predict the neutral temperature.

A review of non-destructive evaluation techniques for axial thermal stress and neutral temperature measurement in rail: Physical phenomena and performance assessment

Thermal buckling of continuous welded rail (CWR) has been a long-standing challenge for the railroad industry because of the high derailment rate and the associated social, economic, and environmental impacts it causes. Rail buckling is generally attributed to excessive thermally-induced axial compressive stress developed in the rail from high temperatures. Knowing the rail thermal stress or its rail neutral temperature (RNT) is critical for safe and efficient rail system operation. There has been great interest and much work on the development of nondestructive evaluation (NDE) techniques to estimate rail thermal stress and RNT in situ. This paper reviews the findings and conclusions from research about NDE approaches for estimating rail thermal stress or RNT, emphasizing the physical phenomena and performance interpretation related to each of the approaches. We identify the type of reference measurements each technique relies on and tabulate this information showing key assumptions, performance, and limitations for each technique.

Vibration-Based Approach to Measure Rail Stress: Modeling and First Field Test.

This paper describes a non-invasive inspection technique for the estimation of longitudinal stress in continuous welded rails (CWR) to infer the rail neutral temperature (RNT), i.e., the temperature at which the net longitudinal force in the rail is zero. The technique is based on the use of finite element method (FEM), vibration measurements, and machine learning (ML). FEM is used to model the relationship between the boundary conditions and the longitudinal stress of any given CWR to the vibration characteristics (mode shapes and frequencies) of the rail. The results of the numerical analysis are used to train a ML algorithm that is then tested using field data obtained by an array of accelerometers polled on the track of interest. In the study presented in this article, the proposed technique was proven in the field during an experimental campaign conducted in Colorado. A commercial FEM software was used to model the rail track as a short rail segment repeated indefinitely and under varying boundary conditions and stress. Three datasets were prepared and fed to ML models developed using hyperparameter search optimization techniques and k-fold cross validation to infer the stress or the RNT. The frequencies of vibration were extracted from the time waveforms obtained from two accelerometers temporarily attached to the rail. The results of the experiments demonstrated that the success of the technique is dependent on the accuracy of the model and the ability to properly identify the modeshapes. The results also proved that the ML was also able to predict successfully the neutral temperature of the tested rail by using only a limited number of experimental data for the training.

Quantification of longitudinal fastener stiffness and the effect on fastening system loading demand

Over the past 20 years, there have been at least 10 derailments due to spike fastener fatigue failures in North America. These fatigue failures have been considered a moderate to severe challenge that require manual walking inspections that are both time and labor intensive. These fatigue failures have been found to result from spike overloading due to lateral and longitudinal loads. To date, there has been limited quantification of the vertical, lateral, and longitudinal fastener forces in track. This paper quantifies the effect of fastener type on fastener load to account for various track types and locations. Laboratory experimentation was performed to quantify the stiffness of multiple fastening systems and this data was input into a previously validated analytical model to quantify the effect of stiffness on fastener loading. Additional laboratory experimentation was performed to quantify the relationships between both fastening system type and vertical loading and spike strain. While the laboratory data indicate a significant variance in stiffness between fastening systems, the model results indicate that the load transferred to the fastening system is less sensitive. However, spike strain data indicate the load path was affected by fastener type and vertical load. The characterization of longitudinal stiffness of multiple fastening systems and the relationship to spike load as presented can be used to advance track mechanistic-empirical design and improve rail neutral temperature prediction and track buckling models.

A non-contacting system for rail neutral temperature and stress measurements: Concept development

Continuous welded rail has become the standard in modern railway track construction around the world because it alleviates well-documented disadvantages of rail joints in a track. Continuous welded rail practice results in long segments of continuous rail in track that will develop significant thermal longitudinal stresses due to the absence of expansion joints. Before a continuous welded rail is laid, the rail is free of thermal stresses; the temperature at that time is known as the rail neutral temperature. The design rail neutral temperature is calculated based on local climate projections. As a continuous welded rail is laid, it may be stretched or compressed if the current temperature is not within the calculated design rail neutral temperature range, prior to anchoring the rail down. Upon anchoring, as temperatures deviate from the rail neutral temperature, significant tensile or compressive longitudinal stresses develop, leading to a track buckling or rail pull-apart that compromise the integrity of the track and the safety of train operation. Existing methods to estimate the rail neutral temperature and determine the state of stress in the rail have significant shortcomings related to the ease of implementation, system complexity, practicality, reliability, simplicity, cost, and instrumentation demands. We propose a novel concept for measuring stress in rail segments and determining the rail neutral temperature. The proposed method is based on measurements of nonuniform deformations of the rail under thermal loading, as observed in computer simulations and laboratory investigations. The implementation uses thermal imaging and three-dimensional stereo-digital image correlation technology to acquire full-field deformations. The acquired data are processed to estimate rail neutral temperature and quantify the longitudinal stress in the rail. This article presents the analytical and experimental work that led to the conception of the method and introduces the systematic approach to develop the method along with verification and validation studies.

Rail neutral temperature monitoring using non-contact photoluminescence Piezospectroscopy: A field study at high-speed rail track

A novel non-contact stress sensing method using the photoluminescence piezospectroscopy (PLPS) method was applied in full-scale field test at high-speed rail tracks. The fluorescence spectra of the aluminum oxide (α-Al2O3) were collected from different locations on bare thermite weld surface for about 1.5 days. Prominent α-Al2O3 signals (i.e., R1/R2 peaks) were observed with the very short measurement time of 0.01 s, which is preferable in mobile sensing applications. The kernel spectrum estimation (KSE) method was developed as a postprocessing technique to determine R1/R2 peak locations accurately. The R1/R2 time histories showed strong correlation with those of strain gauges mounted on the rail near the thermite welds. The piezospectroscopy coefficients were determined at −30.01 cm−1/GPa for R1 and –32.11 cm−1/GPa for R2, calibrated with the strain gauge measurements. Prior to the field test, a laboratory test was conducted to determine the R1/R2 zero-stress states using a pulverized thermite weld sample. The zero-stress state was determined at 14,404.99 cm−1 for R1 and 14,434.09 cm−1 for R2. Using these parameters, rail neutral temperature was determined at full-scale high-speed tracks based on the absolute rail stress measured with PLPS without additional field calibrations.

Nondestructive measurement of neutral temperature in continuous welded rails by nonlinear ultrasonic guided waves.

Modern rail construction uses continuous-welded rail (CWR). The presence of very few joints leads to an increasing concern due to the large longitudinal loads caused by restrained thermal expansion and contraction, following seasonal temperature variations. The knowledge of the current state of thermal stress in the rail or, equivalently, the rail neutral temperature (corresponding to zero net longitudinal force) is a key need within the railroad transportation community in order to properly schedule slow-order mandates and prevent derailments. This paper presents a nondestructive diagnostic system for measurement of the neutral temperature in CWR based on nonlinear ultrasonic guided waves. The theoretical part of the study involved the development of a constitutive model in order to explain the origin of nonlinear effects arising in complex waveguides under constrained thermal expansion. A numerical framework has been implemented to predict internal resonance conditions of nonlinear waves in complex waveguides. This theoretical/numerical phase has led to the development of an experimental prototype (Rail-NT) that was tested both in the laboratory and in the field. The results of these experimental tests are also summarized.

Development of a unified railway track stability management tool to enhance track safety

Track buckling is a serious problem for railways. High longitudinal rail stresses contribute to problems such as track buckling, rail joint failure, rail breakage and failure of turnouts. The direct and indirect costs of track buckling problems are very high. The influences of rail temperature, stress-free temperature (SFT) and lateral misalignment of track on track buckling need comprehensive investigation. In this paper, an experimental design comprising strain gauges, thermocouples and rail stress sensors has been implemented on the Queensland Rail heavy haul 60 kg/m rail network. A new creep measurement technique using internal rail stress has been developed. The changes in rail neutral temperature due to the variation of actual rail temperature and the occurrence of rail creep in straight and curved track is quantified. Modes of differences of SFT in the two rails at a location, and of SFT in straight track and in curved track are discussed. The relationship of SFT to rail temperature is also presented. Daily variation in rail temperature due to ambient air temperature is presented. Field trials showed that SFT can vary by 2–3 ℃ during the day. Based on this finding and the derivation of an equation for change of SFT, an improvement in utilising rail creep measurements for assessing track condition has resulted. This finding suggests that it is possible to determine the SFT throughout a day rather than just a single SFT value. This paper also presents a simple track stability management tool that is based on two major parameters, namely rail stress and track resistance. Each parameter in the tool has been given three levels of value to determine the required preventive measures. Overall, the tool decides the need for speed restriction during hot weather based on the quantified parameters from the field trials and rail standards.

Stress Dependence of Ultrasonic Guided Waves in Rails

Most modern railways use continuous welded rails (CWRs). A major problem in these structures is the almost total absence of expansion joints, which can create severe issues such as buckling in hot weather and breakage or pulling apart in cold weather. To minimize these risks, CWRs are built by connecting track segments that are prestressed before welding. A related critical parameter is the rail neutral temperature (NT), which is defined as the temperature at which the net longitudinal force in the rail is zero. When the ambient temperature is higher or lower than the NT, the rail is under compression and tension, respectively. Knowledge of the NT provides a potential method for indirect measurement of the stress and load in the rail. Unfortunately, the measurement of the in situ stress (or NT) has been a long-standing challenge for railway owners and operators. This paper presents numerical results on the dynamic behavior of CWRs subjected to a static axial stress. The results show how ultrasonic guided waves are sensitive to variations in stress and could potentially be used to estimate the stress level or the NT in rails. The present work represents the initial concept phase of a research and development study funded by the FRA. The ultimate objective of this study is to develop and test a prototype system that uses noncontact dynamic sensing to measure in situ rail stress in motion at speeds up to 30 mph to determine rail NTs and the related incipient buckling risks in CWRs.

Improved Destressing of Continuous Welded Rail for Better Management of Rail Neutral Temperature

Maintaining high, stable rail neutral temperatures helps prevent the buckling of continuous welded rail (CWR) track. Rail neutral temperatures are typically set high during installation (90°F to 110°F), but the large variations that develop during revenue service often lead to buckling-prone conditions. Readjusting or correcting for these variations requires CWR to be destressed with the use of procedures that do not always restore the desired target neutral temperature. As part of the Federal Railroad Administration's Track Systems Research program, the U.S. Department of Transportation's Volpe Center is investigating rail force and neutral temperature influences on track buckling. An analytic model for field applications has been developed to improve destressing and readjustment of CWR in both winter and summer conditions. The model has been validated in several field tests on instrumented CWR test segments under both high tensile and compressive force conditions. Both wood and concrete tie tracks were ...

Improved Destressing of Continuous Welded Rail for Better Management of Rail Neutral Temperature

Maintaining high, stable rail neutral temperatures helps prevent the buckling of continuous welded rail (CWR) track. Rail neutral temperatures are typically set high during installation (90°F to 110°F), but the large variations that develop during revenue service often lead to buckling-prone conditions. Readjusting or correcting for these variations requires CWR to be destressed with the use of procedures that do not always restore the desired target neutral temperature. As part of the Federal Railroad Administration's Track Systems Research program, the U.S. Department of Transportation's Volpe Center is investigating rail force and neutral temperature influences on track buckling. An analytic model for field applications has been developed to improve destressing and readjustment of CWR in both winter and summer conditions. The model has been validated in several field tests on instrumented CWR test segments under both high tensile and compressive force conditions. Both wood and concrete tie tracks were tested, and the rail longitudinal movement, rail gap, rail force distributions after rail cutting and welding, and readjusted neutral temperature were measured and correlated with the model predictions. The model and test results were used to develop a field tool for more effective destressing and readjustment of CWR. The tool provides the required removal lengths of anchors/fasteners, the rail gap size requirements when mechanical loads (rail-pullers) are used to adjust to the desired neutral temperature, and the required amounts of steel removal in summer when cutting rail out for stress relief.