Rolling stock, particularly of freight railroads, is currently maintained using regular preventative and corrective maintenance schedules. This maintenance approach recommends sets of inspections and maintenance procedures based on the average expected wear and tear across their inventory. In practice, however, this approach to scheduling preventative maintenance is not always effective. When scheduled too soon, it results in a loss of operating revenue, whereas when it is scheduled too late, equipment failure could lead to costly and disastrous derailments. Instead, proactive maintenance scheduling based on Big Data Analytics (BDA) could be utilized to replace traditional scheduling, resulting in optimized maintenance cycles for higher train safety, availability, and reliability. BDA could also be used to discover patterns and relationships that lead to train failures, identify manufacturer reliability concerns, and help validate the effectiveness of operational improvements. In this work, we introduce a train inventory simulation platform that enables the modelling of different train components such as wheels, brakes, axles, and bearings. The simulator accounts for the wear and tear in each component and generates a comprehensive data set suitable for BDA that can be used to evaluate the effectiveness of different BDA approaches in discerning patterns and extracting knowledge from the data. It provides the basis for showing that BDA algorithms such as Random Forest [9] and Linear Regression can be utilized to create models for proactive train maintenance scheduling. We also show the capability of BDA to detect hidden patterns and to predict failure of train components with high accuracy.

To illustrate the metro braking performance more appropriately and comprehensively, and to explore the dynamic braking process more accurately, the consistency analysis was used in this paper, which is used for the first time in the metro brake field. Based on the stream data, several characteristic values (braking pressure setting time, maximum braking pressure and stability value of braking pressure, i.e.) were extracted to represent the dynamic process. Under similar braking conditions, such as the similar applied pressure, speed and brake level, etc., the performance of brake system is similar and stable in certain extent. The dynamic data model has been developed in this paper. Large data mining based statistics was used to analyze braking performance and the dynamic prediction during a braking cycle was provided. Investigation presented in this paper contributes to the efforts in the direction of metro braking systems performance and evaluation of service condition. The dynamic evaluation based on the stream data has been proved to provide a new way to explore the influence of metro braking system so that it is reasonable to analyze the service condition of metro brake system. Based on the LabVIEW platform, the system for analyzing stream data was built. Based on stream data, using the characteristic and statistics method, this paper provides a way to describe the service condition. The result enriches the evaluation criteria of metro service condition.

The Next Generation Passenger Brake Equipment Development (RTDF2013-4732) presented the background and development process to be followed in the application of the Association of American Railroads (AAR) S-4200 Electronically Controlled Pneumatic (ECP) brake standards as applied to Federal Railroad Administration (FRA) Tier I passenger equipment. During the ensuing five-year period, significant work has been performed to allow an ECP-equipped passenger train to enter an FRA revenue service demonstration on the Amtrak Keystone Service between Harrisburg, PA, and New York, NY. This paper summarizes the information contained in the 2013 paper and addresses the actual implementation path that was followed. The paper discusses 26C pneumatic brake characterization testing required for emulation operation, performance standard differences for passenger equipment, and the safety analysis performed. The paper covers interoperability testing, equipment modifications, static train testing, and dynamic train testing that were required to enter the revenue service demonstration phase. Finally, emulation service status, the revenue service demonstration status, American Public Transportation Association (APTA) Passenger Rail Equipment Safety Standards (PRESS) issuance, and Code of Federal Regulation 49CFR238 proposed changes are addressed.

Although accidents at Highway-Rail Grade Crossings (HRGCs) have been greatly reduced over the past decades, they continue to be a major problem for the rail industry, causing injuries, loss of life, and loss of revenue. Recently, the Strategic Highway Research Program sponsored a Naturalistic Driving Study, the SHRP2 NDS, which produced a unique opportunity to look at how drivers behave while traversing HRGCs. This research deviates from previous studies by concentrating on day-to-day actions of drivers who traverse the HRGCs without an incident, instead of focusing on the accident events that have formed the foundation most earlier studies. This paper will focus on the effects of the external environment, weather and day/night conditions, on driver behavior at HRGCs. We will present the methodology and data used for the study and provide some early results from the analysis, such as differences in compliance during poor versus clear weather. We will use both a compliance score based on scanning and speed reduction and an analysis of brake and gas pedal usage during the approach to a HRGC. The paper will conclude with a brief discussion of future research concepts.

Wayside hot-box detectors (HBDs) are devices that are currently used to monitor bearing, axle, and brake temperatures as a way of assessing railcar component health and to indicate any possible overheating or abnormal operating conditions. Conventional hot-box detectors are set to alarm whenever a bearing is operating at a temperature that is 94.4°C (170°F) above ambient, or when there is a 52.8°C (95°F) temperature difference between two bearings that share an axle. These detectors are placed adjacent to the railway and utilize an infrared sensor in order to obtain temperature measurements. Bearings that trigger HBDs or display temperature trending behavior are removed from service for disassembly and inspection. Upon teardown, bearings that do not exhibit any discernible defects are labeled as “non-verified”. The latter may be due to the many factors that can affect the measurement of HBDs such as location of the infrared sensor and the class of the bearing among other environmental factors. A field test was performed along a route that is more than 483 km (300 mi) of track containing 21 wayside hot-box detectors. Two freight cars, one fully-loaded and one empty, and one instrumentation car pulled by a locomotive were used in this field test. A total of 16 bearings (14 Class F and 2 Class K) were instrumented with K-type bayonet thermocouples to provide continuous temperature measurement. The data collected from this field test were used to perform a systematic study in which the HBD IR sensor data were compared directly to the onboard thermocouple data. The analyses determined that, in general, HBDs tend to overestimate Class K bearing temperatures more frequently than Class F bearing temperatures. Additionally, the temperatures of some bearings were underestimated by as much as 47°C (85°F). Furthermore, the HBD data exhibited some false trending events that were not seen in temperature histories recorded by the bayonet thermocouples. The findings from the field test suggest that HBDs may inaccurately report bearing temperatures, which may contribute to the increased percentage of non-verified bearing removals. To further investigate the accuracy of the wayside detection systems, a dynamic test rig was designed and fabricated by the University Transportation Center for Railway Safety (UTCRS) research team at the University of Texas Rio Grande Valley (UTRGV). A mobile infrared sensor was developed and installed on the dynamic tester in order to mimic the measurement behavior of a HBD. The infrared temperature measurements were compared to contact thermocouple and bayonet temperature measurements taken on the bearing cup surface. The laboratory-acquired data were compared to actual field test data, and the analysis reveals that the trends are in close agreement. The large majority of temperature measurements taken using the IR sensor have been underestimated with a similar distribution to that of the data collected by the HBDs in field service.

The Train Energy and Dynamics Simulator (TEDS) is state-of-the-art simulation software, developed by the Federal Railroad Administration (FRA), to study train operation safety and performance as affected by a wide variety of rolling stock, track, train handling and operating configurations. As part of developing TEDS, existing and published data on braking, draft systems and train performance were used for initial validation of TEDS. This paper describes two revenue service tests conducted to further validate TEDS. The first test was on a loaded unit train, while the second test was on a mixed train with empty and loaded cars and included distributed power in which the remote brake valve was cut in. Collected test data included throttle position, train speed, locomotive power, brake system pressures and coupler forces. Several events from these tests, representing typical train operating scenarios, were selected for comparison with TEDS predicted results. The TEDS predictions matched the measured test data for all of the scenarios simulated, further validating the performance of the software and offering additional assurance on the use of TEDS for simulating performance and safety critical train dynamic behavior.

IEEE 1698, Guidelines for Safe Braking Distance Calculations - 2009 brakes down the determination of stopping distance for train control system into separate independent events. One of the most misunderstood portions of the model is Parts I and H (train deceleration). This paper looks at the reasoning behind the unique properties of these parts, how they are related as well as how to calculate them. Examples of applications will be shown and optimization techniques will be investigated to shorten the required safe stopping distance and increase capacity.

Crude oil and ethanol unit train derailments sometimes result in the release of large volumes of flammable liquids which ignite and endanger the safety of persons, property, and the environment. Current methods to reduce the probability and mitigate the consequences of High-Hazard Flammable Train (HHFT) derailments include operational speed constraints, enhanced tank car design/build requirements, improved car and track inspection and maintenance, and use of advanced braking systems. The train brake system can dissipate more energy in a derailment scenario if the brake signal propagation rate is increased, the brake force against the wheel tread is increased, or a combined approach is used. This paper describes a simplified energy conservation model used to determine the emergency braking stopping distance and energy dissipation benefits available for three advanced train braking systems. A 3×3 matrix of brake configurations was defined by three brake signal propagation rates and three car net braking ratio (NBR) values. The brake signal propagation rate was modeled for trains with conventional head-end locomotive power, pneumatic car braking, and no two-way end-of-train device (CONV); locomotive distributed power with pneumatic car braking (trailing DP); and locomotive power with electronically-controlled pneumatic (ECP) braking. Car NBR values of 10, 12.8, and 14 percent were selected to reflect the expected brake force range available from older equipment in the existing tank car fleet (10% NBR) to the maximum acceptable value for new or rebuilt cars (14% NBR). Various in-train emergency brake application scenarios for loaded unit trains were modeled while accounting for the gross effects of derailment/collision blockage forces. Empirical data from four trailing distributed power train derailment events were used to estimate an average derailment/collision blockage force (ADF) and simulate the trailing consist braking performance. The ADF results were subsequently used in a more general tank car unit train parametric study to evaluate the effects of train speed, track grade, and in-train derailment position for each brake configuration in the matrix. The simplified energy conservation model was used to 1) quantify the number of trailing consist cars expected to stop short of the derailment location and 2) compare the car-by-car energy state of each car in the trailing consist that was calculated to reach the derailment location. Results for the empirical and parametric study cases are compared graphically and observations are discussed relative to two assumed baseline brake configurations.

Two major components of rolling stock that are always of great interest when it comes to maintenance and safety related issues are car wheels and bearings. Rail car wheels are subjected to a variety of damage types due to their interaction with the track and brakes. It is important for the rail industry to detect these defects and take proper action at an early stage, before more damage can be caused to the train or possibly the track and to prevent possible safety hazards. Different inspection sensors and systems, such as wheel impact monitors, wheel profile detectors, hotbox detectors and acoustic detection technologies, are employed to detect different types of wheel and bearing defects. Usually no single sensor can accurately detect all kinds of damages and hence a combination of different sensors and systems and manual inspection by experts is used for wheel maintenance purposes and to guarantee train safety. The more complete and accurate the automatic defect detections are, the less manual examination is necessary, leading to potential savings in inspection time/resources and rail car maintenance costs. Wayside thermal and visible spectrum cameras are one option for the automatic wheel and bearing inspection. Each of these sensors has their own strengths and weaknesses. There are some types of defects that are not detectable at an early stage in the images taken by a vision camera, however these defects generate a distinctive heat pattern on the wheel or bearing that is clearly visible in the thermal imagery. On the other hand, other damages might be detectable from the visible spectrum image, but not necessarily have a distinguishable heat pattern in the thermal imagery. Since a thermal image is basically built of solely temperature data, it excludes other critical information, such as texture or color. This makes thermal and visible spectrum imagery complementary and if the images are fused the result will benefit from the strengths of both sensors. In this paper, wavelet decomposition is employed to extract the features of the thermal and vision imagery. Then the two images are merged based on their decompositions and a fused image is composed. The resulting fused image contains more information than each individual image and can be used as an input for image-based wheel and bearing defect detection algorithms. To verify the proposed method and to show an example of this application, it is demonstrated on a real data set from a Union Pacific rail line to identify sliding wheels.

Since the adoption of 286,000 lb gross rail load (286K GRL) car service, an increase in wheel thermal damage and shelling has been observed. This is attributed to the increased braking horsepower in 286K GRL service as compared to the 263K GRL service environment. This study investigated possible designs and methods of braking that could lead to reduced heat input to the tread of freight car wheels in order to mitigate this damage and reduce its occurrence to a level closer to that seen with 263K GRL car service. Fifteen potential concepts to lower the thermal input to wheels and/or accelerate heat removal from the tread were identified and evaluated using the following engineering categories: simplicity of design, maintenance requirements, weight considerations, material and manufacturing costs, controllability of braking effort, and market acceptability. Five final concepts — axle-mounted disc, cheek disc, wheel rim, axle-mounted drum, and high convection coating — were developed through preliminary design and thermal analysis to confirm their effectiveness in meeting the objectives. Four concepts for alternative braking methods — axle-mounted disc brakes, cheek disc brakes, wheel rim clasp brakes, and axle-mounted drum brakes — were analyzed in considerable detail. Of the four concepts presented, the first three appear to be feasible and would be potential candidates for further detailed investigations/evaluation. It is shown that as the demand on railway wheels to withstand increased mechanical and thermal loads grows, there are viable braking enhancements that can help manage the stress state in freight car wheels.

Train car wheels are subjected to different types of damages due to their interactions with the brake shoes and track. If not detected early, these defects can worsen, possibly causing damage to the bogie and rail. In the worst-case scenario, this rail damage can possibly lead to later derailments, a serious concern for the rail industry. Therefore, automatic inspection and detection of wheel defects are high priority research areas. An automatic detection system not only can prevent train and rail damage, but also can reduce operating costs as an alternative for tedious and expensive manned inspection. The main contribution of this paper is to develop a computer vision method for automatically detecting the defects of rail car wheels using a wayside thermal camera. We concentrate on identification of flat-spotted/sliding wheels, which is an important issue for both wheel and suspension hardware and also rail and track structure. Flat spots occur when a wheel locks up and slides while the vehicle is still moving. As a consequence, this process heats up local areas on the metal wheel, which can be observed and potentially detected in thermal imagery. Excessive heat buildup at the flat spot will eventually lead to additional wheel and possibly rail damage, reducing the life of other train wheels and suspension components, such as bearings. Furthermore, as a byproduct of our algorithm, we propose a method for detecting hot bearings. A major part of our proposed hot bearing detection algorithm is common with our sliding wheel detection algorithm. In this paper, we first propose an automatic detection and segmentation method that identifies the wheel and bearing portion of the image. We then develop a computer vision method, using Histogram of Oriented Gradients to extract features of these regions. These feature descriptors are input to a Support Vector Machine classifier, a fast classifier with a good detection rate, which can detect abnormalities in the wheel. We demonstrate our methods on several real data sets taken on a Union Pacific rail line, identifying sliding wheels and hot bearings in these images.

Critical derailment incidents associated with crude oil and ethanol transport have led to a renewed focus on improving the performance of tank cars against the potential for puncture under derailment conditions. Proposed strategies for improving puncture performance have included design changes to tank cars as well as operational considerations, such as reduced speeds and upgraded brake systems. In a prior paper on this topic, the authors conceptualized a novel and objective methodology for quantifying and characterizing the reductions in risk that result from changes to tank car design or to the tank car operating environment. This paper describes an extension of that effort to include additional derailment cases, additional operating speeds, considerations for alternate train configurations, such as Distributed Power (DP) and Electrically Controlled Pneumatic (ECP) brakes, as well as options for component level studies. In essence, the developed methodology considers key elements that are relevant to tank car derailment performance and combines these elements into a consistent probabilistic framework to estimate the relative merit of proposed mitigation strategies. The relevant elements considered include variations in the derailment scenarios, chaotic derailment dynamics, the distribution of impact loads and impactor sizes, various operating speeds, brake system differences, and variations in tank car design. The paper also provides an overview of the validation efforts which suggest that the gross dynamics of a tank car train derailment, and the resulting puncture performance of the tank cars, are captured well by this methodology.

Brake heating of Railroad Wheels has been known to accelerate shelling, soften the wheel tread and contribute to stress reversal which is a requisite for generating thermal cracks. The effect brake heating has on shakedown has been investigated to a lesser extent, yet even here brake heating can be very important. This research combines field investigations and laboratory work in an attempt to quantify the thermal effects on wheel performance, especially as it relates to sub-surface shelling, microstructural changes and hardening mechanisms. Cyclic tensile tests were conducted at three temperatures to show the relationship between deformation at elevated temperatures and hardening response. Ductility measurements from both monotonic and cyclic tests were used to estimate residual stresses.

The Los Angeles County Metropolitan Transportation Authority (LA METRO) subway provides service with up to six-car trains at up to 65 mph at five minute headways on weekdays. To reduce energy usage, LA METRO implemented a flywheel-based Wayside Energy Storage Substation (WESS), which reduces energy usage by capturing and reusing braking energy generated by trains when they decelerate and brake approaching the passenger station. The LA METRO WESS has a 2 MW, 15 second capacity, or 8.33 kWh, and can charge and discharge with a 1.5 minute cycle rate. WESS has been in daily full operation since August 2014. WESS has saved 10 to 18% of the traction power energy at the Westlake TPSS, every day.

Longitudinal dynamics influence several measures of train performance, including schedules and energy efficiency, stopping distances, run-in/run-out forces, etc. Therefore, an effective set of tools for studying longitudinal dynamics is essential to improving the safety and performance of train operations. Train Energy and Dynamics Simulator (TEDS) is a state-of-the-art software program designed and developed by the Federal Railroad Administration (FRA), for studying and simulating train safety and performance, and can be used for modeling train performance under a wide variety of equipment, track, and operating configurations [1]. Several case studies and real-world applications of TEDS, including the investigation of multiple train make-up and train handling related derailments, a study of train stopping distances, evaluations of the safety benefits of Electronically Controlled Pneumatic (ECP) brakes, Distributed Power operations, and a study of alternate train handling methodologies are described in this paper. These studies demonstrate the effectiveness of using the appropriate simulation tools to quantify and enhance a better understanding of train dynamics, and the resultant safety benefits.

Friction force oscillations caused by changes in the properties of the contact zone between brake disc and pad are well known from various applications. Resulting effects like brake judder are known phenomena in brake technologies and in the scope of various scientific work. A new potential measure to reduce brake torque oscillations is the active compensation with the use of the control system of a self-energizing hydraulic brake (SEHB). The SEHB for railway application developed at the Institute for Fluid Power Drives and Controls (IFAS) of RWTH Aachen University offers high dynamic properties with its capability to reduce brake torque oscillations actively. New — in comparison to other braking systems — is the fact that the brake torque is measured by sensing the pressure in an additional supporting cylinder. Within this paper the influence of the hydraulic-mechanical system of the supporting cylinder on the oscillation properties of the SEHB is analyzed. The experimental investigation is conducted using a full scale brake test for railway applications. The brake disc is driven by hydraulic motors in secondary control mode. Measurement results will be used to define requirements of a superimposed dynamic pressure control to minimize brake torque oscillations. Future work will be experimental investigation of the influence of the self-energizing effect and the development of measures to compensate brake torque oscillations actively with the hydraulic actuator of the SEHB.

Wireless Sensor Networks have been a focus of research in the North American freight railroad industry to enable on-board real-time sensing of critical railcar parameters. Important railcar aspects like wheel bearing temperature, air pressure, brake failure, and the integrity of transported goods can then be monitored closely and reliably. This enables immediate preventive actions in case of impending failures and also enables trend analysis that can be used to fine-tune maintenance efforts on railcars. These measures increase the safety, efficiency, and dependability of freight railroad operations. In our previous work [1–3] we have presented our Hybrid Technology Networking (HTN) protocol. This protocol provides optimal network performance for railcar monitoring applications. We have also presented HTNMote, a hardware prototyping platform that implements HTN. A deployment of HTNMotes was conducted and evaluated at the TTCI facility in Pueblo, Colorado in the US. The results from our field tests confirm that this approach is an order of magnitude better in performance compared to solutions based on ZigBee alone. In such an application, energy considerations represent a key challenge. These sensors have no readily available continuous energy source, but are expected to operate for years in harsh conditions. Energy harvesting — from vibrations, temperature differences, or solar radiation — may provide a potential solution to the energy scarcity. This also mandates that the HTNMote hardware and HTN protocol both be as energy efficient as possible. In this paper we present detailed measurements of the energy consumed by the HTNMote in various operational situations that are encountered during their operation onboard freight railcars. We introduce an energy consumption model based on our analysis of the measurements. This model demonstrates the energy-efficiency of the HTNMote implementation.

Railway wheels and rails do not achieve full wear life expectancy due to the combination of wear, plastic deformation, and surface, subsurface, and deep subsurface cracks. Sixty-seven percent of wheel replacement and maintenance in North America is associated with tread damage [1]. Spalling and shelling are the two major types of wheel tread damage observed in railroad operations. Spalling and slid flat defects occur due to skidded or sliding wheels caused by, in general, unreleased brakes. Tread shelling (surface or shallow subsurface fatigue) occurs due to cyclic normal and traction loads that can generate rolling contact fatigue (RCF). Shelling comprises about half of tread damage related wheel replacement and maintenance. The annual problem size associated with wheel tread RCF is estimated to be in the tens of millions of dollars. The total cost includes maintenance, replacement, train delays and fuel consumption. To study the conditions under which RCF damage accumulates, a 36-ton axle load aluminum body coal car was instrumented with a high accuracy instrumented wheelset (IWS), an unmanned data acquisition (UDAC) system, and a GPS receiver. This railcar was sent to coal service between a coal mine and power plant, and traveled approximately 1,300 miles in the fully loaded condition on each trip. Longitudinal, lateral, and vertical wheel-rail forces were recorded continuously during four loaded trips over the same route using the same railcar and instrumentation. The first two trips were conducted with non-steering 3-piece trucks and the last two trips were conducted with passive steering M-976 compliant trucks to allow comparison of the wheel load environment and RCF accumulation between the truck types. RCF initiation predictions were made using “Shakedown Theory” [2]. Conducting two trips with each set of trucks allowed for analysis of the effects of imbalance speed conditions (cant deficiency or excess cant) at some curves on which the operating speeds varied significantly between trips.

Worldwide, railways are among the safest transportation services. Nevertheless, every year some serious accidents are reported. A noticeable portion of these accidents are a result of defective wheels, bearings, or brakes. Train wheels are subjected to different types of damage due to their interaction with the brakes and the track and they are required to be periodically inspected to ensure they meet all the safety criteria for proper operation. If the wheel damage remains undetected, it can worsen and result in overheating and severe damage to the wheel and track. There are a variety of wheel damages, classified in different groups based on the type and severity of the defect. The most usual cause of damage is severe braking, which applies directly to the wheel and results in local heating of the wheel. This can stop the wheel from rotating while the train is still moving, producing a defect called a “flat spot” or “hot spot”. Flat-spotted wheels are a serious concern for the railroad industry. Depending on the level of wheel defect, different solutions should be taken. This paper will focus on automatically detecting flat-spotted wheels from thermal imagery using computer vision methods and introduces an algorithm to detect hot bearings. We first extract and segment both the wheel and bearing regions from the whole image, then we introduce a fuzzy inference to categorize the level of wheel damage. The whole process is done automatically and without any need for time consuming and costly manned inspection. Based on the severity of the defect, it can be decided which solution should be taken.

The main objective in optimizing train control is to eliminate the waste associated with the use of “worst case” assumptions to calculate Safe Braking Distances (SBD). Worst case takes the most conservative approach to the determination of train stopping distance in order to provide adequate stopping distance under virtually all conditions. This leads to stopping distances that could be far more than actually required under the circumstances at the time the train is attempting to brake. Several factors are considered in SBD; however one variable that influences a great portion of this distance is adhesion. This paper investigates adhesion to illustrate this influence on SBD and uses empirical data from Light Rail Systems with various test conditions. The interaction of the train control and other vehicle borne systems with SBD calculation will also be investigated including slip/slide control and brake assurance.