LOCATION BASED RAILWAY ANOMALY DETECTION

Abstract

Systems and methods for detecting train rail and railcar anomalies are disclosed herein. In an example, detecting anomalies includes: receiving measurements from a sensor array coupled to a railcar in a train; obtaining baseline measurements from the sensor array; obtaining, in near real time, measurements from the sensor array while the railcar is operating; and detecting a railcar anomaly based a comparison between the baseline and operating measurements. In an example, the comparison of baseline and operating measurements includes evaluating, over a sequence of time data points, inertia sensor measurements (such as caused by side to side railcar movement) to detect abnormal railcar oscillation. In further examples, the data indexed to a GPS location is stored in a database, and respective alerts are transmitted or outputted to an output device (such as a display device) when an anomaly is detected based on the collected measurements from the railcars.

Description

TECHNICAL FIELD

Embodiments described herein generally relate to railway monitoring and more particularly, to detecting anomalies in a railway system occurring either with the railcars or on the rails of the track.

BACKGROUND

The rail industry has 140,000 of miles of track in the US. The trains that run on these tracks may be made of 100 railcars each, with each railcar possibly weighing over 120 tons. Carrying large amounts of weight over so many miles of track may lead to many problems with both the railcars and the rails of the track. Monitoring for anomalies on the rails and the railcars is a necessary task for continuous operation and prevention of train derailments.

BRIEF DESCRIPTION OF THE DRAWINGS

In the drawings, which are not necessarily drawn to scale, like numerals may describe similar components in different views. Like numerals having different letter suffixes may represent different instances of similar components. The drawings illustrate generally, by way of example, but not by way of limitation, various embodiments discussed in the present document.

FIG. 1 is an example of an environment and system for location based railway anomaly detection, according to an embodiment.

FIG. 2 is a block of an example of an anomaly detection system for location based railway anomaly detection, according to an embodiment.

FIG. 3A is an illustration of an example of a level rail track.

FIG. 3B is an illustration of an example of an out of level rail track.

FIG. 4 is an illustration of an example of rail warp.

FIG. 5A is an illustration of an example of a normal rail connected to a tie.

FIG. 5B is an illustration of a example of a rail that has become separated from the tie.

FIG. 6 is an illustration of an example of forces that may occur when a truck has a stuck center pin.

FIG. 7 is an illustration of an example of a railcar wheel with a flat spot.

FIG. 8 is an illustration of an example of a cross section of a rail and railcar wheel with wheel flange wear and rail head wear.

FIG. 9 is an illustration of a flowchart of an example of a method for detecting railroad anomalies, according to various embodiments.

FIG. 10 is a block diagram illustrating an example of a machine upon which one or more embodiments may be implemented.

DETAILED DESCRIPTION

The use of sensors and wireless communications has grown exponentially in recent years providing a means for monitoring and tracking many aspects of our industrialized world. Trains and the railway system is one of the oldest modes of transportation. Because of the thousands of miles of track and the weight carried by each railcar, problems may easily develop, but are not necessarily detectable until it is too late. Utilizing sensors provides a way of monitoring both the rails and railcars for anomalies. By implementing the use of sensors, a communication system, and a means for aggregating the data, it is possible for trains to detect anomalies while in use and operating under normal speeds and conditions. Real time anomaly detection may occur, alerting the train operator and preventing accidents, damage to the rails or railcars, and in the worst case scenario, derailment.

Some of these anomalies impact train performance, may damage track, or even lead to train derailments. By detecting the signs of early failure or “pre-failure” events, an operator may be notified prior to a massive failure, saving downtime and possibly avoiding harm to people. This invention leverages a computer device with sensors attached to collect and analyze data and performance of various components of the railcar. These components include, for example, attributes of the wheel, axle, bearing, truck, center pin, coupling, suspension and brake. When a system begins to fail—it often provides “pre-failure” symptoms that may be captured and characterized. An alert may be provided to the operator pertaining to the discovered symptom for remedial action. Sometimes these will be emergency issues, other times they may indicate maintenance is needed, but any danger is not immediate. This system allows the operator to better plan and manage maintenance activities for a railcar, with data from its previous loaded journeys.

Current anomaly detection solutions may not be deployed while under normal load conditions. The systems and techniques disclosed herein may provide accurate anomaly identification by capturing and characterizing data from the railcar while it is under load, utilizing sensors, and at track speed. An example of being under load would be when the railcar is carrying cargo and is part of a train containing multiple railcars each carrying over 100 tons. Examples of sensors that could be utilized on the railcar include, but are not limited to, an accelerometer, gyroscope, compass, camera, microphone, strain gauge, GPS, and physical contact sensor. In a further example, a position of the sensor within the railcar or within the train within may be captured or tracked by the sensor or within a data system. For instance, knowledge of whether the sensor is located on railcar #23 in the train may be used to identify a position of a data event captured on the railcar relative to a geographic location of the overall train.

The present subject matter may avoid interruption of service and manual intervention. By continuously assessing the health of the track while under normal operating conditions, the track may continuously be monitored and maintenance crews may be more efficiently deployed. Track maintenance may turn from reactive to proactive by catching early warning signs of wear and tear, well before failure. Performing this data acquisition while at “track speed” allows for a true representation of the state of the rail, while loads continue to run on the track and interact with the rail.

The measurement data with GPS location data is collected, through radio frequency communications technique, by manually gathering the information from the unit, etc. in an example, the GPS location data may be collected from a centralized location (e.g., in the front of the train), and the position of the other sensors is calculated based on information that indicates which railcar the sensor is located at, and the length of each railcar or the distance from the centralized location. The data may be uploaded to a cloud application. The data in the cloud application may be integrated with other data from the same geolocations, providing an historic view of the indicators for each railcar while under different loads and at different speeds.

FIG. 1 illustrates an example environment 100 and an anomaly detection system 200 for location based railway anomaly detection, according to an embodiment. A train may comprise one or more engine cars that provide the power for movement to the train. Behind and/or in front of the engine cars are the railcars which carry the cargo for the train. Each of the railcars are connected to each other and are pulled along the rail track by the engines. In the example environment 100 a railcar 105a positioned on a railway track 131 and connected to another railcar 105b. A train typically has many railcars connected together. In the example environment 100, railcar 105a comprises the cargo container 106 and two trucks 110a and 110b. Railcar 105a is an example of any one of the railcars in a train. A cutout of the cargo container 106 shows a truck 110a that the cargo container 106 is seated on. The truck 110a is connected to the cargo container 106 by center pin 121. The center pin 121 pivots so that the truck 110a can rotate. The rotation of truck 110a allows the train and its railcars, such as railcar 105a, to move along curves on the railway track 131. In the example embodiment 100, the truck 110a comprises the center pin 121 for connecting to cargo container 106 and two axles 120, which each have two wheels 125. The wheels 125 roll along the rails 130 of railway track 131. A wheel 125 will have a flange, which, in combination with the other wheels on the truck 110a, secures the truck 110a to the rails 130. The railcars 105a, 105b have no propulsion means of their own, but because of the immense weight of a railcar's cargo combined with the force and momentum for moving the railcar, smooth railcar wheel and rail interaction is essential.

In this example embodiment, the truck 110a has an anomaly detection system 200 comprising at least one sensor and a processor subsystem. Because of the layout of a train, a railcar may be experiencing an issue, but the train operator is unaware. For a train to operate smoothly and continuously move thousands of tons of cargo, the wheels of each railcar need to interface with the rails without issue. The example environment 100 also depicts a break 135 in the rail 130. The break 135 is a type of anomaly which could lead to, as an example, a difference in the height of the two rails 130 that will cause railcar issues. The break 135 is an example of an anomaly located on the rail which could cause an issue for each railcar that passes over it. Some anomalies located on the rails can cause immediate dangerous situations while others are less severe, but can still damage the railcar wheels. The anomaly detection system 200 may be utilized on each railcar 105a, 105b to monitor the condition and operation of its own trucks, wheels, and axles. Additionally, the anomaly detection system 200 may collect data from multiple railcars to find anomalies located on the rails.

FIG. 2 is a block diagram of an example of an anomaly detection system 200 for location based railway anomaly detection according to an embodiment. The anomaly detection system 200 comprises one or more sensors 201 connected to a processor subsystem and memory in the sensor data processing unit 202. The sensor data processing unit 202 is further connected to an anomaly alert unit 203, which may include an output device (e.g., to provide audible output or visible output) for providing an anomaly alert. In an example, the communication of information between the sensor data processing unit 202 arid the anomaly alert unit 203 may be via a wired connection; in another example, the communication may occur via a wireless transmitter 204.

Features of the anomaly alert unit 203 or the sensor data processing unit 202 may also be connected to a remote database 207 via a remote network, such as via the cloud 205. The features of anomaly detection or anomaly output performed by the sensor data processing unit 202 or the anomaly alert unit 203 may be fed from data in the remote database 207, because an anomaly may be specific to the conditions of the track at a particular location.

In an example, the sensor data processing unit 202 and the anomaly alert unit 203 are connected to a local database 209 that is used to temporarily host, persist, or store data. For example, the local database 209 may include data related to the past 10 miles and the next 10 miles of track. In a further example, the data is periodically synchronized from the local database 209 to the back-end database 209. This is used, for example, when the train goes through tunnels or other periods when there are no communications to the back-end services and the remote database 207.

The sensors 201 may be an accelerometer, gyroscope, GPS, tachometer, thermistor, camera, or microphone. The sensors 201 may be configured to provide constant measurements or measurements at set intervals of time to the sensor data processing unit 202. In an example, the sensors are mounted to obtain sensor measurements regardless of the direction of travel of the railcar, as the railcar may be pushed or pulled from either end. In further examples, two separate sets of sensors, such as two separate cameras, are used to obtain sensor measurements from the different directions.

As an example, the sensor data processing unit 202 receives the measurement data from the one or more sensors 201. The sensor data processing unit 202 evaluates the sensor measurement data to determine if an anomaly is present. To detect anomalies, in an embodiment, the sensor data processing unit 202 may collect measurement data to create a set of baseline data. The baseline data set may be created by the train operator indicating that a set of measurement data collected from a particular journey by the train be used as the baseline data set. The baseline data set may also be created by having the sensor data processing unit 202 analyze a gathered set of measurement data. If the gathered set of data is all within a predetermined range of standard deviation, then the set may be considered anomaly free and used as the baseline data. In both scenarios, the sensor data processing unit 202 may take all the readings for each sensor and average them to find the baseline data value for that sensor. The sensor data processing unit 202 may also find the maximum and minimum values recorded for each sensor 201 from the gathered set and use those values as the threshold or guardrail values in the baseline data set.

Once a set of baseline data has been established for the sensors 201 present on a given railcar 105a, the sensor data processing unit 202 may begin collecting operating measurement data from the sensors 201 on a railcar 105a. The sensor data processing unit 202 may evaluate the operating data against the baseline data to determine if the operating data has surpassed a threshold for what is normative operation by a railcar 105a and its components. When this occurs, an anomaly may have been detected.

As an example scenario, a railcar 105a should not oscillate or move from side to side excessively. The momentum created by railcar oscillation may travel to additional railcars and eventually have enough force to cause a railcar to topple and cause a train derailment. If detected early enough, the train operator may slow the train to decrease the oscillation and prevent a derailment. An inertia sensor, such as an accelerometer or gyroscope, may provide a measurement for the degree from center a railcar is leaning to one side. As a railcar travels, especially as it goes around curves, a normative degree of lean is expected. If the degree of lean measurement data reported by the sensor is determined by the sensor data processing unit to exceed a threshold or guardrail degree of lean, then the sensor data processing unit may begin to look for measurement data indicating lean in the other direction. If the sensor data processing unit receives measurement data from an inertia sensor that it determines is a degree of lean in the opposite direction which exceeds the threshold limit, then the railcar may be experiencing oscillation. The sensor data processing unit may continue to receive measurement data from the inertia sensor and denote the time for each measurement in excess of the threshold. Several factors may be indicative of dangerous oscillation the degree of lean in both directions continues to be in excess of the threshold without diminishing, the frequency increases for the lean in both directions in excess of the threshold, or the degree of lean in both directions increases. In this example, when the sensor data processing unit determines dangerous oscillation is occurring, then a warning or alert is generated.

When the sensor data processing unit determines an anomaly is present, a message is sent to the anomaly alert unit 203 to warn a train operator. Examples of embodiments for the anomaly alert unit 203 may be a display located in the lead engine car or a mobile device used by the train operator.

In an example, measurement data is sent from the sensor data processing unit 202 to a wireless transmitter 204 attached to the railcar or train. The wireless transmitter may communicate using various wireless standards and networks, such as Wi-Fi, a cellular data network, satellite communications, or long-range communication networks. The measurement data is transmitted from wireless transmitter 204 to the cloud 205.

At an offsite location, the measurement data for multiple railcars and trains in the cloud 205 is collected at the data collection unit 206. The collected measurement data, indexed by UPS location, is stored in the database 207. The collected measurement data may include sensor measurement data from railcars in a railway network. In addition to being indexed by the UPS location, the stored measurement data may include train related metadata such as the size, length, or weight of the railcars on which the sensors are located, the number of railcars in the train, the position within the train of the railcar on which the sensor is located, or the weight of railcars immediately in front of and behind the railcar with the sensors.

The location processing unit 208 analyzes the measurement data stored in the database 207. As an example, the location processing unit 208 groups the measurement data by UPS location and analyzes the measurement data at each location. In this embodiment, when the location processing unit 208 finds multiple instances of measurement data that exceeds a predetermined threshold for normative operation all occurring at the same GPS location, then a rail anomaly may be detected. Upon detecting a rail anomaly, the location processing unit 208 may transmit a message through the cloud 205 to the anomaly alert unit 203.

To detect anomalies on either a rail or a railcar, many devices and technologies may be utilized. The following are some of those devices and technologies and how they are used for rail and railcar anomaly detection.

Measurements and sensor recordings are stored in a database to track normal and baseline data and detect anomalies that are persistent on a rail location. The database may also include a railcar manufacturer's recommended service information, railcar owner's maintenance and repair records, and previously detected problem areas. The database may store train data, including the train's configuration and semi static conditions, such as: the size, length, and weight of each railcar on which sensors are located, the number of railcars in the train, the direction of travel for each railcar (forwards or backwards), and the position within the train each railcar. Additional environmental factors that may be detected by respective sensors may include factors such as temperature, humidity, barometric pressure, wind velocity, recent rainfall, soil moisture, or seismic data, relating to the track environment, may be collected and stored.

Location and movement sensors are needed to detect and locate most anomalies occurring on the rails or railcars. A GPS and speedometer may be used to derive real-time data on the train's current operation, including the location and velocity. An accelerometer may be used to measure vibrations and absolute orientation of the sensor with respect to the Earth's gravity, such as an inclinometer. A gyroscope may be used to measure minute angular changes. A compass or magnetometer may be used to measure orientation with respect to the Earth's magnetic field.

Cameras, both visible and infrared, are used to capture images. The one or more cameras may be mounted in one of several positions, such as a single camera mounted with a wide-angle lens to allow for viewing the track and the wheels, including the wheel flange or multiple cameras, with one for each rail. One or more microphones may be used to detect the squeal of the wheels on the track.

Most railcars leverage an air brake system where the brakes are applied by default through the use of internal springs. The brakes are held open, or not applied, by compressed air. The compressed air line extends the length of the train. If the compressed air line is broken, the brakes will go into default mode of applied. If the compressed air is restored, all brakes should be released, but brakes may remain stuck in the applied position. A moving train with a stuck brake results in dragging a non-moving axle and stuck wheels. A railcar may also get stuck wheels outside of a stuck brake scenario from ice or other debris preventing axle rotation. This causes wheel flats and excessive heat build-up which may lead to train derailments. Operators may need to look back at the railcars to observe any signs of stuck brakes or wheels. A stuck brake or wheel may be detected by monitoring axle and wheel rotation while the train is moving. One example to detect a stuck brake or wheel is the use of an optical tachometer with GPS or inertia sensor data which reports data indicating movement of the railcar but no corresponding report of rotation of the axle from the tachometer.

A track will have cross level issues when the levelness of the track is uneven. The cross level refers to the precise measurement of the track's evenness and levelness when railcars are traveling on the tracks. Current methods for cross level assessment involve a rail engineer or road aster walking the track with a handheld device or running a large “geometry car” deployed on the track.

FIG. 3A is an illustration of an example of a level rail track. As seen in FIG. 3A, track cross section 305 is a level track, with rail 325 and rail 326 at approximately the same height. Railroad tie 320 is also level. FIG. 3B is an illustration of an example of an out of level rail track. In FIG. 3B, track cross section 310 is an out of level track, with rail 340 at a lower height than rail 341. Railroad tie 335 sits angled downward on one side. Causes of the anomaly may be frost heave, erosion, ballast issues, tie issues or temperature extremes. An example of detecting cross level issues includes using an inertia sensor to determine three of the dip in track as a railcar moves from level track 305 to out of level track 310. The force measurement, in conjunction with a GPS reading indicates an out of level track location. When a truck utilizes inertia sensors on each side, then if a dip is detected on one side, but not the other, then a cross level issue is present.

When rail track is laid, the method for connecting rails together to form a continuous track leverages two main methods: mechanical connectors joining one rail to another rail, and fusing rails with a welded joint. Both of these joint types are offset from each other to allow for a stronger section of track versus placing them directly next to one another. FIG. 4 is an illustration of an example of rail warp. As seen in FIG. 4, this offset may sometimes cause a slight height variation from one rail 405 to the next connected rail 416. Because the joint 410 and joint 411 of the parallel rails are offset, as one side of the railcar is dipping at a joint 411, the other side of the railcar is raised by a rail 415. As the railcar moves along the track, the railcar will then be rocked in the other direction as the side that was raised dips at a joint 410 and the side that was previously dipped, is raised by a rail 416. As the railcar moves along the track with the alternating up and down for each side, the railcar will begin to oscillate from side to side. The oscillation may cause a “rock ‘n’ roll” derailment. The remedy for an oscillation situation is to slow the train to a speed that does not propagate the frequency of joint crossing to the railcars. When a derailment occurs because of oscillation, the documented speeds of the train are between 12 and 24 MPH. Detection of this activity depends on the type of railcar, weight of the railcar, speed of the railcar, and observation of the engineers during train travel. Rail companies have enacted speed policies, adjusted track geometry, and adjusted joint positioning to reduce this issue. An example of detecting the symptoms of one or more oscillating railcars uses an inertia sensor to detect a repeated pattern of significant amplitude in the one or more railcars. The force is recorded in conjunction with the GPS location, train speed, railcar and train weight. The data is then aggregated into a larger dataset in the cloud for that section of track.

FIG. 5A is an illustration of an example of a normal rail connected to a tie. As seen in FIG. 5A, a normal cross section 505 shows rail 520 resting on top of tie 515. FIG. 5B is an illustration of an example of a rail that has become separated from the tie. Abnormal cross section 510 shows rail 540 not resting on tie 535. When a railroad tie 535 becomes worn or ballast 545 is lost, a “pot hole” or a general “dip” in the support of the rail 540 will develop. In normal cross section 505, the rail 520 is placed upon the top plate 516, which is placed on the tie 515. The top plate 516 is tightly secured with lag bolts to the bottom section of rail 520. In abnormal cross section 510, when the tie 535 begins to shrink or rot, or the ballast 545 erodes, a gap 541 occurs between the top plate 536 and the tie 535. This gap 541 may decrease in size when a train rolls along the rail 540 as the rail 540 is pressed onto the tie 535. Because the surrounding ties to the rotted tie 535 are not deteriorating, it is not detectable without manual cross level testing. An example of detecting a tie or ballast issue along the track with railcar sensors uses an inertia sensor to detect the amplitude of a change in conjunction with the speed of train, weight of the railcar and train, and the GPS location. A significant amplitude change occurring for multiple railcars in the same geographic location is an indication of a chuck hole or pot hole.

FIG. 6 is an illustration of an example of forces that may occur when a truck has a stuck center pin. As illustrated in FIG. 6 from a top down view, a railcar's truck 620 includes four wheels 615 that engage with rail 635 and 636. In the center of the truck is the center pin 621 and truck bed 622 which the railcar sits upon. This allows the truck 620 to pivot around curves. If a truck center pivot pin 621 becomes stuck, track gauge may be widened or worse a derailment may occur.

When the truck 620 does not properly pivot to the requirements of the track 630, a lateral force 610 is applied to inside of rail 635 with force 605 and the inside of rail 636 with force 625 by the flange 616 of the wheels 615. One example of detection for a stuck center pin 621 is using a sensor to monitor the truck centerline. If the truck centerline position does not move when train or railcar curve movement is detected by a sensor, such as an accelerometer or gyroscope, then the truck center pin 621 is stuck. Another example is the use of a spring to calculate force of a normal truck turn. If the similar forces from the spring are not recorded at the times as an inertia sensor or GPS is indicating that the train is on a curve, then the truck center pin 621 is stuck. A further example would be the use of an LED attached to the railcar which is pointing at non-uniform reflective tape attached to the truck 620. The non-uniform reflective tape is most reflective in center and less reflective at the edges. A sensor measures light amplitude reflection. If the light amplitude reflection does not change when the inertia sensors or GPS indicate the train is on a curve, then the truck center pin 621 is stuck. Another example uses the physical measure of fixed point on railcar in contact with variable electronic sensor on the truck 620 or a fixed point on the truck 620 with a variable electronic sensor on the railcar. If the sensor does not indicate a change in position when an inertia sensor indicates the train is on a curve, then the truck center pin 621 is stuck.

The bearings of the axle may overheat and cause axle failure leading to downtime and a possible train derailment. An example of detecting overheating bearings uses a thermistor or infrared camera that continually measures the temperature values of the bearing assembly. If the bearing temperature exceeds a predetermined temperature, then the bearings are overheating. Another example is the use of a vibration sensor to detect a failing bearing. A wheel and axle will generate a normative amount of vibration operating under normal conditions. When a hearing fails, the axle will spin abnormally and create abnormal vibration with may be detected when compared to the normative vibration. A further example is the use of a microphone to capture the sound of the bearings and analyze the captured sound for the sound of a failing bearing.

Each truck 621 of a railcar usually includes two axles with two wheels connected to each axle. The wheels are heat and pressure fitted onto the axle. The axle rotates with the wheels as one unit. When the train, and each truck of a railcar goes around a turn there is great pressure exerted on each axle to keep the wheels turning at a consistent rate. This stress may cause an axle to fail, and thus allowing the wheels to rotate independently. Independently rotating wheels may lead to railcar downtime or the possibility of a derailment. One example for detecting a broken or failing axle involves placing tachometer sensors next to each wheel. When the tachometer measurements for two wheels on the same axle are not the same, then the axle has broken. Another example of broken axle detection uses imaging sensors that monitors the alignment of the axle.

Railcar wheels 615 are commonly made from thick steel, but may still lose their balance or become warped. This results in wheels that wobble either vertically or horizontally which then causes uneven flange wear on the rails. One example for detecting unbalanced or warped wheels is by using an inertia sensor that measures the repeating pattern of a wheel. An uneven repeating pattern differing from a baseline wheel repeating patter is indicative of a wobbling wheel, in either a vertical or horizontal direction. Another example uses an imaging analytics system to watch the wheel and rail interface to detect out of balance or warped wheels.

FIG. 7 is an illustration of an example of a railcar wheel with a flat spot. As illustrated in FIG. 7, a wheel 710 may develop a flat spot 711 by moving a railcar with a non-rotating wheel 710. A stuck brake or a frozen wheel 710 or axle 715 are several examples of causes for non-rotating wheels 710. A wheel 710 with a flat spot 711 will make a “ticking” sound when it rotates. When a wheel 710 with a flat spot 711 rolls along a track rail 720, it may seriously damage the rail 720 as the rotation impact, while under heavy load, may mark or dent the rail head 725. Remediation for a wheel 710 with a flat spot 711 is to remove the railcar from service and resurface the wheel 710 or replace the entire axle 715 and wheel 710. An example for detecting a flat spot 711 on a wheel 710 uses an inertia sensor to measure a repeating vibration pattern caused by the wheel flat 711. A wheel 710 with a flat spot 711 will not have a smooth rotation and detected vibrating pattern is indicative of a wheel flat. Another example uses an acoustic sensor to record the repeating signature of the “ticking” noise. A further example uses a camera to capture images of the wheel 710 and then analyze the images of the wheel 710 for its round or out-of-round characteristics. An additional example uses a physical sensor such as an Electronic Drop Indicator, attached to the truck 705 and identify the flat spot 711 through continuous measurements.

FIG. 8 is an illustration of an example of a cross section of a rail and railcar wheel with wheel flange wear and rail head wear. As seen in FIG. 8, a cross section of a wheel 805 is resting on a rail 820. When a wheel 805 wears past industry standards, the railcar may experience unpredictable behaviors including vibrations and potential derailment. Wheel 805 examination is performed manually on a cadence per railroad and railcar leasing company. A wheel 805 may develop a problem between maintenance cycles. Over time, a wheel 805 may develop flange wear 810 or a rail head 820 may develop rail wear 815. Each of these conditions may allow for more side to side movement of the wheel 805 along the rail 820 from the gap that exists as a result of the wheel wear 810 or rail wear 815. An example for detecting wheel wear 810 includes using a camera for collecting images of the wheel 805. The wheel 805 may be continuously or periodically monitored over the course of its life. Due to the variable nature of wheel wear 810 and rail wear 815, it is difficult to attribute one vibration signal specifically to either wheel wear 810 or rail wear 815.

Truck hunting is when the truck of the railcar is pivoting back and forth while the train is on a straight segment of track. Truck hunting typically occurs when the contact point of the wheels on the rails do not find a balance point between the four wheels of the truck. Truck hunting may damage rails and wheel flanges. Excessive truck hunting pushes the gauge of the wheel out of specification. An example of detecting truck hunting use an inertia sensor to detect the oscillating nature of truck which occurs during truck hunting. If measurements show the truck pivoting while the train or railcar is on a straight section of track, then truck hunting is occurring. This may be combined with an acoustic sensor to detect the wheel flange hit between the front wheel on one side of the truck and the back wheel on the other side of the truck.

Railcars that begin to oscillate as a train goes over track with recurring track geometry defects may cause a complete derailment. This typically happens when trains are traveling between 12 MPH and 24 MPH. Causes of this oscillation, or “Rock 'n Roll” train effect are from either recurring joints that have begun to depress in the ties or joints that are offset the same distance for a long run. This causes the oscillation to occur and propagate along the train. An example for detecting railcar oscillation uses an inertia sensor, to measure the oscillation. Abnormal oscillation for a railcar is when side to side sway of the car occurs at such a degree that the railcars become unbalanced and have enough force cause a complete derailment. Oscillation is determined by using a sequence of timed data points. To detect the oscillation, an initial inertia sensor measurement is recorded at one data point for the degree of sway from center for the railcar. When a measurement is recorded at a subsequent time data point for similar sway in the opposite direction, then possible oscillation may be occurring. If the pattern continues, with an increase in sequence repetition or increase in degree of sway, then abnormal oscillation may be occurring. Oscillation guard band limits may be assigned. If the guard band limits are reached, then a notice to the engineering may be sent, as well as braking of the train and counter suspension activities to prevent a railcar from toppling over.

Track rails in North America use a gauge of 56½inches. Gauge is measured from the inside (gauge side) of the rail head to inside of the other rail head. The outside of the rail head is called the field side. Train derailments may occur if the gauge exceeds 58″ or is under 56″. In either case, a wheel may fall off the rail and ride on the wheel flange along the ballast and ties. A derailment such as this will destroy tie plates, lag bolts, and ties. The damage may cause the rail to dislodge from the ties and all subsequent cars will derail. An example of detecting track gauge deflection uses sensors such as an electronic measuring stick and measuring the gauge near the load points or wheels. By using sensors attached to the railcar, the gauge may be recorded at track speed under heavy loads. This data is incorporated with UPS information, the railcar and train weight. Additional sensors could be used such as cameras or laser distance sensors to assess gauge.

Rail head is the section of rail where the wheels touch the rail. The wear occurs on the top but also occurs on the gauge side of the rail head due to wheel flange contact, especially on track curves. Rail head wear may also create a “corrugated” rail, which usually occurs on a track with an incline. Detection of wear or a corrugated rail is performed by a manual inspection or with a geometry car inspection. An example of detecting rail head wear uses an inertia sensor arid a camera. With the information from these sensors, rail wear anomalies may be identified from observation and actual loaded system actions. Rail head wear, such as on the side of the rail head, has symptoms similar to and is detected in the same manner as issues like wheel wear or truck hunting. A corrugated rail would have similar symptoms and detection as a failing bearing. However, if the detection is not continuous for the railcar and its respective trucks and wheels, but the detection is repeated for multiple railcars in the same geographic location, then the issue lies with the rail instead. The sensor data is captured and integrated with GPS data, railcar and train weight, and train speed. The data is uploaded to the cloud system for integration with other data from previous recordings in the same location.

Tracks are welded together using a method called Continuous Weld Track. Each rail section is 144 feet long, and is welded with a thermite welding kit to join two rails together. Sometimes these welds may become corrupt and will fail over time. An example of detecting rail weld anomalies uses magnetic inductance to determine rail continuity by running a current through the rail and detecting changes in the magnetic inductance. When the inductance breaks or becomes weak, the track weld is either failing or close to failing and maintenance is necessary. This is performed while a train is on the track, and the train is under load and at track speed. Data is also collected for the GPS position, the ambient temperature, the induction measurement and other train data. The collected data is uploaded to the cloud system for further analysis and correlation with other data.

FIG. 9 illustrates a flowchart of an example of a method 900 for railway anomaly detection, according to various embodiments. The method 900 may provide similar functionality as described in FIG. 2.

A railcar may have one or more sensors installed on it. Some of the sensors may report information for the railcar itself, while others may be situated to monitor specific parts of the railcar such as the trucks or individual wheels. While the train and railcars are underload and moving at track speed, the sensors may continually or at set intervals record and transmit measurements. At operation 905, a processor subsystem at the sensor data processing unit on the railcar receives the measurement data from a sensor array. A sensor array may include one or more sensors of one or more types of sensors including an accelerometer, gyroscope, GPS, or thermistor. In another example, the processor subsystem receives images or video from a camera attached to the railcar. In a further example, the processor subsystem receives audio recording from a microphone attached to the railcar.

At operation 910, the processor subsystem obtains measurement data to use as baseline measurements.

At operation 915, as the train and railcars continue to operate under normative conditions, the sensors transmit measurements to the processor subsystem. For example, the near real time sensor measurements occur while the railcar is travelling at track speed. As another example, the near real time sensor measurements occur while the railcar is a full carry weight.

At operation 920, the baseline and operating measurement data is compared to determine if a railway anomaly has been detected. For example, a stuck brake or stuck wheel anomaly may be detected when it is determined a wheel or axle is not moving using measurement data from an optical tachometer in combination with either a GPS sensor or an inertia sensor. In another example, a stuck truck center pin may be detected when it is determined the truck is not pivoting. A non-pivoting truck may be determined in any of the following ways in combination with GPS positioning that shows the railcar is turning: using accelerometer or gyroscope measurement data of the truck, force measurement from a spring connected to the truck, measuring light amplitude reflection from a light pointed at the truck with non-uniform reflective tape attached, a physical measure of a fixed point on the railcar in contact with a variable electronic sensor on the truck, or a physical measure of a fixed point on the truck in contact with a variable electronic sensor on the railcar. In another example, an anomaly of overheating bearings may be detected using measurement data from a thermistor sensor, an infrared camera, a vibration sensor, or audio analytics. As another example, a broken axle is detected when the two wheels of the axle are rotating independently based on measurements from tachometers placed next to each wheel on the same axle. A further example, a warped wheel may be detected when a wobble is found in either a vertical or horizontal direction using an inertia sensor. In another example, a flat spot on a wheel may be detected using measurements from one or more sensors including an inertia sensor, an acoustic sensor, an imaging sensor, or a physical sensor. For example, excessive wheel wear may be detected based on images taken from the attached camera. In a further example, excessive truck pivoting while the train is on a straight segment of track may be detected by determining railcar oscillation from an inertia sensor or measuring wheel flange hits with an acoustic sensor.

At operation 925, measurement data from one or more inertia sensors, taken at a sequence of time data points, is evaluated to determine if the railcar is excessively swaying from side to side. Sway may also be referred to as tilt, and may represent the angle or amount of lean from vertical. To determine side to side sway or oscillation, a measurement is taken from the inertia sensors for a lean of the railcar to one side starting at one time data point. Another inertia sensor measurement is then taken at the next data point. The data points are examined to determine the amplitude of the side to side sway or oscillation, and also the instantaneous sway to either side. If the amplitude of the oscillation exceeds a predetermined threshold, or the instantaneous tilt to the left or right exceeds a threshold, then the train may automatically brake or provide a warning provided to the train operator. Note that each type of railcar may have separate thresholds for the maximum permitted oscillation amplitude and maximum instantaneous tilt, and those values may also vary depending on the specific contents of the railcar.

Thresholds may be maintained for both a maximum permitted oscillation amplitude as well as a maximum instantaneous tilt in either direction. As an example, a particular railcar may have a maximum permitted oscillation amplitude of 15 degrees. However, this may not detect scenarios where there is excessive lean in one direction but not the other, such as a 10 degree lean to the left from center, but only a 4 degree lean to the right from center. Thus, the particular railcar may also have a maximum instantaneous tilt threshold, such as 9 degrees, to detect excessive sway even though the total oscillation amplitude has not exceeded the maximum permitted oscillation amplitude.

At operation 930, the measurement data is stored in a database and indexed by the GPS location. As an example, the database may include train's configuration and operating conditions, including at least one of: the size, length, or weight of the railcars on which the sensors are located, the number of railcars in the train, the position within the train of the railcar on which the sensor is located, or the weight of railcars immediately in front of the railcar with the sensors. For example, the measurement data is transmitted to a cloud-based service to collect the data from multiple trains and update the track condition database.

The data indexed UPS location may be further analyzed based on the location to determine anomalies located on the rails. For example, the sensor measurement data and time data points may be integrated with the geographic location information to determine location based anomalies, or anomalies occurring on the rail. A further example, a tie or ballast issue is determined when a pot hole is detected from measurement data including the amplitude of drop from an inertia sensor, the speed of the train, the weight of the railcar and train, and the UPS location. Another example may determine the presence of rail warp from repeated amplitude change occurring at the same UPS location, along with the train speed, and the weight of the railcar and train. Another example may find the presence of rail wear when an inertia sensor or camera captures side to side movement of the truck and wheels along the rails occurring in the same UPS location, along with train speed and the weight of the train and railcar data. In another example, an anomaly may be detected as a cross level issue based on unevenness of the rails, wherein the measurements to determine unevenness include the use of an inertia sensor to detect the force of a dip in the rails combined with the GPS for train location arid speed, and the weight of the railcar and train. Another example may use images taken from a camera attached to the railcar and integrated with GPS location information to detect a track gauge deflection from either a wide or narrow rail gauge. In a further example, rail weld failure is detected by running a current through the rail while the train is under load and at track speed, and collecting the GPS location, ambient temperature, and induction measurement.

At operation 935, if an anomaly is detected, then an alert is transmitted to a display device to notify an operator. As an example, the display device may he located in the lead engine car. The display device may also be a mobile device carried by a train operator.

FIG. 10 illustrates a block diagram of an example machine 1000 upon which any one or more of the techniques (e.g., methodologies) discussed herein may perform. In alternative embodiments, the machine 1000 may operate as a standalone device or may be connected (e.g., networked) to other machines. In a networked deployment, the machine 1000 may operate in the capacity of a server machine, a client machine, or both in server-client network environments. In an example, the machine 1000 may act as a peer machine in peer-to-peer (P2P) (or other distributed) network environment. The machine 1000 may be a personal computer (PC), a tablet PC, a set-top box (STB), a personal digital assistant (PDA), a mobile telephone, a web appliance, a network router, switch or bridge, or any machine capable of executing instructions (sequential or otherwise) that specify actions to be taken by that machine. Further, while only a single machine is illustrated, the term “machine” shall also be taken to include any collection of machines that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein, such as cloud computing, software as a service (SaaS), other computer cluster configurations.

Examples, as described herein, may include, or may operate by, logic or a number of components, or mechanisms. Circuit sets are a collection of circuits implemented in tangible entities that include hardware (e.g., simple circuits, gates, logic, etc.). Circuit set membership may be flexible over time and underlying hardware variability. Circuit sets include members that may, alone or in combination, perform specified operations when operating. In an example, hardware of the circuit set may be immutably designed to carry out a specific operation (e.g., hardwired). In an example, the hardware of the circuit set may include variably connected physical components (e.g., execution units, transistors, simple circuits, etc.) including a computer readable medium physically modified (e.g., magnetically, electrically, moveable placement of invariant massed particles, etc.) to encode instructions of the specific operation. In connecting the physical components, the underlying electrical properties of a hardware constituent are changed, for example, from an insulator to a conductor or vice versa. The instructions enable embedded hardware (e.g., the execution units or a loading mechanism) to create members of the circuit set in hardware via the variable connections to carry out portions of the specific operation when in operation. Accordingly, the computer readable medium is communicatively coupled to the other components of the circuit set member when the device is operating. In an example, any of the physical components may be used in more than one member of more than one circuit set. For example, under operation, execution units may be used in a first circuit of a first circuit set at one point in time and reused by a second circuit in the first circuit set, or by a third circuit in a second circuit set at a different time.

Machine (e.g., computer system) 1000 may include a hardware processor 1002 (e.g., a central processing unit (CPU), a graphics processing unit (GPU), a hardware processor core, or any combination thereof), a main memory 1004 and a static memory 1006, some or all of which may communicate with each other via an interlink (e.g., bus) 1008. The machine 1000 may further include a display unit 1010, an alphanumeric input device 1012 (e.g., a keyboard), and a user interface (UI) navigation device 1014 (e.g., a mouse). In an example, the display unit 1010, input device 1012 and UI navigation device 1014 may be a touch screen display. The machine 1000 may additionally include a storage device (e.g., drive unit) 1016, a signal generation device 1018 (e.g., a speaker), a network interface device 1020, and one or more sensors 1021, such as a global positioning system ((GPS) sensor, compass, accelerometer, or other sensor. The machine 1000 may include an output controller 1028, such as a serial (e.g., Universal Serial Bus (USB), parallel, or other wired or wireless (e.g., infrared (IR), near field communication (NFC), etc.) connection to communicate or control one or more peripheral devices (e.g., a printer, card reader, etc.).

The storage device 1016 may include a machine readable medium 1022 on which is stored one or more sets of data structures or instructions 1024 (e.g., software) embodying or utilized by any one or more of the techniques or functions described herein. The instructions 1024 may also reside, completely or at least partially, within the main memory 1004, within static memory 1006, or within the hardware processor 1002 during execution thereof by the machine 1000. In an example, one or any combination of the hardware processor 1002, the main memory 1004, the static memory 1006, or the storage device 1016 may constitute machine readable media.

While the machine readable medium 1022 is illustrated as a single medium, the term “machine readable medium” may include a single medium or multiple media (e.g., a centralized or distributed database, and/or associated caches and servers) configured to store the one or more instructions 1024.

The term “machine readable medium” may include any medium that is capable of storing, encoding, or carrying instructions for execution by the machine 1000 and that cause the machine 1000 to perform any one or more of the techniques of the present disclosure, or that is capable of storing, encoding or carrying data structures used by or associated with such instructions. Non-limiting machine readable medium examples may include solid-state memories, and optical and magnetic media. In an example, a massed machine readable medium comprises a machine readable medium with a plurality of particles having invariant (e.g., rest) mass. Accordingly, massed machine-readable media are not transitory propagating signals. Specific examples of massed machine readable media may include: non-volatile memory, such as semiconductor memory devices (e.g., Electrically Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM)) and flash memory devices; magnetic disks, such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks.

A processor subsystem may be used to execute the instruction on the machine-readable medium. The processor subsystem may include one or more processors, each with one or more cores. Additionally, the processor subsystem may be disposed on one or more physical devices. The processor subsystem may include one or more specialized processors, such as a graphics processing unit (GPU), a digital signal processor (DSP), a field programmable gate array (FPGA), or a fixed function processor.

The instructions 1024 may further be transmitted or received over a communications network 1026 using a transmission medium via the network interface device 1020 utilizing any one of a number of transfer protocols (e.g., frame relay, Internet protocol (IP), transmission control protocol (TCP), user datagram protocol (UDP), hypertext transfer protocol (HTTP), etc.). Example communication networks may include a local area network (LAN), a wide area network (WAN), a packet data network (e.g., the Internet), mobile telephone networks (e.g., cellular networks), Plain Old Telephone (POTS) networks, and wireless data networks (e.g., Institute of Electrical and Electronics Engineers (IEEE) 802.11 family of standards known as Wi-Fi®, IEEE 802.16 family of standards known as WiMax®), IEEE 802.15.4 family of standards, peer-to-peer (P2P) networks, among others. In an example, the network interface device 1020 may include one or more physical jacks (e.g., Ethernet, coaxial, or phone jacks) or one or more antennas to connect to the communications network 1026. In an example, the network interface device 1020 may include a plurality of antennas to wirelessly communicate using at least one of single-input multiple-output (SIMO), multiple-input multiple-output (MLMO), or multiple-input single-output (MISO) techniques. The term “transmission medium” shall be taken to include any intangible medium that is capable of storing, encoding or carrying instructions for execution by the machine 1000, and includes digital or analog communications signals or other intangible medium to facilitate communication of such software.

ADDITIONAL NOTES AND EXAMPLES

Example 1 is a system to detect railway anomalies, the system comprising: a processor subsystem; and a memory including instructions that, when executed by the processor subsystem, cause the processor subsystem to: receive measurements from a sensor array coupled to a railcar in a train; obtain baseline measurements from the sensor array; obtain, in near real time, measurements from the sensor array while the railcar is operating; detect a railcar anomaly based a comparison between the baseline and operating measurements, wherein the comparison of baseline and operating measurements includes an evaluation, over a sequence of time data points, inertia sensor measurements to detect abnormal railcar oscillation; store the data indexed to a GPS location in a database; and transmit an alert to an output device when an anomaly is detected based on the collected measurements from the railcars.

In Example 2, the subject matter of Example 1 optionally includes wherein the inertia sensor measurements and time data points are correlated with geographic location data to locate a rail anomaly.

In Example 3, the subject matter of any one or more of Examples 1-2 optionally include wherein the inertia sensor measurements and time data points are integrated with environmental data to locate a rail anomaly, the environmental data indicating at least one of: temperature, humidity, barometric pressure, wind velocity, recent rainfall, soil moisture, or seismic data.

In Example 4, the subject matter of any one or more of Examples 1-3 optionally include wherein to detect the railcar anomaly, the processor subsystem is to detect an oscillation of the railcar that exceeds a predetermined threshold, and wherein the processor subsystem is to initiate an automatic braking subsystem to reduce the railcar speed.

In Example 5, the subject matter of any one or more of Examples 1-4 optionally include wherein the memory further includes instructions to detect an anomaly as a tie or ballast issue, wherein the evaluation is based on measurements including the amplitude of a dip from the rails with an inertia sensor, the speed of the train, weight of the railcar and train, and GPS location.

In Example 6, the subject matter of any one or more of Examples 1-5 optionally include wherein the memory further includes instructions to detect an anomaly as rail warp, wherein the evaluation is based on measurements including repeated amplitude change, GPS location, train speed, and railcar and train weight.

In Example 7, the subject matter of any one or more of Examples 1-6 optionally include wherein the memory further includes instructions to detect an anomaly as rail wear, wherein the evaluation is based on captured data from an inertia sensor and a camera is correlated with GPS location, railcar and train weight, and train speed.

In Example 8, the subject matter of any one or more of Examples 1-7 optionally include wherein the data indexed by GPS location and stored in a database is subsequently analyzed, to identify a rail anomaly at a GPS location of the railway.

In Example 9, the subject matter of any one or more of Examples 1-8 optionally include wherein the inertia sensor measurements indicate a side to side movement of the railcar, and wherein the sensor array is coupled via an attachment of the sensor array to respective trucks of the railcar.

In Example 10, the subject matter of any one or more of Examples 1-9 optionally include wherein the output device is a located in a lead engine car.

In Example 11, the subject matter of any one or more of Examples 1-10 optionally include wherein the sensor array comprises: a GPS, accelerometer, or gyroscope.

In Example 12, the subject matter of any one or more of Examples 1-11 optionally include wherein the operating sensor measurements occur while the railcar is at track speed.

In Example 13, the subject matter of any one or more of Examples 1-12 optionally include wherein the operating sensor measurements occur when the railcar is at full carry weight.

In Example 14, the subject matter of any one or more of Examples 1-13 optionally include wherein the database includes the train's configuration and operating conditions, including: the size, length, or weight of the railcars on which the sensor array is located, the number of railcars in the train, the position within the train of the railcar on which the sensor is located, or the weight of railcars immediately in front of the railcar with the sensors.

In Example 15, the subject matter of any one or more of Examples 1-14 optionally include wherein the sensor array includes a camera to produce image data, and wherein the near real time measurements are derived from the image data.

in Example 16, the subject matter of any one or more of Examples 1-15 optionally include wherein the sensor array includes a microphone to produce audio data, and wherein the near real time measurements are derived from the audio data.

In Example 17, the subject matter of any one or more of Examples 1-16 optionally include communications circuitry communicatively coupled to the processor subsystem, and wherein the processor subsystem is to transmit the real time measurement data via the communications circuitry to a remote server.

In Example 18, the subject matter of Example 17 optionally includes wherein the remote server is configured to collect data from multiple trains and maintain a track condition database based on the data from the multiple trains.

In Example 19, the subject matter of any one or more of Examples 1-18 optionally include wherein the memory further includes instructions to detect an anomaly as a cross level issue based on an unevenness of the rails determination, wherein the measurements include the use of an inertia sensor to detect the force of a dip in the rails combined with the GPS for train location and speed, and the weight of the railcar and train.

In Example 20, the subject matter of any one or more of Examples 1-19 optionally include wherein the memory further includes instructions to detect an anomaly as track gauge deflection based on a wide rail gauge determination, wherein the images taken from a camera are used to measure the rail gauge.

In Example 21, the subject matter of any one or more of Examples 1-20 optionally include wherein the memory further includes instructions to detect an anomaly as track gauge deflection based on a narrow rail gauge determination, wherein the images taken from a camera are used to measure the rail gauge.

In Example 22, the subject matter of any one or more of Examples 1-21 optionally include wherein the memory further includes instructions to detect an anomaly as track gauge deflection based on a wide rail gauge determination, wherein the track gauge is measured with an electronic measurement device.

In Example 23, the subject matter of any one or more of Examples 1-22 optionally include wherein the memory further includes instructions to detect an anomaly as track gauge deflection based on a narrow rail gauge determination, wherein the track gauge is measured with an electronic measurement device.

In Example 24, the subject matter of any one or more of Examples 1-23 optionally include wherein the memory further includes instructions to detect an anomaly as a rail weld failure, wherein the evaluation is based on measurements including running a current through the rail while the train is under load and at track speed, and collecting the GPS location, ambient temperature, and induction measurement.

In Example 25, the subject matter of any one or more of Examples 1-24 optionally include wherein the memory further includes instructions to detect an anomaly as a stuck brake based on a non-moving axle or wheel from an unreleased brake determination, wherein measurements are collected from an optical tachometer in combination with either a GPS sensor or an inertia sensor.

In Example 26, the subject matter of any one or more of Examples 1-25 optionally include wherein the memory further includes instructions to detect an anomaly as a stuck wheel based on a non-moving axle or wheel determination, wherein measurements are collected from an optical tachometer in combination with either a GPS sensor or an inertia sensor.

In Example 27, the subject matter of any one or more of Examples 1-26 optionally include wherein the memory further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include accelerometer or gyroscope sensor data with GPS position.

In Example 28, the subject matter of any one or more of Examples 1-27 optionally include wherein the memory further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include calculated force from a spring connected to the truck combined with inertia sensor and GPS sensor data.

In Example 29, the subject matter of any one or more of Examples 1-28 optionally include wherein the memory further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include light amplitude reflection from a light pointed at the truck with non-uniform reflective tape attached.

In Example 30, the subject matter of any one or more of Examples 1-29 optionally include wherein the memory further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include a physical measure of a fixed point on the railcar in contact with a variable electronic sensor on the truck.

In Example 31, the subject matter of any one or more of Examples 1-30 optionally include wherein the memory further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include a physical measure of a fixed point on the truck in contact with a variable electronic sensor on the railcar.

In Example 32, the subject matter of any one or more of Examples 1-31 optionally include wherein the memory further includes instructions to detect an anomaly as overheating bearings, wherein the evaluation is based on measurements that include data from thermistor sensor, an infrared camera, a vibration sensor, or audio analytics.

In Example 33, the subject matter of any one or more of Examples 1-32 optionally include wherein the memory further includes instructions to detect an anomaly as a broken axle based on the two wheels of one axle rotating independently determination, wherein the measurement includes tachometers placed next to each wheel on the same axle.

In Example 34, the subject matter of any one or more of Examples 1-33 optionally include wherein the memory further includes instructions to detect an anomaly as the wheels being out of balance based on a warped wheel determination, wherein the measurements include a wobble in either horizontal or vertical directions using an inertia sensor.

In Example 35, the subject matter of any one or more of Examples 1-34 optionally include wherein the memory further includes instructions to detect an anomaly as a wheel with one or more flat spots, wherein the evaluation is based on measurements from the sensor array including an inertia sensor, an acoustic sensor, an imaging sensor, or a physical sensor.

In Example 36, the subject matter of any one or more of Examples 1-35 optionally include wherein the memory further includes instructions to detect an anomaly as excessive wheel wear, wherein the evaluation is based on measurements that include collected images taken from an attached camera.

In Example 37, the subject matter of any one or more of Examples 1-36 optionally include wherein the memory further includes instructions to detect an anomaly as truck hunting based on the truck pivoting while the train is on a straight segment of track, wherein the measurements include detected railcar oscillation by an inertia sensor.

In Example 38, the subject matter of any one or more of Examples 1-37 optionally include wherein the memory further includes instructions to detect an anomaly as truck hunting based on the truck pivoting while the train is on a straight segment of track, wherein the measurements also comprise measuring wheel flange hits with an acoustic sensor.

Example 39 is at least one machine readable medium including instructions to detect railway anomalies that, when executed by a machine, cause the machine to: receive measurements from a sensor array coupled to a railcar in a train; obtain baseline measurements from the sensor array; obtain, in near real time, measurements from the sensor array while the railcar is operating; detect a railcar anomaly based a comparison between the baseline and operating measurements, wherein the comparison of baseline and operating measurements includes an evaluation, over a sequence of time data points, of inertia sensor measurements to detect abnormal railcar oscillation; store the data indexed to a GPS location in a database; and transmit an alert to an output device when an anomaly is detected based on the collected measurements from the railcars.

In Example 40, the subject matter of Example 39 optionally includes wherein the inertia sensor measurements and time data points are integrated with geographic location data to locate a rail anomaly.

In Example 41, the subject matter of any one or more of Examples 39-40 optionally include wherein the inertia sensor measurements and time data points are integrated with environmental data to locate a rail anomaly, the environmental data indicating at least one of: temperature, humidity, barometric pressure, wind velocity, recent rainfall, soil moisture, or seismic data.

In Example 42, the subject matter of any one or more of Examples 39-41 optionally include wherein to detect the railcar anomaly, the machine is to detect an oscillation of the railcar that exceeds a predetermined threshold, and wherein the machine is to initiate an automatic braking subsystem to reduce the railcar speed.

In Example 43, the subject matter of any one or more of Examples 39-42 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a tie or ballast issue, wherein the evaluation is based on measurements including the amplitude of a dip from the rails with an inertia sensor, the speed of the train, weight of the railcar and train, and GPS location.

In Example 44, the subject matter of any one or more of Examples 39-43 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as rail warp, wherein the evaluation is based on measurements including repeated amplitude change, GPS location, train speed, and railcar and train weight.

In Example 45, the subject matter of any one or more of Examples 39-44 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as rail wear, wherein the evaluation is based on captured data from an inertia sensor and a camera is correlated with GPS location, railcar and train weight, and train speed.

In Example 46, the subject matter of any one or more of Examples 39-45 optionally include wherein the data indexed by GPS location and stored in a database is subsequently analyzed, to identify a rail anomaly at a GPS location of the railway.

In Example 47, the subject matter of any one or more of Examples 39-46 optionally include wherein the inertia sensor measurements indicate a side to side movement of the railcar, and wherein the sensor array is coupled via an attachment of the sensor array to respective trucks of the railcar.

In Example 48, the subject matter of any one or more of Examples 39-47 optionally include wherein the output device is a located in the lead engine car.

In Example 49, the subject matter of any one or more of Examples 39-48 optionally include wherein the sensor array comprises at least one of a GPS, accelerometer, gyroscope, or other inertia sensor.

In Example 50, the subject matter of any one or more of Examples 39-49 optionally include wherein the operating sensor measurements occur while the railcar is at track speed.

In Example 51, the subject matter of any one or more of Examples 39-50 optionally include wherein the operating sensor measurements occur when the railcar is at full carry weight.

In Example 52, the subject matter of any one or more of Examples 39-51 optionally include wherein the database includes the train's configuration and operating conditions, including at least one of: the size, length, or weight of the railcars on which the sensor array is located, the number of railcars in the train, the position within the train of the railcar on which the sensor is located, or the weight of railcars immediately in front of the railcar with the sensors.

In Example 53, the subject matter of any one or more of Examples 39-52 optionally include wherein the sensor array includes a camera to produce image data, and wherein the near real time measurements are derived from the image data.

In Example 54, the subject matter of any one or more of Examples 39-53 optionally include wherein the sensor array includes a microphone to produce audio data, and wherein the near real time measurements are derived from the audio data.

In Example 55, the subject matter of any one or more of Examples 39-54 optionally include wherein the machine transmits the real time measurement data via communications circuitry to a remote server.

In Example 56, the subject matter of Example 55 optionally includes wherein the remote server is configured to collect data from multiple trains and maintain a track condition database based on the data from the multiple trains.

In Example 57, the subject matter of any one or more of Examples 39-56 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a cross level issue based on an unevenness of the rails determination, wherein the measurements include the use of an inertia sensor to detect the force of a dip in the rails combined with the GPS for train location and speed, and the weight of the railcar and train.

In Example 58, the subject matter of any one or more of Examples 39-57 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as track gauge deflection based on a wide rail gauge determination, wherein the images taken from a camera are used to measure the rail gauge.

In Example 59, the subject matter of any one or more of Examples 39-58 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as track gauge deflection based on a narrow rail gauge determination, wherein the images taken from a camera are used to measure the rail gauge.

In Example 60, the subject matter of any one or more of Examples 39-59 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as track gauge deflection based on a wide rail gauge determination, wherein the track gauge is measured with an electronic measurement device.

in Example 61, the subject matter of any one or more of Examples 39-60 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as track gauge deflection based on a narrow rail gauge determination, wherein the track gauge is measured with an electronic measurement device.

In Example 62, the subject matter of any one or more of Examples 39-61 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a rail weld failure, wherein the evaluation is based on measurements including running a current through the rail while the train is under load and at track speed, and collecting the OPS location, ambient temperature, and induction measurement.

In Example 63, the subject matter of any one or more of Examples 39-62 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a stuck brake based on a non-moving axle or wheel from an unreleased brake determination, wherein measurements are collected from an optical tachometer in combination with either a GPS sensor or an inertia sensor.

in Example 64, the subject matter of any one or more of Examples 39-63 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a stuck wheel based on a non-moving axle or wheel determination, wherein measurements are collected from an optical tachometer in combination with either a GPS sensor or an inertia sensor.

in Example 65, the subject matter of any one or more of Examples 39-64 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include accelerometer or gyroscope sensor data with GPS position.

In Example 66, the subject matter of any one or more of Examples 39-65 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include calculated force from a spring connected to the truck combined with inertia sensor and GPS sensor data.

In Example 67, the subject matter of any one or more of Examples 39-66 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include light amplitude reflection from a light pointed at the truck with non-uniform reflective tape attached.

In Example 68, the subject matter of any one or more of Examples 39-67 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include a physical measure of a fixed point on the railcar in contact with a variable electronic sensor on the truck.

In Example 69, the subject matter of any one or more of Examples 39-68 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include a physical measure of a fixed point on the truck in contact with a variable electronic sensor on the railcar.

In Example 70, the subject matter of any one or more of Examples 39-69 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as overheating bearings, wherein the evaluation is based on measurements that include data from thermistor sensor, an infrared camera, a vibration sensor, or audio analytics.

In Example 71, the subject matter of any one or more of Examples 39-70 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a broken axle based on the two wheels of one axle rotating independently determination, wherein the measurement includes tachometers placed next to each wheel on the same axle.

In Example 72, the subject matter of any one or more of Examples 39-71 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as the wheels being out of balance based on a warped wheel determination, wherein the measurements include a wobble in either horizontal or vertical directions using an inertia sensor.

In Example 73, the subject matter of any one or more of Examples 39-72 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as a wheel with one or more flat spots, wherein the evaluation is based on measurements from the sensor array including an inertia sensor, an acoustic sensor, an imaging sensor, or a physical sensor.

In Example 74, the subject matter of any one or more of Examples 39-73 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as excessive wheel wear, wherein the evaluation is based on measurements that include collected images taken from an attached camera.

In Example 75, the subject matter of any one or more of Examples 39-74 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as truck hunting based on the truck pivoting while the train is on a straight segment of track, wherein the measurements include detected railcar oscillation by an inertia sensor.

In Example 76, the subject matter of any one or more of Examples 39-75 optionally include wherein the at least one machine readable medium further includes instructions to detect an anomaly as truck hunting based on the truck pivoting while the train is on a straight segment of track, wherein the measurements also comprise measuring wheel flange hits with an acoustic sensor.

Example 77 is a method for detecting railway anomalies, the method comprising: receiving, by a processor subsystem, measurements from a sensor array coupled to a railcar in a train; obtaining baseline measurements from the sensor array; obtaining, in near real time, measurements from the sensor array while the railcar is operating; detecting a railcar anomaly based a comparison between the baseline and operating measurements, wherein the comparison of baseline and operating measurements includes an evaluation, over a sequence of time data points, of inertia sensor measurements to detect abnormal railcar oscillation; storing the data indexed to a GPS location in a database; and transmitting an alert to an output device when an anomaly is detected based on the collected measurements from the railcars.

In Example 78, the subject matter of Example 77 optionally includes wherein the inertia sensor measurements and time data points are integrated with geographic location data to locate a rail anomaly.

In Example 79, the subject matter of any one or more of Examples 77-78 optionally include wherein the inertia sensor measurements and time data points are integrated with environmental data to locate a rail anomaly, the environmental data indicating at least one of: temperature, humidity, barometric pressure, wind velocity, recent rainfall, soil moisture, or seismic data.

In Example 80, the subject matter of any one or more of Examples 77-79 optionally include wherein to detect the railcar anomaly, the processor subsystem is to detect an oscillation of the railcar that exceeds a predetermined threshold, and wherein the processor subsystem is to initiate an automatic braking subsystem to reduce the railcar speed.

In Example 81, the subject matter of any one or more of Examples 77-80 optionally include wherein the method further includes detecting an anomaly as a tie or ballast issue, wherein the evaluation is based on measurements including the amplitude of a dip from the rails with an inertia sensor, the speed of the train, weight of the railcar and train, and GPS location.

In Example 82, the subject matter of any one or more of Examples 77-81 optionally include wherein the method further includes detecting an anomaly as rail warp, wherein the evaluation is based on measurements including repeated amplitude change, GPS location, train speed, and railcar and train weight.

In Example 83, the subject matter of any one or more of Examples 77-82 optionally include wherein the method further includes detecting an anomaly as rail wear, wherein the evaluation is based on captured data from an inertia sensor and a camera is correlated with GPS location, railcar and train weight, and train speed.

In Example 84, the subject matter of any one or more of Examples 77-83 optionally include wherein the data indexed by GPS location and stored in a database is subsequently analyzed, to identify a rail anomaly at a GPS location of the railway.

In Example 85, the subject matter of any one or more of Examples 77-84 optionally include wherein the inertia sensor measurements indicate a side to side movement of the railcar, and wherein the sensor array is coupled via an attachment of the sensor array to respective trucks of the railcar.

In Example 86, the subject matter of any one or more of Examples 77-85 optionally include wherein the output device is a located in the lead engine car.

In Example 87, the subject matter of any one or more of Examples 77-86 optionally include wherein the sensor array comprises at least one of a GPS, accelerometer, gyroscope, or other inertia sensor.

In Example 88, the subject matter of any one or more of Examples 77-87 optionally include wherein the operating sensor measurements occur while the railcar is at track speed.

In Example 89, the subject matter of any one or more of Examples 77-88 optionally include wherein the operating sensor measurements occur when the railcar is at full carry weight.

In Example 90, the subject matter of any one or more of Examples 77-89 optionally include wherein the database includes the train's configuration and operating conditions, including at least one of: the size, length, or weight of the railcars on which the sensor array is located, the number of railcars in the train, the position within the train of the railcar on which the sensor is located, or the weight of railcars immediately in front of the railcar with the sensors.

In Example 91, the subject matter of any one or more of Examples 77-90 optionally include wherein the sensor array includes a camera to produce image data, and wherein the near real time measurements are derived from the image data.

In Example 92, the subject matter of any one or more of Examples 77-91 optionally include wherein the sensor array includes a microphone to produce audio data, and wherein the near real time measurements are derived from the audio data.

In Example 93, the subject matter of any one or more of Examples 77-92 optionally include communications circuitry communicatively coupled to the processor subsystem, and wherein the, processor subsystem is to transmit the real time measurement data via the communications circuitry to a remote server.

In Example 94, the subject matter of Example 93 optionally includes wherein the remote server is configured to collect data from multiple trains and maintain a track condition database based on the data from the multiple trains.

In Example 95, the subject matter of any one or more of Examples 77-94 optionally include wherein the method further includes detecting an anomaly as a cross level issue based on an unevenness of the rails determination, wherein the measurements include the use of an inertia sensor to detect the force of a dip in the rails combined with the UPS for train location and speed, and the weight of the railcar and train.

In Example 96, the subject matter of any one or more of Examples 77-95 optionally include wherein the method further includes detecting an anomaly as track gauge deflection based on a wide rail gauge determination, wherein the images taken from a camera are used to measure the rail gauge.

In Example 97, the subject matter of any one or more of Examples 77-96 optionally include wherein the method further includes detecting an anomaly as track gauge deflection based on a narrow rail gauge determination, wherein the images taken from a camera are used to measure the rail gauge.

In Example 98, the subject matter of any one or more of Examples 77-97 optionally include wherein the method further includes detecting an anomaly as track gauge deflection based on a wide rail gauge determination, wherein the track gauge is measured with an electronic measurement device.

In Example 99, the subject matter of any one or more of Examples 77-98 optionally include wherein the method further includes detecting an anomaly as track gauge deflection based on a narrow rail gauge determination, wherein the track gauge is measured with an electronic measurement device.

In Example 100, the subject matter of any one or more of Examples 77-99 optionally include wherein the method further includes detecting an anomaly as a rail weld failure, wherein the evaluation is based on the measurements including running a current through the rail while the train is under load and at track speed, and collecting the GPS location, ambient temperature, and induction measurement.

In Example 101, the subject matter of any one or more of Examples 77-100 optionally include wherein the method further includes detecting an anomaly as a stuck brake based on a non-moving axle or wheel from an unreleased brake determination, wherein measurements are collected from an optical tachometer in combination with either a GPS sensor or an inertia sensor.

In Example 102, the subject matter of any one or more of Examples 77-101 optionally include wherein the method further includes detecting an anomaly as a stuck wheel based on a non-moving axle or wheel determination, wherein measurements are collected from an optical tachometer in combination with either a UPS sensor or an inertia sensor.

In Example 103, the subject matter of any one or more of Examples 77-102 optionally include wherein the method further includes detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include accelerometer or gyroscope sensor data with GPS position.

In Example 104, the subject matter of any one or more of Examples 77-103 optionally include wherein the method further includes detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include calculated force from a spring connected to the truck combined with inertia sensor and GPS sensor data.

In Example 105, the subject matter of any one or more of Examples 77-104 optionally include wherein the method further includes detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include light amplitude reflection from a light pointed at the truck with non-uniform reflective tape attached.

In Example 106, the subject matter of any one or more of Examples 77-105 optionally include wherein the method further includes detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include a physical measure of a fixed point on the railcar in contact with a variable electronic sensor on the truck.

In Example 107, the subject matter of any one or more of Examples 77-106 optionally include wherein the method further includes detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include a physical measure of a fixed point on the truck in contact with a variable electronic sensor on the railcar.

In Example 108, the subject matter of any one or more of Examples 77-107 optionally include wherein the method further includes detecting an anomaly as overheating bearings, wherein the evaluation is based on measurements that include data from thermistor sensor, an infrared camera, a vibration sensor, or audio analytics.

In Example 109, the subject matter of any one or more of Examples 77-108 optionally include wherein the method further includes detecting an anomaly as a broken axle based on the two wheels of one axle rotating independently determination, wherein the measurement includes tachometers placed next to each wheel on the same axle.

In Example 110, the subject matter of any one or more of Examples 77-109 optionally include wherein the method further includes detecting an anomaly as the wheels being out of balance based on a warped wheel determination, wherein the measurements include a wobble in either horizontal or vertical directions using an inertia sensor.

In Example 111, the subject matter of any one or more of Examples 77-110 optionally include wherein the method further includes detecting an anomaly as a wheel with one or more flat spots, wherein the evaluation is based on measurements from the sensor array including an inertia sensor, an acoustic sensor, an imaging sensor, or a physical sensor.

In Example 112, the subject matter of any one or more of Examples 77-111 optionally include wherein the method further includes detecting an anomaly as excessive wheel wear, wherein the evaluation is based on measurements that include collected images taken from an attached camera.

In Example 113, the subject matter of any one or more of Examples 77-112 optionally include wherein the method further includes detecting an anomaly as truck hunting based on the truck pivoting while the train is on a straight segment of track, wherein the measurements include detected railcar oscillation by an inertia sensor.

In Example 114, the subject matter of any one or more of Examples 77-113 optionally include wherein the method further includes detecting an anomaly as truck hunting based on the truck pivoting while the train is on a straight segment of track, wherein the measurements also comprise measuring wheel flange hits with an acoustic sensor.

Example 115 is at least one machine readable medium including instructions, which when executed by an electronic device, cause the computing system to perform any of the methods of Examples 77-114.

Example 116 is an apparatus comprising means for performing any of the methods of Examples 77-114.

Example 117 is a system for detecting railway anomalies, the system comprising: means for receiving measurements from a sensor array coupled a railcar in a train; means for obtaining baseline measurements from the sensor array; means for obtaining, in near real time, measurements from the sensor array while the railcar is operating; means for detecting a railcar anomaly based a comparison between the baseline and operating measurements, wherein the comparison of baseline and operating measurements includes an evaluation, over a sequence of time data points, of inertia sensor measurements to detect abnormal railcar oscillation; means for storing the data indexed to a GPS location in a database; and means for transmitting an alert to an output device when an anomaly is detected based on the collected measurements from the railcars.

In Example 118, the subject matter of Example 117 optionally includes wherein the inertia sensor measurements and time data points are integrated with geographic location data to locate a rail anomaly.

In Example 119, the subject matter of any one or more of Examples 117-118 optionally include wherein the inertia sensor measurements and time data points are integrated with environmental data to locate a rail anomaly, the environmental data indicating at least one of: temperature, humidity, barometric pressure, wind velocity, recent rainfall, soil moisture, or seismic data.

In Example 120, the subject matter of any one or more of Examples 117-119 optionally include wherein to detect the railcar anomaly, the system is to detect an oscillation of the railcar that exceeds a predetermined threshold, and wherein the system is to initiate an automatic braking subsystem to reduce the railcar speed.

In Example 121, the subject matter of any one or more of Examples 117-120 optionally include wherein the system further includes a means for detecting an anomaly as a tie or ballast issue, wherein the evaluation is based on measurements including the amplitude of a dip from the rails with an inertia sensor, the speed of the train, weight of the railcar and train, and GPS location.

In Example 122, the subject matter of any one or more of Examples 117-121 optionally include wherein the system further includes a means for detecting an anomaly as rail warp, wherein the evaluation is based on measurements including repeated amplitude change, GPS location, train speed, and railcar and train weight.

In Example 123, the subject matter of any one or more of Examples 117-122 optionally include wherein the system further includes a means for detecting an anomaly as rail wear, wherein the evaluation is based on captured data from an inertia sensor and a camera is correlated with GPS location, railcar and train weight, and train speed.

In Example 124, the subject matter of any one or more of Examples 117-123 optionally include wherein the data indexed by UPS location and stored in a database is subsequently analyzed, to identify a rail anomaly at a UPS location of the railway.

In Example 125, the subject matter of any one or more of Examples 117-124 optionally include wherein the output device is a located in the lead engine car.

In Example 126, the subject matter of any one or more of Examples 117-125 optionally include wherein the sensor array comprises at least one of a UPS, accelerometer, gyroscope, or other inertia sensor.

In Example 127, the subject matter of any one or more of Examples 117-126 optionally include wherein the operating sensor measurements occur while the railcar is at track speed.

In Example 128, the subject matter of any one or more of Examples 117-127 optionally include wherein the operating sensor measurements occur when the railcar is at full carry weight.

In Example 129, the subject matter of any one or more of Examples 117-128 optionally include wherein the database includes the train's configuration and operating conditions, including at least one of: the size, length, or weight of the railcars on which the sensor array is located, the number of railcars in the train, the position within the train of the railcar on which the sensor is located, or the weight of railcars immediately in front of the railcar with the sensors.

In Example 130, the subject matter of any one or more of Examples 117-129 optionally include wherein the sensor array includes a camera to produce image data, and wherein the near real time measurements are derived from the image data.

In Example 131, the subject matter of any one or more of Examples 117-130 optionally include wherein the sensor array includes a microphone to produce audio data, and wherein the near real time measurements are derived from the audio data.

In Example 132, the subject matter of any one or more of Examples 117-131 optionally include means for transmitting the real time measurement data via the communications circuitry to a remote server.

In Example 133, the subject matter of Example 132 optionally includes wherein the remote server is configured to collect data from multiple trains and maintain a track condition database based on the data from the multiple trains.

In Example 134, the subject matter of any one or more of Examples 117-133 optionally include wherein the system further includes a means for detecting an anomaly as a cross level issue based on an unevenness of the rails determination, wherein the measurements include the use of an inertia sensor to detect the force of a dip in the rails combined with the GPS for train location and speed, and the weight of the railcar and train.

In Example 135, the subject matter of any one or more of Examples 117-134 optionally include wherein the system further includes a means for detecting an anomaly as track gauge deflection based on a wide rail gauge determination, wherein the images taken from a camera are used to measure the rail gauge.

In Example 136, the subject matter of any one or more of Examples 117-135 optionally include wherein the system further includes a means for detecting an anomaly as track gauge deflection based on a narrow rail gauge determination, wherein the images taken from a camera are used to measure the rail gauge.

In Example 137, the subject matter of any one or more of Examples 117-136 optionally include wherein the system further includes a means for detecting an anomaly as track gauge deflection based on a wide rail gauge determination, wherein the track gauge is measured with an electronic measurement device.

In Example 138, the subject matter of any one or more of Examples 117-137 optionally include wherein the system further includes a means for detecting an anomaly as track gauge deflection based on a narrow rail gauge determination, wherein the track gauge is measured with an electronic measurement device.

In Example 1.39, the subject matter of any one or more of Examples 117-138 optionally include wherein the system further includes a means for detecting an anomaly as a rail weld failure, wherein the evaluation is based on measurements including running a current through the rail while the train is under load and at track speed, and collecting the UPS location, ambient temperature, and induction measurement.

In Example 140, the subject matter of any one or more of Examples 117-139 optionally include wherein the system further includes a means for detecting an anomaly as a stuck brake based on a non-moving axle or wheel from an unreleased brake determination, wherein measurements are collected from an optical tachometer in combination with either a UPS sensor or an inertia sensor.

In Example 141, the subject matter of any one or more of Examples 117-140 optionally include wherein the system further includes a means for detecting an anomaly as a stuck wheel based on a non-moving axle or wheel determination, wherein measurements are collected from an optical tachometer in combination with either a UPS sensor or an inertia sensor.

In Example 142, the subject matter of any one or more of Examples 117-141 optionally include wherein the system further includes a means for detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include accelerometer or gyroscope sensor data with UPS position.

In Example 143, the subject matter of any one or more of Examples 117-142 optionally include wherein the system further includes a means for detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include calculated force from a spring connected to the truck combined with inertia sensor and UPS sensor data.

In Example 144, the subject matter of any one or more of Examples 117-143 optionally include wherein the system further includes a means for detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include light amplitude reflection from a light pointed at the truck with non-uniform reflective tape attached.

In Example 145, the subject matter of any one or more of Examples 117-144 optionally include wherein the system further includes a means for detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include a physical measure of a fixed point on the railcar in contact with a variable electronic sensor on the truck.

In Example 146, the subject matter of any one or more of Examples 117-145 optionally include wherein the system further includes a means for detecting an anomaly as a stuck truck center pin based on a non-pivoting truck determination, wherein the measurements include a physical measure of a fixed point on the truck in contact with a variable electronic sensor on the railcar.

In Example 147, the subject matter of any one or more of Examples 117-146 optionally include wherein the system further includes a means for detecting an anomaly as overheating hearings, wherein the evaluation is based on measurements that include data from thermistor sensor, an infrared camera, a vibration sensor, or audio analytics.

In Example 148, the subject matter of any one or more of Examples 117-147 optionally include wherein the system further includes a means for detecting an anomaly as a broken axle based on the two wheels of one axle rotating independently determination, wherein the measurement includes tachometers placed next to each wheel on the same axle.

In Example 149, the subject matter of any one or more of Examples 117-148 optionally include wherein the system further includes a means for detecting an anomaly as the wheels being out of balance based on a warped wheel determination, wherein the measurements include a wobble in either horizontal or vertical directions using an inertia sensor.

In Example 150, the subject matter of any one or more of Examples 117-149 optionally include wherein the system further includes a means for detecting an anomaly as a wheel with one or more flat spots, wherein the evaluation is based on measurements from the sensor array including an inertia sensor, an acoustic sensor, an imaging sensor, or a physical sensor.

In Example 151, the subject matter of any one or more of Examples 117-150 optionally include wherein the system further includes a means for detecting an anomaly as excessive wheel wear, wherein the evaluation is based on measurements that include collected images taken from an attached camera.

In Example 152, the subject matter of any one or more of Examples 117-151 optionally include wherein the system further includes a means for detecting an anomaly as truck hunting based on the truck pivoting while the train is on a straight segment of track, wherein the measurements include detected railcar oscillation by an inertia sensor.

In Example 153, the subject matter of any one or more of Examples 117-152 optionally include wherein the system further includes a means for detecting an anomaly as truck hunting based on the truck pivoting while the train is on a straight segment of track, wherein the measurements also comprise measuring wheel flange hits with an acoustic sensor.

The above detailed description includes references to the accompanying drawings, which form a part of the detailed description. The drawings show, by way of illustration, specific embodiments that may be practiced. These embodiments are also referred to herein as “examples.” Such examples may include elements in addition to those shown or described. However, the present inventors also contemplate examples in which only those elements shown or described are provided. Moreover, the present inventors also contemplate examples using any combination or permutation of those elements shown or described (or one or more aspects thereof), either with respect to a particular example (or one or more aspects thereof), or with respect to other examples (or one or more aspects thereof) shown or described herein.

All publications, patents, and patent documents referred to in this document are incorporated by reference herein in their entirety, as though individually incorporated by reference. In the event of inconsistent usages between this document and those documents so incorporated by reference, the usage in the incorporated reference(s) should be considered supplementary to that of this document; for irreconcilable inconsistencies, the usage in this document controls.

In this document, the terms “a” or “an” are used, as is common in patent documents, to include one or more than one, independent of any other instances or usages of “at least one” or “one or more.” in this document, the term “or” is used to refer to a nonexclusive or, such that “A or B” includes “A but not B,” “B but not A,” and “A and B,” unless otherwise indicated. In the appended claims, the terms “including” and “in which” are used as the plain-English equivalents of the respective terms “comprising” and “wherein.” Also, in the following claims, the terms “including” and “comprising” are open-ended, that is, a system, device, article, or process that includes elements in addition to those listed after such a term in a claim are still deemed to fall within the scope of that claim. Moreover, in the following claims, the terms “first,” “second,” and “third,” etc. are used merely as labels, and are not intended to impose numerical requirements on their objects.

The above description is intended to he illustrative, and not restrictive. For example, the above-described examples (or one or more aspects thereof) may be used in combination with each other. Other embodiments may be used, such as by one of ordinary skill in the art upon reviewing the above description. The Abstract is to allow the reader to quickly ascertain the nature of the technical disclosure and is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims. Also, in the above Detailed Description, various features may be grouped together to streamline the disclosure. This should not be interpreted as intending that an unclaimed disclosed feature is essential to any claim. Rather, inventive subject matter may lie in less than all features of a particular disclosed embodiment. Thus, the following claims are hereby incorporated into the Detailed Description, with each claim standing on its own as a separate embodiment. The scope of the embodiments should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims

1. A system to detect railway anomalies, the system comprising:

a processor subsystem; and

a memory including instructions that, when executed by the processor subsystem, cause the processor subsystem to:

receive measurements from a sensor array coupled to a railcar in a train;

obtain baseline measurements from the sensor array;

obtain, in near real time, measurements from the sensor array while the railcar is operating;

detect a railcar anomaly based a comparison between the baseline and operating measurements, wherein the comparison of baseline and operating measurements includes an evaluation, over a sequence of time data points, inertia sensor measurements to detect abnormal railcar oscillation;

store the data indexed to a GPS location in a database; and

transmit an alert to an output device when an anomaly is detected based on the collected measurements from the railcars.

2. The system of claim 1, wherein the inertia sensor measurements and time data points are correlated with geographic location data to locate a rail anomaly.

3. The system of claim 1, wherein to detect the railcar anomaly, the processor subsystem is to detect an oscillation of the railcar that exceeds a predetermined threshold, and wherein the processor subsystem is to initiate an automatic braking subsystem to reduce the railcar speed.

4. The system of claim 1, wherein the memory further includes instructions to detect an anomaly as a tie or ballast issue, wherein the evaluation is based on measurements including the amplitude of a dip from the rails with an inertia sensor, the speed of the train, weight of the railcar and train, and GPS location.

5. The system of claim 1, wherein the memory further includes instructions to detect an anomaly as rail warp, wherein the evaluation is based on measurements including repeated amplitude change, UPS location, train speed, and railcar and train weight.

6. The system of claim 1, wherein the memory further includes instructions to detect an anomaly as rail wear, wherein the determination is based on captured data from an inertia sensor and a camera is correlated with GPS location, railcar and train weight, and train speed.

7. The system of claim 1, wherein the data indexed by UPS location and stored in a database is subsequently analyzed, to identify a rail anomaly at a UPS location of the railway.

8. The system of claim 1, wherein the inertia sensor measurements indicate a side to side movement of the railcar, and wherein the sensor array is coupled via an attachment of the sensor array to respective trucks of the railcar.

9. At least one machine readable medium including instructions to detect railway anomalies that, when executed by a machine, cause the machine to:

receive measurements from a sensor array coupled to a railcar in a train;

obtain baseline measurements from the sensor array;

obtain, in near real time, measurements from the sensor array while the railcar is operating;

detect a railcar anomaly based a comparison between the baseline and operating measurements, wherein the comparison of baseline and operating measurements includes an evaluation, over a sequence of time data points, of inertia sensor measurements to detect abnormal railcar oscillation;

store the data indexed to a GPS location in a database; and

transmit an alert to an output device when an anomaly is detected based on the collected measurements from the railcars.

10. The at least one machine readable medium of claim 9, wherein the inertia sensor measurements and time data points are integrated with geographic location data to locate a rail anomaly.

11. The at least one machine readable medium of claim 9, wherein to detect the railcar anomaly, the machine is further to detect an oscillation of the railcar that exceeds a predetermined threshold, and wherein the machine is further to initiate an automatic braking subsystem to reduce the railcar speed.

12. The at least one machine readable medium of claim 9, wherein the at least one machine readable medium further includes instructions to detect an anomaly as a tie or ballast issue, wherein the evaluation is based on measurements including the amplitude of a dip from the rails with an inertia sensor, the speed of the train, weight of the railcar and train, and GPS location.

13. The at least one machine readable medium of claim 9, wherein the at least one machine readable medium further includes instructions to detect an anomaly as rail warp, wherein the evaluation is based on measurements including repeated amplitude change, GPS location, train speed, and railcar and train weight.

14. The at least one machine readable medium of claim 9, wherein the at least one machine readable medium further includes instructions to detect an anomaly as rail wear, wherein the evaluation is based on captured data from an inertia sensor and a camera is correlated with GPS location, railcar and train weight, and train speed.

15. The at least one machine readable medium of claim 9, wherein the data indexed by UPS location and stored in a database is subsequently analyzed, to identify a rail anomaly at a GPS location of the railway.

16. The at least one machine readable medium of claim 9, wherein the inertia sensor measurements indicate a side to side movement of the railcar, and wherein the sensor array is coupled via an attachment of the sensor array to respective trucks of the railcar.

17. A method for detecting railway anomalies, the method comprising:

receiving, by a processor subsystem, measurements from a sensor array coupled to a railcar in a train;

obtaining baseline measurements from the sensor array;

obtaining, in near real time, measurements from the sensor array while the railcar is operating;

detecting a railcar anomaly based a comparison between the baseline and operating measurements, wherein the comparison of baseline and operating measurements includes an evaluation, over a sequence of time data points, of inertia sensor measurements to detect abnormal railcar oscillation;

storing the data indexed to a GPS location in a database; and

transmitting an alert to an output device when an anomaly is detected based on the collected measurements from the railcars.

18. The method of claim 17, wherein the inertia sensor measurements and time data points are integrated with geographic location data to locate a rail anomaly.

19. The method of claim 17, wherein to detect the railcar anomaly, the processor subsystem is to detect an oscillation of the railcar that exceeds a predetermined threshold, and wherein the processor subsystem is to initiate an automatic braking subsystem to reduce the railcar speed.

20. The method of claim 17, wherein the method further includes detecting an anomaly as a tie or ballast issue, wherein the evaluation is based on measurements including the amplitude of a dip from the rails with an inertia sensor, the speed of the train, weight of the railcar and train, and GPS location.

21. The method of claim 17, wherein the method further includes detecting an anomaly as rail warp, wherein the evaluation is based on measurements including repeated amplitude change, GPS location, train speed, and railcar and train weight.

22. The method of claim 17, wherein the method further includes detecting an anomaly as rail wear, wherein the evaluation is based on captured data from an inertia sensor and a camera is correlated with GPS location, railcar and train weight, and train speed.

23. The method of claim 17, wherein the data indexed by GPS location and stored in a database is subsequently analyzed, to identify a rail anomaly at a GPS location of the railway.

24. The method of claim 17, wherein the inertia sensor measurements indicate a side to side movement of the railcar, and wherein the sensor array is coupled via an attachment of the sensor array to respective trucks of the railcar.