PREDICTION AND IDENTIFICATION OF POTENTIAL RAILWAY SYSTEM ANOMALIES

BACKGROUND

Rail transport transfers people and goods on wheeled vehicles (e.g., locomotives and other rolling stock) running on rails. The rails are incorporated in tracks and the wheeled vehicles are known as railway vehicles. In contrast to automotive vehicle travel, railway vehicles are guided by the rails on which they run.

SUMMARY

Over time, both the railway vehicles and the tracks may experience degradation or other anomalies. For example, the railway vehicle may experience bearing wear, wheel defects, and the like. The tracks themselves may also experience wear, warping, cracking, and/or misalignment. Anomalies in the wheels of the railway vehicle or the tracks may result in damage to the railway vehicle or track.

It would be beneficial if railway vehicle (for example, wheel, axle, chassis, etc.) and track anomalies could be detected early to help avoid damage. Accordingly, examples and aspects described herein provide, among other things, a system and a method for utilizing a machine and deep learning system to predict anomalies of a railway vehicle and track via a vibration sensor and, in some instances, using collected data from a network of railway vehicles.

One example provides a system for predicting anomalies in a wheel system of a railway vehicle and a track upon which the railway vehicle is run. The system includes a vibration sensor positioned to sense vibrations of the wheel system and an electronic processor communicatively coupled to the vibration sensor. The electronic processor configured is to determine a speed of the railway vehicle. The electronic processor configured is to determine whether the speed exceeds a predetermined speed threshold. The electronic processor configured is to obtain, in response to the speed exceeding the predetermined speed threshold, a vibration measurement. The electronic processor configured is to determine, from the vibration measurement, a classified vibration level. The electronic processor configured is to determine whether the classified vibration level is indicative of an anomaly. The electronic processor configured is to derive, in response to the classified vibration level being indicative of the anomaly, a dominant vibration frequency of the vibration measurement. The electronic processor configured is to perform a comparison between the dominant vibration frequency and a predetermined frequency threshold. The electronic processor configured is to identify, based on the comparison and a reference classified vibration level, the anomaly existing within either or both of the railway vehicle wheel system and the track. The electronic processor configured is to perform a mitigation action in response to identifying the anomaly.

Another example provides a method for predicting anomalies in a wheel system of a railway vehicle and a track upon which the railway vehicle is run. The method includes determining, with an electronic processor, a speed of the railway vehicle. The method includes determining whether the speed exceeds a predetermined speed threshold. The method includes obtaining, in response to the speed exceeding the predetermined speed threshold, a vibration measurement from a vibration sensor positioned to sense vibrations of the wheel system. The method includes determining from the vibration measurement, a classified vibration level. The method includes determining whether the classified vibration level is indicative of an anomaly. The method includes deriving, in response to the classified vibration level being indicative of the anomaly, a dominant vibration frequency of the vibration measurement. The method includes performing a comparison between the dominant vibration frequency and a predetermined frequency threshold. The method includes identifying, based on the comparison and a reference classified vibration level, the anomaly existing within either or both of the railway vehicle wheel system and the track. The method includes performing a mitigation action in response to identifying the anomaly.

Other aspects, features, and examples will become apparent by consideration of the detailed description and accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying figures, where like reference numerals refer to identical or functionally similar elements throughout the separate views, together with the detailed description below, are incorporated in and form part of the specification, and serve to further illustrate examples and aspects of concepts that include the claimed subject matter and explain various principles and advantages of various aspects and examples.

FIG. 1 is a block diagram of a railway vehicle wheel anomaly detection system, in accordance with some examples.

FIGS. 2A and 2B illustrate a railway vehicle wheel system of the system of FIG. 1, in accordance with some examples.

FIG. 3 schematically illustrates an electronic controller of the system of FIG. 1, in accordance with some examples.

FIG. 4 is a flowchart of an example method for predicting an anomaly within a railway vehicle of FIG. 1 and a track.

FIG. 5A is a flowchart of an example method for identifying an anomaly performed by the system of FIG. 1.

FIG. 5B is a flowchart of an example method for identifying an anomaly performed by the system of FIG. 1.

FIG. 5C is a flowchart of an example method for identifying an anomaly performed by the system of FIG. 1.

FIG. 6 a vibration signal classifier for predicting a vertical geometric defect in a railway track.

FIG. 7A is an example of a vertical geometric defect in a railway track.

FIG. 7B is an example of a vertical geometric defect in a railway track.

FIG. 7C is an example of a horizontal geometric defect in a railway track.

FIG. 7D is an example of a horizontal geometric defect in a railway track.

Skilled artisans will appreciate that elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. For example, the dimensions of some of the elements in the figures may be exaggerated relative to other elements to help to improve understanding of examples and aspects.

The apparatus and method components have been represented where appropriate by conventional symbols in the drawings, showing only those specific details that are pertinent to understanding the examples so as not to obscure the disclosure with details that will be readily apparent to those of ordinary skill in the art having the benefit of the description herein.

DETAILED DESCRIPTION

Before any embodiments of the invention are explained in detail, it is to be understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways.

For ease of description, some or all of the example systems presented herein are illustrated with a single exemplar of each of its component parts. Some examples may not describe or illustrate all components of the systems. Other examples may include more or fewer of each of the illustrated components, may combine some components, or may include additional or alternative components.

It will be appreciated that some embodiments may be comprised of one or more generic or specialized processors (or “processing devices”) such as microprocessors, digital signal processors, customized processors and field programmable gate arrays (FPGAs) and unique stored program instructions (including both software and firmware) that control the one or more processors to implement, in conjunction with certain non-processor circuits, some, most, or all of the functions of the method and/or apparatus described herein. Alternatively, some or all functions could be implemented by a state machine that has no stored program instructions, or in one or more application specific integrated circuits (ASICs), in which each function or some combinations of certain of the functions are implemented as custom logic. Of course, a combination of the two approaches could be used.

Moreover, an embodiment can be implemented as a computer-readable storage medium having computer readable code stored thereon for programming a computer (e.g., comprising a processor) to perform a method as described and claimed herein. Examples of such computer-readable storage mediums include, but are not limited to, a hard disk, a CD-ROM, an optical storage device, a magnetic storage device, a ROM (Read Only Memory), a PROM (Programmable Read Only Memory), an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically Erasable Programmable Read Only Memory) and a Flash memory. Further, it is expected that one of ordinary skill, notwithstanding possibly significant effort and many design choices motivated by, for example, available time, current technology, and economic considerations, when guided by the concepts and principles disclosed herein will be readily capable of generating such software instructions and programs and ICs with minimal experimentation.

It should be understood that although certain drawings illustrate hardware and software located within particular devices, these depictions are for illustrative purposes only. In some examples, the illustrated components may be combined or divided into separate software, firmware and/or hardware. For example, instead of being located within and performed by a single electronic processor, logic and processing may be distributed among multiple electronic processors. Regardless of how they are combined or divided, hardware and software components may be located on the same computing device or may be distributed among multiple different devices. Accordingly, in the claims, if an apparatus, method, or system is claimed, for example, as including a controller, control unit, electronic processor, computing device, logic element, module, or other element configured in a certain manner, for example, to perform multiple functions, the claim or claim element should be interpreted as meaning one or more of such elements where any one of the one or more elements is configured as claimed, for example, to make any one or more of the recited multiple functions, such that the multiple elements, as a set in a collective nature, perform the multiple functions.

Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of “including,” “comprising” or “having” and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. The terms “mounted,” “connected” and “coupled” are used broadly and encompass both direct and indirect mounting, connecting and coupling. Further, “connected” and “coupled” are not restricted to physical or mechanical connections or couplings, and can include electrical connections or couplings, whether direct or indirect. The term “predetermined” means specified prior to an event. Also, electronic communications and notifications may be performed using any known means including direct connections (e.g., wired or optical), wireless connections, or other communication.

Furthermore, the systems and methods described herein may be used independently (e.g., as alternatives) or in various combinations. The systems and methods described herein may be used with any kind of railway vehicle operated on rails, including both powered and unpowered vehicles. Examples include, a locomotive, a freight car, a passenger car, a caboose, and the others. Powered railway vehicles described herein may be capable of operating autonomously, being controlled manually by a driver, or operate via combination of manual and autonomous control (for example, autonomous control under limited conditions or for limited functions. The term “driver,” as used herein, generally refers to an occupant of who is seated in the driver's position, operates the controls of the vehicle while in a manual mode, or provides control input to the vehicle to influence the operation of the vehicle (for example, remotely controlling the vehicle).

FIG. 1 illustrates a railway vehicle anomaly detection system 100 for detecting/predicting an anomaly in a railway vehicle wheel system 200 of a railway vehicle 102 according to some embodiments. The railway vehicle 102 may be any type of vehicle that is run along a railway track. The railway vehicle 102 may be powered or unpowered.

In the example illustrated, the railway vehicle anomaly detection system 100 includes an electronic controller 300 and a vibration sensor 104. The system 100 may also include one or more additional sensors 106. In some instances, the system 100 also includes a geo-location system 108.

In some instances, the system 100 is wirelessly communicatively coupled to one or more other railway vehicle anomaly detection systems 110 of one or more other railway vehicles (not shown) via a communications network 112. The other railway vehicle anomaly detection systems 110 are configured similarly to the system 100 and perform similar functions to the system 100 as described herein for a respective railway vehicle. The components of the system 100, along with other various modules and components are electrically coupled to each other by or through one or more control or data buses, which enable communication therebetween. For example, in some instances, the components of the system 100 communicate according to a Controller Area Network (CAN™) protocol. In some instances, one or more of the buses include an Ethernet™, a FlexRay™ communications bus, or another suitable wired bus. In alternative instances, some or all of the components of the system 100 may be communicatively coupled using suitable wireless modalities (for example, Bluetooth™). For ease of description, the system 100 illustrated in FIG. 1 includes one of each of the foregoing components. Alternative instances may include one or more of each component or may exclude or combine some components.

The electronic controller 300 (described more particularly below with respect to FIG. 3) evaluates information from the vibration sensor 104 (and the sensors 106) to detect/predict a potential anomaly of the railway vehicle 102 (for example, within the railway vehicle wheelset 200).

In some embodiments, the system 100 also includes a remote server 114 communicatively coupled to the electronic controller 300 of the railway vehicle 102 via the communications network 112. The server 114 may include components similar to those of the electronic controller 300 (described in more detail below with respect to FIG. 3). In some embodiments, some or all of the components and functionality of the electronic controller 300 described herein may be integrated into the remote server 114.

FIGS. 2A and 2B illustrate an example of a railway vehicle wheel system 200. FIG. 2A focuses on a portion of the wheel system 200, in particular a wheelset 202 coupled to a truck 208. Variations of the wheel system 200 include one or more wheelsets 202. Each wheelset 202 includes two wheels 204 mounted using wheel bearings (not shown) on an axle (for example, axle 206) opposite to one another. Each wheelset 202 is carried by a truck 208, which is attached to a railway vehicle (not shown in FIGS. 2A and 2B). The truck 208 may include one or more wheelsets and may be removably attached to the railway vehicle. For example, as illustrated in FIG. 2B, an example truck 222 includes two wheelsets 222, 224, which are coupled by a chassis 226. The chassis 226 includes a suspension 228. The suspension 228 of the truck 220 allows the truck to absorb shocks between the chassis 226 and the railway vehicle. In the example illustrated, the suspension 228 includes coil springs. Other examples of the suspension 228 include leaf springs and airbags. In some aspects, the suspension 228 may be controllably adjustable (for example, via commands issued from the electronic controller 300). The wheel system 200, in some embodiments, may include an electric drive system (for example, a truck with an electric drive), which is not illustrated. In such embodiments, systems and methods described herein may be configured to predict and/or identify anomalies in those electric drives (for example, bearing wear, torque ripple, and the like).

The railway vehicle 102 includes at least two wheelsets for moving along a pair of respective tracks 210. Both wheels of each wheelset may each include at least one vibration sensor 104 mounted at the respective wheel. For ease of description, anomaly detection/prediction methods are described herein in terms of a single wheel 204, respective track 210, and a single vibration sensor 104. It should be understood, however, that the controller 300 may perform similar methods for any other wheel of the railway vehicle 102.

Returning to FIG. 1, the vibration sensor 104 (and, in some instances, the other sensors 106) determine one or more attributes of the railway vehicle 102 and communicate information regarding those attributes to the other components of the system 100 using, for example, electrical signals. The vibration sensor 104 is configured to detect one or more vibrations within the railway vehicle wheel system 200. In some instances, the vibration sensor 104 is a transducer capable of sensing vibrations in a vehicle component, converting the vibrations to electrical signals, and transmitting the electrical signals to the electronic controller 300 as sensor information. The sensor 104 is positioned such that it detects vibrations where a wheel (for example, the wheel 204) and a respective end of the wheel axle (for example, the wheel axle 206) where the wheel and axle are coupled together (for example, as shown in FIG. 2A). In some instances, the vibration sensor 104 may be integrated into the wheel 204 or the wheel axle 206. In some examples, more than one of the vibration sensor 104 may be utilized in the systems and methods described herein. For example, as shown in FIG. 2A, a second vibration sensor 104 may be positioned at the wheel 204. In such instances, the second vibration sensor 104 may be configured to measure vibrations in the same or a similar direction as the other vibration sensor 104. Alternatively or additionally, in some instances, each of the vibration sensors 104 are configured to measure vibrations in different directions. For example, one of the vibration sensors 104 may be positioned to measure vibrations in a vertical direction with respect to the wheel axle 206 whereas a second vibration sensor 104 may be positioned to measure vibrations in a horizontal direction with respect to the wheel axle 206.

In some instances, the vibration sensor 104 is integrated into another sensor of the railway vehicle 102 (for example, one or more of the other sensors 106). For example, the sensor 104 may be integrated into an accelerometer or a speed sensor. In some instances, the sensor 104 is part of a strain gauge, an eddy-current sensor, a gyroscope, a microphone, or another suitable sensor. In some instances, multiple sensors are used, for example, mounted at different points on the railway vehicle 102 (for example, proximate to the wheel 204).

As described herein, the electronic controller 300 processes the electrical signals received from the vibration sensor 104 to produce vibration signal information related to the railway vehicle wheel system 200, which may be analyzed to determine/identify a potential anomaly that is causing the particular vibration. In some instances, the sensor 104 includes on-board signal processing circuitry, which produces and transmits sensor information including vibration measurements to the electronic controller 300 for processing. The electronic controller 300 receives and interpret the signals received from the vibration sensor 104 (and, in some instances, one or more of the other sensors 106) to automatically detect/predict and identify an anomaly of the railway vehicle 102.

In addition to the vibration sensor 104, the system 100 may include one or more other sensors 106. The sensors 106 determine one or more attributes of the railway vehicle 102 and its surrounding environment and communicate information regarding those attributes to the other components of the system 100 using, for example, electrical signals. The railway vehicle attributes include, for example, the position of the vehicle or portions or components of the railway vehicle 102, the movement of the railway vehicle 102 or portions or components of the railway vehicle 102, the forces acting on the railway vehicle 102 or portions or components of the railway vehicle 102, a temperature of one or more components of the railway vehicle 102 or of the environment surrounding the railway vehicle 102, vehicle speed, and the like. The sensors 106 may include, for example, a speed sensor 107A, an ambient temperature sensor 107B, an additional vibration sensor 107C, and a vehicle load sensor 107D, as illustrated in FIG. 1.

In some instances, the system 100 includes, in addition to the sensors 106, a geo-location system 108. The geo-location system 108 receives radio frequency signals from orbiting satellites using one or more antennas and receivers (not shown). The geo-location system 108 determines geo-spatial positioning (i.e., latitude, longitude, altitude, and speed) for the railway vehicle 102 based on the received radio frequency signals. The geo-location system 108 communicates this positioning information to the electronic controller 300. The geo-location system 108 may be or include, for example, a global navigation satellite system (GNSS), a global positioning system (GPS), or another type of satellite-based location system. The electronic controller 300 may use this information in conjunction with or in place of information received from some of the sensors 106 (for example, a speed of the railway vehicle 102 may be determined from information from the geo-location system 108 instead of from a speed sensor).

The communications network 112 is a communications network including wireless connections, wired connections, or combinations of both. The communications network 112 may be implemented using a wide area network, for example, the Internet, a Long-Term Evolution (LTE) network, a 4G network, 5G network and one or more local area networks, for example, a Bluetooth™ network or Wi-Fi network, and combinations or derivatives thereof.

The example illustrated in FIG. 1 provides but one example of the components and connections of the system 100. However, these components and connections may be constructed in other ways than those illustrated and described herein.

FIG. 3 is a block diagram of the electronic controller 300 of the system 100. The electronic controller 300 includes a plurality of electrical and electronic components that provide power, operation control, and protection to the components and modules within the electronic controller 300. The electronic controller 300 includes, among other things, an electronic processor 305 (such as a programmable electronic microprocessor, microcontroller, or similar device), a memory 310 (for example, non-transitory, computer readable memory), an input/output interface 315, and a transceiver 320. The electronic processor 305 is communicatively connected to the memory 310, the input/output interface 315, and the transceiver 320. The input/output interface 315 includes a human machine interface (HMI) 325. The electronic processor 305, in coordination with the memory 310 and the input/output interface 315, is configured to implement, among other things, the methods described herein. It should be understood that some or all of the components, including additional components, of the controller 300 may be remote/dispersed from each other within the railway vehicle 102 and/or remote from the railway vehicle 102 (for example, part of the vibration sensor 104, the communications network 112 and/or a separate electronic computing device).

The memory 310 may be made up of one or more non-transitory computer-readable media and includes at least a program storage area and a data storage area. The program storage area and the data storage area can include combinations of different types of memory, such as read-only memory (“ROM”), random access memory (“RAM”), flash memory, or other suitable memory devices. The electronic processor 305 is coupled to the memory 310 and the input/output interface 315.

The electronic processor 305 sends and receives information (for example, from the memory 310 and/or the input/output interface 315) and processes the information by executing one or more software instructions or modules, capable of being stored in the memory 310, or another non-transitory computer readable medium. The software can include firmware, one or more applications, program data, filters, rules, one or more program modules, and other executable instructions. The electronic processor 305 is configured to retrieve from the memory 310 and execute, among other things, software for automatic detection/prediction of an anomaly within the wheel system 200 and for performing methods as described herein. In the example illustrated, the memory 310 stores, among other things, a vibration classification algorithm 320, which operates as described herein (for example, in regard to the method 400 described in regard to FIG. 4 below) to detect and classify vibration patterns to identify and/or predict anomalies within the wheel system 200.

The input/output interface 315 transmits and receives information from devices external to the electronic controller 300 (for example, over one or more wired and/or wireless connections), for example, components of the system 100 via one or more data buses and/or designated communication channels. The input/output interface 315 receives input (for example, from the vibration sensor 104 and the sensors 106 etc.), provides system output (for example, to the transceiver 325 and/or the HMI 330, etc., or a combination of both). The input/output interface 315 may also include other input and output mechanisms, which for brevity are not described herein and which may be implemented in hardware, software, or a combination of both.

The transceiver 325 includes a radio transceiver communicating data over one or more wireless communications networks (for example, cellular networks, satellite networks, land mobile radio networks, etc.) including the communications network 112. The transceiver 325 may also provide wireless communications within the vehicle 102 using suitable network modalities (for example, Bluetooth™, near field communication (NFC), Wi-Fi™, and the like). Accordingly, the transceiver 325 communicatively couples the electronic controller 300 and other components of the system 100 with networks or electronic devices both inside and outside the railway vehicle 102. For example, the electronic controller 300, using the transceiver 325, can communicate with a one or more devices (for example, the other systems 110) over the communications network 112 to send and receive data, commands, and other information (for example, component anomaly notifications). The transceiver 325 includes other components that enable wireless communication (for example, amplifiers, antennas, baseband processors, and the like), which for brevity are not described herein and which may be implemented in hardware, software, or a combination of both. Some instances include multiple transceivers or separate transmitting and receiving components (for example, a transmitter and a receiver) instead of a combined transceiver.

As mentioned above, the input/output interface 315 includes the HMI 330. The HMI 330 provides visual output, such as, for example, graphical indicators (i.e., fixed, or animated icons), lights, colors, text, images, combinations of the foregoing, and the like. The HMI 330 includes a suitable display mechanism for displaying the visual output, such as, for example, an instrument cluster, a heads-up display, a center console display screen (for example, a touch screen, or other suitable mechanisms. In some instances, the HMI 330 displays a graphical user interface (GUI) (for example, generated by the electronic controller 300 and presented on a display screen) that enables a user to interact with one or more systems (and components thereof) the railway vehicle 102. The HMI 330 may also provide audio output to the user such as a chime, buzzer, voice output, or other suitable sound through a speaker included in the HMI 330 or separate from the HMI 330. In some instances, HMI 330 provides a combination of visual, audio, and haptic outputs. In some examples, the HMI 330 is implemented on a separate electronic device of a user or another third-party (for example, a railway fleet operator or conductor). The electronic device may be any kind of computing device such as a laptop, tablet, or a smart phone.

In some instances, the electronic controller 300 may be implemented in several independent controllers (for example, programmable electronic controllers) each configured to perform specific functions or sub-functions. For example, as mentioned above, one or more components of the controller 300 may be remote from the railway vehicle 102 (for example, part of a remote cloud-based server, which is not shown, of the communications network 112). Additionally, the electronic controller 300 may contain sub-modules that include additional electronic processors, memory, or circuits for handling input/output functions, processing of signals, and application of the methods listed below. In other instances, the electronic controller 300 includes additional, fewer, or different components. In some embodiments, one or more components of the electronic controller 300 are integrated into a dashboard (not shown) of the railway vehicle 102. Thus, the programs may also be distributed among one or more processors.

As described in further detail below, in some instances the memory 310 includes, among other things, computer executable instructions for detecting, predicting, and/or identifying one or more anomalies of the railway vehicle 102 and, in particular, of the railway vehicle wheel system 200 and of the track 210. Anomalies of the railway vehicle wheel system 200 may include, for example, physical defects or damage to one or more of the wheels 204, the axle 206, and the suspension and brake components (not shown) of the wheelset 202 (for example, wheel bearings, the suspension system, etc.). Anomalies in a track (for example, the track 204) may also be detected by the system 100. In some instances, the computer executable instructions include instructions for training a machine or deep learning system to detect/predict one or more anomalies related to the system 200 of the railway vehicle 102.

In some instances, the electronic controller 300 uses one or more machine learning methods to analyze sensor information from the vibration sensor 104 to identify/predict anomalies within the railway vehicle wheel system 200 (as described herein). Machine learning generally refers to the ability of a computer program to learn without being explicitly programmed. In some instances, a computer program (for example, a learning engine) is configured to construct an algorithm based on inputs. Supervised learning involves presenting a computer program with example inputs and their desired outputs. The computer program is configured to learn a general rule that maps the inputs to the outputs from the training data it receives. Example machine learning engines include decision tree learning, association rule learning, artificial neural networks, classifiers, edge computing, inductive logic programming, support vector machines, clustering, Bayesian networks, reinforcement learning, representation learning, similarity and metric learning, sparse dictionary learning, and genetic algorithms. Using these approaches, a computer program can ingest, parse, and understand data and progressively refine algorithms for data analytics.

In some examples, the electronic processor 305 maintains a digital twin (stored in the memory 310) of the railway vehicle 102 or one or more components thereof. For example, in the illustrated example, the memory 310 includes a digital twin 335 representing the railway vehicle wheel system 200. The electronic processor 305 provides detected vehicle attributes from the vibration sensor 104 and the sensors 106 to the digital twin as input. The generated data is used as an output to predict or simulate how one or more physical components of the wheel system 200 (for example, the wheelset 202 or the track 210) or another component or system of the railway vehicle 102 has been (or will be) affected by these inputs.

As mentioned above, in some instances, one or more components, including additional components (for example, additional components similar to those described above in regard to FIG. 3), of the electronic controller 300 may be positioned or distributed throughout the vehicle 102 or remote from the vehicle 102. In some examples the vibration sensor 104 is directly coupled to an electronic controller (for example, the electronic controller 300 or a separate electronic controller, which is not shown, which includes components similar to those of controller 300). The controller may include an electronic processor, memory, input/output interface, transceiver, and/or the like. The controller 300 (and/or the second controller) may include additional components such as a battery. In some examples, the vibration sensor 104 and the electronic controller are positioned on/within a card or circuit board (not shown). In addition, in some instances, the controller 300 wirelessly communicates with one or more other devices. For example, the electronic controller 300 communicates with one or more other electronic communication devices via the communications network 112.

FIG. 4 illustrates an example of a method 400 for predicting an anomaly within a railway vehicle (for example, railway vehicle 102 of FIG. 1). Although the method 400 is described in conjunction with the system 100 as described herein, the method 400 could be used with other systems and vehicles. In addition, the method 400 may be modified or performed differently than the specific example provided. As an example, the method 400 is described as being performed by the electronic controller 300 and, in particular, the electronic processor 305. However, it should be understood that in some instances, portions of the method 400 may be performed by other devices or subsystems of the system 100. For example, in some examples, some or all of the steps of the method 400 are performed between the remote server 114 and the electronic controller 300.

At block 402, the electronic processor 305 determines a speed of the railway vehicle 102. The electronic processor 305 may determine the speed via one or more speed sensors of the vehicle 102 (for example, the speed sensor 107A). The electronic processor 305, for example, may derive the speed of the vehicle 102 according to information from the geo-location system 108.

At block 404, the electronic processor 305 determines whether the speed exceeds a predetermined speed threshold. The particular speed threshold may be dependent on the particular model of the railway vehicle 102, as well as other factors such as the vehicle load weight (for example, resulting from carrying passengers, cargo, or both). The particular predetermined speed threshold corresponds to a speed in which characteristic vibrations for anomaly detection may be measured by the vibration sensor 104.

The electronic processor 305 returns to block 402 in instances where the speed fails to exceed the predetermined speed threshold. Otherwise, in response to the speed exceeding the predetermined speed threshold, the electronic processor 305 obtains from the vibration sensor 104, a vibration measurement (block 406). The vibration measurement is, for example, a time series of measurements sampled at a particular frequency over a predetermined time interval. The electronic processor 305, at block 408, determines, from the vibration measurement, a classified vibration level including one or more labels indicating one or more characteristic features and/or data points of the vibration measurement. The classified vibration level is a categorization of at least one vibration measurement in time (for example, a determined level, as shown in FIG. 6, explained in more detail below). The electronic processor 305, for example, may provide the vibration measurement as input to the digital twin 335 for classification of the vibration measurement.

The electronic processor 305, at block 410, then determines whether the classified vibration level is indicative of an anomaly. For example, as shown in FIG. 6 (explained in more detail below), in instances where the classified vibration level does not exceed a predetermined classification level (for example, a “level 1”) for one or more consecutive vibration measurements, the electronic processor 305 may determine that an anomaly is not present. In instances where one or more classified vibration levels corresponding to a respective vibration measurement within a predetermined time period exceed one or more classified vibration level thresholds, the electronic processor 305 may determine that an anomaly may be present.

Returning to FIG. 4, the electronic processor 305, at block 412, derives a dominant vibration frequency (i.e., a peak frequency) from the vibration measurement and, at block 414, performs a comparison between the dominant vibration frequency to a predetermined frequency threshold. The electronic processor 305, at block 416, identifies an anomaly existing in the railway vehicle wheel system 200 (for example, wheelset 202), the track of the railway vehicle 102 (for example, track 210), or both based on the comparison and one or more of a reference classified vibration level. The reference classified vibration level includes one or more classified vibration levels derived from one or more vibration measurements corresponding a different anomaly case of one or more railway vehicles. The reference classified vibration level may be received from (or derived based on vibration measurements received from) the one or more other railway vehicle anomaly detection systems 110, pre-stored in the memory 310, or some combination of both. The reference classified vibration level may be based on real-time data, simulation data, or both. The electronic processor 305, in some examples, performs the classification of the received vibration measurements according one or more machine-learning methods (for example, according to one or more additional digital twins of the respective other railway vehicles). The reference classified vibration level is stored at the memory 310 and is compared with the classified vibration level for common or comparable labels/features. Based on detected correlations between particular members of the plurality of classified vibration levels, the electronic processor 305 may identify the particular anomaly present within the railway vehicle 102 (in particular, the railway vehicle wheel system 200).

In classifying the vibration measurements, the electronic processor 305, in some examples, is further configured to collect and utilize vehicle attributes from one or more additional railway vehicles (for example, one or more of the other vehicles of the other systems 110) in addition to received vibration measurements. Additional information may include, but is not limited to, location information, historic sensor information, and any information regarding a particular railway vehicle. The electronic processor 305 may further utilize the information in the determination of the potential anomaly within the wheel system 200, the track 210, or both. Such information, in particular, may be used to derive metadata for classification of the vibration measurement. Metadata may include, for example, the railway vehicle speed at the time of the vibration pattern, the model of railway vehicle 102 in which the vibration measurement was sensed, the state of the vehicle 102 at the time of the vibration pattern (for example, braking, accelerating, turning, etc.), thermal and/or electrical characteristics of one or more of the systems of the railway vehicle 102, and environmental conditions at the time of the vibration pattern (for example, ambient temperature, ambient humidity, weather conditions, etc.). Such information may be provided to the controller 300, for example, via the one or more sensors 106 of the railway vehicle 102. In examples where the electronic controller 300 also performs the classification of the vibration measurements from the other railway vehicle anomaly detection systems 110, the electronic controller 300 may also receive respective sensor information from the systems 110 and derive metadata for classification from the received sensor information.

The electronic processor 305 may, in some examples, perform at least partial processing of the received information from the sensors 106 and generate metadata for the classification of vibration measurements. In examples where the electronic controller 300 is partially implemented on the server 114, the electronic controller 300 may provide the vibration measurements, derived information, and metadata to the remote server 114 for classification. Identification of the anomaly may then be performed at the server 114.

In response to identifying the anomaly, the electronic processor 305 performs a mitigation action. In some examples, the mitigation action includes transmitting (e.g., via the transceiver 325) a notification to an electronic communications device (for example, a laptop, a smartphone, or a tablet) of a driver of the railway vehicle 102, a railway manager, and/or a railway maintenance technician. For example, a suitable network message or API may be used to send a notification that indicates an anomaly has occurred, the time and place of the anomaly, the type of the anomaly, and the like.

In some examples, the mitigation action includes producing a visual and/or audio alert on a human machine interface of the railway vehicle 102 (e.g., the HMI 330) to inform the driver (and any other passengers of the vehicle 102) of the anomaly and any other mitigation actions being taken. The alert may be an audio alert (for example, a buzzer or an alarm) or a visual alert (for example, a light or a text generated on a GUI), for example, generated by the HMI 330. For example, a display of the HMI 330 may display a message such as “LOOSE WHEEL” or “WORN BEARINGS.” In some examples, the HMI 330 may speak the alerts aloud to the driver and/or passengers of the vehicle 102. In some examples, a combination of alerts may be used.

In some examples, the mitigation action includes adjusting an operation of the vehicle 102. For example, the mitigation action may include operating the brakes of the vehicle 102, adjusting a suspension of the vehicle 102 (for example, by actuating one or more components of the suspension system), increasing a speed of the vehicle 102, and the like. In some examples, the mitigation action includes adjusting a railroad switch to change tracks (for example, by transmitting a command to one or more remotely-controllable point actuators of the railway). In some examples, the mitigation action includes placing one or more control systems of the vehicle 102 into a fault mode and/or disabling further use by the driver of the vehicle 102 of one or more control system(s) affected by the anomaly. In some instances, multiple mitigation actions are combined.

FIGS. 5A-5C each illustrate a flowchart 500A, 500B and, 500C, respectively, of an example method for identifying a particular type of anomaly within the railway vehicle 102 or the track 210 (for example, performed at block 414 of the method 400 of FIG. 4). Referring to FIG. 5A, which presents an example method for identifying whether the anomaly relates to the wheelset 202 or the track 210, the electronic processor 305 determines, at block 502A, whether the one or more reference classified vibration levels and the classified vibration level are consistent with geometric defects in a track. The electronic processor 305, for example, determines whether the classified vibration level includes a predetermined classified vibration level pattern (i.e., if the classified vibration levels vary in a predetermined amount of time by a particular respective level). For example, FIG. 6 illustrates a vibration signal classifier 600 for predicting a vertical geometric defect in the track 210 in accordance with some aspects. A vertical geometric drop may be determined from a vibration measured by the vibration sensor 104 in the vertical direction with respect to the wheel axle 206 (as described above with respect to FIG. 2A). FIGS. 7A and 7B each illustrate an example of a vertical geometric defect 700A, 700B in the track 210. FIG. 7A illustrates a vertical drop or crack as a vertical geometric defect 700A and FIG. 7B illustrates a vertically worn track 210 as a vertical geometric defect. Returning to FIG. 6, as illustrated, the classified vibration signal over time reaches a first classified vibration level (for example, “level 3”) before dropping to a second, lower level (for example, “level 1”) and then no vibration, which may be consistent with a vertical drop caused by a vertical geometric defect in the track 210. Depending on the particular vibration level pattern, the electronic processor 305 identifies whether and which kind of vertical geometric defect (for example, defects 700A, 700B) that the track 210 has.

Returning to FIG. 5A, in instances where the reference classified vibration level and the classified vibration level are determined to be inconsistent with a track geometry defect (i.e., do not include the predetermined classified vibration level pattern and the dominant frequency exceeds the predetermined frequency threshold), the electronic processor 305, at block 504A, evaluates the wheelset 202 for the anomaly, for example, by performing a structural analysis via a simulation with the digital twin 335 using the vibration measurement (and, in some examples, one or more attributes from the sensors 106). The electronic processor 305 may accordingly identify the anomaly based on the results of the analysis.

For example, in instances where the dominant frequency exceeds the predetermined frequency threshold and the classified vibration level is comparable to the classified reference level of one or more other sensors of the vehicle 102, the anomaly may be related to a wheel anomaly, a bearing anomaly, or both. The electronic processor 305 may then perform an analysis with the digital twin 335 to identify the particular anomaly with the wheel 204 or the bearing (not shown). For example, the anomaly may include a deformation in a wheel, a loose wheel, a worn bearing, a loose bearing, and/or a broken bearing.

In instances where the reference classified vibration levels and the classified vibration level are determined to be consistent with a track geometry defect (i.e., include the predetermined vibration level pattern), the electronic processor 305 may evaluate whether a reference classified vibration level from a second vibration sensor are comparable to the classified vibration level, wherein the second vibration sensor is positioned on a same side of the railway vehicle 102 as the vibration sensor 104 (for example, at a wheel of a second wheelset on the same side as the wheel 204 with respect to the vehicle 102). If the classified vibration level and the reference classified vibration level are not comparable or similar, the electronic processor 305 may determine that the anomaly is a geometric defect of the track 210. In some examples, the electronic processor 305 is further configured to identify the location of the geometric defect on the track 210 and record the location of the geometric defect on the track 210 (for example, in the memory 310). The electronic processor 305 may also provide the location information in the generated alert to a user.

The electronic processor 305, in some examples, may also be configured to identify horizontal geometric defects of the track 210 as an anomaly. FIGS. 7C and 7D each illustrate an example of a horizontal geometric defect 702C, 702D of the track 210. FIG. 7C illustrates a horizontal crack or break as a horizontal geometric defect 702C, for example, and FIG. 7D illustrates a horizontal bend or warp as a horizontal geometric defect 702D. FIG. 5B illustrates the flowchart 500B of an example method for identifying a horizontal geometric defect in a track (for example, the track 210). The electronic processor 305 utilizes a vibration sensor (for example, the vibration sensor 104) configured and positioned to measure in the horizontal direction with respect to the wheel axle 206 (as described above with respect to FIG. 2A). At block 502B, the electronic processor 305 evaluates whether a reference classified vibration level from a respective vibration sensor and the classified vibration level exceed a predetermined horizontal vibration reference level, wherein a second vibration sensor of the reference classified vibration level is positioned on a same side of the railway vehicle 102 as the vibration sensor 104 (for example, at a wheel of a second wheelset on the same side as the wheel 204 with respect to the vehicle 102). If the vibration levels exceed the predetermined classified horizontal vibration level, the electronic processor 305 identifies the anomaly as being that of a horizontal geometric defect with the track 210 (block 504B). The electronic processor 305 may accordingly identify and record the location of the geometric defect on the track 210 (for example, in the memory 310). The electronic processor 305 may provide the location information in the generated alert to a user. Otherwise, the electronic processor 305 may continue evaluating for horizontal geometric defects (block 506B).

Returning to FIG. 5A, if the reference classified vibration level is determined to not be comparable, the electronic processor 305 evaluates whether a reference classified vibration level from a second vibration sensor is comparable to the classified vibration level, wherein the second vibration sensor is positioned on a second wheel system of the railway vehicle 102 (block 508A), separate from the system 200. In instances where the reference classified vibration level is not comparable to the classified vibration level, the electronic processor 305 identifies that the anomaly is a malfunctioning suspension (not shown) of the wheel axle 206 (block 512A). In instances where the reference classified vibration is comparable to the second wheel system, the electronic processor 305 may return to block 402 of the method 400 (FIG. 4) and evaluate one or more additional vibration measurements as described above.

FIG. 5C illustrates an example method for identifying whether the anomaly relates to a braking system of the railway vehicle 102. To begin identifying a brake anomaly, the electronic processor 305 (at block 502C) determines whether the railway vehicle 102 is beginning to move from a parking condition (e.g., beginning to move forward from a stop), is moving steadily at some velocity above zero (e.g., is maintaining a velocity), or is accelerating. For example, the electronic processor 305 may use position and velocity data from the geo-location system 108 to determine a velocity profile for the railway vehicle 102 and analyze the velocity profile using a machine learning pattern recognition algorithm to classify the movement of the railway vehicle. In some embodiments, the velocity profile is determined using the speed sensor 107A, either alone or in conjunction with data from the geo-location system 108. Where the electronic processor 305 determines (at block 502C) that the railway vehicle 102 is either beginning to move, moving steadily, or accelerating, it (at block 504C) determines whether the classified vibration level is comparable to the reference classified vibration level indicative of a seized braking system (e.g., the brakes are clamped and/or frozen). Where the classified vibration level is indicative of a seized braking system (at block 504C), the electronic processor 305 (at block 506C) generates an alert (as described herein) informing a driver of the railway vehicle, a fleet manager, and the like of the seized braking condition.

Where the classified vibration level is not indicative of a seized braking system (at block 504C), the electronic processor 305 (at block 508C) analyzes the velocity profile to determine whether the railway vehicle 102 is braking (for example, using a machine learning classifier). For example, the electronic processer 305 may determine that the velocity of the railway vehicle 102 is slowing at a rate, pattern, or both consistent with the application of the brakes. In some aspects, where the electronic processor 305 determines (at block 508C) that the railway vehicle 102 is not braking (e.g., the deceleration is not the results of the brakes being applied), it (at block 510C) may continue to evaluate sensed vibrations for other types of anomalies, as described herein. Where the electronic processor 305 determines (at block 508C) that the railway vehicle 102 is braking, it determines (at block 512C) whether the classified vibration level is comparable to the reference classified vibration level indicative of a normal brake event. Where the classified vibration level is indicative of a normal brake event, the electronic processor 305 (at block 514C) may continue to iterate the method 500C, analyzing braking events as they are detected. Where the classified vibration level is not indicative of a normal brake event, the electronic processor 305 (at block 516C) generates an alert (as described herein) informing a driver of the railway vehicle, a fleet manager, and the like to check the brake system for worn or maladjusted components (e.g., worn brake pads, discs, drums, and the like). In some instances, the electronic processor 305 may perform the method 500C continuously or periodically to analyze braking events as the railway vehicle 102 operates. In some instances, the electronic processor 305 may execute the method 500C when particular conditions are met (e.g., when the ambient temperature is low enough to cause ice to form). In some aspects, the execution of the method 500C may be based on factors such as vehicle load, vehicle type, and the like.

Thus, the examples described herein provide, among other things, an anomaly detection system for a railway vehicle configured to detect/predict anomalies within a wheel system of the railway vehicle.

In the foregoing specification, specific examples, aspects, and features have been described. However, one of ordinary skill in the art appreciates that various modifications and changes can be made without departing from the scope of the aspects, examples, and features as set forth in the claims below. Accordingly, the specification and figures are to be regarded in an illustrative rather than a restrictive sense, and all such modifications are intended to be included within the scope of present teachings.

The benefits, advantages, solutions to problems, and any element(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential features or elements of any or all the claims. The invention is defined solely by the appended claims including any amendments made during the pendency of this application and all equivalents of those claims as issued.

In this document, relational terms such as first and second, top and bottom, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. The terms “comprises,” “comprising,” “has,” “having,” “includes,” “including,” “contains,” “containing” or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises, has, includes, contains a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by “comprises . . . a,” “has . . . a,” “includes . . . a,” or “contains . . . a” does not, without more constraints, preclude the existence of additional identical elements in the process, method, article, or apparatus that comprises, has, includes, contains the element. The terms “a” and “an” are defined as one or more unless explicitly stated otherwise herein. The terms “substantially,” “essentially,” “approximately,” “about” or any other version thereof, are defined as being close to as understood by one of ordinary skill in the art, and in one non-limiting embodiment the term is defined to be within 10%, in another embodiment within 5%, in another embodiment within 1% and in another embodiment within 0.5%. The term “coupled” as used herein is defined as connected, although not necessarily directly and not necessarily mechanically. A device or structure that is “configured” in a certain way is configured in at least that way but may also be configured in ways that are not listed.

Various features, aspects, advantages, and examples are set forth in the following claims.

PREDICTION AND IDENTIFICATION OF POTENTIAL RAILWAY SYSTEM ANOMALIES

Information

Publication Number

Date Filed

Date Published

Inventors

CPC

International Classifications

Abstract

Description

Claims