System and method for preventing axle locking

Abstract

To solve the problem of causing a loss of revenue while disrupting traffic of a rail system.SOLUTION: Methods are provided which may include determining one or more of a vibration characteristic and a fluid characteristic of one or more components of a vehicle and determining one or more expected characteristics for the one or more of the vibration characteristic and the fluid characteristic. The methods may also include determining whether the one or more of the vibration characteristic and the fluid characteristic deviate from the one or more expected respective characteristics, and implementing one or more responsive actions in response to determining that the one or more of the vibration characteristic and the fluid characteristic deviate from the one or more expected characteristics.SELECTED DRAWING: Figure 7

Description

Technical Field

[0001] Cross - Reference to Related Applications This application claims the priority of U.S. Provisional Application No. 63 / 152,157, filed on February 22, 2021, the entire disclosure of which is incorporated herein by reference.

Background Art

[0002] The described subject matter relates to systems and methods for a control system for early detection of drive - train failures in vehicles.

[0003] Technical Considerations Vehicle systems such as railway vehicles are used for both passenger transportation and / or cargo transportation. A railway vehicle is typically a vehicle system that includes a number of vehicles that can be mechanically, logically, virtually, etc., coupled to each other to travel along a railway. Vehicles within a vehicle system include their own wheel sets and optionally corresponding axles. Some vehicles, such as locomotives, include traction motors to provide power to the axles, and other vehicles are simply pushed or pulled along the railway while their wheels rotate accordingly. Such railway vehicles travel along a number of tracks of a railway system, and multiple railway vehicles utilize the railway system at any given time.

[0004] When a railway vehicle gets stuck on the main line of a railway system, excessive delays occur. As a result, the railway vehicle experiences a loss of revenue and at the same time disrupts the traffic of the railway system. Removing a stuck railway vehicle requires a lot of labor and cost. For example, if a railway vehicle stops because an axle is locked, it is necessary to either replace the locked axle, cut the pinion, or perform other temporary repairs on the track. Sometimes when a remote stop occurs, it may be difficult to reach a location where repair facilities are not available, causing further difficulties.

Summary of the Invention

[0005] In one or more embodiments, a method is provided which may include determining one or more vibration characteristics or fluid characteristics of one or more components of a vehicle, and determining one or more expected characteristics for one or more of the vibration characteristics or fluid characteristics. The method may also include determining whether one or more of the vibration characteristics or fluid characteristics deviate from one or more expected characteristics, and taking one or more response actions in response to determining that one or more of the vibration characteristics or fluid characteristics deviate from one or more expected characteristics.

[0006] In one or more embodiments, a method is provided which may include determining one or more of the temperature, vibration signature, or fluid lubrication amount of a first axle of a vehicle, and determining one or more of the temperature, vibration signature, or fluid lubrication amount of at least a second axle of a vehicle. The method may also include determining the integrity of the first axle based on comparing one or more of the temperature, vibration signature, or fluid lubrication amount determined for the first axle with one or more of the temperature, vibration, or fluid lubrication amount determined for at least a second axle, and comparing one or more of the temperature, vibration signature, or fluid lubrication amount determined for the first axle with one or more of the temperature, vibration signature, or fluid lubrication amount determined for at least a second axle.

[0007] In one or more embodiments, a system is provided which may include one or more sensors capable of measuring one or more vibration characteristics or fluid lubrication amounts of one or more components of a vehicle. The system may also include one or more processors capable of determining one or more expected characteristics for one or more of the vibration characteristics or fluid lubrication amounts. The one or more expected characteristics may be based at least in part on ambient conditions. One or more processors may determine whether one or more of the vibration characteristics or lubrication characteristics deviate from one or more expected characteristics. In response to determining that one or more of the vibration characteristics or lubrication characteristics deviate from one or more expected characteristics, one or more processors may perform one or more response actions.

[0008] In one or more embodiments, a method is provided which may include determining one or more vibration characteristics or fluid lubrication amounts of one or more components of a vehicle. The method may also include determining one or more expected characteristics for one or more vibration characteristics or fluid lubrication amounts. The one or more expected characteristics may be based at least in part on ambient conditions. The method may also include determining whether one or more vibration characteristics or lubrication characteristics deviate from one or more expected characteristics, and determining a health score for each of the vehicle's multiple axles based on the deviation between one or more vibration characteristics or lubrication characteristics and one or more expected characteristics. The method may also include choosing to stop the vehicle from moving, slow the vehicle from moving, complete the current run of the vehicle, and prevent further runs of the vehicle until the vehicle is repaired, or scheduling an inspection of one or more components of the vehicle in a future previously scheduled maintenance of the vehicle. [Brief explanation of the drawing]

[0009] The subject matter of the present invention can be understood by referring to the accompanying drawings and reading the following description of non-limiting embodiments.

[0010] [Figure 1] This example illustrates a block diagram of a vehicle system. [Figure 2] This section provides an example of a block diagram illustrating a control system. [Figure 3] This example illustrates a flow block diagram for determining the probability of a combination of events. [Figure 4A] Let's illustrate this with a graph showing oil health over a short period. [Figure 4B] Let's illustrate this with an example graph showing the probability of an axle being locked. [Figure 5A] Let's illustrate this with a graph showing oil health over a short period. [Figure 5B] Let's illustrate this with an example graph showing the probability of an axle being locked. [Figure 6] A schematic block diagram of the system will be used as an example. [Figure 7] This example illustrates a flow block diagram for determining which axles in a vehicle system are locked. [Figure 8] Let's illustrate the algorithm's outline block diagram with an example. [Figure 9] Let's illustrate the algorithm's outline block diagram with an example. [Figure 10] Let's illustrate the algorithm's outline block diagram with an example. [Figure 11] Let's illustrate the algorithm's outline block diagram with an example. [Figure 12] This example illustrates a schematic block diagram of the period algorithm. [Figure 13] This section provides an example of a flow chart block diagram illustrating a method for determining the likelihood of a vehicle's drivetrain failure. [Modes for carrying out the invention]

[0011] Embodiments of the subject matter described herein relate to systems and methods for the early detection of drivetrain failures in railway vehicles. Based on the detection and likelihood of failure over a given period, actions are taken to reduce the probability or prevent the drivetrain failure. By making a determination and taking actions to mitigate the drivetrain failure, the disruption caused by a stalled vehicle system is reduced.

[0012] Figure 1 illustrates a schematic diagram of an example of a vehicle system 100 including a control system 102. The vehicle system may travel along a route 104 in a journey from a starting or departure point to a destination or arrival point. The vehicle system includes at least one propulsion-generating vehicle 108 and optionally at least one non-propulsion-generating vehicle 110, which are mechanically interconnected with each other to travel together along the route. Alternatively, the vehicle system may consist of only a single propulsion-generating vehicle or only two or more propulsion-generating vehicles.

[0013] A propulsion-generating vehicle can generate a traction force to propel (e.g., pull or push) the vehicle system along a path. A propulsion-generating vehicle includes a propulsion subsystem, e.g., an engine, one or more traction motors, and / or similar, which operate to generate a traction force to propel the vehicle system. A propulsion-generating vehicle also includes a braking system that generates a braking force to slow or stop the movement of the vehicle system. While one propulsion-generating vehicle and one non-propulsion-generating vehicle are shown in Figure 1, the vehicle system may include multiple propulsion-generating vehicles and / or multiple non-propulsion-generating vehicles. In yet another embodiment, the vehicles within the vehicle system are logically or virtually coupled together, but not mechanically coupled together.

[0014] In the example in Figure 1, each vehicle in the vehicle system includes a plurality of wheels 120 that engage with the path, and at least one axle 122 that connects the left and right wheels together (only the left wheel is shown in Figure 1). Optionally, the wheels and axles are located on one or more freight cars or bogies 118. Optionally, the freight cars may be fixed-axle cars such that the wheels are rotatably fixed to the axles, and thus the left wheel rotates at the same speed, amount, and time as the right wheel. In one embodiment, the vehicle system may not include axles, such as some mining vehicles, electric vehicles, etc.

[0015] The control system may further include a wireless communication system 126 that enables wireless communication between vehicles within the vehicle system and / or with a remote location such as a remote location 128 (e.g., a vehicle dispatch location). The communication system may include a receiver and a transmitter, or a transceiver that performs both a receiving function and a transmitting function. The communication system may also include an antenna and associated circuitry.

[0016] In one example, the control system further includes a controller 129 that includes a driving characteristic evaluation element 130 that may include a characteristic evaluation circuit that communicates with one or more processors. The driving characteristic evaluation element may provide information regarding the driving of the vehicle system along a route. The driving information may include route characteristics, designated locations, designated stopping locations, scheduled times, encounter events, directions along the route, and the like.

[0017] In an embodiment, the controller may also include a vehicle characteristic evaluation element 134. The vehicle characteristic evaluation element may provide information regarding the configuration of the vehicle system, such as the type of non-propulsion power generation vehicle (e.g., manufacturer, product number, material, etc.), the number of non-propulsion power generation vehicles, the weight of the non-propulsion power generation vehicles, whether the non-propulsion power generation vehicles are consistent (meaning that the weight and distribution are relatively the same over the entire length of the vehicle system) or inconsistent, the type and weight of the cargo, the total weight of the vehicle system, the number of propulsion power generation vehicles, the position and arrangement of the propulsion power generation vehicles relative to the non-propulsion power generation vehicles, the type of propulsion power generation vehicle (including manufacturer, product number, output capacity, available notch settings, fuel utilization rate, etc.).

[0018] The vehicle characteristic evaluation element may be an electronic storage device or a database stored in a memory. The information within the vehicle characteristic evaluation element may be input by an operator using an input / output (I / O) device (referred to as a user interface device), automatically uploaded, or received remotely via the communication system. At least some sources of the information within the vehicle characteristic evaluation element may be a vehicle manifest, a log, and the like.

[0019] Figure 2 provides a schematic illustrative diagram of a controller 200 that can control the operation of a propulsion-generating vehicle. In one example, the controller represents the controller in Figure 1. The controller may be a device comprising one or more processors 202 (such as a microprocessor, integrated circuit, or field-programmable gate array). The controller may also optionally include a controller memory 204, which may be an electronic computer-readable storage device or medium. The controller memory may be located within the housing of the controller, or alternatively, on a separate device that can be communicatively coupled to the controller and one or more processors therein. "Communicatively coupled" means that two devices, systems, subsystems, assemblies, modules, components, etc., are coupled by one or more wired or wireless link, such as one or more conductive (e.g., copper) wires, cables, or buses, wireless networks, or fiber optic cables. The controller memory may include a tangible non-temporary computer-readable storage medium that stores data temporarily or permanently for use by one or more processors. Memory may include one or more volatile and / or non-volatile memory devices such as random access memory (RAM), static random access memory (SRAM), dynamic RAM (DRAM), other types of RAM, read-only memory (ROM), flash memory, magnetic storage devices (e.g., hard disks, floppy disks, or magnetic tapes), and optical disks. Memory may be used to store information related to vehicle parameters, route characteristics, driving parameters, etc. Vehicle parameters may include vehicle weight, wheel diameter, tachometer readings, throttle settings, brake settings, speed, acceleration, etc. Route characteristics may include route gradient, route weather, route curvature, etc. Driving parameters may include destination, area speed limits, traffic congestion, rest stops, tunnel locations, etc.

[0020] The controller may also include a transceiver 206 capable of communicating with a remote device 208. The transceiver may be a single unit or may consist of separate receiver and transmitter components. In one example, the transceiver may only transmit signals. The remote device 208 may be a dispatch controller, a controller for another vehicle, a second controller coupled to a vehicle system, a controller in a roadside device, and so on.

[0021] The controller may also include, or communicate with, one or more sensors 210 coupled to the vehicle system to detect vehicle parameters, path characteristics, driving parameters, etc. In one embodiment, at least one sensor is a temperature sensor that measures the temperature of components of the vehicle system. For example, the temperature sensor may be an axle bearing temperature sensor that can measure either a pinion end (PE) bearing or other bearing, the temperature of oil used in relation to the axle and axle bearing, or an oil temperature sensor that can or cannot be used in relation to an ambient temperature sensor, engine temperature sensor, motor winding temperature sensor, etc. In another embodiment, at least one sensor may be a vibration sensor related to components of the vehicle system. For example, the vibration sensor may be a bearing vibration sensor, a gear vibration sensor, a gear case vibration sensor, a gear noise sensor, or a vibration sensor that detects the acceleration of a wheel caused by a flat spot. Specifically, the vibration sensor may detect amplitude, frequency, etc., within a specific frequency range of interest. Alternatively, at least one sensor may be a vehicle motion sensor including an oil fluid level sensor for the engine or gear case, a sensor that detects ambient conditions of the vehicle system or components of the vehicle system, or a speed sensor. Each of these sensors may also be provided in any combination. Specifically, each sensor may detect or monitor operation to acquire data that can be used for the early detection of drivetrain failures. For example, a determination may be made as to whether it is likely or unlikely for a particular axle to lock up.

[0022] The controller may also include an input device 212 and an output device 214. Specifically, the input device may be an interface between the operator and one or more processors. The input device may include a display or touchscreen, input buttons, ports for receiving memory devices, etc. In this way, the operator may manually provide parameters to the controller, including vehicle parameters, route characteristics, driving parameters, etc. Similarly, the output device may present or prompt the operator for information and data. The output device may also be a display or touchscreen. In this way, the display or touchscreen may be both an input device and an output device. The controller may also receive input from both the vehicle characteristic evaluation element 218 and the driving characteristic evaluation element 220.

[0023] Figure 3 illustrates a method 300 for determining the combined probability of an axle locking over a determined period. In other exemplary embodiments, other conditions may be determined, but in the example of Figure 3, an axle locking is determined. The methodology of Figure 3 can be used as the basis for the methodology of Figure 7 (this methodology can determine a health score and take proactive action to prevent a vehicle system shutdown as a result of the conditions determined in Figure 3). Specifically, by utilizing the strategy and methodology of Figure 3, a health score can ultimately be determined for each axle in the vehicle system based on the difference between expected vehicle characteristics and measured vehicle characteristics.

[0024] In step 302, a first characteristic is monitored. The first characteristic may be vibration characteristics, fluid characteristics, temperature characteristics, lubricant contaminant characteristics, etc. The first characteristic may be monitored by one or more sensors. In exemplary embodiments, one or more sensors may include thermometers, acoustic sensors, optical sensors, infrared sensors, vibration sensors, accelerometers, gear case sensors, axle bearing temperature sensors, oil temperature sensors, motor winding temperature sensors, etc. By monitoring the first characteristic with sensors, readings, data, information, etc., related to the first characteristic are obtained. In one example, the vibration characteristic is a vibration signature from a vibration sensor, the fluid characteristic is the amount of fluid lubricant contaminants, and the temperature characteristic is temperature.

[0025] In step 304, a first probability is determined that an axle will lock up during a given period. The given period could be a duration, such as 3 days. In another example, the period could be related to a number of miles (or kilometers). Specifically, the probability of when an axle will lock up can be determined by monitoring the first characteristic.

[0026] In an exemplary embodiment, the oil level in the gear case is a monitored fluid characteristic. Figure 4A is a graph illustrating short-term oil health 402 as a function of oil level 404. Specifically, as the oil percentage decreases, the oil health state also decreases. In this way, a first probability based on the first characteristic can be determined. Figure 4B illustrates the first probability of an axle locking up based on the oil level 406 within a determined period, such as one week, based on the short-term oil health state. Similarly, the first probability may be a determined distance traveled, such as within 100 miles. As illustrated by the graph, as oil health decreases, the first probability of an axle locking up increases.

[0027] In step 306, a second characteristic is monitored. The second characteristic may be vibration characteristics, fluid characteristics, temperature characteristics, lubricant contaminant characteristics, etc. The characteristic may be monitored by one or more sensors. In exemplary embodiments, one or more sensors may include thermometers, acoustic sensors, optical sensors, infrared sensors, vibration sensors, accelerometers, gear case sensors, axle bearing temperature sensors, oil temperature sensors, motor winding temperature sensors, etc. By monitoring the second characteristic with sensors, readings, data, information, etc., related to the second characteristic are obtained. In one example, the vibration characteristic is a vibration signature from a vibration sensor, the fluid characteristic is the amount of fluid lubricant, and the temperature characteristic is temperature.

[0028] In step 308, a second probability is determined that an axle will lock up within a given period. The given period may be a duration such as 3 days. In another example, the period may be related to a number of miles (or kilometers). Specifically, by monitoring a second characteristic, a second probability regarding when an axle will lock up can be determined. Figure 5A illustrates a graph of short-term oil health 408 as a function of oil temperature 410. As the oil temperature in the gear case rises, short-term oil health decreases. Figure 5B illustrates a second probability of an axle locking up based on the oil temperature 412 within a determined period such as one week, based on the short-term oil health state. Similarly, the second probability may be a determined distance traveled, such as within 100 miles. As illustrated by the graph, as the oil temperature rises, the second probability of an axle locking up also increases.

[0029] In step 310, the first probability and the second probability are combined to form a combined probability that an axle will be locked during a given period. Further, additional probabilities may be included to determine the combined probability that an axle will be locked during a given period. Figure 6 illustrates a schematic block diagram of system 600 for determining the probability of a locked axle. In this example, inputs 602A, 602B, 602C, and 602D are provided as a result of monitoring characteristics. From the monitored inputs, the first probability 604A, the second probability 604B, the third probability 604C, and the fourth probability 604D are all determined based on the individual inputs. In one example, the first probability 604A is the probability of an axle being locked based on the oil level. In one embodiment, the probability of an axle being locked based on the oil level is determined as shown by the graphs in Figures 4A and 4B. In another embodiment, the second probability is the probability of an axle being locked based on the oil temperature. In one example, the probability of an axle locking based on oil temperature is determined as shown in graphs 5A and 5B. In another example, a third probability is the probability of an axle locking based on gear case vibration, and a fourth probability is the probability of an axle locking based on wheel defect. The first, second, third, and fourth probabilities are then used to determine the combined probability of an axle locking over a given period. The combined probability of an axle locking can be determined by averaging the probabilities, giving more or less weight to specific probabilities, or selecting the highest probability. By utilizing combinations of variables, accuracy can be improved, and axle locking can be prevented through maintenance and other preventative measures.

[0030] Figure 7 illustrates a method 700 for determining the possibility of a rail vehicle drivetrain failure for a vehicle system as a result of a locked axle. In one example, the vehicle system is the vehicle system of Figure 1. In another example, the method is carried out by the controller of Figure 2. In one embodiment, the vehicle system is a rail vehicle system. In other embodiments, the vehicle system is one of the following: off-road vehicles, trolleys, aircraft fleets, ship fleets, etc.

[0031] In step 702, one or more of the vibration, fluid, or temperature characteristics of one or more components of the vehicle are determined. In the example, the determination may be performed by a controller. In the example, the vehicle is part of a vehicle system. Vibration characteristics may be any readings, data, information, signals, currents, slips, amplitudes, frequencies, vibration amplitudes, etc., within a specific frequency range that is related to or associated with vibration. Readings, data, information, signals, currents, slips, amplitudes, frequencies, vibration amplitudes, etc., within a specific frequency range that is related to or associated with vibration may be related to or associated with vibration if they provide vibration of any component of the vehicle, or can be used to obtain vibration of any component of the vehicle. Vibration characteristics may be provided by readings, data, information, signals, currents, slips, capacitances, amplitudes, frequencies, vibration amplitudes, etc., within a specific frequency range that is related to vibration. Vibration characteristics may be obtained using algorithms, mathematical equations, mathematical models, functions, lookup tables, decision trees, etc. The determined vibration characteristics may include one or more of the following: bearing vibration of one or more bearings on the vehicle's axle, gear vibration of one or more gears on the vehicle, gear transmission vibration generated by one or more gears on the vehicle, air transmission vibration, structural transmission vibration, case vibration of the vehicle's gear case, acceleration of one or more wheels on the vehicle, vibration signature, etc.

[0032] Temperature characteristics may include any readings, data, information, signals, currents, slips, capacitances, etc., that are related to or associated with temperature. Readings, data, information, signals, currents, slips, capacitances, etc., may be related to or associated with temperature if they provide the temperature of any component of the vehicle. Temperature characteristics may be obtained using algorithms, mathematical equations, mathematical models, functions, lookup tables, decision trees, etc. Determined temperature characteristics may include one or more of the following: temperature, fluid temperature, lubricant temperature, axle bearing temperature (including bearing elements, raceways, tubes, etc.), including PE bearing temperature and CE bearing temperature, motor winding temperature, etc.

[0033] In an exemplary embodiment, the characteristics may be two or more of the following: vibration characteristics, fluid characteristics, and temperature characteristics. For example, the temperature of a fluid, such as oil in a gear case, may be both a fluid characteristic and a temperature characteristic.

[0034] Similarly, fluid properties may include any readings, data, information, signals, currents, slips, capacitances, etc., that are related to or associated with the fluid. Readings, data, information, signals, currents, slips, capacitances, etc., may be related to or associated with the fluid if they provide or can be used to obtain the fluid properties of the vehicle. Fluid properties may be provided by readings, data, information, signals, currents, slips, capacitances, etc. Fluid properties may be obtained using algorithms, mathematical equations, mathematical models, functions, lookup tables, decision trees, etc. Determined fluid properties may include one or more of the following: fluid volume, axle bearing temperature of one or more bearings on the vehicle's axles, motor bearing temperature of one or more traction motor bearings, combination bearing temperature of one or more combination bearings, journal bearing temperature of one or more journal bearings, fluid temperature of one or more lubricants in the vehicle's gear case or engine, motor winding temperature of one or more conductive windings in the vehicle's motor, lubricant or fluid level, etc.

[0035] In step 704, one or more expected characteristics are determined for one or more of the vibration characteristics, fluid characteristics, or temperature characteristics. Expected characteristics are those based on calculations or determinations, as opposed to actual characteristics that can be directly measured or determined based on direct measurements. Specifically, calculations or determinations can be made for each vibration characteristic, fluid characteristic, and temperature characteristic, assuming no malfunctions occur in the components, based on the forces acting on the vehicle, vehicle components, etc. For example, if a railway vehicle travels 100 miles at an average vehicle speed of 50 mph and an average axle output of 500 HP, a calculation can be made for the expected lubricating fluid temperature for the axle gear case.

[0036] On the other hand, the actual fluid temperature can also be directly determined by a sensor immersed in oil. In this way, the fluid temperature calculated from variables such as distance traveled and vehicle speed is the expected fluid characteristic, while the fluid characteristic directly measured by the temperature sensor is not the expected fluid characteristic. Specifically, in one example, a vehicle includes multiple axles, and one or more of the following—vibration characteristics, fluid characteristics such as lubrication oil level, or temperature characteristics—are determined for the first axle, while one or more expected vibration characteristics, fluid characteristics, or temperature characteristics are determined at least partially based on the operation of a set of axles that do not include the first axle. Specifically, the expected vibration characteristics and fluid characteristics do not need to utilize direct measurement of the axles.

[0037] In one example, expected characteristics may be determined using information from a remote location. In particular, in embodiments of railway vehicles, a positive vehicle controller (PVC) may be used to restrict and control the movement of a railway vehicle on a railway system in relation to other railway vehicles using the system. Such continuously acquired information may be used to make determinations and calculations related to expected characteristics. Furthermore, a controller equipped with a running optimizer software that continuously acquires data, information, etc., related to the railway vehicle may be used. Specifically, the running optimizer software functions to achieve objectives related to any individual run. For example, a particular run may be desired to occur within a certain amount of time under conditions where a certain amount of pollution has occurred. The running optimizer software then operates the railway vehicle in a manner that satisfies these objectives. Nevertheless, information, data, etc., are continuously collected to achieve such optimization. Such information, data, etc., may be used to make determinations and calculations related to expected characteristics.

[0038] In exemplary embodiments, one or more expected characteristics may be based at least in part on ambient conditions. Such ambient conditions may be used in combination with direct measurements relating to the expected characteristics, or without utilizing direct measurements of the expected characteristics. Ambient conditions may include one or more of the following: rotational speed of one or more wheels of the vehicle, ambient temperature, ambient noise, path characteristics of the path the vehicle is traveling, vehicle characteristics of the vehicle, accelerator torque applied to one or more axles of the vehicle, one or more of the soundness or irregularity of the path or rail, the presence of one or more gaps in the path, curvature of the path, position of the vehicle on the path, or inconsistencies in the path as path characteristics, weight of the vehicle, vehicle speed, size of the vehicle's wheels, estimated dynamic axle weight as a vehicle characteristic, etc.

[0039] In another example, one or more expected characteristics for one or more of the vibration characteristics, fluid lubrication, or thermal lubrication amount may also be based on the vehicle's operation. Specifically, one or more expected characteristics may be based on the vehicle's operation, including vehicle speed, axle torque acting on the vehicle's axles, ambient temperature, one or more characteristics of the path the vehicle is traveling, and one or more characteristics of the vehicle.

[0040] In step 706, one or more of the vibration, fluid, or temperature characteristics are determined to deviate from one or more expected characteristics. In one example, the vibration, fluid, or temperature characteristics may be compared to each expected characteristic. In other examples, mathematical functions, mathematical models, lookup tables, decision trees, functions, algorithms, etc., may be used to determine the deviation. In other examples, tolerances may be determined such that no deviation determination is provided if the difference between the vibration, fluid, or temperature characteristics is not greater than the tolerance. In another embodiment, tolerances may be based on the determined quantity. For example, if the fluid characteristic is the oil temperature of the oil lubricating the bearings in a bearing assembly, the determined tolerance may be 5 degrees Fahrenheit (or about 2 degrees Celsius). In this way, if the expected fluid characteristic of the oil is 297 degrees Fahrenheit (149 degrees Celsius), and the fluid characteristic of the oil is 300 degrees Fahrenheit (150 degrees Celsius), the fluid characteristic is determined not to deviate from each expected oil temperature. If the expected fluid properties of the oil are at 290 degrees Fahrenheit (143 degrees Celsius), then the fluid properties deviate from the expected fluid properties. In other embodiments, a percentage difference may be used instead of a determined quantity. In this way, as vibration properties, fluid properties, or temperature properties increase or decrease, the tolerance range may also increase or decrease. In yet another example, the fluid properties may indicate the amount of lubricating oil contaminants. Such contaminants may include water, soot, dirt, metal particles, etc. Such properties can indicate the need for fluid replacement and / or repair of seals or gaskets. Thus, the amount of contaminants outside the determined range may indicate such a need.

[0041] If, in step 706, it is determined that there is no deviation between vibration characteristics, fluid characteristics, or temperature characteristics, the vibration characteristics, fluid characteristics, or temperature characteristics continue to be evaluated in 704. If, in step 706, it is determined that there is a deviation between vibration characteristics, fluid characteristics, or temperature characteristics, the soundness score for each of the vehicle's multiple axles is determined in 708 based on the deviation between one or more of the vibration characteristics, lubrication characteristics, or temperature characteristics and one or more expected characteristics. In one example, the method shown in Figure 3 is used to determine the scored soundness. Specifically, an evaluation of axle soundness can be performed.

[0042] By utilizing vibration, fluid, and temperature characteristics, the integrity of an axle and the probability of drivetrain failure can be determined. For example, a reduced oil level in the gear case may increase the probability of an axle locking up. A rise in oil temperature within the gear case may increase the probability of an axle locking up. Increased vibration in the gear case may increase the probability of an axle locking up. A rise in axle bearing temperature or bearing raceway temperature may increase the probability of an axle locking up. Gear or wheel defects may also increase the probability of an axle locking up. In yet another example, motor winding temperature may be an indicator of an increased probability of an axle locking up.

[0043] In many of the provided examples and cases, while an increase in temperature or vibration is expected during operation, an increase exceeding expectations, based on the forces acting on the vehicle system and in addition to historical information related to such characteristics, indicates an increased probability of an axle locking. Therefore, by monitoring one of these characteristics, the probability associated with the possibility of an axle locking can be determined. In some embodiments, more than one probability can be used to determine the overall probability of an axle locking. For example, both the gear case temperature and the gear case fluid level can be monitored, and the probability of an axle locking can be determined for both. In one example, the probabilities can be averaged to determine the overall probability of an axle locking. In another embodiment, the determined probabilities can be weighted based on the determined factors, the severity of the deviation from expected values, etc. In yet another embodiment, an algorithm can be used to make such probability determinations. Overall, based on vibration characteristics, fluid characteristics, and / or temperature characteristics, the health of an axle can be determined based on the determined probabilities. Based on the probabilities, a Heath score can be provided. Health scores can be determined using lookup tables, decision trees, algorithms, mathematical functions, mathematical models, and other methods.

[0044] In step 710, response actions may be taken based on the health score. Specifically, different response actions may occur based on the health score if the health score indicates a high probability of an axle locking up. Response actions may include sending an alert to the driver, sending an alert to an external system including a PVC device or parking device, sending an alert to another vehicle or vehicle system, generating a notification to an external facility to determine whether vehicle repairs or changes to the vehicle's scheduled dispatch should occur, communicating vibration, fluid, or temperature characteristics to an external system, changing the axle torque applied to one or more axles to reduce the output of at least one axle and increase the axle torque applied to one or more other axles, stopping the vehicle's movement, slowing the vehicle's movement, preventing further travel of the vehicle until repairs are made, or scheduling inspections of one or more components. Alerts may include visual messages, auditory messages, sounds, flashing lights, etc. Notifications may include electronic messages (email), text messages, automated voice messages, etc. Specifically, based on the health score, response actions may be taken to prevent an axle from locking up before it actually happens.

[0045] Figure 8 shows an illustrative algorithm 800 used to implement the method of Figure 7. In this example, a health score is determined based on characteristics related to the vehicle system that relate to the health of the axle bearings. More specifically, the determination is based on the operation of the vehicle system.

[0046] The first characteristic 802, the second characteristic 804, the third characteristic 806, the fourth characteristic 808, and the fifth characteristic 810 are used when determining the expected temperature of the bearing 812. The bearing may be a PE bearing, a CE bearing, etc. For example, the first characteristic is the rotational speed of the axle, the second characteristic is the axle torque, the third characteristic is the ambient temperature of the bearing's surrounding environment, the fourth characteristic is the raceway characteristics such as gradient and curvature, and the fifth characteristic is the vehicle characteristics such as the material of the bearing elements and the size of the axle. Each of the characteristics is a temperature characteristic, as it can be used to determine the expected temperature of the bearing. Specifically, each can be used to calculate friction and other temperature-based characteristics. Characteristics 802-810 may be received from sensors, vehicle characteristic evaluation elements, driving characteristic evaluation elements, external devices, driving plans, external PTC controllers, etc. The determination of the expected temperature can be made based on mathematical calculations, models, functions, etc.

[0047] After determining the expected temperature, both the expected temperature 814 and the measured temperature 816 are provided to the health determination algorithm 818. The health determination algorithm uses both the expected and measured temperatures to make a determination related to the deterioration of the bearing as a result of operation 820. In this example, a tolerance 822 is provided for typical variation between such measurements. Next, a bearing health score 824 is provided based on the expected temperature compared to the measured temperature. Depending on the health score, response actions may be taken, including providing an alert, notifying the driver or a third party, scheduling a maintenance appointment, reducing the accelerator torque on the bearing, stopping the vehicle system, or reducing the vehicle speed of the vehicle system. In this way, the bearing health score is used to mitigate the risks associated with a locked axle.

[0048] Figure 9 illustrates yet another exemplary algorithm 900 used to implement the method of Figure 7. In this example, the axle health score is determined based on a comparison with another axle. Specifically, each axle in the vehicle system must exhibit relatively similar characteristics to the other axles. As a result, if one axle operates significantly worse than another, or performs poorly, it provides an indication of the possibility of the axle being locked. Furthermore, if sensors are coupled to the axles, vibration signals common to multiple axles may indicate poor track surface conditions. This information can be used to promote infrastructure working range or operating speed limits.

[0049] When determining the integrity of each axle, in the exemplary algorithm shown in Figure 9, the first characteristic 902 and the second characteristic 904 relate to the first axle, while the third characteristic 906 and the fourth characteristic 908 relate to another axle, or the second axle. Each characteristic is input to the determination 910 of the expected bearing temperature. The bearings may be PE bearings, CE bearings, etc. For example, the first characteristic may be the bearing temperature of the first axle, and the second characteristic may be a duty cycle characteristic such as the axle torque requirement of the first axle. The third characteristic may be the bearing temperature of the second axle, and the fourth characteristic may be the axle torque requirement of the second axle. Based on these inputs, a determination may be made in relation to the expected temperature.

[0050] Similar to the embodiment in Figure 8, after determining the expected temperature, both the expected temperature 914 and the measured temperature 916 are provided to the health determination algorithm 918. The health determination algorithm uses both the expected and measured temperatures to make a determination related to the deterioration of the bearing as a result of operation 920. In this example, a tolerance 922 is provided for typical variation between such measurements. Next, a bearing health score 924 is provided based on the expected temperature compared to the measured temperature. Depending on the health score, response actions may be taken, including providing an alert, notifying the driver or a third party, scheduling a maintenance appointment, reducing the torque of the axle with bearings having a poor health score, stopping the vehicle system, or reducing the vehicle speed of the vehicle system.

[0051] Figure 10 illustrates yet another exemplary algorithm 1000 used to implement the method of Figure 7. In this example, the axle health score is determined based on sensors in the drivetrain. Specifically, the expected temperature can be determined by utilizing vibration characteristics, fluid characteristics, and / or temperature characteristics associated with the drivetrain.

[0052] In the exemplary algorithm in Figure 10, the first characteristic 1002 may relate to the bearing structure, such as the axle bearing temperature. Meanwhile, the second characteristic 1004 and the third characteristic 1006 may relate to the drivetrain. In one example, the second characteristic is engine oil temperature and the third characteristic is motor temperature. Each characteristic is input to the determination 1010 of the expected bearing temperature. The bearing may be a PE bearing, a CE bearing, etc. Similar to the embodiments in Figures 8-9, the determination of the expected temperature may be performed using mathematical functions, mathematical models, computer-generated modes, lookup tables, decision trees, etc. In yet another example, additional drivetrain characteristics may be utilized. In such embodiments, three drivetrain-related characteristics may be compared to determine whether one of the measured characteristics provides a fault in the measurement. Thus, by utilizing comparisons, one or more response actions to be performed may be limited based on comparisons to eliminate false detections of vehicle faults.

[0053] Similar to the embodiments shown in Figures 8-9, after determining the expected temperature, both the expected temperature 1014 and the measured temperature 1016 are provided to the health determination algorithm 1018. The health determination algorithm uses both the expected and measured temperatures to make a determination relating to the deterioration of the bearing as a result of operation 1020. In this example, a tolerance 1022 is provided for typical variation between such measurements. Next, a bearing health score 1024 is provided based on the expected temperature compared to the measured temperature. Depending on the health score, response actions may be taken, including providing an alert, notifying the driver or a third party, scheduling a maintenance appointment, reducing the torque of the axle with bearings having a poor health score, stopping the vehicle system, or reducing the output of the vehicle system or the vehicle speed.

[0054] Figure 11 illustrates yet another exemplary algorithm 1100 used to implement the method of Figure 7. In this example, the axle health score is determined based on a comparison of delta temperature against expected delta. Specifically, the expected temperature can be determined by utilizing vibration characteristics, fluid characteristics, and / or temperature characteristics related to the drive system.

[0055] In the exemplary algorithm in Figure 11, the first characteristic 1102 may relate to the temperature of the PE bearing, and the second characteristic 1104 may relate to the temperature of the CE bearing. Each characteristic is input to the determination 1010 of the expected bearing temperature. The bearing in this exemplary embodiment is a PE bearing. As with the embodiments in Figures 8-10, the determination of the expected temperature can be performed using mathematical functions, mathematical models, computer-generated modes, lookup tables, decision trees, etc.

[0056] Similar to the embodiments in Figures 8-10, after determining the expected temperature, both the expected temperature 1114 and the measured temperature 1116 are provided to the health determination algorithm 1118. The health determination algorithm uses both the expected and measured temperatures to make a determination related to the deterioration of the bearing as a result of operation 1120. In this example, a tolerance 1122 is provided for typical variation between such measurements. Next, a bearing health score 1124 is provided based on the expected temperature compared to the measured temperature. Depending on the health score, response actions may be taken, including providing an alert, notifying the driver or a third party, scheduling a maintenance appointment, reducing the torque of the axle with bearings having a poor health score, stopping the vehicle system, or reducing the vehicle speed of the vehicle system.

[0057] Figures 8–11 all demonstrate exemplary embodiments in which the expected temperature of an axle bearing is determined and compared to the measured temperature of the axle bearing to determine the axle bearing's health score. Furthermore, a similar methodology may be used to determine the health scores of other characteristics and utilize them to determine the axle's health score, which may then be used to take action to mitigate potential axle locking.

[0058] Figure 12 illustrates a power reduction algorithm 1200 provided for mitigating potential locked axles. As illustrated, first, second, third, fourth, fifth, and sixth health scores 1202A-F are illustrated. In this embodiment, six health scores are utilized, but in other embodiments, only one health score may be utilized. Alternatively, more than six health scores may be utilized. Specifically, one or more health scores may be utilized, and more health scores utilized provide more accurate results. In the embodiment shown in Figure 12, in one example, the first health score is the lubricant level of the gear case health score, the second health score is the lubricant temperature of the gear case health score, the third health score is the vibration of the gear case health score, the fourth health score is the gear health score, the fifth health score is the PE temperature health score, and the sixth health score is the wheel defect health score. In the exemplary embodiment, the PE temperature health score is determined using the algorithms shown in Figures 8-11. As shown, by using vibration characteristics, fluid characteristics, and / or temperature characteristics, a number of health scores can be determined to improve the accuracy of determining the health of the axle.

[0059] A weight 1204A-F is assigned to each health score. The weight can be determined based on the health score. For example, the worse the health score, the greater the weight assigned to that health score. In this way, if the wheel defect score indicates that an axle is expected to lock up in the next 50 miles of travel, while all other scores indicate that an axle is not expected to lock up until more than 10,000 hours have passed, the wheel defect score will be assigned a significantly greater weight so that an accurate result is presented.

[0060] Alternatively, the weights may depend on the accuracy of the health score. In one embodiment, the health score is on a scale of 1 to 100. In an example of such an embodiment, when the health score based on lubricant temperature is below 20, there is a 95% probability that the axle will lock during the next 100 miles of driving, but when the health score based on bearing vibration is below 20, there is only a 50% probability that the axle will lock. Specifically, in such an embodiment, the type of vibration sensor measuring the bearing may not be as accurate as the lubricant temperature system. As a result, more weight is given to the health score based on lubricant temperature below 20 than to the health score based on bearing vibration below 20. Alternatively, in other examples, the weights may be calculated using mathematical models, functions, lookup tables, decision trees, etc.

[0061] Based on the weights, maximum scores 1206A-F are provided. The maximum score can be determined by one of the methods for making the determination discussed earlier. Then, based on the maximum score, a minimum score 1208 for each axle is determined, and axle health scores 1210A-C for each axle may be provided. From the axle health scores 1210A-C for each axle, a response action 1212 may be taken to prevent an axle from locking up during driving. For example, based on the axle health score for each axle, a minimum time to a predicted failure or a high-probability failure may be provided. A high probability of failure may include a probability of more than 50% that an axle will lock up during the period. Alternatively, a high probability of failure may include a probability of more than 20% that an axle will lock up during the minimum period. The minimum period may be provided in terms of time, distance, etc. For example, the minimum period may be 10 drives exceeding 3 hours, 500 miles, 300 miles, etc., at the current vehicle speed.

[0062] In this way, information is provided that can be used when taking response actions to prevent axle locking during operation. Such response actions may include reducing the output of identified axles to decrease their axle torque while increasing the axle torque of healthier axles, reducing the vehicle speed of the vehicle system, preventing the vehicle system from leaving a station or stop after a stop until repairs are needed, scheduling repairs, skipping pre-planned stops for unplanned stops where maintenance may be needed, informing a third-party device of the axle health score, and changing the driver of the axle health score.

[0063] Figure 13 illustrates an exemplary embodiment of a flow block diagram of Method 1300 for detecting drive system failures in railway vehicles. In one example, the vehicle system is the vehicle system of Figure 1. In another example, the method is carried out by the controller of Figure 2. In one embodiment, the vehicle system is a railway vehicle system. In other embodiments, the vehicle system is one of the following: off-road vehicles, trolleys, aircraft fleets, ship fleets, etc.

[0064] In step 1302, one or more of the vibration, fluid, or temperature characteristics of one or more components of the vehicle are acquired. In an example, characteristics may be acquired by sensors, in-vehicle controllers, driving plans, etc. In an example, the vehicle is part of a vehicle system. Vibration characteristics may be any readings, data, information, signals, currents, slips, capacitances, etc. that are related to or associated with vibration. Readings, data, information, signals, currents, slips, capacitances, etc. may be related to or associated with vibration if they provide vibration of any component of the vehicle, or can be used to acquire vibration of any component of the vehicle. Vibration characteristics may be provided by readings, data, information, signals, currents, slips, capacitances, etc. Vibration characteristics may be acquired using algorithms, mathematical equations, mathematical models, functions, lookup tables, decision trees, etc. In an example, vibration characteristics are vibration signatures.

[0065] Temperature characteristics can be any readings, data, information, signals, currents, slips, capacitances, etc. that are related to or associated with temperature. Readings, data, information, signals, currents, slips, capacitances, etc., can be related to or associated with temperature if they provide the temperature of any component of the vehicle. Temperature characteristics can be obtained using algorithms, mathematical equations, mathematical models, functions, lookup tables, decision trees, etc. In one example, the temperature characteristic is temperature.

[0066] Similarly, fluid properties can be any readings, data, information, signals, currents, slips, capacitances, etc., related to or associated with the fluid. Readings, data, information, signals, currents, slips, capacitances, etc., can be related to or associated with the fluid if they provide or can be used to obtain the fluid properties of the vehicle. Fluid properties can be provided by readings, data, information, signals, currents, slips, capacitances, etc. Fluid properties can be obtained using algorithms, mathematical equations, mathematical models, functions, lookup tables, decision trees, etc. For example, a fluid property could be a fluid quantity.

[0067] In step 1304, vibration characteristics, fluid characteristics, or temperature characteristics are communicated to an external system. The external system may be a PTC, station controller, maintenance controller, client controller, customer controller, etc. The external system may include additional historical information and data related to the vehicle system, the vehicle system's route, route characteristics, other similar vehicle systems, other similar vehicle systems that have previously traveled the route, etc. Based on the characteristics, vibration characteristics, fluid characteristics, or temperature characteristics may be determined by the external system. Determined vibration characteristics may include one or more of the following: bearing vibration of one or more bearings on the vehicle's axle, gear vibration of one or more gears on the vehicle, gear transmission vibration generated by one or more gears on the vehicle, air transmission vibration, structural transmission vibration, case vibration of the vehicle's gear case, acceleration of one or more wheels on the vehicle, etc. Determined temperature characteristics may include one or more of the following: fluid temperature, lubricant temperature, axle bearing temperature (including temperature of bearing elements, raceways, tubes, etc.), including PE bearing temperature and CE bearing temperature, winding temperature, etc. In exemplary embodiments, the characteristics may be two or more of vibration characteristics, fluid characteristics, and temperature characteristics. For example, the temperature of a fluid such as oil in a gear case may be both a fluid characteristic and a temperature characteristic. The determined fluid characteristics may include one or more of the following: axle bearing temperature of one or more bearings on the vehicle's axle, motor bearing temperature of one or more traction motor bearings, combined axle bearing temperature of one or more combined bearings, journal bearing temperature of one or more journal bearings, fluid temperature of one or more lubricants in the vehicle's gear case or engine, winding temperature of one or more conductive windings in a motor in the vehicle, lubricant or fluid level, etc.

[0068] In step 1306, vibration characteristics, fluid characteristics, or temperature characteristics are determined by comparing the vehicle's characteristics with similar characteristics measured for another vehicle. For example, if vibration characteristics include axle size, vehicle size, vehicle weight, vehicle speed, local terrain, wheel wear, etc., the characteristics may also include the characteristics of a vehicle with the same axle size traveling at the same speed on the same route terrain, as found in the external controller's database. Measured vibration characteristics from other vehicles can be compared with the measured vibrations of the vehicle.

[0069] In step 1308, optionally, vibration characteristics, fluid characteristics, or temperature characteristics are determined by examining the trends of these characteristics. Specifically, the trend of vibration characteristics may involve examining the vibration characteristics at similar vehicle speeds at numerous points during travel to determine whether the vibration characteristics are steadily increasing, remaining the same, decreasing, or fluctuating.

[0070] In step 1310, optionally, vibration characteristics, fluid characteristics, or temperature characteristics are determined by comparing characteristics measured under a first condition with characteristics measured under a different second condition. For example, vibration characteristics are measured when the vehicle system is traveling at a first vehicle speed, such as 40 mph, and then when the vehicle system is traveling at a faster second vehicle speed, such as 60 mph.

[0071] In step 1312, optionally, vibration characteristics, fluid characteristics, or temperature characteristics are determined by comparing the characteristics measured for the vehicle at a first age with the characteristics measured for the vehicle at a second age. Thus, in the example, the vibration characteristics of the vehicle system may be determined when the vehicle system is 10 years old and traveling on a specific section of the route, and when the same train travels the same route on the same course after 3 years. Next, the vibration characteristics of the vehicle system after 3 years are compared with those of the vehicle system after 10 years.

[0072] In one or more embodiments, a method is provided which may include determining one or more vibration characteristics or fluid characteristics of one or more components of a vehicle, and determining one or more expected characteristics for one or more of the vibration characteristics or fluid characteristics. The method may also include determining whether one or more of the vibration characteristics or fluid characteristics deviate from one or more expected characteristics, and taking one or more response actions in response to determining that one or more of the vibration characteristics or fluid characteristics deviate from one or more expected characteristics.

[0073] Optionally, fluid properties may be determined as one or more of the following: bearing temperature of one or more bearings on a vehicle's axle, motor bearing temperature measured as bearing temperature, oil temperature, motor winding temperature of one or more traction motor bearings, combined bearing temperature of one or more combined bearings, journal bearing temperature of one or more journal bearings, fluid temperature of one or more lubricants in the vehicle's gear case or engine, or winding temperature of one or more conductive windings of motors within the vehicle. In one embodiment, one or more expected temperature properties to be determined may include expected temperature differences between bearings on different axles. In another embodiment, one or more expected temperature properties to be determined may include expected motor temperatures of motors that rotate at least one axle.

[0074] Optionally, a vehicle may include multiple axles, one or more vibrational or fluidic properties are determined for a first axle of the axles, and one or more expected vibrational, fluidic, or temperature properties are determined at least in part based on the operation of a set of axles that do not include the first axle. In one embodiment, the one or more expected vibrational and fluidic properties determined may be expected axle temperatures, which are at least in part based on the torque and temperature of each axle in the set. In one example, the axles may be in the same vehicle within a multi-vehicle system. In another example, the axles may be in different vehicles within a multi-vehicle system.

[0075] Optionally, vibration characteristics may be determined and include one or more of the following: bearing vibration of one or more bearings of the vehicle's axles, gear vibration of one or more gears of the vehicle, gear transmission vibration generated by one or more gears of the vehicle, air transmission vibration, structural transmission vibration, case vibration of the vehicle's gear case, or acceleration of one or more wheels of the vehicle. In one embodiment, fluid characteristics may be lubrication amount and include the amount of oil in one or more of the engine, gear case, or drivetrain housing within the vehicle. In one embodiment, one or more expected characteristics may be at least in part ambient conditions. In one example of an embodiment, ambient conditions may include one or more of the following: rotational speed of one or more wheels of the vehicle, ambient temperature, ambient noise, path characteristics of the path the vehicle is traveling, vehicle characteristics of the vehicle, torque applied to one or more axles of the vehicle, path integrity, presence of one or more gaps in the path, curvature of the path, position of the vehicle on the path, or misalignment in the path as a path characteristic, weight of the vehicle, speed of travel of the vehicle, size of the vehicle's wheels, or estimated dynamic axle weight as a vehicle characteristic.

[0076] Optionally, one or more expected characteristics for one or more of the vibration characteristics or fluid lubrication amounts may also be based on the vehicle's operation. In one embodiment, one or more expected characteristics may be based on the vehicle's operation, including one or more of the vehicle's speed, the torque acting on the vehicle's axles, the ambient temperature, one or more characteristics of the path the vehicle is traveling, or one or more characteristics of the vehicle. In one embodiment, one or more response actions performed may include generating a notification to an external facility that determines whether to repair the vehicle or change the vehicle's scheduled dispatch based on one or more deviations of the vibration characteristics or lubrication characteristics from the one or more expected characteristics.

[0077] Optionally, the method may also include communicating vibration characteristics to an external system that performs one or more of the following: comparing the vibration characteristics of a vehicle with vibration characteristics measured for another vehicle, or examining trends in vibration characteristics. The method may also include communicating vibration characteristics to an external system that performs one or more of the following: comparing vibration characteristics measured under first conditions with vibration characteristics measured under different second conditions, or comparing vibration characteristics measured for a vehicle at a first age with vibration characteristics measured for a vehicle at a second age later in its history. In one embodiment, one or more response actions that may be performed include reducing the output of a first axle of the vehicle while one or more other axles of the vehicle continue to be rotated by one or more of the vehicle's motors or engines to propel the vehicle.

[0078] Optionally, the method also includes determining the health score of each of the vehicle's axles based on a deviation between one or more vibration characteristics or lubrication characteristics and one or more expected characteristics. In one embodiment, one or more response actions that may be taken include choosing to change the torque applied to one or more of the axles to reduce the output of at least one of the axles, increase the torque applied to one or more of the other axles based on the health score, or stop the vehicle from moving. One or more response actions that may be taken may also include choosing to slow the vehicle from moving, complete the vehicle's current run, and prevent the vehicle from running further until the vehicle is repaired, or schedule the inspection of one or more components of the vehicle in a future previously scheduled maintenance of the vehicle. In another embodiment, one or more response actions taken may be selected based on the different effects of one or more response actions on the possibility of completing the run. For example, one or more response actions taken may include preventing the change in the torque applied to one or more of the axles in response to determining that the vehicle would stall on a gradient when the torque is changed.

[0079] In one or more embodiments, a method is provided which may include determining one or more of the temperature, vibration signature, or fluid lubrication amount of a first axle of a vehicle, and determining one or more of the temperature, vibration signature, or fluid lubrication amount of at least a second axle of a vehicle. The method may also include determining the integrity of the first axle based on comparing one or more of the temperature, vibration signature, or fluid lubrication amount determined for the first axle with one or more of the temperature, vibration, or fluid lubrication amount determined for at least a second axle, and comparing one or more of the temperature, vibration signature, or fluid lubrication amount determined for the first axle with one or more of the temperature, vibration signature, or fluid lubrication amount determined for at least a second axle.

[0080] Optionally, vibration characteristics may be determined by multiple sensors coupled to different axles of the vehicle, and the method may also include comparing vibrations measured by sensors coupled to different axles. The method may also include identifying damaged portions of a path based on vibrations measured by sensors coupled to different axles and compared with each other. In another embodiment, one or more of the vibration characteristics or fluid lubrication amounts may be determined by direct measurement. In one example, one or more of the vibration characteristics or fluid lubrication amounts are determined by indirect measurement. In another example, one or more of the vibration characteristics or fluid lubrication amounts are determined from measurements of another different characteristic. In another embodiment, the other different characteristic may include one or more of wheel slip, current conducted within the vehicle, or capacitance measured within the vehicle. In one embodiment, the vibration characteristics are determined and include one or more of the frequency or amplitude of the vibration.

[0081] Optionally, the method may also include determining a comparison between one or more vibration characteristics or fluid lubrication quantities determined for a vehicle. The method may also include determining a comparison between one or more vibration characteristics or fluid lubrication quantities measured for a vehicle under different conditions, or measured for another vehicle. The method may also include restricting one or more response actions performed based on the comparison in order to eliminate false detection of faults with the vehicle.

[0082] In one or more embodiments, a system is provided that includes one or more sensors capable of measuring one or more vibration characteristics or fluid lubrication amounts of one or more components of a vehicle. The system may also include one or more processors capable of determining one or more expected characteristics for one or more of the vibration characteristics or fluid lubrication amounts. The one or more expected characteristics may be based at least in part on ambient conditions. The one or more processors may also determine whether one or more of the vibration characteristics or lubrication characteristics deviate from one or more expected characteristics. The one or more processors may also perform one or more response actions in response to determining that one or more of the vibration characteristics or lubrication characteristics deviate from one or more expected characteristics.

[0083] In one or more embodiments, a method is provided which may include determining one or more vibration characteristics or fluid lubrication amounts of one or more components of a vehicle. The method may also include determining one or more expected characteristics for one or more vibration characteristics or fluid lubrication amounts. The one or more expected characteristics may be based at least in part on ambient conditions. The method may also include determining whether one or more vibration characteristics or lubrication characteristics deviate from one or more expected characteristics, and determining a health score for each of the vehicle's multiple axles based on the deviation between one or more vibration characteristics or lubrication characteristics and one or more expected characteristics. The method may also include choosing to stop the vehicle from moving, slow the vehicle from moving, complete the current run of the vehicle, and prevent further runs of the vehicle until the vehicle is repaired, or scheduling an inspection of one or more components of the vehicle in a future previously scheduled maintenance of the vehicle.

[0084] As used herein, the terms “processor” and “computer,” as well as related terms, such as “processing device,” “computing device,” and “controller,” may refer to microcontrollers, microcomputers, programmable logic controllers (PLCs), field-programmable gate arrays, and application-specific integrated circuits, as well as other programmable circuits, but are not limited to those integrated circuits referred to in the art as computers. Suitable memory may include, for example, computer-readable media. Computer-readable media may be, for example, computer-readable non-volatile media such as random-access memory (RAM) and flash memory. The term “non-temporary computer-readable media” refers to tangible computer-based devices implemented for short-term and long-term storage of information such as computer-readable instructions, data structures, program modules and submodules, or other data in any device. Thus, the methods described herein may be encoded as executable instructions embodied in tangible non-temporary computer-readable media, including but not limited to storage devices and / or memory devices. When such instructions are executed by a processor, they cause the processor to execute at least a portion of the methods described herein. Thus, this term includes tangible computer-readable media, including volatile and non-volatile media, as well as removable and non-removable media such as firmware, physical and virtual memory devices, CD-ROMs, DVDs, and other digital sources such as networks or the internet, but not limited to these, including non-temporary computer memory devices.

[0085] In one embodiment, the control system may deploy a local data collection system that can use machine learning to enable derivation-based learning outcomes. The control system may learn from a set of data (including data provided by various sensors) and make decisions by performing data-driven predictions and adapting according to the set of data. In embodiments, machine learning may include performing multiple machine learning tasks by a machine learning system, such as supervised learning, unsupervised learning, and reinforcement learning. Supervised learning may include presenting a set of exemplary inputs and desired outputs to the machine learning system. Unsupervised learning may include learning algorithms that structure its inputs by methods such as pattern detection and / or feature learning. Reinforcement learning may include the machine learning system running in a dynamic environment and then providing feedback on correct and incorrect decisions. In examples, machine learning may include several other tasks based on the output of the machine learning system. In examples, the tasks may be machine learning problems such as classification, regression, clustering, density estimation, dimensionality reduction, and anomaly detection. In examples, machine learning may include several mathematical and statistical techniques. In the examples, many types of machine learning algorithms may include decision tree-based learning, correlation rule learning, deep learning, artificial neural networks, gene learning algorithms, inductive logic programming, support vector machines (SVMs), Bayesian networks, reinforcement learning, representation learning, rule-based machine learning, sparse dictionary learning, similarity metric learning, learning classifier systems (LCS), logistic regression, random forests, K-means algorithms, gradient boosting, K-nearest neighbors (KNNs), and a priori algorithms. In the embodiments, specific machine learning algorithms may be used (for example, to solve both constrained and unconstrained optimization problems that can be based on natural selection). In the examples, an algorithm may be used to address a mixed-integer programming problem where some components are restricted to integer values. Algorithms and machine learning techniques and systems may be used in computational intelligence systems, computer vision, natural language processing (NLP), recommendation systems, reinforcement learning, graphical model building, and more.For example, machine learning can be used for vehicle performance and behavioral analysis.

[0086] In one embodiment, the control system may include a policy engine capable of applying one or more policies. These policies may be based at least in part on the characteristics of a given item of equipment or environment. With respect to the control policy, the neural network may receive inputs of several environmental and task-related parameters. These parameters may include the identification of a determined driving plan for a group of vehicles, data from various sensors, and location and / or positional data. The neural network may be trained to generate an output based on these inputs, the output representing an action or sequence of actions that the group of vehicles should take to achieve the driving plan. During operation in one embodiment, a determination may occur by processing the input through the parameters of the neural network to generate a value at the output node that designates that action as the desired action. This action may be translated into a signal that operates the vehicle. This may be achieved via backpropagation, a feedforward process, closed-loop feedback, or open-loop feedback. Alternatively, instead of using backpropagation, the controller's machine learning system may use evolutionary strategy techniques to tune various parameters of the artificial neural network. The maintenance system may use a neural network architecture with functions that may not always be solvable using backpropagation, such as non-convex functions. In one embodiment, the neural network has a set of parameters representing the weights of its node connections. Several copies of this network are generated, and then various adjustments are made to the parameters, and simulations are performed. Once outputs are obtained from the various models, they can be evaluated based on their performance using a determined success metric. The best model is selected, and the vehicle controller executes a plan to achieve the desired input data in order to mirror the predicted best-case outcome scenario. Furthermore, the success metric may be a combination of optimized results that can be weighted against each other.

[0087] The singular forms “a,” “an,” and “the” include multiple references unless the context explicitly indicates otherwise. “Optional” or “optionally” means that the event or situation described therein may or may not occur, and that the description may include both the case in which the event occurs and the case in which it does not. The approximation language used herein throughout this specification and the claims may be applied to modify any quantitative expression that may change acceptablely without resulting in a change of the underlying function to which it may relate. Thus, values ​​modified by terms such as “about,” “substantially,” and “approximately” should not be limited to a specified exact value. In at least some examples, the approximation language may correspond to the precision of an instrument for measuring a value. Throughout this specification and the claims, range limitations may be combined and / or interchangeable, and such ranges may be identified, including all subranges contained therein unless the context or language otherwise indicates.

[0088] This written description, using examples, discloses embodiments including best modes, enabling those skilled in the art to practice embodiments including fabricating and using any device or system and performing any incorporated methods. The claims define the patentable scope of this disclosure and include other examples conceived to those skilled in the art. Such other examples are intended to be within the scope of the claims if they have structural elements different from the literal language of the claims, or if they include equivalent structural elements that are substantially different from the literal language of the claims.

Claims

1. A step of determining the vibration characteristics, fluid characteristics and temperature characteristics of the vehicle components as characteristics to be monitored, The process involves calculating the vibration characteristics, fluid characteristics, and temperature characteristics of the components of the vehicle as expected characteristics, based on the vehicle's speed, the torque applied to the vehicle's axles, the ambient temperature, or the characteristics of the path the vehicle is traveling along. A step of determining whether the determined vibration characteristics, fluid characteristics, and temperature characteristics deviate from the calculated vibration characteristics, fluid characteristics, and temperature characteristics, respectively. A step of determining the soundness score for the vehicle's axle when the determined vibration characteristics, fluid characteristics, or temperature characteristics deviate from the calculated vibration characteristics, fluid characteristics, or temperature characteristics, respectively. The steps include: performing one or more response actions when the health score indicates a predetermined probability that an axle will lock up; A method including, The one or more response actions performed are: Send an alert to the driver. To stop the movement of the aforementioned vehicle, To slow down the movement of the vehicle, Prevent the vehicle from being driven until it is repaired, and Schedule the inspection of one or more components. A method characterized by including one or more of the following.

2. The method according to claim 1, wherein the alert includes a visual message, an auditory message, an auditory sound, and / or a flashing light.

3. The method according to claim 1, wherein the health score is determined by a lookup table, a decision tree, an algorithm, a mathematical function, and / or a mathematical model.

Images