Chassis wear prediction based on machine vibration data

Abstract

A system may include an apparatus. The apparatus may be configured to receive machine vibration data that identifies a measure of machine vibration. The machine vibration may be caused by a combination of a first vibration caused by the movement of components of the machine's chassis and a second vibration unrelated to the first vibration. The apparatus may be configured to identify segments of the machine vibration data corresponding to the first vibration; transform the segments into signals in the frequency domain using a Fast Fourier Transform (FFT); and analyze the signals to identify a characteristic spectrum associated with the movement of the components. The apparatus may be configured to predict the amount of wear on the components based on the characteristic spectrum. The apparatus may be configured to cause actions to be performed based on the amount of wear on the components.

Description

Technical Field

[0001] This disclosure generally relates to monitoring the wear of a machine chassis, and for example to predicting chassis wear based on machine vibration data. Background Technology

[0002] Components of a machine's chassis (e.g., track links, bushings, and / or pins) wear down over a period of time. One technique for detecting component wear involves obtaining manual measurements of the component's dimensions. These manual measurements can be compared to specified dimensions of the component. To obtain these manual measurements, the machine needs to pause at the work site to perform a task. Because obtaining manual measurements requires the machine to pause and is a time-consuming process (e.g., due to travel time and / or the amount of time spent obtaining the manual measurements), it can negatively impact productivity at the work site. In this case, the task (the task to be performed by the machine) can be paused for an extended period (e.g., the time required to obtain the manual measurements).

[0003] Additionally, such manual measurements may be inaccurate. Inaccurate measurements of part dimensions can, in turn, lead to incorrect predictions about the amount of wear on the part. Due to such incorrect predictions, parts may fail prematurely or be repaired or replaced prematurely (e.g., because the part may not be worn to the point of needing replacement or repair). Such premature failure or premature replacement or repair of parts can also adversely affect workplace productivity.

[0004] International Patent Publication No. WO2020049526 ('562 publication document) discloses a tracked system including an attachment assembly comprising at least one of a first pivot defining a roll pivot axis, a second pivot defining a pitch pivot axis, and a third pivot defining a yaw pivot axis. '562 publication document also discloses that the tracked system further includes at least one actuator and at least one monitoring sensor. The at least one actuator is used to pivot a frame assembly about at least one of the roll pivot axis and the yaw pivot axis, and the at least one monitoring sensor is used to at least indirectly determine at least one of the tracked system's state and ground surface conditions. '562 publication document discloses that the at least one monitoring sensor communicates with a tracked system controller to control the operation of the at least one actuator based on at least one of the tracked system's state and ground surface conditions.

[0005] Although the '562 publication discloses that the at least one monitoring sensor is communicating with the track system controller, the '562 publication does not disclose that the track system controller receives vibration data from the at least one monitoring sensor, nor does it disclose that the track system controller takes into account factors affecting machine vibration in determining the amount of wear on the track system.

[0006] The apparatus disclosed herein solves one or more of the problems mentioned above and / or other problems in the art. Summary of the Invention

[0007] A method performed by a device includes receiving machine vibration data that identifies a measure of vibration of a machine over a period of time; segmenting the machine vibration data to obtain time-domain signals, including time-domain signals associated with vibrations related to the machine's chassis; transforming the time-domain signals into spectral-domain signals using a Fast Fourier Transform (FFT); identifying characteristic spectra associated with motion of components of the machine's chassis from the spectral-domain signals; predicting the amount of wear on the components based on the amplitude of the characteristic spectra; and causing an action to be performed based on the amount of wear on the components.

[0008] A system includes means configured to: receive machine vibration data identifying a vibration measure of a machine, wherein the machine vibration is caused by a combination of a first vibration caused by movement of a component of the machine's chassis and a second vibration unrelated to the first vibration; identify segments of the machine vibration data corresponding to the first vibration; transform the segments into signals in the frequency domain using a Fast Fourier Transform (FFT); analyze the signals to identify a characteristic spectrum associated with the movement of the component; predict the amount of wear on the component based on the characteristic spectrum; and execute actions based on the amount of wear on the component.

[0009] An apparatus includes one or more memories; and one or more processors configured to: receive machine vibration data identifying a vibration measure of a machine, wherein the machine vibration is caused by a combination of a first vibration caused by movement of a component of the machine's chassis and a second vibration unrelated to the first vibration; identify segments of the machine vibration data corresponding to the first vibration; transform the segments into signals in the frequency domain using a Fast Fourier Transform (FFT); analyze the signals to identify a characteristic spectrum associated with the movement of the component; predict the amount of wear on the component based on the characteristic spectrum; and cause an action to be performed based on the amount of wear on the component. Attached Figure Description

[0010] Figure 1 These are illustrations of the exemplary implementation described herein.

[0011] Figure 2 This is a diagram illustrating an example described in this article.

[0012] Figure 3 This is a flowchart of an example process associated with chassis wear prediction using a machine learning model. Detailed Implementation

[0013] This disclosure relates to an apparatus that predicts the amount of wear on one or more components of a machine's chassis based on machine vibration data. The term "machine" can refer to any machine that performs operations associated with industries such as mining, construction, agriculture, transportation, or other sectors. Furthermore, one or more implements can be attached to the machine.

[0014] Figure 1 This is a diagram illustrating the exemplary implementation 100 described herein. For example... Figure 1 As shown, example embodiment 100 includes a machine 105 and a wear detection device 190. The machine 105 is embodied as an earthmoving machine, such as a bulldozer. Alternatively, the machine 105 can be another type of tracked machine, such as an excavator.

[0015] like Figure 1 As shown, machine 105 includes engine 110, sensor system 120, operator cabin 130, operator controls 135, controller 140, rear attachment 150, front attachment 160, and ground engagement member 170.

[0016] Engine 110 may include an internal combustion engine, such as a compression-ignition engine, a spark-ignition engine, a laser-ignition engine, a plasma-ignition engine, etc. Engine 110 provides power to machine 105 and / or a set of loads associated with machine 105 (e.g., components that absorb power and / or use power to operate). For example, engine 110 may provide power to one or more control systems (e.g., controller 140), sensor system 120, operator compartment 130, and / or ground connection member 170.

[0017] Engine 110 can power implements of machine 105, such as implements used in mining, construction, agriculture, transportation, or any other industry. For example, engine 110 can power components (e.g., one or more hydraulic pumps, one or more actuators, and / or one or more electric motors) to control the rear attachment 150 and / or front attachment 160 of machine 105.

[0018] Sensor system 120 may include sensor devices capable of generating signals (described in more detail below) that can be used to predict the amount of wear on one or more components of the chassis of machine 105. The types of sensor devices in sensor system 120 are listed below. Figure 2 To provide a more detailed description.

[0019] The operator compartment 130 includes an integrated display (not shown) and operator controls 135. The operator controls may include one or more input components (e.g., an integrated joystick, buttons, levers, and / or steering wheel) to control the operation of machine 105. For example, operator controls 135 may be used to control the operation of one or more appliances of machine 105 (e.g., rear attachment 150 and / or front attachment 160) and / or control the operation of ground engagement member 170.

[0020] For autonomous machines, operator controls may not be designed for operator use, but rather can be designed to operate independently of an operator. In this case, for example, operator controls may include one or more input components that provide input signals to another component without any operator input.

[0021] Controller 140 (e.g., electronic control module (ECM)) can control and / or monitor the operation of machine 105. For example, controller 140 can control and / or monitor the operation of machine 105 based on signals from operator controls 135, sensor system 120, and / or wear detection device 190. Controller 140 can predict the amount of wear on one or more components of the chassis based on signals from sensor system 120, operator controls 135, and / or wear detection device 190, as described in more detail below.

[0022] The rear attachment 150 may include a ripper assembly, a winch assembly, and / or a drawbar assembly. The front attachment 160 may include a blade assembly. The ground engagement member 170 may be configured to propel the machine 105. The ground engagement member 170 may include wheels, tracks, rollers, and / or similar components for propulsing the machine 105. The ground engagement member 170 may include a chassis including tracks (such as...). Figure 1 (As shown). The track may include track links. Track links may include track link bushings and track link pins. As an example, the track may include a first track link 172 and a second track link 174. The first track link 172 includes a first track link bushing 176 and a first track link pin 178. The second track link 174 includes a second track link pin 180.

[0023] Sprocket 182 may include one or more segments 184 (referred to herein solely as "segment 184" and collectively as "segment 184"). Sprocket 182 may be configured to engage and drive ground engagement member 170. For example, segment 184 may be configured to engage track link bushings (e.g., track link bushings of the track of ground engagement member 170) and rotate to propel the track into machine 105. In some instances, one or more idler pulleys 186 may guide the track as it rotates to propel machine 105.

[0024] Wear detection device 190 may include one or more means capable of predicting the amount of wear on one or more components of the chassis (e.g., one or more tracks, one or more track links such as first track link 172 and / or second track link 174, one or more track link bushings such as first track link bushing 176, one or more track link pins such as first track link pin 178 and / or second track link pin 180, one or more sprockets 182, one or more segments 184 and / or one or more idler pulleys 186). Based on the amount of wear, wear detection device 190 may predict the remaining life of one or more components.

[0025] In some instances, the wear detection device 190 can predict the wear rate of one or more components and predict the amount of wear based on that wear rate. The wear detection device 190 can use a machine learning model to predict the wear rate and / or amount of wear of one or more components, as described in more detail below. The wear detection device 190 may be located within machine 105 (e.g., as part of controller 140), outside machine 105, or partially within and partially outside machine 105.

[0026] As mentioned above, Figure 1 Provided as an instance. Other instances can be combined. Figure 1 The descriptions are different.

[0027] Figure 2 This is a diagram of the example system 200 described in this article. For example... Figure 2 As shown, system 200 includes sensor system 120, operator controls 135, wear detection device 190, and measuring device 210. Wear detection device 190 and / or measuring device 210 may be part of a field management system (e.g., a work site associated with machine 105). Alternatively, wear detection device 190 and / or measuring device 210 may be part of a back-end system or part of machine 105.

[0028] Wear detection device 190 and / or measuring device 210 may be included in the same device. Alternatively, wear detection device 190 and / or measuring device 210 may be separate devices.

[0029] Sensor system 120 may include sensor devices that generate sensor data associated with the amount of wear on one or more components of the chassis. The one or more components may include one or more tracks, one or more track links such as first track link 172 and / or second track link 174, one or more track link bushings such as first track link bushing 176, one or more track link pins such as first track link pin 178 and / or second track link pin 180, one or more sprockets 182, one or more segments 184, and / or one or more idler pulleys 186. The sensor data may be used (e.g., by wear detection device 190) to predict the amount of wear on one or more components. The sensor data may include information identifying the time and / or date the sensor data was generated.

[0030] Sensor system 120 can provide sensor data to predict the amount of wear on one or more components, as described in more detail below. For example, sensor system 120 can provide sensor data to wear detection device 190 periodically (e.g., hourly, every other hour, and / or per work shift). Additionally or alternatively, sensor system 120 can provide sensor data to wear detection device 190 based on triggering events (e.g., requests from wear detection device 190, requests from controller 140, and / or requests from the operator of machine 105 (e.g., via integrated display and / or operator controls)).

[0031] The sensor device may include a vibration sensor device, a motion sensor device, and / or another sensor device that provides sensor data that can be used to predict the amount of wear on one or more components. The vibration sensor device may include one or more devices that sense the vibration of machine 105 and generate machine vibration data based on that vibration. As an example, the vibration sensor device may include one or more inertial measurement units (IMUs). The machine vibration data can indicate the magnitude of vibration of machine 105 over a period of time.

[0032] The motion sensor device may include one or more devices that sense the speed associated with machine 105 (e.g., engine speed of engine 110 and / or track speed of chassis) and generate speed data identifying the speed associated with machine 105. In some implementations, the motion sensor device may also sense the acceleration of machine 105 and generate acceleration data identifying the acceleration of machine 105. The motion sensor device may also sense the direction of travel of machine 105 and generate direction data identifying the direction of travel of machine 105. The motion sensor device may include an accelerometer, a tachometer, a speedometer, and / or an IMU.

[0033] Operator control 135 may include one or more devices capable of generating operator control data for controlling the operation of machine 105. For example, operator control 135 may be used to control the operation of one or more appliances of machine 105 (e.g., rear attachment 150 and / or front attachment 160) and / or control the operation of ground engagement member 170.

[0034] Operator control data may include appliance command data identifying commands used to control one or more appliances, including steering command data identifying steering commands for machine 105 and / or gear setting data identifying gear settings for machine 105. Operator control 135 may provide operator control data periodically and / or based on triggering events (e.g., to wear detection device 190).

[0035] In some instances, operator control data and / or sensor data can form appliance data. When predicting the amount of wear on one or more components of the chassis, the wear detection device 190 can use the appliance data to determine whether one or more appliances are engaging the ground surface (and thus causing the machine 105 to vibrate), as described in more detail below.

[0036] Wear detection device 190 may include one or more devices (e.g., server devices or a group of server devices) configured to predict the amount of wear on one or more components of the chassis, as described in more detail below. Wear detection device 190 may include one or more processors 220 (hereinafter referred to as “processor 220”) and one or more memories 230 (hereinafter referred to as “memory 230”).

[0037] Processor 220 is implemented in hardware, firmware, and / or a combination of hardware and software. Processor 220 includes a central processing unit (CPU), graphics processing unit (GPU), accelerated processing unit (APU), microprocessor, microcontroller, digital signal processor (DSP), field-programmable gate array (FPGA), application-specific integrated circuit (ASIC), or another type of processing unit. Processor 220 can be programmed to perform functions.

[0038] Memory 230 includes random access memory (RAM), read-only memory (ROM), and / or another type of dynamic or static storage device (e.g., flash memory, magnetic storage, and / or optical storage) that stores information and / or instructions for use by processor 220 to perform functions. For example, when performing the function, wear detection device 190 can acquire sensor data (e.g., from sensor system 120), operator control data (e.g., from operator control 135), and / or historical wear data (e.g., from measuring device 210), and can predict the amount of wear of one or more components based on the sensor data, operator control data, and / or historical wear data.

[0039] In some implementations, the wear detection device 190 can be implemented using one or more computing resources in a cloud computing environment. For example, the wear detection device 190 can be hosted in a cloud computing environment. Alternatively, the wear detection device 190 can be non-cloud-based or partially cloud-based.

[0040] The measuring device 210 may include one or more devices capable of providing historical wear data regarding the amount of wear on a historical component. The historical component may be similar to or identical to one or more components of the chassis (e.g., similar or identical type of component, similar or identical specifications, and / or similar or identical type of function performed). The historical wear data may include vibration data of vibrations caused by movement of the historical component and the amount of wear on the historical component corresponding to the vibration data. For example, the historical wear data may include first vibration data associated with a first amount of wear on the historical component, second vibration data associated with a second amount of wear on the historical component, etc. As an example, the historical wear data may include a data structure that includes the above information.

[0041] Historical wear data may include information obtained based on measurements of historical components (e.g., manual measurements associated with historical inspections). For example, the amount of wear on a historical component may be determined based on measurements of that component. Historical wear data may be used by wear detection device 190 to predict the amount of wear on one or more components of the chassis, as described in more detail below. Measurement device 210 may provide historical wear data periodically and / or based on triggering events (e.g., to wear detection device 190).

[0042] like Figure 2 As shown, the wear detection device 190 can receive sensor data from the sensor system 120, operator control data from the operator control 135, and / or historical wear data from the measuring device 210. The sensor data may include machine vibration data received from the vibration sensor device. The machine vibration data can indicate the degree of vibration of the machine 105 over a period of time.

[0043] Vibrations in machine 105 may include a first vibration caused by the movement of one or more components (e.g., cyclic movement of one or more components causing movement of machine 105) and a second vibration unrelated to the first vibration. The second vibration may be caused by one or more implements of machine 105 (e.g., front attachment 150 and / or rear attachment 150) engaging with a ground surface (e.g., the ground surface on which machine 105 is traveling). The second vibration may also be caused by one or more other components of machine 105, such as engine 110.

[0044] The wear detection device 190 can analyze machine vibration data to identify a first portion of the machine vibration data corresponding to a first vibration and a second portion of the machine vibration data corresponding to a second vibration. For example, the wear detection device 190 can use device data indicating whether one or more devices of the machine 105 are engaging with the ground surface to analyze the machine vibration data.

[0045] The wear detection device 190 can identify a first portion of machine vibration data based on instrument data indicating that one or more instruments of machine 105 are not engaged with the ground surface. The wear detection device 190 can identify a second portion of machine vibration data based on instrument data indicating that one or more instruments of machine 105 are engaged with the ground surface. The wear detection device 190 can remove the second portion of machine vibration data from consideration.

[0046] The equipment data may include velocity data, acceleration data, and / or orientation data generated during the time period (e.g., included in sensor data) and / or may include equipment command data, steering command data, and / or gear setting data generated during the time period (e.g., included in operator control data). The equipment data may include information identifying one or more first time portions corresponding to a first vibration (one or more first time portions of the time period associated with the machine vibration data) and information identifying one or more second time portions corresponding to a second vibration (one or more second time portions of the time period).

[0047] For example, one or more first time segments may correspond to the time during which one or more appliances do not cause machine 105 to vibrate (e.g., because one or more appliances are not engaging the ground surface). One or more second time segments may correspond to the time during which one or more appliances cause machine 105 to vibrate (e.g., because one or more appliances are engaging the ground surface).

[0048] Wear detection device 190 can determine whether one or more implements are engaging with a ground surface based on machine speed data, acceleration data, and / or direction data. For example, when machine speed data indicates that the speed (associated with machine 105) is constant, when acceleration data indicates that the acceleration of machine 105 is constant, and / or when direction data indicates that machine 105 is traveling in a forward direction, wear detection device 190 can determine that one or more implements are not engaging with a ground surface (e.g., a first vibration).

[0049] Alternatively, when machine speed data indicates that the speed is variable (e.g., because machine 105 is performing a task associated with moving material using one or more tools), when machine speed data indicates that the speed meets a threshold speed (e.g., the speed associated with moving material using one or more tools), when acceleration data indicates that the acceleration of machine 105 is variable, and / or when direction data indicates that machine 105 is traveling in a rearward direction, the wear detection device 190 may determine that one or more tools are engaging the ground surface (e.g., a second vibration).

[0050] In some cases, the wear detection device 190 can determine whether one or more appliances are engaging the ground surface based on appliance command data, steering command data, and / or gear setting data. For example, when appliance command data indicates that the operator of machine 105 has not issued a command to use one or more appliances, when steering command data indicates that the operator has not issued a steering command associated with the use of one or more appliances, and / or when gear setting data indicates that machine 105 is in a gear unrelated to the use of one or more appliances, the wear detection device 190 can determine that one or more appliances are not engaging the ground surface.

[0051] Alternatively, when appliance command data indicates that the operator has issued a command to use one or more appliances, when steering command data indicates that the operator has issued a steering command associated with the use of one or more appliances, and / or when gear setting data indicates that machine 105 is in a gear associated with the use of one or more appliances, wear detection device 190 may determine that one or more appliances are engaging the ground surface.

[0052] The wear detection device 190 can analyze machine vibration data in conjunction with instrument data to isolate the first vibration from the second vibration based on information identifying one or more first time segments corresponding to a first vibration (for the machine vibration data) and information identifying one or more second time segments corresponding to a second vibration (e.g., included in the instrument data). In some cases, the wear detection device 190 can identify segments of the machine vibration data as time-domain signals based on information identifying one or more first time segments and one or more second time segments of the time.

[0053] As an example, the wear detection device 190 can segment machine vibration data into time-domain signals based on information identifying one or more first time segments of the time period and information identifying one or more second time segments of the time period. The time-domain signals may include a first time-domain signal corresponding to a first portion (or first segment) of the machine vibration data and a second time-domain signal corresponding to a second portion (or second segment) of the machine vibration data. The wear detection device 190 can identify and focus on the first portion (or first segment) of the machine vibration data corresponding to a first vibration, and can identify and discard the second portion (or second segment) of the machine vibration data corresponding to a second vibration.

[0054] In some implementations, the first time-domain signal may include a first time-domain portion corresponding to the motion (e.g., a first vibration) of one or more components and a second time-domain portion corresponding to random noise associated with the terrain conditions at the location of machine 105. As examples, the random noise may be caused by machine 105 traveling on an uneven ground surface, by machine 105 traveling over potholes, by machine 105 impacting rocks on the ground surface, and / or by another event affecting the movement of machine 105.

[0055] The wear detection device 190 can transform a first time-domain signal into a spectral-domain signal to amplify (or maximize) a first time-domain portion of the first time-domain signal and reduce (or minimize) a second time-domain portion of the first time-domain signal. In some instances, the wear detection device 190 can use a Fast Fourier Transform (FFT) to perform the transformation of the first time-domain signal. Alternatively, the wear detection device 190 can perform the transformation of the first time-domain signal by determining (or calculating) the power spectral density (PSD) of the first time-domain signal. The PSD can be determined using an FFT.

[0056] The wear detection device 190 can determine the PSD because the frequency resolution associated with the PSD exceeds the frequency resolution associated with the transform using FFT. For example, the PSD can be compared between a first time domain portion and a second time domain portion, which exceeds the comparison between the first time domain portion and the second time domain portion enabled by FFT.

[0057] The wear detection device 190 can analyze spectral domain signals to identify characteristic spectra associated with the motion of one or more components. For example, the wear detection device 190 can analyze spectral domain signals to identify spectral features associated with a first time domain portion. The wear detection device 190 can identify characteristic spectra in a portion of the power spectral density corresponding to the first time domain portion.

[0058] Wear detection device 190 can predict the amount of wear of one or more components based on the amplitude of a characteristic spectrum (e.g., the amplitude of a portion of the characteristic spectrum). As an example, wear detection device 190 can predict the amount of wear of one or more components based on the amplitude of the characteristic spectrum and historical wear data. As explained above, historical wear data may include first vibration data associated with a first amount of wear of a historical component, second vibration data associated with a second amount of wear of a historical component, etc.

[0059] The first vibration data may include information identifying a first characteristic spectrum associated with vibrations caused by the movement of a historical component (e.g., when the wear amount of the historical component is a first wear amount) and information identifying the amplitude of the first characteristic spectrum. The second vibration data may include information identifying a second characteristic spectrum associated with vibrations caused by the movement of a historical component (e.g., when the wear amount of the historical component is a second wear amount) and information identifying the amplitude of the second characteristic spectrum.

[0060] Wear detection device 190 can predict the wear amount of one or more components by analyzing historical wear data to identify specific vibration data (e.g., a specific amplitude of a specific characteristic spectrum) corresponding to the amplitude of a characteristic spectrum. For example, suppose wear detection device 190 determines that the amplitude of a characteristic corresponds to the amplitude of a second characteristic spectrum. In this case, wear detection device 190 can predict the wear amount of one or more components corresponding to the second wear amount of the historical components.

[0061] In addition to predicting the wear amount of one or more components, as an alternative to predicting the wear amount of one or more components, the wear detection device 190 may predict the wear rate of one or more components and / or the date and / or time when one or more components will be replaced and / or repaired. The predicted wear rate of one or more components, the predicted wear amount of one or more components, and / or the predicted date and / or time may be referred to below as "predicted component wear information".

[0062] In some implementations, the wear detection device 190 may use a machine learning model to determine predicted component wear information. For example, the wear detection device 190 may input sensor data and / or operator control data into the machine learning model, and the machine learning model may output the predicted component wear information. The wear detection device 190 may use historical data associated with machine 105 and / or with one or more other machines similar to machine 105 to train the machine learning model. The one or more machines may include components similar to machine 105, similar dimensions, and / or similar uses.

[0063] Similar components may include similar ground engagement members 170, similar tracks, similar track links such as first track link 172 and / or second track link 174, similar track link bushings such as first track link bushing 176, similar track link pins such as first track link pin 178 and / or second track link pin 180, similar sprockets 182, similar segments 184 and / or similar idler pulleys 186. Historical data may include historical sensor data, historical operator control data, and / or historical wear data. Historical sensor data may include sensor data received from sensor system 120, historical operator control data may include operator control data received from operator control 135, and / or historical wear data may include historical wear data received from measuring device 210.

[0064] When training a machine learning model, the wear detection device 190 can divide the training data into a training set (e.g., a set of data used to train the machine learning model), a validation set (e.g., a set of data used to evaluate the fit of the machine learning model and / or fine-tune the machine learning model), a test set (e.g., a set of data used to evaluate the final fit of the machine learning model), etc. The wear detection device 190 can preprocess and / or perform dimensionality reduction to reduce the training data to a minimum feature set. The wear detection device 190 can train the machine learning model on the minimum feature set, thereby reducing the processing required for training the machine learning model, and classification techniques can be applied to the minimum feature set.

[0065] Wear detection device 190 can use classification techniques such as logistic regression, random forest, and gradient boosting machine learning (GBM) to determine the classification result (e.g., the amount of wear on one or more parts). In addition to using classification techniques, or as an alternative, wear detection device 190 can use a Naive Bayes classifier. In this case, wear detection device 190 can perform binary recursive partitioning to divide the training data into partitions and / or branches with a minimal feature set, and use these partitions and / or branches to perform predictions (e.g., the wear rate and / or amount of wear on one or more parts). Based on the use of recursive partitioning, wear detection device 190 can reduce the utilization of computational resources compared to manual linear sorting and analysis of data items, thereby enabling the model to be trained using thousands, millions, or billions of data items, which can produce a more accurate model than using fewer data items.

[0066] The wear detection device 190 can use a supervised training procedure to train a machine learning model. This supervised training procedure includes receiving input to the machine learning model from subject matter experts (e.g., one or more operators associated with machine 105 and / or one or more machines). This can reduce the amount of time, processing resources, etc., required to train the machine learning model compared to an unsupervised training procedure. The wear detection device 190 can use one or more other model training techniques, such as neural network techniques, latent semantic indexing techniques, etc.

[0067] For example, the wear detection device 190 can perform artificial neural network processing techniques (e.g., using a two-layer feedforward neural network architecture, a three-layer feedforward neural network architecture, etc.) to perform pattern identification of patterns of different wear amounts on one or more components. In this case, using artificial neural network processing techniques can improve the accuracy of the machine learning model generated by the wear detection device 190 by making it more robust to noisy, inaccurate, or incomplete data, and by enabling the wear detection device 190 to use less sophisticated techniques to detect patterns and / or trends that human analysts or systems cannot detect.

[0068] After training, the machine learning model can be used to determine (or predict) the predicted wear information of a component. In other words, after training the machine learning model, the wear detection device 190 can receive sensor data and / or operator control data from the machine 105, and input the received sensor data and operator control data into the machine learning model, and the machine learning model can output data related to the wear rate and / or wear amount of one or more components. The output of the machine learning model may include a score for the predicted wear information of the component. The score for the predicted wear information of the component can represent a measure of the confidence level of the predicted wear information of the component.

[0069] Different devices, such as server devices, can generate and train machine learning models. Different devices can provide machine learning models for use by the wear detection device 190. Different devices can update the machine learning model and provide it to the wear detection device 190 (e.g., on a predetermined basis, on an on-demand basis, on a triggered basis, on a periodic basis, etc.). In some cases, the wear detection device 190 can receive additional training data (e.g., additional historical sensor data, additional historical operator control data, and / or additional historical wear data) and retrain the machine learning model. Alternatively, the wear detection device 190 can provide additional training data to different devices to train the machine learning model. The machine learning model can be retrained periodically and / or based on triggered events.

[0070] In some implementations, the wear detection device 190 may provide a machine learning model to the controller 140, enabling the controller 140 to determine predicted component wear information. Alternatively, the wear detection device 190 may receive a request from the controller 140 to determine predicted component wear information. The request may include sensor data from the machine 105.

[0071] Wear detection device 190 (and / or controller 140) can perform actions based on predicted component wear information. For example, actions may include wear detection device 190 adjusting the operation of machine 105 based on predicted wear of one or more components (e.g., when the predicted wear meets a threshold wear level). For example, wear detection device 190 may cause changes in the speed of machine 105, changes in the acceleration of machine 105, changes in the direction of travel of machine 105, changes in implement commands, changes in steering commands, changes in gear settings, and / or other actions that may reduce the wear rate of one or more components and extend the time until one or more components must be repaired or replaced.

[0072] The wear detection device 190 can navigate the machine 105 to different work sites and perform one or more tasks at those sites in an effort to extend the life of one or more components. For example, different work sites can be associated with a wear rate (the wear rate of one or more components) that is less than the wear rate (the wear rate of one or more components) associated with the work site where the machine 105 is currently located. Additionally or alternatively, the wear detection device 190 can cause the machine 105 to perform different tasks in an effort to extend the life of one or more components. For example, different tasks can be associated with a wear rate (the wear rate of one or more components) that is less than the wear rate (the wear rate of one or more components) associated with the task that the machine 105 is currently performing.

[0073] This action may include wear detection device 190 sending wear information to one or more devices that monitor the wear of components of multiple machines (e.g., including machine 105). In some instances, wear detection device 190 may send wear information when the wear amount (the wear amount of one or more components) meets a threshold wear amount. The wear information may indicate the wear amount of one or more components, indicate the wear rate of one or more components, indicate the wear quantity of one or more components, and / or a quote associated with repairing and / or replacing the one or more components. The one or more devices may include devices of a field management system, devices of a back-end system, devices associated with the operator of machine 105, devices associated with technicians, and / or controller 140.

[0074] Wear detection device 190 can send wear information to cause one or more devices to command one or more replacement parts. In some cases, the wear information may include information identifying one or more parts and / or one or more replacement parts.

[0075] Wear detection device 190 can send wear information to cause one or more devices (e.g., controller 140) to autonomously navigate machine 105 to a repair facility. Additionally or alternatively, wear detection device 190 can send wear information to cause one or more devices to load calendar events into a technician's calendar for inspection and / or repair of one or more components. Additionally or alternatively, wear detection device 190 can send wear information to cause one or more devices (e.g., controller 140) to activate an alarm. The alarm can indicate that one or more components need to be repaired or replaced.

[0076] In some cases, the wear detection device 190 may send wear information to cause one or more devices to generate a service request to repair and / or replace one or more components. As part of generating the service request, the one or more devices may perform one or more of the actions described herein.

[0077] In some instances, the action may include the wear detection device 190 causing the first autonomous device to deliver one or more replacement parts to a location associated with machine 105. The location may include the current location of machine 105, the location of the work site where machine 105 performs multiple tasks, the location of machine 105 when machine 105 is not performing tasks, and / or the location of machine 105 when machine 105 is undergoing repair and / or replacement. In some cases, wear information may include information identifying the location associated with machine 105.

[0078] In some instances, the action may include the wear detection device 190 navigating the second autonomous device to a location associated with the machine 105 to verify predicted component wear information. The second autonomous device may generate verification information based on the verified component wear information and may send the verification information to the wear detection device 190. The wear detection device 190 may use the verification information to retrain a machine learning model.

[0079] In some cases, the wear detection device 190 can (e.g., based on predicted component wear information) determine whether a failure of one or more components is imminent. If the wear detection device 190 determines that a failure is imminent, it can perform one or more of the actions described above. If the wear detection device 190 determines that a failure is not imminent, it may not perform any action.

[0080] Figure 2The number and arrangement of the devices and networks shown are provided as examples. In practice, there may be additional devices, fewer devices, different devices, or devices connected to... Figure 2 The apparatus shown is arranged in different ways. Furthermore, Figure 2 The two or more devices shown can be implemented within a single device, or Figure 2 The single device shown can be implemented as multiple distributed devices. Additionally or alternatively, a group of devices (e.g., one or more devices) of system 200 can perform one or more functions described as being performed by another group of devices of system 200.

[0081] Figure 3 This is a flowchart of an example process 300 associated with chassis wear prediction based on machine vibration data. Figure 3 One or more process frames can be executed by a device (e.g., wear detection device 190). Figure 3 One or more process frames may be executed by another device or a group of devices that are separate from or include the device, such as a controller (e.g., controller 140).

[0082] like Figure 3 As shown, process 300 may include receiving machine vibration data (block 310) that identifies the vibration measure of the machine over a period of time. For example, as described above, the device may receive machine vibration data that identifies the vibration measure of the machine over a period of time.

[0083] As an example, receiving machine vibration data may include receiving machine vibration data from one or more first sensor devices of the machine. Receiving device data may include at least one of the following: receiving sensor data indicating whether one or more devices are engaging with the ground surface from one or more second sensor devices of the machine, or receiving operator control data indicating whether one or more devices are engaging with the ground surface from one or more operator controls of the machine.

[0084] like Figure 3 As further shown, process 300 may include segmenting machine vibration data to obtain time-domain signals, including time-domain signals related to vibrations associated with the machine's chassis (block 320). For example, as described above, the apparatus may segment machine vibration data to obtain time-domain signals, including time-domain signals related to vibrations associated with the machine's chassis.

[0085] As an example, machine vibration can be caused by the movement of components and by one or more machine fixtures engaging with a ground surface, wherein the method further includes receiving fixture data indicating whether one or more fixtures are engaging with the ground surface during a certain period of time. Segmenting machine vibration data includes segmenting the machine vibration data based on fixture data to identify time-domain signals.

[0086] The time-domain signal can be a first time-domain signal. The time-domain signal can also include a second time-domain signal related to vibration, which is independent of the vibration associated with the machine's chassis.

[0087] like Figure 3 As further shown, process 300 may include transforming a time-domain signal into a spectral-domain signal using a Fast Fourier Transform (FFT) (block 330). For example, as described above, the apparatus may use a Fast Fourier Transform (FFT) to transform a time-domain signal into a spectral-domain signal.

[0088] Transforming a time-domain signal into a spectral-domain signal amplifies the first portion of the time-domain signal corresponding to the motion of the component based on the power spectral density of the time-domain signal, and reduces the second portion of the time-domain signal corresponding to random noise associated with the terrain conditions at the machine's location based on the power spectral density of the time-domain signal. The power spectral density can be determined using FFT. The characteristic spectrum can be identified in a portion of the power spectral density corresponding to the first portion of the time-domain signal.

[0089] like Figure 3 As further shown, process 300 may include identifying a characteristic spectrum (block 340) from a spectral domain signal that is associated with the motion of components of the machine's chassis. For example, as described above, the device may identify a characteristic spectrum associated with the motion of components of the machine's chassis from a spectral domain signal.

[0090] like Figure 3 As further shown, process 300 may include predicting the wear amount of a component based on the amplitude of a characteristic spectrum (box 350). For example, as described above, the device may predict the wear amount of a component based on the amplitude of a characteristic spectrum.

[0091] Predicting the wear of a component can include using amplitude of a characteristic spectrum and historical wear data associated with the component.

[0092] like Figure 3 As further shown, process 300 may include performing an action based on the amount of wear on a component (block 360). For example, as described above, the apparatus may perform an action based on the amount of wear on a component.

[0093] The actions performed may include at least one of the following: repairing a component, replacing a component, adjusting the operation of the machine, or providing an alarm to the operator's device.

[0094] Although Figure 3 Example boxes for process 300 are shown, but in some embodiments, process 300 may include additional boxes, fewer boxes, different boxes, or boxes with... Figure 3The boxes depicted in the diagram are arranged in different ways. Additionally or alternatively, two or more boxes in process 300 can be executed in parallel.

[0095] Industrial applicability

[0096] This disclosure relates to an apparatus that predicts the amount of wear on one or more components of a machine chassis based on sensor data from a sensor device associated with machine vibrations. The disclosed apparatus for predicting the amount of wear on one or more components can prevent problems associated with manual measurement of the machine's tracks (to determine the amount of track wear) and erroneous predictions of track wear.

[0097] Manual measurement of tracks can waste machine resources that are already in use to prevent machine movement while obtaining manual measurements. Additionally, erroneous manual measurements of tracks and / or inaccurate predictions of track wear can waste computational resources that are used to remedy problems associated with these errors (e.g., premature track failure, premature track repair, and / or premature track replacement).

[0098] The disclosed apparatus for predicting the wear of one or more chassis components solves the aforementioned problems regarding manual measurement and erroneous predictions of wear. Several advantages can be associated with the disclosed apparatus. For example, by predicting the wear of one or more components based on machine vibration data, the apparatus can limit any interruption to machine operation and / or limit machine immobilization, thereby enabling the machine to remain operational for an extended period (e.g., until the tracks require repair or replacement). By predicting the wear of one or more components based on machine vibration data, the apparatus enables the tracks to be repaired or replaced when necessary (as opposed to premature repair or replacement).

[0099] By predicting the wear of one or more components based on machine vibration data, this device can reduce erroneous predictions of wear. By reducing erroneous predictions of track wear, the device can decrease the likelihood of track failure before it needs repair and / or replacement. Furthermore, by reducing erroneous predictions of track wear, the device can conserve computational or machine resources that could be used to remedy problems associated with inaccurate manual measurements and erroneous predictions of track wear (e.g., premature track failure, premature track repair, and / or premature track replacement).

[0100] The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the embodiments to the precise forms disclosed. Modifications and variations can be made based on the foregoing disclosure, or can be derived from practice of the embodiments. Furthermore, any embodiments described herein can be combined unless the foregoing disclosure expressly provides for reasons why one or more embodiments cannot be combined. Even if specific combinations of features are recited in the claims and / or disclosed in the specification, these combinations are not intended to limit the disclosure of various embodiments. Although each dependent claim listed below may be directly subordinate to only one claim, the disclosure of various embodiments includes each dependent claim in combination with all other claims in the group of claims.

[0101] As used herein, “a,” “one,” and “a group” are intended to include one or more things and are interchangeable with “one or more.” Furthermore, as used herein, the article “the” is intended to include one or more things referred to in conjunction with the article “the” and is interchangeable with “the one or more.” Additionally, the phrase “based on” is intended to mean “at least partially based on,” unless otherwise expressly stated. Furthermore, as used herein, the term “or,” when used serially, is intended to be inclusive and is interchangeable with “and / or,” unless otherwise expressly stated (e.g., if used in combination with “either” or “only one”).

Claims

1. A method for predicting wear of the chassis of an earthmoving machine (105), the chassis comprising chassis components, the chassis components including one or more of the following: tracks, one or more track links, one or more track link bushings, one or more track link pins, one or more sprockets, one or more segments, and one or more idler pulleys. The earthmoving machine also includes one or more devices, including front attachments and / or rear attachments. The method is performed by the apparatus (190), and the method includes: Vibration data of a soil transport machine (105) is received, the vibration data of the soil transport machine (105) identifying the vibration magnitude of the soil transport machine (105) over a period of time, wherein the causes of the vibration of the soil transport machine include: The movement of the chassis components; and The one or more implements of the earthmoving machine (105) engage with the ground surface; The receiver receives device data, which indicates whether the one or more devices are engaging the ground surface during the time period. The vibration data of the earthmoving machine (105) is segmented to obtain a time-domain signal, the time-domain signal including a time-domain signal related to the vibration associated with the chassis of the earthmoving machine (105), wherein segmenting the vibration data of the earthmoving machine (105) includes segmenting the vibration data of the earthmoving machine (105) based on the equipment data to identify the time-domain signal. The time-domain signal is transformed into a spectral-domain signal using a Fast Fourier Transform (FFT); The characteristic spectrum associated with the motion of the components (172, 174, 176, 178, 180, 182, 184, 186) of the chassis of the earthmoving machine (105) is identified from the spectral domain signal. The wear amount of the components (172, 174, 176, 178, 180, 182, 184, 186) is predicted based on the amplitude of the characteristic spectrum; and The action is performed based on the wear amount of the components (172, 174, 176, 178, 180, 182, 184, 186).

2. The method of claim 1, wherein receiving vibration data of the earthmoving machine (105) comprises receiving vibration data of the earthmoving machine (105) from one or more first sensor devices of the earthmoving machine (105); and The receipt of the device data includes at least one of the following: Sensor data is received from one or more second sensor devices of the earthmoving machine (105), the sensor data indicating whether the one or more implements (150, 160) are engaging the ground surface; or Operator control data is received from one or more operator controls (135) of the earthmoving machine (105), the operator control data indicating whether the one or more implements (150, 160) are engaging the ground surface.

3. The method according to any one of claims 1 to 2, wherein transforming the time-domain signal into the spectral-domain signal comprises: The first portion of the motion of the time-domain signal corresponding to the components (172, 174, 176, 178, 180, 182, 184, 186) is amplified based on the power spectral density of the time-domain signal, and The second portion of the time-domain signal corresponding to the random noise associated with the terrain conditions at the location of the earthmoving machine (105) is reduced based on the power spectral density of the time-domain signal. The power spectral density is determined using the FFT; and The characteristic spectrum is identified in a portion of the power spectral density corresponding to the first part of the time-domain signal.

4. The method according to any one of claims 1 to 2, wherein performing the action comprises at least one of the following: To enable the repair of the aforementioned components (172, 174, 176, 178, 180, 182, 184, 186); The components (172, 174, 176, 178, 180, 182, 184, 186) are replaced; The operation of the earthmoving machine (105) is adjusted; or An alarm is provided to the device (190) of the operator of the earthmoving machine (105).

5. A soil-moving machine, comprising: Device (190), the device (190) is configured to: Vibration data of the earthmoving machine (105) is received, wherein the vibration data of the earthmoving machine (105) identifies the vibration magnitude of the earthmoving machine (105). The vibration of the earthmoving machine (105) is caused by a combination of a first vibration caused by the movement of the components (172, 174, 176, 178, 180, 182, 184, 186) of the chassis of the earthmoving machine (105) and a second vibration unrelated to the first vibration; The chassis described herein includes one or more of the following: tracks, one or more track links, one or more track link bushings, one or more track link pins, one or more sprockets, one or more segments, and one or more idler pulleys; and The earthmoving machine includes one or more devices that cause the second vibration, the devices including front attachments and / or rear attachments; The device is further configured to: The segment corresponding to the first vibration is identified by the vibration data of the earthmoving machine (105), wherein identifying the segment includes receiving instrument data and identifying the segment based on the instrument data, the instrument data indicating whether the instrument of the earthmoving machine (105) is engaging the ground surface; The segments are transformed into signals in the frequency domain using a Fast Fourier Transform (FFT). The signals are analyzed to identify the characteristic spectrum associated with the motion of the components (172, 174, 176, 178, 180, 182, 184, 186); The wear amount of the components (172, 174, 176, 178, 180, 182, 184, 186) is predicted based on the characteristic spectrum; and The action is performed based on the wear amount of the components (172, 174, 176, 178, 180, 182, 184, 186).

6. The earthmoving machine according to claim 5, wherein when predicting the amount of wear of the components (172, 174, 176, 178, 180, 182, 184, 186), the device (190) is configured to: The wear amount of the components (172, 174, 176, 178, 180, 182, 184, 186) is predicted based on the amplitude of the characteristic spectrum.

7. The earthmoving machine according to any one of claims 5 to 6, wherein when the action is performed, the device (190) is configured to: To enable the repair of the aforementioned components (172, 174, 176, 178, 180, 182, 184, 186); The components (172, 174, 176, 178, 180, 182, 184, 186) are replaced; The operation of the earthmoving machine (105) is adjusted; Information about the amount of wear is provided to the device (190), which monitors the wear of components (172, 174, 176, 178, 180, 182, 184, 186) of the earthmoving machine (105); or An alarm is provided to the device (190) of the operator of the earthmoving machine (105).

8. The earthmoving machine according to any one of claims 5 to 6, wherein the equipment data includes at least one of the following: Speed ​​data of the earthmoving machine (105), wherein the speed data of the earthmoving machine (105) indicates the speed of the earthmoving machine (105); Appliance command data, wherein the appliance command data identifies commands used to control the appliances (150, 160); Steering command data, wherein the steering command data identifies the steering command of the earthmoving machine (105); or Gear setting data, which identifies the gear setting of the earthmoving machine (105).

Images