Chassis wear prediction based on machine vibration data

Abstract

Provides chassis wear prediction based on machine vibration data. The system may include a device. The device may be configured to receive machine vibration data identifying vibration measurements of the machine. The machine vibration may be caused by a combination of a first vibration caused by movement of a component of a chassis of the machine and a second vibration independent of the first vibration. The device may be configured to identify a segment of the machine vibration data corresponding to the first vibration, convert the segment into a frequency domain signal using a fast Fourier transform (FFT), and analyze the signal to identify a signature spectrum associated with the movement of the component. The device may be configured to predict an amount of wear on the component based on the signature spectrum. The device may be configured to perform an action based on the amount of wear on the component.

Description

[Technical field]

[0001] FIELD OF THE DISCLOSURE This disclosure relates generally to monitoring wear on a machine chassis, for example, to predicting chassis wear based on machine vibration data. [Background technology]

[0002] Machine chassis components (e.g., track links, bushings, and / or pins) wear over time. Techniques for detecting wear on components include obtaining manual measurements of component dimensions of such components. The manual measurements may be compared to specified dimensions of the components. To obtain the manual measurements, the machine must pause performance of a task on the work site. Because obtaining manual measurements requires the machine to pause performance of a task and is a time-consuming process (e.g., due to travel time to obtain the manual measurements and / or time to obtain the manual measurements), obtaining manual measurements may adversely affect productivity on the work site. In this regard, a task (a task performed by a machine) may be interrupted for an extended period of time (e.g., the period to obtain the manual measurements).

[0003] Additionally, such manual measurements may be inaccurate. Conversely, inaccurate measurements of part dimensions may lead to inaccurate predictions of the amount of wear on the part. Such inaccurate predictions may result in premature failure or repair or replacement of the part (e.g., the part may not have worn down enough to require replacement or repair). Such premature part failure or premature replacement or repair of the part may also have a negative impact on work site productivity.

[0004] International Patent Publication No. WO2020049526 (the "562" disclosure) discloses a mounting assembly including 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 of a track system. The '562 publication also discloses that the track system further includes at least one actuator for pivoting the frame assembly about at least one of the roll and yaw pivot axes, and at least one monitoring sensor for at least indirectly determining at least one of a state of the track system and ground conditions. The '562 publication discloses that the at least one monitoring sensor is in communication with a track system controller to control action of the at least one actuator based on at least one of the state of the track system and ground conditions.

[0005] Although the '562 publication discloses that at least one monitoring sensor is in communication with a track system controller, the '562 publication does not disclose that the track system controller receives vibration data from the at least one monitoring sensor, or that the track system controller takes into account factors that affect machine vibration in determining the amount of wear in the track system.

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

[0007] A method performed by an apparatus includes receiving machine vibration data identifying vibration measurements of a machine over a period of time; segmenting the machine vibration data to obtain a time domain signal including a time domain signal related to vibration associated with a chassis of the machine; converting the time domain signal to a spectral domain signal using a Fast Fourier Transform (FFT); identifying a signature spectrum from the spectral domain signal related to movement of a component of the chassis of the machine; predicting an amount of wear of the component based on an amplitude of the signature spectrum; and performing an action based on the amount of wear of the component.

[0008] The system includes an apparatus configured to receive machine vibration data identifying a measurement of vibration of the machine, the vibration of the machine being caused by a combination of a first vibration caused by movement of the part of a chassis of the machine and a second vibration independent of the first vibration, identify a segment in the machine vibration data corresponding to the first vibration, convert the segment into a frequency domain signal using a Fast Fourier Transform, analyze the signal to identify a signature spectrum associated with movement of the part, predict an amount of wear of the part based on the signature spectrum, and perform an action based on the amount of wear of the part.

[0009] The apparatus includes one or more memories and one or more processors configured to receive the machine vibration data identifying a measurement of vibration of the machine, the vibration of the machine being caused by a combination of a first vibration caused by movement of the part of the chassis of the machine and a second vibration independent of the first vibration, identify a segment in the machine vibration data corresponding to the first vibration, convert the segment into a frequency domain signal using a Fast Fourier Transform, analyze the signal to identify a signature spectrum associated with movement of the part, predict an amount of wear of the part based on the signature spectrum, and perform an action based on the amount of wear of the part. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram of an exemplary implementation described herein. [Diagram 2] FIG. 1 is a diagram of an example described herein. [Diagram 3] 1 is a flowchart of an example process related to chassis wear prediction using a machine learning model. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] The present disclosure relates to an apparatus for predicting the amount of wear of one or more components of a chassis of a machine based on mechanical vibration data of the machine. The term "machine" may refer to any machine that performs an operation related to an industry, such as mining, construction, agriculture, transportation, or other industry. Furthermore, one or more tools may be connected to the machine.

[0012] 1 is a diagram of an example implementation 100 described herein. As shown in FIG. 1, the example implementation 100 includes a machine 105 and a wear detection device 190. The machine 105 is embodied as an earth-moving machine, such as a dozer. Alternatively, the machine 105 may be another type of track-type machine, such as an excavator.

[0013] As shown in FIG. 1, machine 105 includes an engine 110 , a sensor system 120 , an operator cabin 130 , operator controls 135 , a controller 140 , a rear mounting 150 , a front mounting 160 , and ground engaging members 170 .

[0014] Engine 110 may include a compression ignition engine, a spark ignition engine, a laser ignition engine, a plasma ignition engine, and / or an internal combustion engine. Engine 110 powers machine 105 and / or a set of loads associated with machine 105 (e.g., components that absorb and / or operate using electrical power). For example, engine 110 may power one or more control systems (e.g., controller 140), sensor system 120, operator cabin 130, and / or ground engaging members 170.

[0015] Engine 110 may power tools of machine 105 used in mining, construction, agriculture, transportation, or other industries. For example, engine 110 may provide power to components (e.g., one or more hydraulic pumps, one or more actuators, and / or one or more electric motors) to facilitate control of rear mount 150 and / or front mount 160 of machine 105.

[0016] Sensor system 120 may include sensor devices (described in more detail below) capable of generating signals that may be used to predict the amount of wear on one or more components of a chassis of machine 105. Types of sensor devices in sensor system 120 are described in more detail below in connection with FIG.

[0017] Operator cabin 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, control levers, and / or a steering wheel) for controlling the action of machine 105. For example, operator controls 135 may be used to control the action of one or more tools (e.g., rear mount 150 and / or front mount 160) of machine 105 and / or to control the action of ground engaging members 170.

[0018] For an autonomous machine, the operator controls may not be designed for use by an operator, but rather may be designed to operate independently of an operator, for example, the operator controls may include one or more input components that provide input signals used by other components without operator input.

[0019] Controller 140 (e.g., electronic control module (ECM)) may control and / or monitor actions of machine 105. For example, controller 140 may control and / or monitor actions of machine 105 based on signals from operator controls 135, signals from sensor system 120, and / or signals from wear detection device 190. Controller 140 may predict an 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.

[0020] The rear attachment 150 may include a ripper assembly, a winch assembly, and / or a tow bar assembly. The front attachment 160 may include a blade assembly. The ground engaging member 170 may be configured to propel the machine 105. The ground engaging member 170 may include wheels, tracks, rollers, and / or similar components for propelling the machine 105. The ground engaging member 170 may include a chassis including tracks (as shown in FIG. 1 ). The tracks may include track links. The track links may include track link bushings and track link pins. By way of example, the tracks 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.

[0021] Sprocket 182 may include one or more sections 184 (referred to herein simply as “sections 184” and collectively as “sections 184”). Sprocket 182 may be configured to engage and drive ground engaging members 170. For example, segments 184 may be configured to engage track link bushings (e.g., tracks of ground engaging members 170) and rotate such that the tracks propel the machine 105. In some examples, one or more idlers 186 may guide the tracks as they rotate to advance the machine 105.

[0022] The wear detection device 190 can include one or more devices capable of predicting an amount of wear on one or more components of the chassis (e.g., one or more tracks, one or more track links, e.g., first track link 172 and / or second track link 174, one or more track bushings, e.g., first track link bushing 176, one or more track link pins, e.g., 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 idlers 186). The wear detection device 190 can predict a remaining life of the one or more components based on the amount of wear.

[0023] In some examples, the wear detection device 190 can predict a wear rate of one or more components and predict an amount of wear based on the wear rate. The wear detection device 190 can predict a wear rate and / or amount of wear of one or more components using machine learning models, as described in more detail below. The wear detection device 190 can be located within the machine 105 (e.g., as part of the controller 140), external to the machine 105, partially located within the machine 105, or partially located external to the machine 105.

[0024] As noted above, Figure 1 is provided as an example. Other examples may differ from the example described with reference to Figure 1.

[0025] Figure 2 is a diagram of an example system 200 described herein. As shown in Figure 2, the system 200 includes a sensor system 120, an operator control 135, a wear detection device 190, and a measurement device 210. The wear detection device 190 and / or the measurement device 210 may be part of a site management system (e.g., a work site associated with the machine 105). Alternatively, the wear detection device 190 and / or the measurement device 210 may be part of a back office system or may be part of the machine 105.

[0026] The wear detection device 190 and / or the measurement device 210 may be included in the same device. Alternatively, the wear detection device 190 and / or the measurement device 210 may be separate devices.

[0027] The sensor system 120 may include a sensor device that generates sensor data related to an 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, e.g., first track link 172 and / or second track link 174, one or more track link bushings, e.g., first track link bushing 176, one or more track link pins, e.g., first track link pin 178 and / or second track link pin 180, one or more sprockets 182, one or more sections 184, and / or one or more idlers 186. The sensor data may be used (e.g., by wear detection device 190) to predict an amount of wear on the one or more components. The sensor data may include information identifying a time and / or date that the sensor data was generated.

[0028] The sensor system 120 can provide sensor data for predicting the amount of wear of one or more components, as described in more detail below. For example, the sensor system 120 can provide the sensor data to the wear detection device 190 periodically (e.g., hourly, every time interval, and / or every work shift). Additionally or alternatively, the sensor system 120 can provide the sensor data to the wear detection device 190 based on a trigger event (e.g., a request from the wear detection device 190, a request from the controller 140, and / or a request from an operator of the machine 105 (e.g., via an integrated display and / or operator controls).

[0029] The sensor devices may include vibration sensor devices, motion sensor devices, and / or other sensor devices that provide sensor data that may be used to predict the amount of wear of one or more components. The vibration sensor devices may include one or more devices that sense vibrations of the machine 105 and generate machine vibration data based on the vibrations. As an example, the vibration sensor devices may include one or more inertial measurement units (IMUs). The machine vibration data may be indicative of vibration measurements of the machine 105 over a period of time.

[0030] The motion sensor device may include one or more devices that sense a speed associated with the machine 105 (e.g., engine speed of the engine 110 and / or track speed of the chassis) and generate speed data identifying a speed associated with the machine 105. In some implementations, the motion sensor device may further sense an acceleration of the machine 105 and generate acceleration data identifying the acceleration of the machine 105. The motion sensor device may also sense a direction of travel of the machine 105 and generate directional data identifying the direction of travel of the machine 105. The motion sensor device may include an accelerometer, a tachometer, a speedometer, and / or an IMU.

[0031] Operator controls 135 may include one or more devices capable of generating operator control data for controlling the action of machine 105. For example, operator controls 135 may be used to control the action of one or more tools (e.g., rear mount 150 and / or front mount 160) of machine 105 and / or to control the action of ground engaging members 170.

[0032] The operator control data may include tool command data identifying commands for controlling one or more tools, steering command data identifying steering commands for the machine 105, and / or gear setting data identifying gear settings for the machine 105. The operator control 135 may provide the operator control data (e.g., to the wear detection device 190) periodically and / or based on a trigger event.

[0033] In some examples, the operator control data and / or the sensor data may form tool data. In predicting the amount of wear on one or more components of the chassis, as described in more detail below, the wear detection device 190 may use the tool data to determine whether one or more tools are engaging the ground (and thus whether the machine 105 is vibrating).

[0034] Wear detection device 190 may include one or more devices (e.g., a server device or set of server devices) configured to predict an amount of wear on one or more components of a chassis, as described in more detail below. Wear detection device 190 may include one or more processors 220 (also referred to herein as “processor 220” and collectively referred to as “processor 220”) and one or more memories 230 (also referred to herein as “memory 230” and collectively referred to as “memory 230”).

[0035] Processor 220 may be implemented in hardware, firmware, and / or a combination of hardware and software. Processor 220 may include a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), a microprocessor, a microcontroller, a digital signal processor (DSP), a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or other types of processing components. Processor 220 may be programmed to perform functions.

[0036] Memory 230 may include random access memory (RAM), read only memory (ROM), and / or other types of dynamic or static storage (e.g., flash memory, magnetic memory, and / or optical memory) that store information and / or instructions used by processor 220 to perform functions. For example, when this function is performed, wear detection device 190 may obtain 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 measurement device 210) and may predict the amount of wear of one or more components based on the sensor data, operator control data, and / or historical wear data.

[0037] In some implementations, the wear detection device 190 may be implemented by one or more computing resources of a cloud computing environment. The wear detection device 190 may be hosted in the cloud computing environment, for example. Alternatively, the wear detection device 190 may be non-cloud based or partially cloud based.

[0038] The measurement device 210 may include one or more devices capable of providing historical wear data relating to a historical amount of wear of a part. The historical part may be similar or identical to one or more parts of the chassis (e.g., a similar or identical type of part, similar or identical specifications, and / or performing a similar or identical function). The historical wear data may include vibration data of vibrations caused by a movement of the historical part and a historical amount of wear of the part corresponding to the vibration data. For example, the historical wear data may include first vibration data associated with a first amount of wear of the part in the past, second vibration data associated with a second amount of wear of the part in the past, etc. As an example, the historical wear data may include a data structure including the above information.

[0039] The historical wear data may include information obtained based on past component measurements (e.g., manual measurements associated with past inspections). For example, the amount of past component wear may be determined based on the past component measurements. The wear detection device 190 may use the historical wear data to predict the amount of wear on one or more components of the chassis, as described in more detail below. The measurement device 210 may provide the historical wear data (e.g., to the wear detection device 190) periodically and / or based on a trigger event.

[0040] 2, the wear detection device 190 may receive sensor data from the sensor system 120, operator control data from the operator control 135, and / or historical wear data from the measurement device 210. The sensor data may include machine vibration data received from a vibration sensor device. The machine vibration data may be indicative of vibration measurements of the machine 105 over a period of time.

[0041] The vibrations of the machine 105 may include a combination of a first vibration caused by the movement of one or more parts (e.g., periodic movement of one or more parts that causes the machine 105 to move) and a second vibration that is independent of the first vibration. The second vibration may be caused by one or more tools of the machine 105 (e.g., the front mounting 150 and / or the rear mounting 150) engaging the ground (e.g., the ground on which the machine 105 travels). The second vibration may be caused by one or more other parts of the machine 105, such as the engine 110.

[0042] The wear detection system 190 may analyze the machine vibration data to identify a first portion of the machine vibration data that corresponds to the first vibration and to identify a second portion of the machine vibration data that corresponds to the second vibration. For example, the wear detection system 190 may analyze the machine vibration data using tool data that indicates whether one or more tools of the machine 105 are engaged with the ground.

[0043] The wear detection system 190 may identify a first portion of the machine vibration data based on the tool data indicating that one or more tools of the machine 105 are not engaged with the ground. The wear detection system 190 may identify a second portion of the machine vibration data based on the tool data indicating that one or more tools of the machine 105 are engaged with the ground. The wear detection system 190 may not consider the second portion of the machine vibration data.

[0044] The tool data may include speed, acceleration, and / or direction data generated during a period of time (e.g., included in the sensor data), and / or tool command, steering command, and / or gear setting data generated during a period of time (e.g., included in the operator control data). The tool data may include information identifying one or more first time portions (of a period associated with the machine vibration data) corresponding to a first vibration, and information identifying one or more second time portions (of the period of time) corresponding to a second vibration.

[0045] For example, the one or more first time portions may correspond to when one or more tools are not vibrating the machine 105 (e.g., because one or more tools are not engaged with the ground). The one or more second time portions may correspond to when one or more tools are vibrating the machine 105 (e.g., because one or more tools are engaged with the ground).

[0046] Wear detection system 190 may determine whether one or more tools are engaged with the ground based on the machine speed data, acceleration data, and / or directional data. For example, if the machine speed data indicates that the speed (associated with machine 105) is constant, if the acceleration data indicates that the acceleration of machine 105 is constant, and / or if the directional data indicates that machine 105 is moving forward, wear detection system 190 may determine that one or more tools are not engaged with the ground (e.g., first vibration).

[0047] Alternatively, if the machine speed data indicates that the speed is variable (e.g., because the machine 105 is performing a task associated with moving material using one or more tools), if the machine speed data indicates that the speed meets a threshold speed (e.g., a speed associated with moving material using one or more tools), if the acceleration data indicates that the acceleration of the machine 105 is variable, and / or if the direction data indicates that the machine 105 is reversing, the wear detection device 190 can determine that one or more tools are engaging the ground (e.g., a second vibration).

[0048] In some examples, wear detection device 190 may determine whether one or more tools are engaged with the ground based on the tool command data, the steering command data, and / or the gear setting data. For example, if the tool command data indicates that an operator of machine 105 has not issued a command to use one or more tools, the steering command data indicates that the operator has not issued a steering command associated with the use of one or more tools, and / or the gear setting data indicates that machine 105 is in a gear position not associated with the use of one or more tools, wear detection device 190 may determine that one or more tools are not engaged with the ground.

[0049] Alternatively, if the tool command data indicates that the operator has issued a command to use one or more tools, the steering command data indicates that the operator has issued a steering command associated with the use of one or more tools, and / or the gear setting data indicates that the machine 105 is in a gear position associated with the use of one or more tools, the wear detection device 190 can determine that the one or more tools are engaged with the ground.

[0050] Based on information identifying one or more first time portions of a time period (for the mechanical vibration data) corresponding to a first vibration and information (e.g., included in the tool data) identifying one or more second time portions of a time period corresponding to a second vibration, the wear detection device 190 can analyze the mechanical vibration data in combination with the tool data to separate the first vibration from the second vibration. In some cases, the wear detection device 190 can identify a segment of the mechanical vibration data as a time domain signal based on information of one or more first time portions of a particular time period and information of one or more second time portions of a particular time period.

[0051] As an example, the wear detection device 190 can divide the mechanical vibration data into time domain signals based on information identifying one or more first time portions of the particular time period and information identifying one or more second time portions of the particular time period. The time domain signals can include a first time domain signal corresponding to a first portion (or a first segment) of the mechanical vibration data and a second time domain signal corresponding to a second portion (or a second segment) of the mechanical vibration data. The wear detection device 190 can identify and focus on the first portion (or the first portion) of the mechanical vibration data corresponding to the first vibration and can identify and discard the second portion (or the second portion) of the mechanical vibration data corresponding to the second vibration.

[0052] In some implementations, the first time domain signal can include a first time domain portion corresponding to a movement of one or more parts (e.g., a first vibration) and a second time domain portion of the time domain signal corresponding to random noise associated with terrain conditions at the location of the machine 105. By way of example, the random noise can be caused by the machine 105 traveling over uneven ground, the machine 105 traveling over a depression, the machine 105 hitting a rock in the ground, and / or other events affecting the movement of the machine 105.

[0053] The wear detection device 190 may convert the 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 examples, the wear detection device 190 may perform the conversion of the first time domain signal using a Fast Fourier Transform (FFT). Alternatively, the wear detection device 190 may perform the conversion of the first time domain signal by determining (or calculating) a power spectral density (PSD) of the first time domain signal. The PSD may be determined using an FFT.

[0054] The wear detection device 190 can determine the PSD because the frequency resolution associated with the PSD exceeds the frequency resolution associated with a transform using an FFT. For example, the PSD can provide a higher contrast between a first time domain portion and a second time domain portion than the contrast between the first time domain portion and the second time domain portion enabled by an FFT.

[0055] The wear detection device 190 can analyze the spectral domain signal to identify a signature spectrum associated with the movement of one or more components. For example, the wear detection device 190 can analyze the spectral domain signal to identify a spectral signature associated with a first time domain portion. The wear detection device 190 can identify a signature spectrum of a portion of the power spectral density that corresponds to the first time domain portion.

[0056] The wear detection device 190 can predict an amount of wear of one or more components based on the amplitude of the signature spectrum (e.g., the amplitude of a portion of the signature spectrum). As an example, the wear detection device 190 can predict an amount of wear of one or more components based on the amplitude of the signature spectrum and based on historical wear data. As described above, the historical wear data can include first vibration data associated with a first amount of past wear of the component, second vibration data associated with a second amount of past wear of the component, etc.

[0057] The first vibration data can include information identifying a first signature spectrum associated with vibrations caused by past part movements (e.g., when the past wear amount of the part is a first wear amount) and information identifying an amplitude of the first signature spectrum. The second vibration data can include information identifying a second signature spectrum associated with vibrations caused by past part movements (e.g., when the past wear amount of the part is a second wear amount) and information identifying an amplitude of the second signature spectrum.

[0058] The wear detection device 190 can predict the amount of wear of one or more components by analyzing the historical wear data to identify particular vibration data (e.g., a particular amplitude of a particular signature spectrum) that corresponds to an amplitude of the signature spectrum. As an example, assume that the wear detection device 190 determines that the amplitude of the signature corresponds to an amplitude of a second signature spectrum. In this case, the wear detection device 190 can predict that the amount of wear of the one or more components corresponds to the second amount of wear of the components in the past.

[0059] In addition to predicting the amount of wear of one or more components, the wear detection device 190 may alternatively predict a rate of wear of one or more components and / or predict a date and time when one or more components will be replaced and / or repaired. The predicted rate of wear of one or more components, the predicted amount of wear of one or more components, and / or the predicted date and time may hereinafter be referred to as "predicted component wear information."

[0060] In some implementations, the wear detection device 190 can use a machine learning model to determine the predicted part 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 part wear information. The wear detection device 190 can train the machine learning model using historical data related to the machine 105 and / or related to one or more other machines similar to the machine 105. The one or more machines may include similar parts, similar dimensions, and / or similar uses to the machine 105.

[0061] The similar parts may include similar ground engaging members 170, similar tracks, similar track links, e.g., first track link 172 and / or second track link 174, similar track link bushings, e.g., first track link bushing 176, similar track link pins, e.g., first track link pin 178 and / or second track link pin 180, similar sprockets 182, similar segments 184, and / or similar idlers 186. The historical data may include historical sensor data, historical operator control data, and / or historical wear data. The historical sensor data may include sensor data received from sensor system 120, the historical operator control data may include operator control data received from operator control 135, and / or the historical wear data may include historical wear data received from measurement device 210.

[0062] In training the machine learning model, the wear detection device 190 can separate the training data into a training set (e.g., a data set for training the machine learning model), a validation set (e.g., a data set for evaluating the fit of the machine learning model and / or a data set for fine-tuning the machine learning model), a test set (e.g., a data set for evaluating the final fit of the machine learning model), and / or the like. The wear detection device 190 can pre-process 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 this minimum feature set, so that processing for training the machine learning model can be reduced and classification techniques can be applied to the minimum feature set.

[0063] The wear detector 190 may use classification techniques such as logistic regression classification techniques, random forest classification techniques, gradient boosting machine learning (GBM) techniques, and / or similar techniques to determine a classification result (e.g., wear amount of one or more parts). The wear detector 190 may use a naive Bayes classifier technique in addition to or as an alternative to using a classification technique. In this case, the wear detector 190 may perform a bipartite recursive partitioning to partition the training data of a minimum feature set into regions and / or branches and use the partitions and / or branches to perform a prediction (e.g., wear rate and / or wear amount of one or more parts). Based on the use of recursive partitioning, the wear detector 190 may reduce the use of computational resources associated with manual linear classification and analysis of data items, allowing a model to be trained using thousands, millions, or billions of data items, resulting in a more accurate model than using fewer data items.

[0064] The wear detection device 190 may train the machine learning model using a supervised training process that includes receiving input to the machine learning model from a subject matter expert (e.g., one or more operators associated with the machine 105 and / or one or more machines), which may reduce an amount of time, processing resources, and / or the like to train the machine learning model as compared to an unsupervised training process. The wear detection device 190 may use one or more other model training techniques, such as neural network techniques, latent semantic indexing techniques, and / or the like.

[0065] For example, the wear detection device 190 may implement 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 recognition on the patterns of different amounts of wear on one or more components. In this case, the use of artificial neural network processing techniques may improve the accuracy of the machine learning model generated by the wear detection device 190 by being more robust to noisy, inaccurate, or incomplete data, and by allowing the wear detection device 190 to detect patterns and / or trends that cannot be detected by a human analyst or system, or by less complex techniques.

[0066] Once trained, the machine learning model may be used to determine (or predict) predicted part wear information. In other words, the wear detection device 190 may receive sensor data and / or operator control data from the machine 105 after training the machine learning model, input the received sensor data and / or operator control data into the machine learning model, and the machine learning model may output data regarding the wear rate and / or amount of wear of one or more parts. The output of the machine learning model may include a score of the predicted part wear information. In the case of predicted part wear information, the score may represent a measure of the reliability of the predicted part wear information.

[0067] Various devices, such as a server device, can generate and train the machine learning models. Different devices can provide the machine learning models for use with the wear detection device 190. Different devices can update the machine learning models and provide the machine learning models to the wear detection device 190 (e.g., on a scheduled basis, on-demand basis, triggered basis, periodic basis, and / or the like). In some examples, the wear detection device 190 can receive additional training data (e.g., additional past sensor data, additional past operator control data, and / or additional past wear data) and retrain the machine learning models. Alternatively, the wear detection device 190 can provide additional training data to another device to train the machine learning models. The machine learning models may be retrained periodically and / or based on a trigger event.

[0068] In some implementations, the wear detection device 190 can provide the machine learning model to the controller 140 to enable the controller 140 to determine predicted part wear information. Alternatively, the wear detection device 190 can receive a request from the controller 140 to determine predicted part wear information. The request can include sensor data of the machine 105.

[0069] The wear detector 190 (and / or the controller 140) can take action based on the predicted component wear information. For example, the action can include the wear detector 190 adjusting an action of the machine 105 based on the predicted amount of wear of one or more components (e.g., if the predicted amount of wear meets a threshold amount of wear). For example, the wear detector 190 can cause a change in the speed of the machine 105, a change in the acceleration of the machine 105, a change in the direction of travel of the machine 105, a change in tool commands, a change in steering commands, a change in gear settings, and / or other action that can reduce the rate of wear of one or more components and extend the time before one or more components must be repaired or replaced.

[0070] The wear detection device 190 may include having the machine 105 navigate to a different work location and perform one or more tasks at the different work location to extend the life of the one or more components. For example, the different work location may be associated with a lower wear rate (of the one or more components) than a wear rate (of the one or more components) associated with a work location in which the machine 105 is currently located. Additionally or alternatively, the wear detection device 190 may cause the machine 105 to perform a different task to extend the life of the one or more components. For example, the different task may be associated with a lower wear rate (of the one or more components) than a wear rate (of the one or more components) associated with a task currently being performed by the machine 105.

[0071] The action may include the wear detection device 190 transmitting the wear information to one or more devices that monitor the amount of wear on parts of the plurality of machines (e.g., including the machine 105). In some examples, the wear detection device 190 may transmit the wear information if the amount of wear (on one or more parts) meets a threshold amount of wear. The wear information may indicate the amount of wear on one or more parts, the rate of wear on one or more parts, the amount of wear on one or more parts, and / or suggestions related to repair and / or replacement of the one or more parts. The one or more devices may include devices of a location management system, devices of a back office system, devices associated with an operator of the machine 105, devices associated with a technician, and / or the controller 140.

[0072] The wear detection device 190 may transmit wear information to cause one or more devices to order 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.

[0073] The wear detection device 190 may transmit the wear information to cause one or more devices (e.g., the controller 140) to autonomously navigate the machine 105 to a repair facility. Additionally or alternatively, the wear detection device 190 may transmit the wear information to cause one or more devices to populate a technician's calendar with calendar events to inspect and / or repair one or more parts. Additionally or alternatively, the wear detection device 190 may transmit the wear information to cause one or more devices (e.g., the controller 140) to activate an alarm. The alarm may suggest repairing or replacing one or more parts.

[0074] In some cases, the wear detection device 190 may transmit wear information to cause one or more devices to generate a service request to repair and / or replace one or more parts. As part of generating a service request, the one or more devices may perform one or more actions described herein.

[0075] In some examples, the action may include causing the wear detection device 190 to cause the first autonomous device to deliver one or more replacement parts to a location associated with the machine 105. The location may include a current location of the machine 105, a location of a work site where the machine 105 performs multiple tasks, a location where the machine 105 is located when the machine 105 is not performing a task, and / or a location where the machine 105 is located when the machine 105 is being repaired and / or replaced. In some examples, the wear information may include information identifying a location associated with the machine 105.

[0076] In some examples, the action may include causing the wear detection device 190 to navigate to a location associated with the machine 105 to verify the predicted part wear information. The second autonomous device may generate validation information based on the validation of the part wear information and transmit the validation information to the wear detection device 190. The wear detection device 190 may use the validation information to retrain the machine learning model.

[0077] In some examples, the wear detector 190 may determine whether a failure of one or more components is imminent (e.g., based on predicted component wear information). If the wear detector 190 determines that a failure is about to occur, the wear detector 190 may perform one or more of the actions described above. If the wear detector 190 determines that a failure is not imminent, no action may be taken.

[0078] The number and arrangement of devices and networks shown in Figure 2 are provided as an example. In practice, there may be additional devices, fewer devices, different devices, or a different arrangement of devices compared to those shown in Figure 2. Furthermore, two or more devices shown in Figure 2 may be implemented within a single device, or a single device shown in Figure 2 may be implemented as multiple distributed devices. Additionally or alternatively, a set of devices (e.g., one or more devices) of system 200 may perform one or more functions that are described as being performed by another set of devices of system 200.

[0079] 3 is a flow chart of an example process 300 related to predicting chassis wear based on mechanical vibration data. One or more of the process blocks of FIG. 3 may be performed by an apparatus (e.g., wear detection apparatus 190). One or more of the process blocks of FIG. 3 may be performed by another apparatus or by a set of apparatus separate from or including the apparatus, such as a controller (e.g., controller 140).

[0080] 3, process 300 may include receiving machine vibration data identifying vibration measurements over a period of time of a machine, block 310. For example, as described above, an apparatus may receive machine vibration data identifying vibration measurements over a period of time of a machine.

[0081] As an example, receiving the machine vibration data can include receiving machine vibration data from one or more first sensor devices of the machine, and receiving the tool data can include at least one of receiving sensor data from one or more second sensor devices of the machine indicating whether the one or more tools are engaged with the ground, or receiving operator control data from one or more operator controls of the machine indicating whether the one or more tools are engaged with the ground.

[0082] 3, the process 300 may include splitting the machine vibration data to obtain a time domain signal that includes a time domain signal related to vibrations associated with a chassis of the machine (Block 320). For example, as described above, the device may split the machine vibration data to obtain a time domain signal that includes a time domain signal related to vibrations associated with a chassis of the machine.

[0083] As an example, the machine vibrations may be caused by movement of a part and may be caused by one or more tools of the machine engaging the ground, and the method further includes receiving tool data indicative of whether the one or more tools are engaging the ground over a time period. Segmenting the machine vibration data includes segmenting the machine vibration data based on the tool data to identify a time domain signal.

[0084] The time domain signal may be a first time domain signal. The time domain signal may include a second time domain signal related to vibrations independent of vibrations relative to a chassis of the machine. 3, the process 300 may include converting the time domain signal to a spectral domain signal using a Fast Fourier Transform (FFT) (Block 330). For example, as described above, the device may convert the time domain signal to a spectral domain signal using a Fast Fourier Transform (FFT).

[0085] By converting the time domain signal to a spectral domain signal, a first portion of the time domain signal corresponding to part motion is amplified based on the power spectral density of the time domain signal, and a second portion of the time domain signal corresponding to random noise associated with terrain conditions at the machine's location is reduced based on the power spectral density of the time domain signal. The power spectral density may be determined using an FFT. A signature spectrum may be identified in a portion of the power spectral density corresponding to the first portion of the time domain signal.

[0086] 3, the process 300 may include identifying a signature spectrum from the spectral domain signal associated with a movement of a part of a chassis of the machine (Block 340). For example, as described above, the apparatus may identify a signature spectrum from the spectral domain signal associated with a movement of a part of a chassis of the machine.

[0087] 3, the process 300 may include predicting an amount of wear on the part based on the amplitude of the signature spectrum (Block 350). For example, as described above, an amount of wear on the part may be predicted based on the amplitude of the signature spectrum.

[0088] Predicting the amount of wear of the part may include predicting the amount of wear of the part based on the amplitude of the signature spectrum and historical wear data associated with the part.

[0089] 3, the process 300 may include causing an action to be taken based on the amount of wear on the part (block 360). For example, as described above, an action may be taken based on the amount of wear on the part.

[0090] Taking this action may include at least one of causing a repair of the part, causing a replacement of the part, causing an adjustment of a machine action, or providing an alarm to a machine operator device.

[0091] Although Figure 3 illustrates example blocks of process 300, in some implementations, process 300 may include additional, fewer, different, or differently arranged blocks than those illustrated in Figure 3. Additionally or alternatively, two or more blocks of process 300 may be performed in parallel. [Industrial Applicability]

[0092] The present invention relates to a device for predicting the amount of wear of one or more parts of a chassis of a machine based on sensor data from a sensor device related to mechanical vibrations of the machine. The device for predicting the amount of wear of one or more parts can avoid problems associated with manual measurements of the tracks of the machine (to determine the amount of wear of the tracks) and inaccurate prediction of the amount of wear of the tracks.

[0093] When manual measurements are taken, the manual measurements of the track may waste machine resources to prevent machine movement. Additionally, erroneous manual measurements of the track and / or erroneous predictions of track wear may waste computational resources to remedy problems associated with the erroneous manual measurements of track wear and / or erroneous predictions of track wear (e.g., premature track failure, premature track repair, and / or premature track replacement).

[0094] The apparatus for predicting the amount of wear of one or more components of a chassis can solve the above problems of manual measurements and inaccurate prediction of the amount of wear. Several advantages may be associated with the disclosed apparatus. For example, predicting the amount of wear of one or more components based on machine vibration data can limit interruptions during the action of the machine and / or limit immobilization of the machine, allowing the machine to maintain the action for an extended period of time (e.g., until repair or replacement of the track is required). Predicting the amount of wear of one or more components based on machine vibration data can allow repair or replacement of the track when repair or replacement of the track is required (as opposed to repair or replacement of the track earlier).

[0095] By predicting the amount of wear of one or more components based on the machine vibration data, inaccurate prediction of the amount of wear can be reduced. By reducing inaccurate prediction of the amount of wear of the track, the device can reduce the likelihood of the track failing before the track can be repaired and / or replaced. By reducing inaccurate prediction of the amount of wear of the track, the device can preserve computational or machine resources that would otherwise be used to ameliorate problems associated with inaccurate manual measurements and inaccurate prediction of the amount of wear of the track (e.g., premature failure of the track, premature repair of the track, and / or premature replacement of the track).

[0096] The above disclosure provides illustrations and descriptions, but is not intended to be exhaustive or to limit the embodiments to the precise forms disclosed. Modifications and variations may be made based on the above disclosure, or may be acquired from practice of the embodiments. Furthermore, any of the embodiments described herein may be combined, unless the above disclosure clearly provides a reason why one or more embodiments cannot be combined. Although specific combinations of features may be recited in the claims and / or disclosed in the specification, these combinations are not intended to limit the disclosure of the various embodiments. Each dependent claim listed below may be directly dependent on only one claim, but the disclosure of the various embodiments includes each dependent claim in combination with all other claims in the claim set. As used herein, the terms "a," "one," and "a set" are intended to include one or more and may be used interchangeably with "one or more." Additionally, as used herein, the article "the" is intended to include one or more of the items recited in conjunction with the article "the" and may be used interchangeably with "one or more." Additionally, the phrase "based on" is intended to mean "based at least in part on," unless otherwise indicated. Additionally, as used herein, the term "or" when used in a series is intended to be inclusive and may be used interchangeably with "and / or" unless otherwise indicated (e.g., when used in conjunction with "any" or "only one").

Claims

1. A method performed by an apparatus (190), comprising: receiving machine (105) vibration data identifying a measurement of vibration of the machine (105) over a period of time; segmenting the machine (105) vibration data to obtain time domain signals, the time domain signals including time domain signals related to vibrations associated with a chassis (170) of the machine (105); converting the time domain signal into a spectral domain signal using a Fast Fourier Transform (FFT); identifying a signature spectrum from the spectral domain signal associated with movement of components (172, 174, 176, 178, 180, 182, 184, 186) of the chassis (170) of the machine (105); predicting an amount of wear of the component (172, 174, 176, 178, 180, 182, 184, 186) based on an amplitude of the signature spectrum; and performing an action based on the amount of wear on the component (172, 174, 176, 178, 180, 182, 184, 186).

2. the vibration of the machine (105) being caused by the movement of the parts (172, 174, 176, 178, 180, 182, 184, 186) and caused by one or more tools (150, 160) of the machine (105) engaging the ground; The method comprises: receiving tool data indicative of whether the one or more tools (150, 160) engaged the ground within the period of time; The step of dividing the machine (105) vibration data further comprises: The method of claim 1 , comprising partitioning the machine (105) vibration data based on the tool data to identify the time domain signal.

3. receiving the machine (105) vibration data includes receiving the machine (105) vibration data from one or more first sensor devices (120) of the machine (105); The step of receiving the tool data includes: receiving sensor data from one or more second sensor devices (120) of the machine (105) indicative of whether the one or more tools (150, 160) are engaged with the ground; or 3. The method of claim 2, comprising at least one of the steps of receiving operator control data from one or more operator controls (135) of the machine (105) indicating whether the one or more tools (150, 160) are engaged with the ground.

4. Transforming the time domain signal into the spectral domain signal comprises: amplifying a first portion of the time domain signal corresponding to the movement of the component (172, 174, 176, 178, 180, 182, 184, 186) based on a power spectral density of the time domain signal; reducing a second portion of the time domain signal corresponding to random noise associated with terrain conditions at a location of the machine (105) based on a power spectral density of the time domain signal; The power spectral density is determined using an FFT; The method of any one of claims 1 to 3, wherein the signature spectrum is identified in a portion of a power spectral density that corresponds to the first portion of the time domain signal.

5. The step of performing the action includes: repairing said components (172, 174, 176, 178, 180, 182, 184, 186); replacing said parts (172, 174, 176, 178, 180, 182, 184, 186); adjusting the operation of the machine (105); The method of any one of claims 1 to 4, further comprising at least one of the steps of: providing an alert to an operator's device (190) of the machine (105).

6. 1. A system comprising: receiving machine (105) vibration data identifying a measurement of vibration of the machine (105); the vibration of the machine (105) is caused by a combination of a first vibration caused by movement of parts (172, 174, 176, 178, 180, 182, 184, 186) of a chassis (170) of the machine (105) and a second vibration independent of the first vibration; identifying a segment in the machine (105) vibration data corresponding to the first vibration; Transforming the segments into frequency domain signals using a Fast Fourier Transform (FFT); analyzing the signals to identify a signature spectrum associated with the movement of the components (172, 174, 176, 178, 180, 182, 184, 186); predicting an amount of wear of the components (172, 174, 176, 178, 180, 182, 184, 186) based on the signature spectrum; A system comprising: an apparatus (190) configured to cause an action to be taken based on the amount of wear on the parts (172, 174, 176, 178, 180, 182, 184, 186).

7. In predicting the amount of wear of the components (172, 174, 176, 178, 180, 182, 184, 186), the apparatus (190) comprises: The system of claim 6 , configured to predict the amount of wear of the component (172, 174, 176, 178, 180, 182, 184, 186) based on an amplitude of the signature spectrum.

8. In performing the action, the device (190): Repairing said parts (172, 174, 176, 178, 180, 182, 184, 186); replacing said parts (172, 174, 176, 178, 180, 182, 184, 186); Coordinating the operation of said machine (105); providing information about the amount of wear of a device (190) for monitoring wear of said parts (172, 174, 176, 178, 180, 182, 184, 186) of said machine (105); or The system of any one of claims 6 to 7, configured to provide an alert to an operator's device (190) of the machine (105).

9. In identifying the segment, the device (190) receiving tool data indicative of whether a tool (150, 160) of the machine (105) is engaged with the ground; A system according to any one of claims 6 to 8, configured to identify the segments based on the tool data.

10. The tool data includes machine (105) speed data identifying a speed of the machine (105); tool command data identifying commands for controlling the tools (150, 160); steering command data identifying a steering command for the machine (105); and gear setting data identifying a gear setting of the machine (105).

Images