System, device and method for estimating remaining useful life of a bearing

Abstract

The present invention provides a system (100), an apparatus (110), and a method for estimating the remaining useful life of a bearing. In one embodiment, the method includes receiving a request from a source (115, 125, 130) for analyzing a defect in a bearing. The request includes operational data associated with the bearing. The method includes monitoring the effect of the defect on the bearing over a period of time. The method includes using a machine learning model to determine a time period during which the effect of the defect on the bearing is above a threshold range. The method includes calculating the severity of the effect associated with the defect during the time period. The method includes determining the remaining useful life of the bearing based on the severity and the operational data during the time period.

Description

Technical Field

[0001] The present invention relates to the field of bearing monitoring systems and more particularly to estimating the remaining useful life of a bearing. Background Art

[0002] Bearings are used in a variety of machines in industry with the purpose of reducing the friction between two rotating parts. These bearings also limit the relative motion between the rotating parts to the desired motion. However, bearings may fail unexpectedly due to factors such as poor lubrication and contamination within the bearing structure. For example, due to excessive temperatures, the lubrication within the bearing may fail during the operating phase of the bearing. Contamination may occur due to the ingress of foreign particles, moisture, etc. into the structure of the bearing. The above-mentioned factors lead to failure modes of the bearing such as corrosion, spalling, pitting, galvanic corrosion, plastic deformation, etc. As a result, the expected fatigue life of the bearing assembly is reduced and eventually failure occurs. Therefore, bearing failure may lead to unexpected machine downtime, resulting in production and economic losses. In safety-critical applications, bearing failure may also put human lives at risk.

[0003] In view of the foregoing, there exists a need for estimating the remaining useful life of a bearing. Summary of the invention

[0004] It is therefore an object of the present invention to provide a system, apparatus and method for estimating the remaining useful life of a bearing.

[0005] The object of the invention is achieved by a method for estimating the remaining useful life of a bearing.

[0006] The method includes receiving a request for analyzing a defect in a bearing. The term "defect" as used herein refers to any structural deformation within a bearing that causes abnormal operation of the bearing. The request includes operational data associated with the bearing. In one embodiment, the operational data includes real-time output of at least one sensing unit associated with the bearing. It must be understood that the term "sensing unit" as used herein includes transducers and sensors. In addition to the above, the request may also specify one or more bearing parameters.

[0007] Advantageously, the present invention facilitates estimating the remaining useful life of a bearing of any size based on corresponding operating data.

[0008] The method includes monitoring the effect of the defect on the bearing over a period of time. As used herein, the term "effect" refers to a deviation from normal operation of the bearing due to the defect. In one embodiment, when monitoring the effect of the defect, the method includes monitoring an anomaly in the output of at least one sensing unit.

[0009] Advantageously, the present invention facilitates continuous monitoring of the effects due to defects in a bearing in real time.

[0010] The method includes using a machine learning model to determine a time period during which the effect of the defect on the bearing is above a threshold range. In one embodiment, determining the time period includes analyzing operational data associated with the bearing using a machine learning model to determine the time period.

[0011] Advantageously, the calculation of the time periods is performed using a machine learning model. As a result, even shorter time periods caused by low-energy impacts can be detected.

[0012] The method includes calculating a severity of an impact associated with the defect during the time period. In one embodiment, in calculating the severity, the method includes calculating a defect size corresponding to the defect based on the duration of the time period. The term "defect size" as used herein refers to a distance traveled by a rolling element between entering and exiting a defect during operation of the bearing.

[0013] Advantageously, the present invention uses the defect size to calculate the severity of the impact associated with the defect.

[0014] The method includes determining a remaining useful life of the bearing based on the severity and the operational data during the time period. The term "remaining useful life" as used herein refers to the duration between the initiation of a detectable failure mode and a functional failure of the bearing. In a preferred embodiment, in determining the remaining useful life, the method includes calculating a dynamic parameter associated with the bearing based on the defect size and the operational data using a virtual model of the bearing. In one embodiment, the dynamic parameter is a dynamic load on the bearing. The dynamic parameter can also be associated with a contact force of a rolling element when in contact with the defect or with a contact stress.

[0015] In one embodiment, the virtual model is established based on simulation data, experimental data and mathematical models associated with multiple other bearings. In addition, a remaining service life model of the bearing is configured based on the dynamic parameters. The remaining service life model is a dynamic model that relates the dynamic parameters to the life of the bearing. In addition, the remaining service life of the bearing is calculated based on the configured remaining service life model and the operating data.

[0016] Advantageously, the present invention facilitates the use of defect size to determine dynamic parameters that influence bearing degradation.

[0017] The method includes generating a notification on an output device indicating the remaining useful life of the bearing. In addition to the remaining useful life, the notification may also include diagnostic information associated with the bearing. For example, the diagnostic information may indicate a defect size, a RUL curve, and an indication of a degradation state on the RUL curve.

[0018] The object of the present invention is achieved by a device for estimating the remaining useful life of a bearing. The device includes one or more processing units, and a memory unit communicatively coupled to the one or more processing units. The memory unit includes a bearing management module stored in the form of machine-readable instructions executable by the one or more processing units. The bearing management module is configured to perform the above-mentioned method steps. The execution of the condition management module can also be performed using a coprocessor such as a graphics processing unit (GPU), a field programmable gate array (FPGA), or a neural processing / computing engine.

[0019] According to one embodiment of the present invention, the apparatus may be an edge computing device. As used herein, "edge computing" refers to a computing environment that can be executed on an edge device (e.g., one end is connected to a sensing unit in an industrial device, and the other end is connected to (one or more) remote servers, such as (one or more) computing servers or (one or more) cloud computing servers), which may be a compact computing device with a small form factor and resource limitations in terms of computing power. A network of edge computing devices may also be used to implement the apparatus. Such a network of edge computing devices is referred to as a fog network.

[0020] In another embodiment, the device is a cloud computing system having a cloud computing-based platform, which is configured to provide a cloud service for analyzing defects in bearings. "Cloud computing" as used herein refers to a processing environment that includes configurable computing physical and logical resources (e.g., networks, servers, storage, applications, services, etc.) and data distributed on a network (e.g., the Internet). The cloud computing platform can be implemented as a service for analyzing defects in bearings. In other words, the cloud computing system provides on-demand network access to a shared pool of configurable computing physical and logical resources. The network is, for example, a wired network, a wireless network, a communication network, or a network formed by any combination of these networks.

[0021] Furthermore, the object of the present invention is achieved by a system for estimating the remaining useful life of a bearing. The system comprises one or more sources capable of providing operational data associated with the bearing and a device as described above, which is communicatively coupled to the one or more sources. The term "source" as used herein refers to an electronic device configured to obtain operational data and transmit it to the device. Non-limiting examples of sources include sensing units, controllers, and edge devices.

[0022] The object of the present invention is also achieved by a computer-readable medium, on which a program code segment of a computer program is stored. The program code segment can be loaded into a processor and / or can be executed by the processor. When the program code portion is executed, the processor executes the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The above-mentioned properties, features and advantages of the present invention and the ways to achieve them will become more obvious and easy to understand (clear) through the following description of the embodiments of the present invention in conjunction with the corresponding drawings. The described embodiments are intended to illustrate rather than limit the present invention.

[0024] Figure 1A A block diagram of a system for estimating the remaining useful life of a bearing according to an embodiment of the present invention is illustrated;

[0025] Figure 1B A block diagram of an apparatus for estimating the remaining useful life of a bearing according to an embodiment of the present invention is illustrated;

[0026] Figure 2A The diagram shows the structure of a ball bearing;

[0027] Figure 2B The figure shows a defect in the outer ring of a ball bearing;

[0028] Figure 3 An experimental test device for establishing a virtual model of a bearing according to an embodiment of the present invention is illustrated;

[0029] Figure 4A is a graphical user interface view showing an example of simulation results generated for a bearing according to an embodiment of the present invention; Figure 4B is a graphical user interface diagram according to an embodiment of the present invention, showing experimental values ​​of maximum impact force as a function of angular velocity;

[0030] Figure 5 is a graphical user interface view showing a representation of the relationship between maximum impact forces and acceleration values ​​associated with three bearings according to an embodiment of the present invention;

[0031] Figure 6 depicts a flow chart of a method for estimating the remaining useful life of a bearing according to an embodiment of the present invention;

[0032] Figure 7 is an exemplary web-based interface according to an embodiment of the present invention that enables a user to provide values ​​of one or more bearing parameters from a client device;

[0033] Figure 8 is a graphical user interface view showing an example of an acceleration signal in the time domain according to an embodiment of the present invention; and

[0034] Fig. 9 is a graphical user interface view showing the degradation of remaining useful life as defect size increases according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] Hereinafter, embodiments for implementing the present invention are described in detail. Various embodiments are described with reference to the accompanying drawings, wherein the same reference numerals are used throughout to refer to the same elements. In the following description, for the purpose of explanation, many specific details are set forth in order to provide a thorough understanding of one or more embodiments. It may be apparent that such embodiments may be practiced without these specific details.

[0036] Figure 1A A block diagram of a system 100 for estimating the remaining useful life of a bearing 105 according to an embodiment of the present invention is illustrated. For example, the bearing 105 may be part of a rotating device such as an industrial motor (not shown). Non-limiting examples of bearings include deep groove ball bearings, cylindrical roller bearings, tapered roller bearings, thrust bearings, angular contact ball bearings, needle bearings, and the like. In the present embodiment, the bearing 105 includes an inner ring, an outer ring, and a plurality of rolling elements arranged in a gap between the inner ring and the outer ring. The bearing 105 also includes a cage positioned between the inner ring and the outer ring for maintaining a symmetrical radial clearance between the rolling elements. Reference will be made later Figure 2A Examples of bearings are described in this disclosure.

[0037] The system 100 includes an apparatus 110 communicatively coupled to one or more edge devices 115. The one or more edge devices 115 are connected to the apparatus 110 via a network 120 (e.g., a local area network (LAN), a wide area network (WAN), WiFi, etc.). Each edge device 115 is configured to receive sensor data from at least one sensing unit 125 associated with the bearing 105. The at least one sensing unit 125 may include, for example, a vibration sensor, a velocity sensor, an acceleration sensor, and a force sensor. The sensor data corresponds to the output of the at least one sensing unit 125. For example, the output from the at least one sensing unit 125 may be in the form of vibration data, velocity data, acceleration data, or force data. In one embodiment, the sensor data is acquired through a data acquisition interface on the edge device 115. The edge device 115 provides the sensor data to the apparatus 110 in real time.

[0038] In addition, the edge device 115 is further configured to provide the apparatus 110 with one or more bearing parameters associated with the bearing 105. The one or more bearing parameters include, but are not limited to, a standard bearing number, ball size, bearing static load, material density, angular velocity, internal clearance, bearing diameter, number of rolling elements, radius of rolling elements, inner ring diameter, outer ring diameter, fatigue load limit, and type of lubricant used in the bearing 105. It must be understood that the standard bearing number indicates certain specifications, such as ball size, bearing static load, material density, internal clearance, bearing diameter, number of rolling elements, radius of rolling elements, inner ring diameter, outer ring diameter, fatigue load limit, bearing width, etc., as provided by the manufacturer of the bearing 105. Therefore, in one embodiment, a standard bearing number associated with the bearing 105 may be provided in place of the above-mentioned bearing parameters.

[0039] One or more bearing parameters may be stored in a memory of the edge device 115 or may be input to the edge device 115 by an operator. For example, the edge device 115 may be communicatively coupled to the client device 130. Non-limiting examples of client devices include personal computers, workstations, personal digital assistants, and human-machine interfaces. The client device 130 may enable a user to enter values ​​for one or more bearing parameters through a web-based interface. After receiving the one or more bearing parameters from the user, the edge device 115 transmits a request for analyzing defects in the bearing 105 to the device 110. Defects occur due to the initiation of a failure mode in the bearing 105. Non-limiting examples of failure modes include spalling, pitting, plastic deformation, wear, galvanic erosion, and corrosion, typically on the outer ring of the bearing 105. Defects may occur due to the presence of contaminants or due to the characteristics of the lubricant used in the bearing 105. The request includes one or more bearing parameters and sensor data.

[0040] In the present embodiment, the device 110 is deployed in a cloud computing environment. As used herein, a "cloud computing environment" refers to a processing environment that includes configurable computing physical and logical resources (e.g., networks, servers, storage, applications, services, etc.) and data distributed on a network 120 (e.g., the Internet). The cloud computing environment provides on-demand network access to a shared pool of configurable computing physical and logical resources. The device 110 may include a module for estimating the remaining useful life of a given bearing based on corresponding sensor data and one or more bearing parameters. In addition, the device 110 may include a network interface for communicating with one or more edge devices 115 via the network 120.

[0041] like Figure 1BAs shown in FIG. 1 , the device 110 includes a processing unit 135, a memory unit 140, a storage unit 145, a communication unit 150, a network interface 155, and a standard interface or bus 160. The device 110 may be a computer, a workstation, a virtual machine running on host hardware, a microcontroller, or an integrated circuit. Alternatively, the device 110 may be a real or virtual computer group (the technical term for a real computer group is a "cluster" and the technical term for a virtual computer group is a "cloud").

[0042] As used herein, the term "processing unit" 135 refers to any type of computing circuit, such as, but not limited to, a microprocessor, a microcontroller, a complex instruction set computing microprocessor, a reduced instruction set computing microprocessor, a very long instruction word microprocessor, an explicitly parallel instruction computing microprocessor, a graphics processor, a digital signal processor, or any other type of processing circuit. The processing unit 135 may also include an embedded controller, such as a general-purpose or programmable logic device or array, an application-specific integrated circuit, a single-chip computer, etc. In general, the processing unit 135 may include hardware elements and software elements. The processing unit 135 can be configured as multi-threaded, that is, the processing unit 135 can host different computing processes simultaneously, execute in parallel, or switch between active and passive computing processes.

[0043] The memory unit 140 may be a volatile memory and a non-volatile memory. The memory unit 140 may be coupled for communication with the processing unit 135. The processing unit 135 may execute instructions and / or codes stored in the memory unit 140. Various computer-readable storage media may be stored in and accessed from the memory unit 140. The memory unit 140 may include any suitable element for storing data and machine-readable instructions, such as a read-only memory, a random access memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, a hard drive, a removable media drive for processing optical disks, digital video disks, floppy disks, magnetic tape cassettes, memory cards, and the like.

[0044] The memory unit 140 includes a bearing management module 165 in the form of machine-readable instructions on any of the above-mentioned storage media, and can be in communication with and executed by the processing unit 135. The bearing management module 165 includes a pre-processing module 170, an impact monitoring module 175, a severity calculation module 180, a life estimation module 185, and a notification module 190. The pre-processing module 170 is configured to receive a request for analyzing a defect in the bearing 105. The request includes operating data associated with the bearing 105 and one or more bearing parameters. The impact monitoring module 175 is configured to monitor the impact of the defect on the bearing 105 over a period of time. The impact monitoring module 175 is also configured to determine a time period during which the impact of the defect on the bearing 105 is above a threshold range by using a machine learning model. The severity calculation module 180 is configured to calculate the severity of the impact associated with the defect during the time period. The life estimation module 185 is configured to determine the remaining useful life of the bearing 105 based on the severity and operating data during the time period. The notification module 190 is configured to generate a notification on an output device indicating the remaining useful life of the bearing 105. In this embodiment, the output device may be the client device 130.

[0045] The storage unit 145 includes a non-volatile memory that stores default bearing parameters associated with standard bearing numbers. The storage unit 145 includes a database 195 that includes default values ​​for bearing parameters and one or more lookup tables that include predetermined values ​​for factors that vary with the operating conditions of the bearing 105. The bus 160 serves as an interconnect between the processing unit 135, the memory unit 140, the storage unit, and the network interface 155. The communication unit 150 enables the device 110 to receive requests from one or more edge devices 115. The communication module can support different standard communication protocols, such as transmission control protocol / Internet protocol (TCP / IP), Profinet, Profibus, and Internet Protocol version (IPv4).

[0046] It will be understood by those of ordinary skill in the art that Figure 1A and 1B The hardware depicted in the can vary for different implementations. For example, other peripheral devices, such as optical drives, local area network (LAN) / wide area network (WAN) / wireless (e.g., Wi-Fi) adapters, graphics adapters, disk controllers, input / output (I / O) adapters, network connection devices, can be used in addition to or in place of the hardware depicted. The depicted examples are provided for purposes of explanation only and are not meant to imply architectural limitations with respect to the present disclosure.

[0047] Figure 2AThe figure shows the structure of a ball bearing 200. The ball bearing 200 includes an outer ring 210 with a diameter D, a plurality of balls 215 (each ball has a radius r ball ), a cage 220 and an inner ring 225 having a diameter d. A plurality of balls 215 are arranged in a gap between the outer ring 210 and the inner ring 225. The cage 220 maintains symmetrical radial spacing between the balls 215. Figure 2B The figure shows that the defect size d on the wall of the outer ring 210 defect The defect is in contact with the ball 215. In this example, the defect is an indentation in the housing due to pitting. The defect size can be defined as the distance that a ball of the ball bearing 200 travels between entering and leaving the defect. As shown, the defect at the center of the ball 215A in the plurality of balls 215 subtends an angle θ center The angular velocity of the shaft on which the ball bearing 200 is mounted is denoted as ω.

[0048] Figure 3 The experimental test device 300 for establishing a virtual model of a bearing according to an embodiment of the present invention is illustrated. In this embodiment, the virtual model corresponds to a ball bearing. It must be understood by those skilled in the art that virtual models can be established for other types of bearings in a similar manner.

[0049] The virtual model can be based on one or more of a physics-based model, a computer-aided design (CAD) model, a computer-aided engineering (CAE) model, a one-dimensional (1D) model, a two-dimensional (2D) model, a three-dimensional (3D) model, a finite element (FE) model, a descriptive model, a metamodel, a stochastic model, a parametric model, a reduced-order model, a statistical model, a heuristic model, a predictive model, an aging model, a machine learning model, an artificial intelligence model, a deep learning model, a system model, an agent model, and the like.

[0050] In the present embodiment, the virtual model is established based on simulation data, experimental data and mathematical data associated with multiple bearings under multiple operating conditions. Multiple operating conditions can be generated based on experimental design (DOE) for changing the values ​​of load, angular velocity and defect size. Here, the term "load" represents the static load of the bearing. For example, the defect size can be selected as one of 0.1mm, 0.2mm, 1mm, 2mm, 3mm, 4mm and 5mm. The load can be one of 400N and 500N. The angular velocity can be one of 1000rad / s, 1200rad / s, 1400rad / s and 1600rad / s. In the present embodiment, simulation data, experimental data and mathematical data associated with three standard ball bearings are used. For example, the first bearing of the three bearings may have a standard bearing number 6205, the second bearing may have a standard bearing number 6213, and the third bearing may have a standard bearing number 6319.

[0051] For example, the experimental device 300 includes a bearing 305, at least one force sensor 310 attached to the bearing 305, and at least one vibration sensor 315. The force sensor 310 and the vibration sensor 315 are communicatively coupled to a device 320 similar to the device 110. In one embodiment, the device 320 may include a data acquisition interface for receiving signals from the force sensor 310 and the vibration sensor 315. The bearing 305 is mounted on a rotating shaft. In one example, the rotating shaft is part of a rotating device. In addition, one or more defects are artificially introduced into the outer ring of the bearing 305. Each defect is associated with a known defect size specified in the operating conditions. The force sensor 310 is configured to measure the impact force caused by the ball of the bearing passing through the defect. The term "impact force" as used herein refers to the contact force experienced by the ball when entering the edge of the defect. In one example, the force sensor 310 is a triaxial piezoelectric crystal. The vibration sensor 315 is configured to measure the vibration or acceleration value caused by the ball passing through the defect. In the example, the vibration sensor 315 is an accelerometer. In addition, the experimental data is recorded. For each operating condition, the experimental data includes the impact force measured by the force sensor 310 and the corresponding acceleration value measured by the vibration sensor 315. Similarly, the experimental data corresponding to each of the three bearings are recorded.

[0052] Simulation data is generated by simulating the behavior of the bearing based on a multi-physics simulation model. The operating conditions can be provided as input to the multi-physics simulation model in the simulation environment. For example, the simulation environment can be provided by a computer-aided simulation tool on the device 320. The simulation model includes a finite element model of an outer ring, a cage, a plurality of rolling elements, an inner ring, and a shield associated with the bearing. In this example, the rolling elements are balls. In addition, the simulation model corresponding to each standard ball bearing is configured to model defects in the outer ring of the defect size specified in the operating conditions. For example, defects may be associated with one or a combination of spalling, pitting, plastic deformation, wear, galvanic erosion, or corrosion.

[0053] Based on the configured simulation model, a simulation instance is generated. The simulation instance is executed in a simulation environment to generate simulation data for a bearing corresponding to each simulation data generated for the same operating condition used to generate the experimental data. The simulation data includes a value of a simulated maximum impact force corresponding to each operating condition. Similarly, simulation data is generated for each of the three bearings. Figure 4A 4 is a GUI view 405 showing an example of simulation results for a bearing. The simulation results indicate that for the simulation example, the impact force (in Figure 4AIn particular, the solid line indicates the impact force at the leading edge of the defect, while the discrete line indicates the impact force at the trailing edge of the defect. Here, the leading edge corresponds to the point on the outer race at which the rolling element of the bearing enters the defect. The trailing edge corresponds to the point on the outer race at which the rolling element of the bearing leaves the defect. Similarly, Figure 4B 4 is a GUI view 410 showing defect profiles indicating experimental values ​​of maximum impact force as a function of angular velocity. The defect profiles correspond to different defect sizes of 0.2 mm, 0.5 mm, 4 mm, 6 mm, 7 mm, and 8 mm. As shown, the maximum impact force increases with increasing defect size and also with increasing angular velocity.

[0054] The mathematical data are generated by the device 320 based on a mathematical model of the bearing. The mathematical model is in the form of:

[0055]

[0056] It can be rearranged as:

[0057]

[0058] Among them, x imax is the maximum deflection of the ball, in mm;

[0059]

[0060] where ν is the Poisson's ratio associated with the material of the bearing, E is the Young's modulus associated with the material, and E eq is the equivalent stiffness of the material. For example, if the bearing is made of EN31 steel, the Poisson's ratio is 0.3 and the Young's modulus is 210 GPa.

[0061] r eq =0.5098*r ball (5)

[0062] where r ball is the radius of the ball

[0063]

[0064] where rbearing is the radius of the bearing and is given by:

[0065] r bearing =d m +internal clearance / 2 (9)

[0066]

[0067] where ω is the angular velocity and is the angular frequency with which the ball rotates around the edge of the defect.

[0068] In the above equation, the parameters p1 and p2 are adjusted based on the bearing parameters. In one embodiment, the parameters are adjusted using a trained machine learning model. The trained machine learning model is an evolutionary algorithm. For x imax Further solve the mathematical model. In addition, the maximum impact force F on the ball max By x imax Calculated according to the following equation:

[0069]

[0070] Here F max represents the value of the maximum impact force on the ball calculated mathematically. Thus, the maximum impact force is calculated for the bearing corresponding to each operating condition. Similarly, the maximum impact force is calculated for each of the three bearings under each operating condition.

[0071] The device 320 further establishes a virtual model of the bearing based on the simulation data, the experimental data, and the mathematical data. In one embodiment, the virtual model is a surrogate model. The virtual model is further verified based on test data generated using the experimental device 300. The test data includes a set of operating conditions for the bearing, which includes known values ​​of defect size, angular velocity, and load. The output of the force sensor is compared with the output of the virtual model to detect errors associated with the virtual model. In addition, the parameters of the virtual model are adjusted to minimize the errors. The adjusted virtual model can be used to predict the maximum impact force associated with any bearing for any given set of operating conditions.

[0072] Figure 5 5 is a graphical user interface (GUI) view 500 according to an embodiment of the present invention, which shows a representation of the relationship between the maximum impact force and acceleration values ​​associated with three bearings. The relationship is obtained based on the behavior of the three bearings under a similar set of operating conditions. The standard bearing numbers of the three bearings are 6205, 6213 and 6319.

[0073] Figure 6 A flow chart of a method 600 for estimating the remaining useful life of a bearing according to an embodiment of the present invention is depicted.

[0074] At step 605, a request for analyzing a defect in a bearing is received by processing unit 135. The request includes one or more bearing parameters associated with the bearing received from a client device similar to client device 130 together with sensor data received from at least one sensing unit attached to the bearing. The sensor data may include an output of at least one sensing unit associated with the bearing. The client device and the at least one sensing unit are both communicatively coupled to an edge device similar to edge device 115. In this embodiment, the at least one sensing unit includes an accelerometer mounted on a bearing housing associated with the bearing. The output of the at least one sensing unit is an acceleration signal in the time domain.

[0075] In one implementation, a user may initiate a request by providing one or more bearing parameters through a web-based interface provided on a client device. For example, the one or more bearing parameters include ball radius, bearing static load, material density, angular velocity, and internal clearance. Figure 7 The figure illustrates an exemplary web-based interface 700 that enables a user to provide values ​​for one or more bearing parameters from a client device, in accordance with an embodiment of the present invention.

[0076] In one embodiment, one or more bearing parameters are specified by a standard bearing number based on an international standard such as the ISO size series. For example, if the standard bearing number is 6213, the bearing size (in mm) is 65×120×23, where 65 mm is the inner ring diameter, 120 mm is the outer ring diameter, and 23 mm is the bearing width. The web-based interface may provide a drop-down menu for selecting a bearing number from a plurality of bearing numbers. In this example, the bearing number may be selected as 6319.

[0077] Based on the bearing number, dimensions such as ball radius and bearing diameter may be automatically populated on the web-based interface. Similarly, if the bearing number is not displayed in the drop-down menu, the web-based interface may also provide the user with the option to manually enter the bearing parameters. In addition, the user may confirm the value of one or more bearing parameters by pressing a "Submit" button on the web-based interface to initiate the request. In a preferred embodiment, the sensor data corresponds to the real-time operating condition of the bearing.

[0078] At step 610, the effect of the defect on the bearing is monitored over a period of time. The effect of the defect is monitored based on the presence of anomalies in the sensor data. The anomalies may be indicated by characteristics of a signal generated by at least one sensing unit. Characteristics may include, but are not limited to, amplitude, frequency, harmonics, spectral energy, RMS velocity, presence of shock pulses or transients, repetitive pulses, etc. In this embodiment, the sensor data includes acceleration signals in the time domain. Figure 8A GUI 800 showing an example of an acceleration signal in the time domain is illustrated according to an embodiment of the present invention. An influence may be identified from the acceleration signal using envelope analysis. More specifically, an envelope of the acceleration signal is generated by amplitude demodulation. For example, the amplitude of the envelope may indicate a periodic influence of a defect. Upon identifying the presence of such an influence, step 615 is performed.

[0079] At step 615, a time period during which the effect of the defect on the bearing is above a threshold range is determined using a machine learning model. The time period indicates the time it takes for a rolling element of the bearing to pass through the defect. More specifically, the time period indicates the duration in which the rolling element enters and leaves the defect. The threshold range may be predefined by an operator or may be based on specifications provided by a bearing manufacturer. For example, the threshold may be defined as 2 mm. 2 / s. Based on the results of envelope analysis, if the amplitude of the acceleration signal crosses the threshold, the envelope amplitude is greater than 2mm during the period 2 / s time period is identified.

[0080] In one embodiment, the machine learning model can be a convolutional neural network model. The CNN model is a trained CNN configured to identify a duration. More specifically, the CNN is configured to identify a start and end time associated with the duration. The duration corresponds to the time it takes for the ball to pass through the defect. For example, in Figure 8 In the time domain acceleration signal, the duration is from t1 = 0.53221 to t2 = 0.5749. Therefore, the duration t defect The calculation is as follows: defect =t2-t1=0.0427 seconds.

[0081] At step 620, the severity of the impact associated with the defect during the time period is calculated. The severity of the defect is determined as the defect size associated with the defect. The defect size is determined based on the duration calculated in step 615. In one embodiment, the defect size can be calculated based on a predefined mathematical relationship. For example, the duration t defect The angular distance θ traveled by the ball is calculated using the following mathematical relationship center :

[0082] θ center =ω cage *t defect (12)

[0083] in,

[0084]

[0085] Where ω is the angular velocity of the rotating shaft on which the bearing is mounted. ω can also be considered as the angular velocity of the inner ring of the bearing during operation. D is the outer ring diameter, d is the inner ring diameter, and α is the contact angle between the ball and the outer ring. For example, if the cage speed is calculated as 63.227 degrees / second by equation (13),

[0086] θ center =63.227deg / s*0.0427s=2.67 degrees.

[0087] Additionally, the defect size is calculated as follows:

[0088] d defect =θ center *R (14)

[0089] Where R is the radius of the outer ring, i.e. R = D / 2. For the case of the 6319 bearing, the outer ring diameter D is 200 mm. Therefore, in this example,

[0090]

[0091] The accuracy of the defect size value calculated here depends on the defect It should be understood that the experimental test apparatus 300 can be used to train and / or validate a machine learning model to accurately predict the duration t based on the acceleration signal. defect More specifically, time series data associated with acceleration signals corresponding to known defect sizes are used to train and validate the machine learning model. During validation, defect sizes calculated based on the output of the machine learning model are compared to actual or known defect sizes to determine errors in the model. If the calculated defect size deviates from the known defect size, the machine learning model is adjusted to minimize the error.

[0092] At step 625, the remaining useful life of the bearing is determined based on the severity and operating data during the time period. In one example, the remaining useful life may be expressed as the number of revolutions before failure. In another example, the remaining useful life is expressed as the number of hours of operation at a constant speed before failure. In a preferred embodiment, the remaining useful life model of the bearing is configured based on the following rating life model:

[0093]

[0094] Among them, a iso is the life correction factor based on the systems approach to life calculation and is given by:

[0095]

[0096] Where a1 is the reliability life correction factor, C u is the fatigue load limit in Newtons, e c is the contamination factor specific to the defect size, P is the dynamic equivalent radial load in Newtons, C is the dynamic equivalent radial load rating, and K is the viscosity ratio. Here, the dynamic parameter used to configure the rating life model is the dynamic equivalent radial load.

[0097] The reliability life correction factor is a predefined value specified in ISO 281:2007 for a given reliability value. For example, if the reliability may be considered to be 90%, a1 is taken as 1. The value of reliability is taken as 90% by default. In one embodiment of the present disclosure, the value of reliability can be modified by an operator through a client device. Based on the value of reliability, the reliability life correction factor can be further obtained from a first lookup table stored in the database 195. The fatigue load limit and the dynamic equivalent radial load rating are obtained from one or more bearing parameters associated with the bearing.

[0098] The contamination factor is determined based on the defect size. This is because defects in the bearing result in the removal of small, discrete particles of material from the bearing structure. These discrete particles increase the concentration of contaminants inside the bearing. As the defect size increases, the concentration of contaminants increases further. The defect size is provided as an input to a trained classification model, which classifies the defect size into one of multiple severity levels. For example, multiple severity levels may correspond to "normal cleanliness", "mild to typical contamination", "severe contamination", and "very severe contamination". Based on the defect size, the trained classification model outputs a severity level. The severity level so determined is further used to select a suitable contamination factor from a second lookup table stored in the database 195. The second lookup table may include contamination factor values ​​corresponding to each severity level. It must be understood by those skilled in the art that the classification model can be trained to classify the defect size into one of any number of severity levels.

[0099] The viscosity ratio indicates the lubrication condition of the bearing during operation. The viscosity ratio is calculated as the ratio of the operating viscosity of the lubricant to the rated viscosity of the lubricant. The operating viscosity is calculated based on the viscosity grade of the lubricant and the operating temperature. The viscosity grade of the oil can be obtained from a third lookup table that includes viscosity grades corresponding to different types of lubricants. In one embodiment, the operating temperature of the bearing can be obtained from a temperature sensor associated with the bearing. In another embodiment, a virtual model of the bearing can be used to determine the thermal profile of the bearing based on the sensor data. The rated viscosity is obtained from a fourth lookup table based on the size of the bearing and the angular velocity of the bearing.

[0100] Equation (15) for the rating life model can be rearranged as follows:

[0101]

[0102] Here, the dynamic equivalent radial load P is the same as the maximum impact force calculated by the virtual model based on the sensor data. Therefore, the remaining service life is a function of the dynamic equivalent radial load P. Since the virtual model determines the maximum impact force based on the mathematical model, experimental data, and simulation data,

[0103]

[0104] Therefore, the remaining useful life model is configured as follows:

[0105]

[0106] Based on equation (19), the remaining useful life (RUL) of the bearing is calculated.

[0107] At step 630, a notification is generated on an output device indicating the remaining useful life of the bearing. The output may be a notification indicating the remaining useful life of the bearing as a dynamically changing parameter based on real-time sensor data. For example, the notification may include a message "The remaining useful life of bearing 6319 is 56 hours." The RUL value may be further displayed in a manner such as Fig. 9 Indicated on the RUL curve shown in . Fig. 9 900 is a GUI view showing the degradation of remaining useful life (shown as predicted life in hours) with increasing defect size according to an embodiment of the present disclosure. The GUI view 900 also shows the degradation of remaining useful life for different severity levels of contamination, as shown. The graphs corresponding to the different severity levels are generated by substituting the contamination factor corresponding to each severity level in the rated life model of equation (15) as the defect size gradually increases.

[0108] The present invention facilitates accurate calculation of the remaining useful life of a bearing based on defect size calculated from real-time sensor data.

[0109] The present invention is not limited to a specific computer system platform, processing unit, operating system or network. One or more aspects of the present invention can be distributed in one or more computer systems, which are, for example, servers configured to provide one or more services to one or more client computers or perform complete tasks in a distributed system. For example, one or more aspects of the present invention can be performed on a client-server system, which includes components distributed in one or more server systems that perform multiple functions according to various embodiments. These components include, for example, executable code, intermediate code or interpreted code, which communicate through a network using a communication protocol. The present invention is not limited to being executable on any specific system or system group, and is not limited to any specific distributed architecture, network or communication protocol.

[0110] Although the present invention has been shown and described in detail by means of preferred embodiments, the present invention is not limited to the disclosed examples. Other variations may be derived by those skilled in the art without departing from the scope of protection of the present invention as claimed.

Claims

1. A computer-implemented method for estimating the remaining useful life of a bearing (105), the method comprising: receiving, by a processing unit (135), from a source (115, 125, 130) a request for analyzing a defect in a bearing, wherein the request includes operational data associated with the bearing; Monitor the effects of defects on the bearing over a period of time; using a machine learning model to determine a time period during which the effect of the defect on the bearing was above a threshold range; calculating a severity of an impact associated with the defect during the time period by calculating a defect size corresponding to the defect based on a duration of the time period; and determining a remaining useful life of the bearing based on the severity and the operating data during the time period; and generating a notification on an output device (130) indicating a remaining useful life of the bearing; Wherein determining the remaining useful life of the bearing based on the severity and operational data during the time period comprises: Using a virtual model of the bearing, dynamic parameters associated with the bearing are calculated based on defect size and operating data; Remaining useful life models for bearing configurations based on dynamic parameters; and Calculate the remaining service life of the bearing based on the configured remaining service life model and operating data; wherein the virtual model is generated based on one or more of simulation data, experimental data, and mathematical data associated with a plurality of other bearings; Wherein the time period indicates the duration in which the rolling element enters and leaves the defect.

2. The method of claim 1, wherein the operational data comprises a real-time output of at least one sensing unit (125) associated with the bearing.

3. The method of any one of claims 1 or 2, wherein monitoring the effect of the defect on the bearing over a period of time comprises: Anomalies in the output of at least one sensing unit (125) are monitored.

4. The method of claim 1 , wherein using the machine learning model to determine a time period during which the effect of the defect on the bearing is above a threshold range comprises: The operational data associated with the bearing is analyzed using a machine learning model to determine the time period.

5. A device (110) for estimating the remaining useful life of a bearing, the device (110) comprising: one or more processing units (135); and A memory unit (140) communicatively coupled to one or more processing units (135), wherein the memory unit (140) includes a bearing management module (165) stored in the form of machine-readable instructions, and the one or more processing units (135) can execute the machine-readable instructions, wherein the bearing management module (165) is configured to perform the method steps according to any one of claims 1 to 4.

6. A system (100) for estimating the remaining useful life of a bearing, the system (100) comprising: one or more sources (115, 125, 130) configured to provide operational data associated with the bearing; and The apparatus (110) of claim 5, being communicatively coupled to the one or more sources (115, 125, 130), wherein the apparatus (110) is configured for estimating a remaining useful life of a bearing based on operational data according to any one of claims 1 to 4.

7. A computer program product having machine-readable instructions stored therein which, when executed by one or more processing units (135), cause the processing units (135) to perform the method according to any one of claims 1 to 4.

8. A computer-readable storage medium comprising instructions which, when executed by a data processing system, cause the data processing system to perform the method according to any one of claims 1 to 4.

Images