Bearing fault detection method and device based on scaling bearing and computer equipment

Abstract

The invention relates to a bearing fault detection method and device based on a scaled bearing, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a first bearing structure parameter of a target bearing and respective second bearing structure parameters of a plurality of scaling bearings matched with the target bearing; performing correlation analysis on the first bearing structure parameter and the second bearing structure parameter based on a bearing dynamic model for each scaled bearing to obtain a bearing correlation degree between the target bearing and the scaled bearing; based on the correlation degree of each bearing, determining a target scaled bearing matched with the target bearing from the plurality of scaled bearings, and training based on the bearing fault data set of the target scaled bearing to obtain a fault diagnosis model for the target bearing; and analyzing the operation data of the target bearing based on the fault diagnosis model, and determining a fault detection result of the target bearing. By adopting the method, the fault detection cost can be reduced, and the bearing fault detection accuracy can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of industrial equipment, and in particular to a bearing fault detection method, device, computer equipment, computer-readable storage medium and computer program product based on scaled-down bearings. Background Art

[0002] Bearings are a widely used key basic component that supports the rotation of machinery during the operation of many industrial equipment. Bearing failure will not only cause significant economic losses, but may also endanger the life safety of operators and cause serious accidents. Therefore, the reliability of bearings is crucial to ensuring the safe and stable operation of industrial equipment. Real-time fault diagnosis of bearings is a focus of attention in the industrial field.

[0003] In traditional technology, bearing fault intelligent diagnosis methods based on deep learning can be used to detect bearing faults. However, the accuracy of deep learning-based methods is based on a large amount of test, operation and fault data. For many emerging industrial fields, such as large wind turbines, there is very little test and fault data on bearing operation, and the cost of conducting bearing reliability tests is extremely high. In this case, due to insufficient data, the effect of deep learning methods is not good, and the fault detection rate for bearings is not high. Summary of the invention

[0004] Based on this, it is necessary to provide a bearing fault detection method, device, computer equipment, computer-readable storage medium and computer program product based on scaled bearings, which can reduce fault detection costs and improve bearing fault detection accuracy, in order to address the above technical problems.

[0005] In the first aspect, the present application provides a bearing fault detection method based on a scaled-down bearing. The method comprises: obtaining a first bearing structural parameter of a target bearing, and a second bearing structural parameter of each of a plurality of scaled-down bearings matching the target bearing; for each scaled-down bearing, based on a bearing dynamics model, performing a correlation analysis on the first bearing structural parameter and the second bearing structural parameter to obtain a bearing correlation between the target bearing and the scaled-down bearing; based on each bearing correlation, determining a target scaled-down bearing matching the target bearing from the plurality of scaled-down bearings, and training a fault diagnosis model for the target bearing based on a bearing fault data set of the target scaled-down bearing; analyzing the operating data of the target bearing based on the fault diagnosis model to determine a fault detection result of the target bearing.

[0006] In one embodiment, based on the bearing dynamics model, the first bearing structural parameter and the second bearing structural parameter are subjected to correlation analysis to obtain the bearing correlation between the target bearing and the scaled bearing, including: inputting the first bearing structural parameter into the bearing dynamics model to obtain a first simulation signal, and converting the first simulation signal to obtain a first frequency domain signal; inputting the second bearing structural parameter into the bearing dynamics model to obtain a second simulation signal, and converting the second simulation signal to obtain a second frequency domain signal; and performing correlation analysis on the first frequency domain signal and the second frequency domain signal to obtain the bearing correlation between the target bearing and the scaled bearing.

[0007] In one embodiment, the first frequency domain signal includes a first sub-frequency domain signal of the target bearing in a normal state, and a second sub-frequency domain signal of the target bearing in a fault state; the second frequency domain signal includes a third sub-frequency domain signal of the scaled bearing in a normal state, and a fourth sub-frequency domain signal of the scaled bearing in a fault state; the correlation analysis of the first frequency domain signal and the second frequency domain signal to obtain the bearing correlation between the target bearing and the scaled bearing includes: determining a first autopower spectral density corresponding to each first frequency band in the first sub-frequency domain signal, and a second autopower spectral density corresponding to each first frequency band in the third sub-frequency domain signal, and based on the first sub-frequency domain signal The first mutual power spectral density corresponding to each first frequency band in a normal state is determined based on the first auto-power spectral density, the second auto-power spectral density and the first mutual power spectral density corresponding to each first frequency band; the first sub-correlation between the target bearing and the scaled bearing in a normal state in each first frequency band is determined based on the second sub-frequency domain signal and the fourth sub-frequency domain signal; the second sub-correlation between the target bearing and the scaled bearing in a faulty state in each second frequency band is determined based on the first sub-correlation corresponding to each first frequency band and the second sub-correlation corresponding to each second frequency band.

[0008] In one embodiment, determining a target scaled-down bearing matching the target bearing from the multiple scaled-down bearings based on the correlations of each bearing includes: for each scaled-down bearing, counting a first comparison result between a first sub-correlation corresponding to each first frequency band and a first set threshold, and a second comparison result between a second sub-correlation corresponding to each second frequency band and a second set threshold; when the first comparison result corresponding to the first frequency band indicates that the first sub-correlation is greater than the first set threshold, and the second comparison result corresponding to the second frequency band indicates that the second sub-correlation is greater than the second set threshold, determining the first frequency band as the first target frequency band, and determining the second frequency band as the second target frequency band; when the number of the first target frequency bands exceeds a first number threshold, and the number of the second target frequency bands exceeds a second number threshold, determining the scaled-down bearings corresponding to the first target frequency band and the second target frequency band as candidate scaled-down bearings matching the target bearing; and determining a target scaled-down bearing matching the target bearing based on the candidate scaled-down bearings.

[0009] In one embodiment, there are multiple candidate scaled bearings; determining a target scaled bearing matching the target bearing based on the candidate scaled bearings includes: determining a candidate scaled bearing with the smallest outer diameter from the multiple candidate scaled bearings as the target scaled bearing matching the target bearing.

[0010] In one embodiment, the method further includes: determining, from multiple candidate scaled bearings, the candidate scaled bearing with the largest number of the first target frequency bands and / or the candidate scaled bearing with the largest number of the second target frequency bands as the target scaled bearing that matches the target bearing.

[0011] In the second aspect, the present application also provides a bearing fault detection device based on a scaled-down bearing, the device comprising: an acquisition module, used to acquire a first bearing structural parameter of a target bearing, and a second bearing structural parameter of each of a plurality of scaled-down bearings matching the target bearing; an analysis module, used to perform a correlation analysis on the first bearing structural parameter and the second bearing structural parameter for each of the scaled-down bearings based on a bearing dynamics model, and obtain a bearing correlation between the target bearing and the scaled-down bearing; a processing module, used to determine a target scaled-down bearing matching the target bearing from the plurality of scaled-down bearings based on each of the bearing correlations, and train a fault diagnosis model for the target bearing based on a bearing fault data set of the target scaled-down bearing; a diagnosis module, used to analyze the operating data of the target bearing based on the fault diagnosis model, and determine a fault detection result of the target bearing.

[0012] In a third aspect, the present application also provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program: obtaining a first bearing structural parameter of a target bearing, and a second bearing structural parameter of each of a plurality of scaled-down bearings matching the target bearing; for each scaled-down bearing, based on a bearing dynamics model, performing a correlation analysis on the first bearing structural parameter and the second bearing structural parameter to obtain a bearing correlation between the target bearing and the scaled-down bearing; based on each bearing correlation, determining a target scaled-down bearing matching the target bearing from the plurality of scaled-down bearings, and training a fault diagnosis model for the target bearing based on a bearing fault data set of the target scaled-down bearing; analyzing the operating data of the target bearing based on the fault diagnosis model to determine a fault detection result of the target bearing.

[0013] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented: obtaining a first bearing structural parameter of a target bearing, and a second bearing structural parameter of each of a plurality of scaled-down bearings matching the target bearing; for each scaled-down bearing, based on a bearing dynamics model, performing a correlation analysis on the first bearing structural parameter and the second bearing structural parameter to obtain a bearing correlation between the target bearing and the scaled-down bearing; based on each bearing correlation, determining a target scaled-down bearing matching the target bearing from the plurality of scaled-down bearings, and training a fault diagnosis model for the target bearing based on a bearing fault data set of the target scaled-down bearing; analyzing the operating data of the target bearing based on the fault diagnosis model to determine a fault detection result of the target bearing.

[0014] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program, which implements the following steps when executed by a processor: obtaining a first bearing structural parameter of a target bearing and a second bearing structural parameter of each of a plurality of scaled-down bearings matching the target bearing; for each scaled-down bearing, based on a bearing dynamics model, performing a correlation analysis on the first bearing structural parameter and the second bearing structural parameter to obtain a bearing correlation between the target bearing and the scaled-down bearing; based on each bearing correlation, determining a target scaled-down bearing matching the target bearing from the plurality of scaled-down bearings, and training a fault diagnosis model for the target bearing based on a bearing fault data set of the target scaled-down bearing; analyzing the operating data of the target bearing based on the fault diagnosis model to determine a fault detection result of the target bearing.

[0015] The above-mentioned bearing fault detection method, device, computer equipment, storage medium and computer program product based on scaled-down bearings obtain the first bearing structural parameters of the target bearing and the second bearing structural parameters of each of the multiple scaled-down bearings matching the target bearing, and for each scaled-down bearing, based on the bearing dynamics model, perform correlation analysis on the first bearing structural parameters and the second bearing structural parameters to obtain the bearing correlation between the target bearing and the scaled-down bearing, and based on each bearing correlation, determine the target scaled-down bearing matching the target bearing from the multiple scaled-down bearings, and based on the bearing fault data set of the target scaled-down bearing, train a fault diagnosis model for the target bearing, and then analyze the operating data of the target bearing based on the fault diagnosis model to determine the fault detection result of the target bearing. Therefore, in the case where the target bearing has less operating data and fault data, a fault diagnosis model for the target bearing can be trained based on the bearing fault data set of the target scaled bearing from the perspective of the target scaled bearing that matches the target bearing, and fault detection of the target bearing can be realized by the fault diagnosis model. Therefore, by avoiding the situation where the target bearing has less operating data and fault data and the fault diagnosis model is trained based on the operating data and fault data of the target bearing, the fault detection accuracy can be improved, and there is no need to conduct reliability experiments on the target bearing, which can reduce the fault detection cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of a bearing fault detection method based on a scaled-down bearing in one embodiment;

[0017] Figure 2 A schematic diagram of a process for performing a correlation analysis on a first bearing structure parameter and a second bearing structure parameter based on a bearing dynamics model in an embodiment to obtain a bearing correlation between a target bearing and a scaled bearing;

[0018] Figure 3 A schematic diagram of a flow chart of performing correlation analysis on a first frequency domain signal and a second frequency domain signal to obtain a bearing correlation between a target bearing and a scaled bearing in an embodiment;

[0019] Figure 4 A schematic diagram of a process for determining a target scaled-down bearing matching a target bearing from a plurality of scaled-down bearings based on the correlation of each bearing in an embodiment;

[0020] Figure 5 A schematic flow chart of a bearing fault detection method based on a scaled-down bearing in another embodiment;

[0021] Figure 6 is a schematic diagram of a bearing correlation between a target bearing and a scaled bearing in one embodiment;

[0022] Figure 7 is a schematic diagram of a fault detection result of a target bearing in one embodiment;

[0023] Figure 8 is a structural block diagram of a bearing fault detection device based on a scaled-down bearing in one embodiment;

[0024] Fig. 9 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0026] In one embodiment, Figure 1 As shown, a bearing fault detection method based on a scaled bearing is provided, and the method is applied to a server in a bearing fault detection system as an example for description, and includes the following steps:

[0027] S102, obtaining a first bearing structural parameter of a target bearing and a second bearing structural parameter of each of a plurality of scaled-down bearings matching the target bearing.

[0028] Among them, the first bearing structure parameter refers to the parameter related to the bearing structure of the target bearing, which is used to characterize the physical properties and geometric dimensions of the target bearing, including but not limited to: inner diameter, outer diameter, steel ball diameter, number of steel balls, inner ring groove curvature radius, inner ring groove curvature radius, inner ring channel radius, outer ring channel radius, radial clearance, elastic modulus, Poisson's ratio, and load, etc.

[0029] A scaled-down bearing refers to a bearing used in a model or scaled-down test bench, whose size and specifications are reduced relative to the bearing in the actual application, so as to simulate the performance and behavior of the actual bearing in a smaller space or model. The second bearing structure parameter refers to the parameter related to the bearing structure of the scaled-down bearing. The type of the parameter in the second bearing structure parameter is not limited. For example, the second bearing structure parameter can refer to the parameter adaptation description of the first bearing structure parameter.

[0030] S104, for each scaled-down bearing, based on the bearing dynamics model, performing a correlation analysis on the first bearing structural parameter and the second bearing structural parameter to obtain a bearing correlation between the target bearing and the scaled-down bearing.

[0031] There is no limit to the construction method of the bearing dynamics model. For example, a nonlinear Hertz contact stiffness model of the bearing can be established based on the basic bearing theory and nonlinear Hertz contact theory, and a mathematical model of vibration generated by the bearing structure itself can be established based on nonlinear mechanics theory. The mathematical model of vibration generated by the bearing structure itself is used to reflect the influence of multiple bearing structure parameters (such as radial clearance, groove curvature radius, number of steel balls, steel ball diameter, bearing model, etc.) and parameters such as lubrication damping on bearing vibration. Then, when introducing multiple fault modes, a bearing dynamics model is constructed based on the nonlinear Hertz contact stiffness model of the bearing and the mathematical model of vibration generated by the bearing structure itself.

[0032] Among them, considering the fatigue damage of the bearing caused by long-term service, the accuracy of the constructed bearing dynamics model can be improved by introducing multiple fault types such as inner ring fault, outer ring fault and roller fault.

[0033] Among them, based on the bearing dynamics model, the correlation analysis is performed on the first bearing structural parameters and the second bearing structural parameters, and the implementation method of obtaining the bearing correlation between the target bearing and the scaled bearing is not limited. The following is an example of possible implementation methods.

[0034] In one embodiment, the first bearing structure parameters are input into the bearing dynamics model to obtain a first simulation signal, which includes a corresponding simulation signal including vibration, acceleration and other parameters that match the target bearing; the second bearing structure parameters are input into the bearing dynamics model to obtain a second simulation signal, which includes a corresponding simulation signal including vibration, acceleration and other parameters that match the scaled bearing; by performing correlation analysis on the first simulation signal and the second simulation signal, the bearing correlation between the target bearing and the scaled bearing can be obtained.

[0035] There is no limitation on the method of performing correlation analysis on the first simulation signal and the second simulation signal. For example, correlation analysis may be performed based on a cross-correlation analysis method or a joint probability distribution analysis method.

[0036] S106, based on the correlation of each bearing, determine a target scaled-down bearing that matches the target bearing from multiple scaled-down bearings, and train a fault diagnosis model for the target bearing based on the bearing fault data set of the target scaled-down bearing.

[0037] There is no limitation on the implementation method of determining a target scaled-down bearing that matches the target bearing from a plurality of scaled-down bearings based on the correlation of each bearing, and several of the methods are described below as examples.

[0038] In one embodiment, from among the bearing correlations, the scaled-down bearing corresponding to the largest bearing correlation is determined as the target scaled-down bearing that matches the target bearing.

[0039] In one embodiment, the bearing correlations are sorted in order from large to small, and the top N bearing correlations are used; a bearing correlation is randomly selected from the N bearing correlations, and the scaled bearing corresponding to the randomly selected bearing correlation is determined as the target scaled bearing that matches the target bearing.

[0040] In the case of determining a target scaled bearing that matches the target bearing, a corresponding reliability test can be carried out for the target scaled bearing. For example, the signal of the target scaled bearing under different fault types can be obtained by artificially injecting the fault type into the target scaled bearing, so as to obtain a bearing fault data set. The bearing fault data set may include normal signals and fault signals. Alternatively, the historical operation data and fault data of the target scaled bearing can be combined to obtain a bearing fault data set.

[0041] In one embodiment, the bearing fault data set can be divided into a training set, a validation set and a test set, and then a fault diagnosis model is obtained by training based on the training set, the test set and the validation set using a deep learning method (such as a convolutional neural network, etc.). The validation set is a sample set set aside during the training of the fault diagnosis model, which is used to adjust the hyperparameters of the fault diagnosis model during the training process and to perform a preliminary evaluation of the capabilities of the fault diagnosis model; the test set is used to evaluate the generalization capability of the fault diagnosis model. When the accuracy of the fault diagnosis model for the bearing is greater than a preset value, it can be considered that the trained fault diagnosis model meets the requirements, thereby completing the construction of the fault diagnosis model.

[0042] In some cases, when the reliability test of the target scaled-down bearing still faces high costs and it is impossible to carry out a large number of tests, the bearing operation data obtained by the simulation model corresponding to the target scaled-down bearing can be added to the bearing fault data set as supplementary data to update the bearing fault data set. In this way, a fault diagnosis model can be trained based on the updated bearing fault data set to improve the training accuracy of the fault diagnosis model.

[0043] S108, analyzing the operating data of the target bearing based on the fault diagnosis model to determine the fault detection result of the target bearing.

[0044] Specifically, the operating data of the target bearing is input into the fault diagnosis model to determine the fault detection result of the target bearing. For example, when the fault detection result indicates that the target bearing has a fault, the fault detection result may include the fault type of the target bearing, and the fault type may be other fault types such as inner ring fault, outer ring fault, and roller fault.

[0045] based on Figure 1The method shown, by obtaining the first bearing structural parameter of the target bearing and the second bearing structural parameter of each of the multiple scaled bearings matching the target bearing, for each scaled bearing, based on the bearing dynamics model, the first bearing structural parameter and the second bearing structural parameter are analyzed for correlation, the bearing correlation between the target bearing and the scaled bearing is obtained, based on the bearing correlation, the target scaled bearing matching the target bearing is determined from the multiple scaled bearings, and based on the bearing fault data set of the target scaled bearing, a fault diagnosis model for the target bearing is trained to obtain the fault detection result of the target bearing, and then the operating data of the target bearing is analyzed based on the fault diagnosis model, so that the fault detection result of the target bearing can be determined. Therefore, for the case where the operating data and fault data of the target bearing are less, the fault diagnosis model for the target bearing can be trained based on the bearing fault data set of the target scaled bearing from the perspective of the target scaled bearing matching the target bearing, and the fault detection of the target bearing is realized by the fault diagnosis model, so that the fault detection accuracy can be improved by avoiding the case where the operating data and fault data of the target bearing are less, and the reliability experiment of the target bearing can be avoided, and the fault detection cost can be reduced.

[0046] In one embodiment, based on the bearing dynamics model, the first bearing structural parameter and the second bearing structural parameter are subjected to correlation analysis to obtain the bearing correlation between the target bearing and the scaled bearing (i.e., S104) as follows: Figure 2 As shown, the following steps are included:

[0047] S202: Input the first bearing structural parameter into the bearing dynamics model to obtain a first simulation signal, and perform conversion processing on the first simulation signal to obtain a first frequency domain signal.

[0048] Wherein, the first simulation signal is a time domain signal, and the first frequency domain signal can be obtained by transforming the first simulation information. For example, the first simulation information can be transformed by a method such as Fast Fourier Transform (FFT) to obtain the first frequency domain signal.

[0049] In one embodiment, the bearing dynamics model can be represented by a function. When the bearing dynamics model introduces inner race fault, outer race fault and roller fault, the bearing dynamics model satisfies:

[0050]

[0051] Where m is the bearing mass, c is the lubrication damping, is the horizontal speed, is the horizontal acceleration, is the horizontal speed, is the horizontal acceleration, k is the shaft stiffness, K n is the contact stiffness, is the angle between the center line of the jth rolling element and the line of action of the radial load, x j is the horizontal displacement of the jth rolling element, y j is the vertical displacement of the jth rolling element; is the contact angle, is the radial clearance, u j is the instantaneous deformation released when the jth rolling element passes through the defect, F x is the horizontal contact load, F y is the contact load in the vertical direction; x and y represent the displacement in the horizontal and vertical directions respectively, that is, the first simulation signal of the bearing in a normal state or a fault state includes the displacement x in the horizontal direction and the displacement y in the vertical direction.

[0052] In some embodiments, the first frequency domain signal includes a first sub-frequency domain signal of the target bearing in a normal state, and a second sub-frequency domain signal of the target bearing in a faulty state. That is, the first simulation signal includes a first sub-simulation signal of the target bearing in a normal state, and a second sub-simulation signal of the target bearing in a faulty state. Thus, by transforming the first sub-simulation signal, the first sub-frequency domain signal can be obtained, and by transforming the second sub-simulation signal, the second sub-frequency domain signal can be obtained.

[0053] Specifically, the deformation in the bearing dynamics model (i.e., u j ) is set to 0, the first bearing structural parameters are input into the bearing dynamics model, and the first sub-simulation signal of the target bearing in a normal state can be obtained. The first sub-simulation signal includes the corresponding simulation signal including the vibration, acceleration and other parameters matching the target bearing in the normal state; while retaining the deformation in the bearing dynamics model (i.e., u j ), by inputting the first bearing structural parameters into the bearing dynamics model, a second sub-simulation signal of the target bearing in a fault state can be obtained, and the second sub-simulation signal includes a corresponding simulation signal including parameters such as vibration and acceleration that match the target bearing in the fault state.

[0054] S204, inputting the second bearing structural parameters into the bearing dynamics model to obtain a second simulation signal, and converting the second simulation signal to obtain a second frequency domain signal.

[0055] Among them, the second frequency domain signal includes the third sub-frequency domain signal of the scaled bearing in a normal state, and the fourth sub-frequency domain signal of the scaled bearing in a faulty state. That is to say, the second simulation signal includes the third sub-simulation signal of the scaled bearing in a normal state, and the fourth sub-simulation signal of the scaled bearing in a faulty state. Therefore, after transforming the third sub-simulation signal, the third sub-frequency domain signal can be obtained. Similarly, after transforming the fourth sub-simulation signal, the fourth sub-frequency domain signal can be obtained.

[0056] Specifically, the deformation in the bearing dynamics model (i.e., u j ) is set to 0, the second bearing structural parameters are input into the bearing dynamics model, and the third sub-simulation signal of the scaled bearing in a normal state can be obtained. The third sub-simulation signal includes the corresponding simulation signal including the vibration, acceleration and other parameters matching the scaled bearing in the normal state; while retaining the deformation in the bearing dynamics model (i.e., u j ), by inputting the second bearing structure parameters into the bearing dynamics model, a fourth sub-simulation signal of the scaled-down bearing in a faulty state can be obtained. The fourth sub-simulation signal includes corresponding simulation signals including parameters such as vibration and acceleration that match the scaled-down bearing in the faulty state.

[0057] S206, performing correlation analysis on the first frequency domain signal and the second frequency domain signal to obtain a bearing correlation between the target bearing and the scaled bearing.

[0058] There is no limitation on the implementation method of performing correlation analysis on the first frequency domain signal and the second frequency domain signal to obtain the bearing correlation between the target bearing and the scaled bearing. For example, the first frequency domain signal and the second frequency domain signal can be subjected to correlation analysis based on a cross-correlation analysis method or a joint probability distribution analysis method to obtain the bearing correlation between the target bearing and the scaled bearing.

[0059] based on Figure 2 According to the method described, the frequency domain signal can better reflect the characteristics of the bearing. Therefore, by determining the correlation between the first frequency domain signal and the second frequency domain signal in the frequency domain from the perspective of the frequency domain signal, the accuracy of the determined bearing correlation can be improved.

[0060] In one embodiment, the first frequency domain signal includes a first sub-frequency domain signal of the target bearing in a normal state and a second sub-frequency domain signal of the target bearing in a faulty state, and the second frequency domain signal includes a third sub-frequency domain signal of the scaled bearing in a normal state and a fourth sub-frequency domain signal of the scaled bearing in a faulty state. Specifically, the first frequency domain signal and the second frequency domain signal are subjected to correlation analysis to obtain the bearing correlation between the target bearing and the scaled bearing (i.e., S206), which can be implemented as follows: Figure 3 As shown, the following steps are included:

[0061] S302, determine the first auto power spectral density corresponding to each first frequency band in the first sub-frequency domain signal, and the second auto power spectral density corresponding to each first frequency band in the third sub-frequency domain signal, and determine the first cross power spectral density corresponding to each first frequency band in a normal state based on the first sub-frequency domain signal and the third sub-frequency domain signal.

[0062] S304: Determine the first sub-correlation of the target bearing and the scaled bearing in the normal state in each first frequency band based on the first auto-power spectral density, the second auto-power spectral density and the first cross-power spectral density corresponding to each first frequency band.

[0063] Among them, the first sub-correlation is between [0,1], and can be used to evaluate the frequency linear correlation between the first sub-frequency domain signal and the third sub-frequency domain signal in the first frequency band; when the difference between the first sub-correlation and 1 is smaller, it can be explained that the first sub-frequency domain signal and the third sub-frequency domain signal are more correlated.

[0064] Specifically, for each first frequency band, the first product between the first auto-power spectrum density and the second auto-power spectrum density corresponding to the first frequency band is determined; the square of the first cross-power spectrum density corresponding to the first frequency band and the ratio of the first product are determined as the first sub-correlation of the target bearing and the scaled bearing in the first frequency band under normal conditions. The first sub-correlation may refer to the homology of the target bearing and the scaled bearing in the first frequency band, that is, the correlation between the target bearing and the scaled bearing in the first frequency band.

[0065] In some embodiments, the first sub-correlation between the target bearing and the scaled bearing in the normal state in the first frequency band can be calculated based on the mscohere function in MATLAB.

[0066] For example, f represents the first frequency band, represents the first sub-frequency domain signal, represents the third sub-frequency domain signal, It represents the first sub-correlation between the target bearing and the scaled bearing in the normal state in the first frequency band. represents the first autopower spectral density corresponding to the first frequency band in the first sub-frequency domain signal, represents the second autopower spectrum density corresponding to the first frequency band in the third sub-frequency domain signal, then satisfy:

[0067]

[0068] S306: Determine the second sub-correlation between the target bearing and the scaled bearing in the fault state in each second frequency band based on the second sub-frequency domain signal and the fourth sub-frequency domain signal.

[0069] Specifically, determine the third auto power spectral density corresponding to each second frequency band in the second sub-frequency domain signal and the fourth auto power spectral density corresponding to each second frequency band in the fourth sub-frequency domain signal, and determine the second mutual power spectral density corresponding to each second frequency band in the fault state based on the second sub-frequency domain signal and the fourth sub-frequency domain signal; based on the third auto power spectral density, the fourth auto power spectral density and the second mutual power spectral density corresponding to each second frequency band, determine the second sub-correlation of the target bearing and the scaled bearing in each second frequency band in the fault state.

[0070] The implementation method of determining the third auto power spectrum density, the fourth auto power spectrum density and the second cross power spectrum density may refer to the description of the contents of S302 to S304 and will not be described in detail here.

[0071] S308: Obtain a bearing correlation between the target bearing and the scaled bearing based on the first sub-correlations corresponding to the first frequency bands and the second sub-correlations corresponding to the second frequency bands.

[0072] Among them, based on the first sub-correlation corresponding to each first frequency band and the second sub-correlation corresponding to each second frequency band, there is no limitation on the implementation method of obtaining the bearing correlation between the target bearing and the scaled bearing. The following is an example of possible implementation methods.

[0073] In one embodiment, the first sub-correlation corresponding to each first frequency band and the second sub-correlation corresponding to each second frequency band can be combined to obtain the bearing correlation between the target bearing and the scaled bearing, that is, the bearing correlation between the target bearing and the scaled bearing includes the first sub-correlation corresponding to each first frequency band and the second sub-correlation corresponding to each second frequency band.

[0074] based on Figure 3 The content shown can improve the accuracy of the determined bearing correlation by determining the bearing correlation between the target bearing and the scaled bearing from the perspective of auto-power spectral density and cross-power spectral density.

[0075] In one embodiment, based on the correlation of each bearing, the method for determining a target scaled-down bearing matching the target bearing from a plurality of scaled-down bearings can be as follows: Figure 4 As shown, the following steps are included:

[0076] S402: For each scaled-down bearing, statistics are collected on a first comparison result between a first sub-correlation corresponding to each first frequency band and a first set threshold, and a second comparison result between a second sub-correlation corresponding to each second frequency band and a second set threshold.

[0077] S404, when a first comparison result corresponding to the first frequency band indicates that the first sub-correlation is greater than a first set threshold, and a second comparison result corresponding to the second frequency band indicates that the second sub-correlation is greater than a second set threshold, the first frequency band is determined as a first target frequency band, and the second frequency band is determined as a second target frequency band.

[0078] When the first sub-correlation corresponding to the first frequency band is greater than the first set threshold, it can be said that the first sub-frequency domain signal and the third sub-frequency domain signal are correlated in the first frequency band. Similarly, when the second sub-correlation corresponding to the second frequency band is greater than the second set threshold, it can be said that the second sub-frequency domain signal and the fourth sub-frequency domain signal are correlated in the second frequency band.

[0079] S406: When the number of the first target frequency bands exceeds a first number threshold and the number of the second target frequency bands exceeds a second number threshold, the scaled-down bearings corresponding to the first target frequency bands and the second target frequency bands are determined as candidate scaled-down bearings matching the target bearing.

[0080] When the number of the first target frequency band exceeds the first number threshold, it can be said that the first sub-frequency domain signal is similar to the third sub-frequency domain signal. Similarly, when the number of the second target frequency band exceeds the second number threshold, it can be said that the second sub-frequency domain signal is similar to the fourth sub-frequency domain signal.

[0081] Among them, the first quantity threshold and the second quantity threshold may be the same or different. For example, the first quantity threshold may be determined based on the total number of the first frequency band, and the first quantity threshold may be greater than or equal to half of the total number of the first frequency band. That is, when the proportion of the number of the first target frequency band to the total number of the first frequency band exceeds the first preset proportion, it can be said that the first sub-frequency domain signal and the third sub-frequency domain signal are similar. Similarly, when the proportion of the number of the second target frequency band to the total number of the second frequency band exceeds the second preset proportion, it can be said that the second sub-frequency domain signal and the fourth sub-frequency domain signal are similar. The first preset proportion and the second preset proportion may be the same or different, for example, the first preset proportion may be 60% or other values.

[0082] S408: Determine a target scaled-down bearing that matches the target bearing based on the candidate scaled-down bearings.

[0083] There is no limitation on the implementation method of determining the target scaled-down bearing that matches the target bearing based on the candidate scaled-down bearings, and several of the methods are described below as examples.

[0084] In one embodiment, when the number of candidate scaled-down bearings is one, the candidate scaled-down bearing is determined as a target scaled-down bearing that matches the target bearing.

[0085] In one embodiment, when there are multiple candidate scaled bearings, the candidate scaled bearing with the smallest outer diameter is determined as the target scaled bearing that matches the target bearing from among the multiple candidate scaled bearings; or, the candidate scaled bearing with the largest number of first target frequency bands and / or the largest number of second target frequency bands is determined as the target scaled bearing that matches the target bearing from among the multiple candidate scaled bearings; or, a randomly selected candidate scaled bearing is determined as the target scaled bearing that matches the target bearing from among the multiple candidate scaled bearings.

[0086] In some cases, when there are multiple candidate scaled bearings, for each candidate scaled bearing, determine the first similarity between the contact angle, Poisson's ratio and lubrication condition parameters of the candidate scaled bearing and the contact angle, Poisson's ratio and lubrication condition parameters of the target bearing; determine the candidate scaled bearing corresponding to the largest first similarity as the target scaled bearing that matches the target bearing; or, for each candidate scaled bearing, determine the second similarity between the geometric structure parameters and mechanical structure parameters of the candidate scaled bearing and the geometric structure parameters and mechanical structure parameters of the target bearing; determine the candidate scaled bearing corresponding to the largest second similarity as the target scaled bearing that matches the target bearing. The mechanical structure parameters may include parameters such as the load on the bearing.

[0087] In some embodiments, when there are multiple candidate scaled bearings, a candidate scaled bearing that is easy to carry out reliability experiments can also be determined as a target scaled bearing that matches the target bearing. Thus, the accuracy of the bearing fault data set obtained for the target scaled bearing can be improved, and the accuracy of the obtained fault diagnosis model can be further improved, so as to further improve the accuracy of the fault detection result for the target bearing.

[0088] based on Figure 4 The method shown can determine the target scaled-down bearing that matches the target bearing. In the case where the target bearing has less operating data and fault data, fault detection of the target bearing can be achieved from the perspective of the target scaled-down bearing that matches the target bearing, thereby improving the accuracy of fault detection.

[0089] In combination with the above, in one embodiment, Figure 5 As shown, a bearing fault detection method based on a scaled bearing is provided. Taking the method applied to a server in a bearing fault detection system as an example, the method includes the following steps:

[0090] S502, obtaining a first bearing structural parameter of a target bearing and a second bearing structural parameter of each of a plurality of scaled-down bearings matching the target bearing.

[0091] S504, input the first bearing structure parameters into the bearing dynamics model to obtain a first simulation signal, and convert the first simulation signal to obtain a first frequency domain signal, wherein the first frequency domain signal includes a first sub-frequency domain signal of the target bearing in a normal state and a second sub-frequency domain signal of the target bearing in a fault state.

[0092] S506, for each scaled-down bearing, input the second bearing structural parameters into the bearing dynamics model to obtain a second simulation signal, and convert the second simulation signal to obtain a second frequency domain signal, the second frequency domain signal includes a third sub-frequency domain signal of the scaled-down bearing in a normal state, and a fourth sub-frequency domain signal of the scaled-down bearing in a faulty state.

[0093] S508, determine the first auto power spectral density corresponding to each first frequency band in the first sub-frequency domain signal, and the second auto power spectral density corresponding to each first frequency band in the third sub-frequency domain signal, and determine the first cross power spectral density corresponding to each first frequency band in a normal state based on the first sub-frequency domain signal and the third sub-frequency domain signal.

[0094] S510, determining a first sub-correlation between a target bearing and a scaled bearing in a normal state in each first frequency band based on a first auto-power spectral density, a second auto-power spectral density and a first cross-power spectral density corresponding to each first frequency band.

[0095] S512: Determine the second sub-correlation between the target bearing and the scaled bearing in the fault state in each second frequency band based on the second sub-frequency domain signal and the fourth sub-frequency domain signal.

[0096] S514, for each scaled-down bearing, statistics are collected on a first comparison result between a first sub-correlation corresponding to each first frequency band and a first set threshold, and a second comparison result between a second sub-correlation corresponding to each second frequency band and a second set threshold.

[0097] S516, when the first comparison result corresponding to the first frequency band indicates that the first sub-correlation is greater than the first set threshold, and the second comparison result corresponding to the second frequency band indicates that the second sub-correlation is greater than the second set threshold, the first frequency band is determined as the first target frequency band, and the second frequency band is determined as the second target frequency band.

[0098] S518, when the number of the first target frequency bands exceeds the first number threshold and the number of the second target frequency bands exceeds the second number threshold, the scaled-down bearings corresponding to the first target frequency bands and the second target frequency bands are determined as candidate scaled-down bearings matching the target bearings.

[0099] S520, based on the candidate scaled-down bearings, determine a target scaled-down bearing that matches the target bearing, and based on the bearing fault data set of the target scaled-down bearing, train a fault diagnosis model for the target bearing.

[0100] S522, analyzing the operating data of the target bearing based on the fault diagnosis model to determine the fault detection result of the target bearing.

[0101] Among them, the specific content of S502-S522 can be adapted to the description of the above content and will not be repeated here.

[0102] based on Figure 5 The content shown in the figure establishes a fault diagnosis model for the target bearing based on the bearing fault data set of the scaled bearing, and dynamically analyzes the operating status of the target bearing, which can solve the problem of on-site fault diagnosis of bearings in large industrial equipment due to the lack of test and fault data, so as to support the operation and management of the bearings during actual operation. Moreover, by constructing a fault diagnosis model for the target bearing using the data of the scaled bearing, which is easy to obtain fault data, real-time fault diagnosis of large bearings can be achieved, avoiding reliability tests of large bearings, reducing test costs, and improving economic efficiency.

[0103] Combination Figure 5 The content shown takes the SKF-6206 bearing as the target bearing. The bearing mass m of the SKF-6206 bearing is 0.173kg, the lubrication damping c is 30Ns / mm, and the angular velocity of the rolling element is Taking 145.2 rad / s and contact angle α as 0 as an example, as shown in Table 1, a bearing structure parameter of a target bearing is provided, where:

[0104]

[0105] The bearing dynamics model satisfies:

[0106]

[0107] Then in u j When u is set to 0, the parameters in Table 1 are input into the bearing dynamics model to obtain the first sub-simulation signal of the SKF-6206 bearing in the normal state; j In this case, the parameters in Table 1 are input into the bearing dynamics model, and the second sub-simulation signal of the SKF-6206 bearing in the fault state can be obtained.

[0108] Further, combined with the above analysis, such as Figure 6As shown in FIG. 2 , a schematic diagram of the bearing correlation between the SKF-6205 bearing and the SKF-6206 bearing is provided. It can be seen that the SKF-6205 bearing has a high similarity with the target bearing, so the SKF-6205 bearing can be determined as the target scaled bearing matching the SKF-6206 bearing. Table 2 provides a bearing structural parameter of the SKF-6205 bearing, where:

[0109]

[0110] Furthermore, by conducting reliability experiments or simulation experiments on the SKF-6205 bearing, a bearing fault data set can be obtained. Based on the bearing fault data set, a fault diagnosis model for the target bearing SKF-6206 is trained using methods such as convolutional neural networks. Then, based on the fault diagnosis model, fault detection is performed on the SKF-6206 bearing, and the following results can be obtained: Figure 7 The fault detection results are shown in Figure 2. It can be found that the fault diagnosis model based on the scaled bearing construction can accurately identify whether the SKF-6206 bearing is faulty and the fault type.

[0111] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0112] Based on the same inventive concept, the embodiment of the present application also provides a bearing fault detection device based on a scaled-down bearing for implementing the bearing fault detection method based on a scaled-down bearing involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more bearing fault detection devices based on scaled-down bearings provided below can refer to the limitations of the bearing fault detection method based on scaled-down bearings above, and will not be repeated here.

[0113] In one embodiment, Figure 8 As shown, a bearing fault detection device based on a scaled bearing is provided, comprising: an acquisition module 802, an analysis module 804, a processing module 806 and a diagnosis module 808, wherein:

[0114] An acquisition module 802 is used to acquire the first bearing structural parameters of a target bearing and the second bearing structural parameters of each of a plurality of scaled-down bearings matching the target bearing; an analysis module 804 is used to perform a correlation analysis on the first bearing structural parameters and the second bearing structural parameters for each of the scaled-down bearings based on a bearing dynamics model, so as to obtain a bearing correlation between the target bearing and the scaled-down bearing; a processing module 806 is used to determine a target scaled-down bearing matching the target bearing from the plurality of scaled-down bearings based on the bearing correlations, and to train a fault diagnosis model for the target bearing based on a bearing fault data set of the target scaled-down bearing; and a diagnosis module 808 is used to analyze the operating data of the target bearing based on the fault diagnosis model, so as to determine a fault detection result of the target bearing.

[0115] In one embodiment, the analysis module is also used to: input the first bearing structure parameter into the bearing dynamics model to obtain a first simulation signal, and convert the first simulation signal to obtain a first frequency domain signal; input the second bearing structure parameter into the bearing dynamics model to obtain a second simulation signal, and convert the second simulation signal to obtain a second frequency domain signal; perform correlation analysis on the first frequency domain signal and the second frequency domain signal to obtain the bearing correlation between the target bearing and the scaled bearing.

[0116] In one embodiment, the first frequency domain signal includes a first sub-frequency domain signal of the target bearing in a normal state, and a second sub-frequency domain signal of the target bearing in a fault state; the second frequency domain signal includes a third sub-frequency domain signal of the scaled bearing in a normal state, and a fourth sub-frequency domain signal of the scaled bearing in a fault state; the analysis module is further used to: determine the first autopower spectral density corresponding to each first frequency band in the first sub-frequency domain signal, and the second autopower spectral density corresponding to each first frequency band in the third sub-frequency domain signal, and determine the first autopower spectral density of each first frequency band in the normal state based on the first sub-frequency domain signal and the third sub-frequency domain signal. The first mutual power spectral density corresponding to the first frequency band is determined based on the first autopower spectral density, the second autopower spectral density and the first mutual power spectral density corresponding to each of the first frequency bands; the first sub-correlation of the target bearing and the scaled bearing in a normal state in each of the first frequency bands is determined based on the second sub-frequency domain signal and the fourth sub-frequency domain signal; the bearing correlation between the target bearing and the scaled bearing is obtained based on the first sub-correlation corresponding to each of the first frequency bands and the second sub-correlation corresponding to each of the second frequency bands.

[0117] In one embodiment, the processing module is also used to: for each of the scaled-down bearings, count the first comparison result between the first sub-correlation corresponding to each of the first frequency bands and the first set threshold, and the second comparison result between the second sub-correlation corresponding to each of the second frequency bands and the second set threshold; when the first comparison result corresponding to the first frequency band indicates that the first sub-correlation is greater than the first set threshold, and the second comparison result corresponding to the second frequency band indicates that the second sub-correlation is greater than the second set threshold, determine the first frequency band as the first target frequency band, and determine the second frequency band as the second target frequency band; when the number of the first target frequency bands exceeds the first number threshold, and the number of the second target frequency bands exceeds the second number threshold, determine the scaled-down bearings corresponding to the first target frequency band and the second target frequency band as candidate scaled-down bearings matching the target bearing; and determine a target scaled-down bearing matching the target bearing based on the candidate scaled-down bearings.

[0118] In one embodiment, there are multiple candidate scaled-down bearings; the processing module is further used to: determine the candidate scaled-down bearing with the smallest outer diameter from the multiple candidate scaled-down bearings as the target scaled-down bearing that matches the target bearing.

[0119] In one embodiment, the processing module is also used to: determine, from multiple candidate scaled bearings, the candidate scaled bearing with the largest number of the first target frequency bands and / or the candidate scaled bearing with the largest number of the second target frequency bands as the target scaled bearing that matches the target bearing.

[0120] Each module in the above-mentioned bearing fault detection device based on scaled bearing can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each of the above modules.

[0121] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig. 9As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as the first bearing structural parameters of the target bearing and the second bearing structural parameters of each of a plurality of scaled bearings matching the target bearing. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a bearing fault detection method based on scaled bearings is implemented.

[0122] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0123] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0124] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0125] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0126] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0127] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0128] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0129] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A bearing fault detection method based on a scaled-down bearing, characterized in that: The method comprises: Acquire a first bearing structural parameter of a target bearing and a second bearing structural parameter of each of a plurality of scaled-down bearings matching the target bearing; For each of the scaled-down bearings, based on a bearing dynamics model, a correlation analysis is performed on the first bearing structural parameters and the second bearing structural parameters to obtain a bearing correlation between the target bearing and the scaled-down bearing; Based on the correlation of each bearing, determine a target scaled-down bearing matching the target bearing from the multiple scaled-down bearings, and train a fault diagnosis model for the target bearing based on a bearing fault data set of the target scaled-down bearing; The operating data of the target bearing is analyzed based on the fault diagnosis model to determine a fault detection result of the target bearing.

2. The method according to claim 1, characterized in that The step of performing a correlation analysis on the first bearing structural parameter and the second bearing structural parameter based on the bearing dynamics model to obtain a bearing correlation between the target bearing and the scaled bearing includes: Inputting the first bearing structural parameter into a bearing dynamics model to obtain a first simulation signal, and converting the first simulation signal to obtain a first frequency domain signal; Inputting the second bearing structural parameters into a bearing dynamics model to obtain a second simulation signal, and converting the second simulation signal to obtain a second frequency domain signal; A correlation analysis is performed on the first frequency domain signal and the second frequency domain signal to obtain a bearing correlation between the target bearing and the scaled bearing.

3. The method according to claim 2, characterized in that The first frequency domain signal includes a first sub-frequency domain signal of the target bearing in a normal state, and a second sub-frequency domain signal of the target bearing in a faulty state; the second frequency domain signal includes a third sub-frequency domain signal of the scaled bearing in a normal state, and a fourth sub-frequency domain signal of the scaled bearing in a faulty state; The performing correlation analysis on the first frequency domain signal and the second frequency domain signal to obtain the bearing correlation between the target bearing and the scaled bearing includes: Determine a first auto power spectral density corresponding to each first frequency band in the first sub-frequency domain signal, and a second auto power spectral density corresponding to each first frequency band in the third sub-frequency domain signal, and determine a first cross power spectral density corresponding to each first frequency band in a normal state based on the first sub-frequency domain signal and the third sub-frequency domain signal; Determine, based on the first auto-power spectral density, the second auto-power spectral density, and the first cross-power spectral density corresponding to each of the first frequency bands, a first sub-correlation between the target bearing and the scaled bearing in a normal state in each of the first frequency bands; Determine, based on the second sub-frequency domain signal and the fourth sub-frequency domain signal, a second sub-correlation of the target bearing and the scaled bearing in each second frequency band in a fault state; Based on the first sub-correlations corresponding to the first frequency bands and the second sub-correlations corresponding to the second frequency bands, a bearing correlation between the target bearing and the scaled bearing is obtained.

4. The method according to claim 3, characterized in that: The step of determining a target scaled-down bearing matching the target bearing from the plurality of scaled-down bearings based on the correlations of the bearings comprises: For each of the scaled-down bearings, a first comparison result between a first sub-correlation corresponding to each of the first frequency bands and a first set threshold value, and a second comparison result between a second sub-correlation corresponding to each of the second frequency bands and a second set threshold value are counted; When a first comparison result corresponding to the first frequency band indicates that a first sub-correlation is greater than a first set threshold, and a second comparison result corresponding to the second frequency band indicates that a second sub-correlation is greater than a second set threshold, the first frequency band is determined as a first target frequency band, and the second frequency band is determined as a second target frequency band; When the number of the first target frequency bands exceeds a first number threshold and the number of the second target frequency bands exceeds a second number threshold, the scaled bearings corresponding to the first target frequency band and the second target frequency band are determined as candidate scaled bearings matching the target bearing; Based on the candidate scaled-down bearings, a target scaled-down bearing matching the target bearing is determined.

5. The method according to claim 4, characterized in that The number of the candidate scaled bearings is multiple; and determining a target scaled bearing matching the target bearing based on the candidate scaled bearings includes: From a plurality of candidate scaled-down bearings, a candidate scaled-down bearing with the smallest outer diameter is determined as a target scaled-down bearing matching the target bearing.

6. The method according to claim 5, characterized in that The method further comprises: From a plurality of candidate scaled-down bearings, the candidate scaled-down bearing having the largest number of the first target frequency bands and / or the candidate scaled-down bearing having the largest number of the second target frequency bands is determined as the target scaled-down bearing matching the target bearing.

7. A bearing fault detection device based on a scaled-down bearing, characterized in that: The device comprises: An acquisition module, used to acquire a first bearing structural parameter of a target bearing and a second bearing structural parameter of each of a plurality of scaled-down bearings matching the target bearing; An analysis module, for performing a correlation analysis on the first bearing structural parameter and the second bearing structural parameter for each of the scaled-down bearings based on a bearing dynamics model, to obtain a bearing correlation between the target bearing and the scaled-down bearing; A processing module, configured to determine a target scaled-down bearing matching the target bearing from the multiple scaled-down bearings based on the correlation of each bearing, and to train a fault diagnosis model for the target bearing based on a bearing fault data set of the target scaled-down bearing; A diagnosis module is used to analyze the operating data of the target bearing based on the fault diagnosis model to determine the fault detection result of the target bearing.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images