A method, device, terminal and equipment for determining a bearing fault diagnosis model

Abstract

The present invention provides a method, device, terminal, and equipment for determining a bearing fault diagnosis model, designed in the field of mechanical engineering technology. The method includes: obtaining first domain data and second domain data; optimizing a first bearing fault diagnosis model based on the first domain data and the second domain data to obtain a target bearing fault diagnosis model; wherein the first bearing fault diagnosis model is a bearing fault diagnosis model corresponding to the first operating condition, and the target bearing fault diagnosis model is capable of detangling the first domain data and the second domain data to obtain a first fault common feature corresponding to the first domain data and a second fault common feature corresponding to the second domain data, so as to be applicable to bearing fault diagnosis under different operating conditions. This solves the problem in the prior art that the bearing fault diagnosis model cannot be applied to bearing fault diagnosis under different operating conditions due to the significant differences in the distribution of bearing vibration signal data under different operating conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical engineering, and in particular to a method, device, terminal and equipment for determining a bearing fault diagnosis model. Background Art

[0002] Bearings are one of the most important components in rotating machinery, making accurate fault diagnosis crucial. The most common method for bearing fault diagnosis is to analyze the vibration signal data during the bearing's operation to diagnose the bearing's fault condition.

[0003] A bearing fault diagnosis model developed through deep learning can be used to analyze vibration signal data. Current bearing fault diagnosis models only have good diagnostic capabilities for vibration signal data under specific operating conditions. However, the distribution of bearing operating vibration signal data varies significantly under different operating conditions, making bearing fault diagnosis models inapplicable to bearing faults under these conditions. Summary of the Invention

[0004] The embodiments of the present invention provide a method, device, terminal and equipment for determining a bearing fault diagnosis model, which is used to solve the problem in the prior art that the bearing fault diagnosis model cannot be applied to bearing fault diagnosis under different working conditions due to the large difference in the distribution of bearing vibration signal data under different working conditions.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] An embodiment of the present invention provides a method for determining a bearing fault diagnosis model, comprising:

[0007] Acquire first domain data and second domain data, wherein the first domain data is data collected of the bearing under a first working condition, and the second domain data is data collected of the bearing under at least one second working condition;

[0008] Optimizing a first bearing fault diagnosis model according to the first domain data and the second domain data to obtain a target bearing fault diagnosis model;

[0009] Among them, the first bearing fault diagnosis model is the bearing fault diagnosis model corresponding to the first working condition, and the target bearing fault diagnosis model can de-entangle the first domain data and the second domain data to obtain the first fault common characteristics corresponding to the first domain data and the second fault common characteristics corresponding to the second domain data, so as to be suitable for bearing fault diagnosis under different working conditions.

[0010] Furthermore, the de-entanglement of the first domain data and the second domain data to obtain a first common fault feature corresponding to the first domain data and a second common fault feature corresponding to the second domain data includes:

[0011] Deentangle the first domain data and the second domain data according to a first loss function to obtain deentangled first domain data and deentangled second domain data;

[0012] Conditional constraints are performed on the deentangled first domain data and the deentangled second domain data according to the second loss function to obtain the first operating condition characteristic features and the first common fault features corresponding to the first domain data, as well as the second operating condition characteristic features and the second common fault features corresponding to the second domain data.

[0013] Furthermore, the first loss function is:

[0014]

[0015] is the transpose of the matrix composed of the common features of the first faults; A matrix composed of the first operating condition characteristic features; is the transpose of the matrix composed of the common characteristics of the second fault; is a matrix composed of the second operating condition characteristic features; I(X; Y) represents the mutual information calculation, and:

[0016]

[0017] Furthermore, the second loss function is:

[0018]

[0019] Among them, N s is the number of samples of the first domain data; N t is the number of samples of the second domain data; d i is the domain label of sample i, and its value is 0 or 1.

[0020] Furthermore, the method further comprises:

[0021] Reconstructing the first domain data to obtain third domain data, and / or reconstructing the second domain data to obtain fourth domain data;

[0022] The target bearing fault diagnosis model is verified based on the first domain data and the third domain data, and / or the target bearing fault diagnosis model is verified based on the second domain data and the fourth domain data.

[0023] Furthermore, reconstructing the first domain data to obtain the third domain data includes:

[0024] Recombining the first operating condition characteristic feature and the second fault common feature to obtain third domain data;

[0025] The reconstructing the second domain data to obtain the fourth domain data includes:

[0026] The second operating condition characteristic feature and the first common fault feature are recombined to obtain fourth domain data.

[0027] Furthermore, verifying the target bearing fault diagnosis model based on the first domain data and the third domain data includes:

[0028] Comparing the third domain data with the first domain data, and verifying the target bearing fault diagnosis model based on the similarity between the third domain data and the first domain data;

[0029] The verifying the target bearing fault diagnosis model according to the second domain data and the fourth domain data includes:

[0030] The fourth domain data is compared with the second domain data, and the target bearing fault diagnosis model is verified according to the similarity between the fourth domain data and the second domain data.

[0031] An embodiment of the present invention further provides a device for determining a bearing fault diagnosis model, comprising:

[0032] an acquisition module, configured to acquire first domain data and second domain data, wherein the first domain data is data collected from a bearing under a first operating condition, and the second domain data is data collected from the bearing under at least one second operating condition;

[0033] a determination module, configured to optimize a first bearing fault diagnosis model according to the first domain data and the second domain data to obtain a target bearing fault diagnosis model;

[0034] Among them, the first bearing fault diagnosis model is the bearing fault diagnosis model corresponding to the first working condition, and the target bearing fault diagnosis model can de-entangle the first domain data and the second domain data to obtain the first fault common characteristics corresponding to the first domain data and the second fault common characteristics corresponding to the second domain data, so as to be suitable for bearing fault diagnosis under different working conditions.

[0035] An embodiment of the present invention further provides a terminal for determining a bearing fault diagnosis model, comprising: a transceiver and a processor;

[0036] The transceiver is used to obtain first domain data and second domain data, wherein the first domain data is data collected from the bearing under a first working condition, and the second domain data is data collected from the bearing under at least one second working condition;

[0037] The processor is configured to optimize a first bearing fault diagnosis model according to the first domain data and the second domain data to obtain a target bearing fault diagnosis model;

[0038] Among them, the first bearing fault diagnosis model is the bearing fault diagnosis model corresponding to the first working condition, and the target bearing fault diagnosis model can de-entangle the first domain data and the second domain data to obtain the first fault common characteristics corresponding to the first domain data and the second fault common characteristics corresponding to the second domain data, so as to be suitable for bearing fault diagnosis under different working conditions.

[0039] An embodiment of the present invention also provides a device for determining a bearing fault diagnosis model, comprising: a transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; when the processor executes the program or instruction, the method for determining the bearing fault diagnosis model as described above is implemented.

[0040] The beneficial effects of the present invention are:

[0041] The method for determining a bearing fault diagnosis model in an embodiment of the present invention optimizes an existing bearing fault diagnosis model to obtain a target bearing fault diagnosis model capable of disentangling domain data. This model is therefore applicable to bearing fault diagnosis under different operating conditions, improving the generalization capability of the bearing fault diagnosis model. This overcomes the problem in the prior art where the bearing fault diagnosis model is not applicable to bearing fault diagnosis under different operating conditions due to significant differences in the distribution of bearing vibration signal data under different operating conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A schematic diagram showing the steps of a method for determining a bearing fault diagnosis model according to an embodiment of the present invention;

[0043] Figure 2 A schematic diagram showing the structure of a bearing fault diagnosis model according to an embodiment of the present invention;

[0044] Figure 3 A schematic diagram showing a module of a device for determining a bearing fault diagnosis model according to an embodiment of the present invention;

[0045] Figure 4 A schematic diagram showing the structure of a determination terminal of a bearing fault diagnosis model according to an embodiment of the present invention;

[0046] Figure 5A schematic structural diagram of a device for determining a bearing fault diagnosis model according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0047] To make the technical problems, technical solutions, and advantages to be solved by the present invention more apparent, a detailed description will be given below with reference to the accompanying drawings and specific embodiments. In the following description, specific details such as specific configurations and components are provided solely to facilitate a comprehensive understanding of the embodiments of the present invention. Therefore, it should be clear to those skilled in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. In addition, for the sake of clarity and brevity, descriptions of known functions and configurations have been omitted.

[0048] It should be understood that references throughout this specification to "one embodiment" or "an embodiment" mean that a particular feature, structure, or characteristic associated with the embodiment is included in at least one embodiment of the present invention. Therefore, the appearances of "in one embodiment" or "in an embodiment" throughout this specification do not necessarily refer to the same embodiment. Furthermore, these particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0049] The present invention addresses the problem in the prior art that the bearing fault diagnosis model cannot be applied to bearing fault diagnosis under different working conditions due to the large differences in the distribution of bearing vibration signal data under different working conditions. A method, device, terminal and equipment for determining a bearing fault diagnosis model are provided.

[0050] like Figure 1 As shown, an embodiment of the present invention provides a method for determining a bearing fault diagnosis model, comprising the following steps:

[0051] Step 101: Acquire first domain data and second domain data, wherein the first domain data is data collected of a bearing under a first working condition, and the second domain data is data collected of the bearing under at least one second working condition.

[0052] Optionally, the first domain data is a vibration signal of the bearing under a first working condition, and the second domain data is a vibration signal of the bearing under at least one second working condition.

[0053] In one embodiment of the present invention, the first domain data is implicit data, including operating condition characteristic data and fault characteristic data; the second domain data is implicit data, including operating condition characteristic data and fault characteristic data.

[0054] Step 102: Optimize the first bearing fault diagnosis model based on the first domain data and the second domain data to obtain a target bearing fault diagnosis model;

[0055] Among them, the first bearing fault diagnosis model is the bearing fault diagnosis model corresponding to the first working condition, and the target bearing fault diagnosis model can de-entangle the first domain data and the second domain data to obtain the first fault common characteristics corresponding to the first domain data and the second fault common characteristics corresponding to the second domain data, so as to be suitable for bearing fault diagnosis under different working conditions.

[0056] The target bearing fault diagnosis model can disentangle the first domain data and the second domain data, which can be understood as:

[0057] The target bearing fault diagnosis model has the ability to parse the implicit data in the first domain data and the second domain data to obtain displayed data, that is, to de-entangle the domain data to obtain common fault feature data and working condition characteristic feature data.

[0058] Optionally, the target bearing fault diagnosis model de-entangles the first domain data and the second domain data, so that the common fault feature data in the first domain data and the second domain data exist independently of the operating condition characteristic feature data.

[0059] The method for determining a bearing fault diagnosis model in an embodiment of the present invention optimizes an existing bearing fault diagnosis model to obtain a target bearing fault diagnosis model capable of disentangling domain data. This model is therefore applicable to bearing fault diagnosis under different operating conditions, improving the generalization capability of the bearing fault diagnosis model. This overcomes the problem in the prior art where the bearing fault diagnosis model is not applicable to bearing fault diagnosis under different operating conditions due to significant differences in the distribution of bearing vibration signal data under different operating conditions.

[0060] Optionally, the de-entanglement of the first domain data and the second domain data to obtain a first common fault feature corresponding to the first domain data and a second common fault feature corresponding to the second domain data includes:

[0061] Deentangle the first domain data and the second domain data according to a first loss function to obtain deentangled first domain data and deentangled second domain data;

[0062] Conditional constraints are performed on the deentangled first domain data and the deentangled second domain data according to the second loss function to obtain the first operating condition characteristic features and the first common fault features corresponding to the first domain data, as well as the second operating condition characteristic features and the second common fault features corresponding to the second domain data.

[0063] Optionally, the first operating condition characteristic feature is used to indicate operating condition data of the bearing under the first operating condition, such as speed, pressure, etc.;

[0064] The first common fault feature is used to indicate fault data of the bearing under the first working condition;

[0065] The second operating condition characteristic feature is used to indicate the operating condition data of the bearing under the second operating condition, such as speed, pressure, etc.;

[0066] The second common fault feature is used to indicate fault data of the bearing under the second working condition.

[0067] The solution of the embodiment of the present invention optimizes the existing bearing fault diagnosis model, and the obtained target bearing fault diagnosis model can de-entangle the domain data, so that the common fault characteristics and the operating condition characteristic characteristics are independent of each other, so that the target bearing fault diagnosis model can perform fault diagnosis calculations on bearings under different operating conditions.

[0068] Optionally, the first loss function is:

[0069]

[0070] is the transpose of the matrix composed of the common features of the first faults; A matrix composed of the first operating condition characteristic features; is the transpose of the matrix composed of the common characteristics of the second fault; is a matrix composed of the second operating condition characteristic features; I(X; Y) represents the mutual information calculation, and:

[0071]

[0072] Optionally, the second loss function is:

[0073]

[0074] Among them, N s is the number of samples of the first domain data; N t is the number of samples of the second domain data; d i is the domain label of sample i, and its value is 0 or 1.

[0075] Optionally, the method further includes:

[0076] Reconstructing the first domain data to obtain third domain data, and / or reconstructing the second domain data to obtain fourth domain data;

[0077] The target bearing fault diagnosis model is verified based on the first domain data and the third domain data, and / or the target bearing fault diagnosis model is verified based on the second domain data and the fourth domain data.

[0078] The solution of the embodiment of the present invention verifies the existence of common fault feature data independent of operating condition characteristic feature data through data reconstruction, proves the correctness of the de-entanglement process, and ensures the reliability of the target bearing fault diagnosis model.

[0079] Optionally, reconstructing the first domain data to obtain third domain data includes:

[0080] Recombining the first operating condition characteristic feature and the second fault common feature to obtain third domain data;

[0081] The reconstructing the second domain data to obtain the fourth domain data includes:

[0082] The second operating condition characteristic feature and the first common fault feature are recombined to obtain fourth domain data.

[0083] Optionally, the first operating condition characteristic feature and the second fault common feature are recombined, and the second operating condition characteristic feature and the first fault common feature are recombined through a third loss function.

[0084] Optionally, the third loss function is:

[0085]

[0086] Among them, N s is the number of samples of the first domain data; N t is the number of samples of the second domain data; is a vector composed of the first domain data; is a vector composed of the third domain data; is a vector composed of the second domain data; is a vector composed of the fourth field data;

[0087] and

[0088] l k It is a preset vector, and the preset vector has the same length as the first domain data sample or the preset vector has the same length as the second domain data sample. The preset vector is used to store the influence of each sample point in the vector on the result.

[0089] The solution of the embodiment of the present invention interacts the working condition characteristic features and common fault features of the first domain data and the second domain data to obtain reorganized third domain data and fourth domain data, further verifying the theory that the fault characteristic feature data exists independently of the working condition characteristic feature data.

[0090] Optionally, verifying the target bearing fault diagnosis model based on the first domain data and the third domain data includes:

[0091] Comparing the third domain data with the first domain data, and verifying the target bearing fault diagnosis model based on the similarity between the third domain data and the first domain data;

[0092] The verifying the target bearing fault diagnosis model according to the second domain data and the fourth domain data includes:

[0093] The fourth domain data is compared with the second domain data, and the target bearing fault diagnosis model is verified according to the similarity between the fourth domain data and the second domain data.

[0094] In one embodiment of the present invention, if the similarity between the third domain data and the first domain data is greater than a preset threshold, the target bearing fault diagnosis model passes the verification; and / or,

[0095] If the similarity between the fourth domain data and the second domain data is greater than a preset threshold, the target bearing fault diagnosis model passes the verification.

[0096] The solution of the present invention can verify the correctness of disentanglement by comparing the third domain data obtained by data reconstruction with the first domain data, and / or comparing the fourth domain data obtained by data reconstruction with the second domain data, thereby proving that the common characteristics of faults are independent of the characteristic characteristics of the operating conditions.

[0097] like Figure 2 As shown, the structure of the target bearing fault diagnosis model of the embodiment of the present invention includes:

[0098] Input module, encoding module, disentanglement module, classifier, switching module and verification module;

[0099] Wherein, the input module is used to input the acquired first domain data and second domain data into the encoding module;

[0100] The encoding module is used to parse and process the received first domain data and the second domain data, and abstract them into low-latitude functions;

[0101] The de-entanglement module is used to de-entangle the first domain data and the second domain data to obtain a first operating condition characteristic feature and the first common fault feature corresponding to the first domain data, and a second operating condition characteristic feature and the second common fault feature corresponding to the second domain data;

[0102] The exchange module is used to exchange the first common fault feature in the first domain data with the second common fault feature in the second domain data to perform data reconstruction;

[0103] The verification module is used to compare the third domain data obtained based on data reconstruction with the first domain data, and / or to compare the fourth domain data obtained based on data reconstruction with the second domain data, thereby verifying the reliability of the structure of the target bearing fault diagnosis model.

[0104] like Figure 3 As shown, an embodiment of the present invention further provides a device 300 for determining a bearing fault diagnosis model, comprising:

[0105] An acquisition module 301 is configured to acquire first domain data and second domain data, wherein the first domain data is data collected from a bearing under a first operating condition, and the second domain data is data collected from the bearing under at least one second operating condition;

[0106] A determination module 302 is configured to optimize a first bearing fault diagnosis model based on the first domain data and the second domain data to obtain a target bearing fault diagnosis model;

[0107] Among them, the first bearing fault diagnosis model is the bearing fault diagnosis model corresponding to the first working condition, and the target bearing fault diagnosis model can de-entangle the first domain data and the second domain data to obtain the first fault common characteristics corresponding to the first domain data and the second fault common characteristics corresponding to the second domain data, so as to be suitable for bearing fault diagnosis under different working conditions.

[0108] The device for determining a bearing fault diagnosis model in an embodiment of the present invention optimizes an existing bearing fault diagnosis model to obtain a target bearing fault diagnosis model capable of disentangling domain data. This model is therefore applicable to bearing fault diagnosis under different operating conditions, improving the generalization capability of the bearing fault diagnosis model. This addresses the existing problem of bearing fault diagnosis models being unable to diagnose bearing faults under different operating conditions due to significant differences in the distribution of bearing vibration signal data under different operating conditions.

[0109] Optionally, the determining module is further configured to:

[0110] Deentangle the first domain data and the second domain data according to a first loss function to obtain deentangled first domain data and deentangled second domain data;

[0111] Conditional constraints are performed on the deentangled first domain data and the deentangled second domain data according to the second loss function to obtain the first operating condition characteristic features and the first common fault features corresponding to the first domain data, as well as the second operating condition characteristic features and the second common fault features corresponding to the second domain data.

[0112] Optionally, the device for determining the bearing fault diagnosis model further includes:

[0113] a reconstruction module, configured to reconstruct the first domain data to obtain third domain data, and / or reconstruct the second domain data to obtain fourth domain data;

[0114] A verification module is used to verify the target bearing fault diagnosis model based on the first domain data and the third domain data, and / or to verify the target bearing fault diagnosis model based on the second domain data and the fourth domain data.

[0115] Optionally, the reconstruction module is further configured to:

[0116] Recombining the first operating condition characteristic feature and the second fault common feature to obtain third domain data; and / or,

[0117] The second operating condition characteristic feature and the first common fault feature are recombined to obtain fourth domain data.

[0118] Optionally, the verification module is further configured to:

[0119] Comparing the third domain data with the first domain data, and verifying the target bearing fault diagnosis model based on the similarity between the third domain data and the first domain data;

[0120] The verifying the target bearing fault diagnosis model according to the second domain data and the fourth domain data includes:

[0121] The fourth domain data is compared with the second domain data, and the target bearing fault diagnosis model is verified according to the similarity between the fourth domain data and the second domain data.

[0122] like Figure 4 As shown, a mobile terminal 400 according to an embodiment of the present invention includes a processor 410 and a transceiver 420, wherein:

[0123] The transceiver is used to obtain first domain data and second domain data, wherein the first domain data is data collected from the bearing under a first working condition, and the second domain data is data collected from the bearing under at least one second working condition;

[0124] The processor is configured to optimize a first bearing fault diagnosis model according to the first domain data and the second domain data to obtain a target bearing fault diagnosis model;

[0125] Among them, the first bearing fault diagnosis model is the bearing fault diagnosis model corresponding to the first working condition, and the target bearing fault diagnosis model can de-entangle the first domain data and the second domain data to obtain the first fault common characteristics corresponding to the first domain data and the second fault common characteristics corresponding to the second domain data, so as to be suitable for bearing fault diagnosis under different working conditions.

[0126] Optionally, the processor is further configured to:

[0127] Deentangle the first domain data and the second domain data according to a first loss function to obtain deentangled first domain data and deentangled second domain data;

[0128] Conditional constraints are performed on the deentangled first domain data and the deentangled second domain data according to the second loss function to obtain the first operating condition characteristic features and the first common fault features corresponding to the first domain data, as well as the second operating condition characteristic features and the second common fault features corresponding to the second domain data.

[0129] Optionally, the processor is further configured to:

[0130] Reconstructing the first domain data to obtain third domain data, and / or reconstructing the second domain data to obtain fourth domain data;

[0131] The target bearing fault diagnosis model is verified based on the first domain data and the third domain data, and / or the target bearing fault diagnosis model is verified based on the second domain data and the fourth domain data.

[0132] The processor is further configured to:

[0133] Recombining the first operating condition characteristic feature and the second fault common feature to obtain third domain data; and / or,

[0134] The second operating condition characteristic feature and the first common fault feature are recombined to obtain fourth domain data.

[0135] The processor is further configured to:

[0136] Comparing the third domain data with the first domain data, and verifying the target bearing fault diagnosis model based on the similarity between the third domain data and the first domain data;

[0137] The verifying the target bearing fault diagnosis model according to the second domain data and the fourth domain data includes:

[0138] The fourth domain data is compared with the second domain data, and the target bearing fault diagnosis model is verified according to the similarity between the fourth domain data and the second domain data.

[0139] A mobile terminal according to another embodiment of the present invention, such as Figure 5 As shown, it includes a transceiver 510, a processor 500, a memory 520, and a program or instruction stored in the memory 520 and executable on the processor 500; when the processor 500 executes the program or instruction, the above-mentioned determination method applied to the bearing fault diagnosis model is implemented.

[0140] The transceiver 510 is configured to receive and send data under the control of the processor 500 .

[0141] Among them, Figure 5 In the embodiment, the bus architecture may include any number of interconnected buses and bridges, specifically various circuits of one or more processors represented by processor 500 and memory represented by memory 520, which are linked together. The bus architecture may also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface. The transceiver 510 may be a plurality of components, i.e., a transmitter and a receiver, providing a unit for communicating with various other devices on a transmission medium. For different user devices, the user interface 530 may also be an interface capable of connecting external or internal devices as required, and the connected devices include but are not limited to a keypad, a display, a speaker, a microphone, a joystick, etc.

[0142] The processor 500 is responsible for managing the bus architecture and general processing, and the memory 520 can store data used by the processor 500 when performing operations.

[0143] A readable storage medium according to an embodiment of the present invention stores a program or instruction thereon. When the program or instruction is executed by a processor, the steps in the method for determining the bearing fault diagnosis model as described above are implemented, and the same technical effect can be achieved. To avoid repetition, they will not be described here.

[0144] The processor is the processor in the device for determining the bearing fault diagnosis model described in the above embodiment. The readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0145] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary personnel in this technical field, several improvements and modifications can be made without departing from the principles described in the present invention. These improvements and modifications are also within the scope of protection of the present invention.

Claims

1. A method for determining a bearing fault diagnosis model, characterized in that: include: Acquire first domain data and second domain data, wherein the first domain data is data collected of the bearing under a first working condition, and the second domain data is data collected of the bearing under at least one second working condition; Optimizing a first bearing fault diagnosis model according to the first domain data and the second domain data to obtain a target bearing fault diagnosis model; The first bearing fault diagnosis model is a bearing fault diagnosis model corresponding to the first working condition, and the target bearing fault diagnosis model can de-entangle the first domain data and the second domain data to obtain a first fault common feature corresponding to the first domain data and a second fault common feature corresponding to the second domain data, so as to be applicable to bearing fault diagnosis under different working conditions; The step of detangling the first domain data and the second domain data to obtain a first common fault feature corresponding to the first domain data and a second common fault feature corresponding to the second domain data includes: detangling the first domain data and the second domain data according to a first loss function to obtain the detangled first domain data and the detangled second domain data; and performing conditional constraints on the detangled first domain data and the detangled second domain data according to a second loss function to obtain a first operating condition characteristic feature and the first common fault feature corresponding to the first domain data, and a second operating condition characteristic feature and the second common fault feature corresponding to the second domain data. The first loss function is: is the transpose of the matrix composed of the common features of the first faults; A matrix composed of the first operating condition characteristic features; is the transpose of the matrix composed of the common characteristics of the second fault; is a matrix composed of the second operating condition characteristic features; I(X; Y) represents the mutual information calculation, and: The second loss function is: Among them, N s is the number of samples of the first domain data; N t is the number of samples of the second domain data; d i is the domain label of sample i, and its value is 0 or 1.

2. The method for determining a bearing fault diagnosis model according to claim 1, characterized in that: The method further comprises: Reconstructing the first domain data to obtain third domain data, and / or reconstructing the second domain data to obtain fourth domain data; The target bearing fault diagnosis model is verified based on the first domain data and the third domain data, and / or the target bearing fault diagnosis model is verified based on the second domain data and the fourth domain data.

3. The method for determining a bearing fault diagnosis model according to claim 2, wherein: The reconstructing the first domain data to obtain the third domain data includes: Recombining the first operating condition characteristic feature and the second fault common feature to obtain third domain data; The reconstructing the second domain data to obtain the fourth domain data includes: The second operating condition characteristic feature and the first common fault feature are recombined to obtain fourth domain data.

4. The method for determining a bearing fault diagnosis model according to claim 2, wherein: The verifying the target bearing fault diagnosis model according to the first domain data and the third domain data includes: Comparing the third domain data with the first domain data, and verifying the target bearing fault diagnosis model based on the similarity between the third domain data and the first domain data; The verifying the target bearing fault diagnosis model according to the second domain data and the fourth domain data includes: The fourth domain data is compared with the second domain data, and the target bearing fault diagnosis model is verified according to the similarity between the fourth domain data and the second domain data.

5. A device for determining a bearing fault diagnosis model, characterized in that: include: an acquisition module, configured to acquire first domain data and second domain data, wherein the first domain data is data collected from a bearing under a first operating condition, and the second domain data is data collected from the bearing under at least one second operating condition; a determination module, configured to optimize a first bearing fault diagnosis model according to the first domain data and the second domain data to obtain a target bearing fault diagnosis model; The first bearing fault diagnosis model is a bearing fault diagnosis model corresponding to the first working condition, and the target bearing fault diagnosis model can de-entangle the first domain data and the second domain data to obtain a first fault common feature corresponding to the first domain data and a second fault common feature corresponding to the second domain data, so as to be applicable to bearing fault diagnosis under different working conditions; The determination module is further configured to: de-entangle the first domain data and the second domain data according to a first loss function to obtain the de-entangled first domain data and the de-entangled second domain data; and conditionally constrain the de-entangled first domain data and the de-entangled second domain data according to a second loss function to obtain the first operating condition characteristic feature and the first common fault feature corresponding to the first domain data, and the second operating condition characteristic feature and the second common fault feature corresponding to the second domain data; The first loss function is: is the transpose of the matrix composed of the common features of the first faults; A matrix composed of the first operating condition characteristic features; is the transpose of the matrix composed of the common characteristics of the second fault; is a matrix composed of the second operating condition characteristic features; I(X; Y) represents the mutual information calculation, and: The second loss function is: Among them, N s is the number of samples of the first domain data; N t is the number of samples of the second domain data; d i is the domain label of sample i, and its value is 0 or 1.

6. A determination terminal of a bearing fault diagnosis model, characterized in that: include: transceivers and processors; The transceiver is used to obtain first domain data and second domain data, wherein the first domain data is data collected from the bearing under a first working condition, and the second domain data is data collected from the bearing under at least one second working condition; The processor is configured to optimize a first bearing fault diagnosis model according to the first domain data and the second domain data to obtain a target bearing fault diagnosis model; The first bearing fault diagnosis model is a bearing fault diagnosis model corresponding to the first working condition, and the target bearing fault diagnosis model can de-entangle the first domain data and the second domain data to obtain a first fault common feature corresponding to the first domain data and a second fault common feature corresponding to the second domain data, so as to be applicable to bearing fault diagnosis under different working conditions; The processor is further configured to: de-entangle the first domain data and the second domain data according to a first loss function to obtain the de-entangled first domain data and the de-entangled second domain data; and conditionally constrain the de-entangled first domain data and the de-entangled second domain data according to a second loss function to obtain the first operating condition characteristic feature and the first fault common feature corresponding to the first domain data, and the second operating condition characteristic feature and the second fault common feature corresponding to the second domain data; The first loss function is: is the transpose of the matrix composed of the common features of the first faults; A matrix composed of the first operating condition characteristic features; is the transpose of the matrix composed of the common characteristics of the second fault; is a matrix composed of the second operating condition characteristic features; I(X; Y) represents the mutual information calculation, and: The second loss function is: Among them, N s is the number of samples of the first domain data; N t is the number of samples of the second domain data; d i is the domain label of sample i, and its value is 0 or 1.

7. A device for determining a bearing fault diagnosis model, comprising: A transceiver, a processor, a memory, and a program or instruction stored in the memory and executable on the processor; characterized in that when the processor executes the program or instruction, the method for determining the bearing fault diagnosis model according to any one of claims 1 to 4 is implemented.

Images