Rolling bearing fault diagnosis method and system

Abstract

The invention discloses a rolling bearing fault diagnosis method and system, and belongs to the technical field of fault diagnosis. The method comprises the following steps: acquiring a to-be-diagnosed rolling bearing vibration signal; inputting the to-be-diagnosed rolling bearing vibration signal into a pre-trained fault diagnosis model, and outputting a rolling bearing vibration fault diagnosis result; wherein the step of training the fault diagnosis model by using the improved sparrow search algorithm comprises the following steps of: obtaining an initial flight position of each sparrow in a sparrow population, and calculating an initial fitness value of each sparrow; in a preset search space, the flight positions of the sparrows are updated according to the Levy flight step length, and new fitness values of the sparrows after the flight positions are updated are calculated; and comparing the initial fitness value with the new fitness value, and taking the flight position corresponding to the sparrow with the optimal fitness value as a parameter of a fault diagnosis model to obtain a trained fault diagnosis model. The optimal network structure parameter can be accurately and quickly found, and the fault diagnosis performance is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a rolling bearing fault diagnosis method and system. Background Art

[0002] Rolling bearings, one of the most common components in mechanical equipment, are widely used in industries such as aviation, shipbuilding, and automotive. However, because mechanical equipment often operates in harsh environments and bears heavy loads for extended periods, the probability of rolling bearing failure increases significantly. Rolling bearing failure not only reduces the efficiency of the entire machine but can also result in significant economic losses and even casualties. Therefore, timely and effective diagnosis of the operating status of rolling bearings is of vital engineering importance in industrial production.

[0003] With the continuous development of deep learning technology, it is now widely used in rolling bearing fault diagnosis. Among deep learning models, convolutional neural networks (CNNs) and Transformer networks offer new solutions for rolling bearing fault diagnosis due to their powerful feature extraction capabilities and ability to process sequential data. However, the selection of network structure parameters directly impacts the performance of the diagnostic model. Therefore, strategic improvements to swarm intelligence optimization algorithms are crucial for accurately and rapidly finding optimal network structure parameters and improving fault diagnosis performance. Summary of the Invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a rolling bearing fault diagnosis method and system that can accurately and quickly find the optimal network structure parameters and improve the performance of fault diagnosis.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0006] In one aspect, the present invention provides a rolling bearing fault diagnosis method, comprising:

[0007] Acquire vibration signals of rolling bearings to be diagnosed;

[0008] Inputting the rolling bearing vibration signal to be diagnosed into a pre-trained fault diagnosis model, and outputting a rolling bearing vibration fault diagnosis result;

[0009] The fault diagnosis model is trained using an improved sparrow search algorithm, including:

[0010] Obtain the initial flight position of each sparrow in the sparrow population and calculate the initial fitness value of each sparrow;

[0011] The sparrow with the best initial fitness value is used as the discoverer sparrow for global search, and the remaining sparrows are used as joiner sparrows to follow the discoverer sparrow for local search. Some sparrows are randomly selected as sentinel sparrows to avoid predators.

[0012] In the preset search space, the flight positions of the discoverer sparrow, joiner sparrow, and guard sparrow are updated according to the Levy flight step length, and the new fitness value of each sparrow after the flight position update is calculated;

[0013] The initial fitness value is compared with the new fitness value, and the flight position corresponding to the sparrow with the best fitness value is used as the parameter of the fault diagnosis model to obtain a trained fault diagnosis model.

[0014] Optionally, before inputting the vibration signal of the rolling bearing to be diagnosed into a pre-trained fault diagnosis model, the method further includes:

[0015] Decomposing the vibration signal of the rolling bearing to be diagnosed by using a variational mode decomposition algorithm to obtain an IMF component;

[0016] The time-frequency domain features of the IMF components are extracted to obtain a preprocessed rolling bearing vibration signal.

[0017] Optionally, obtaining the initial flight position of each sparrow in the sparrow population includes:

[0018] ;

[0019] ;

[0020] in, 、 They represent the chaotic variables generated by the n+1th Tent chaotic mapping and the nth Tent chaotic mapping respectively; represents the control parameters of the Tent chaotic map; represents the number of particles in the chaotic variable; Represents a random number; represents the initial flight position of each sparrow in the sparrow population; 、 They respectively represent the lower and upper bounds of the sparrow population.

[0021] Optionally, the sparrow's fitness value is calculated as:

[0022] ;

[0023] in, represents the fitness value of the i-th sparrow; Represents the objective function value of the i-th sparrow.

[0024] Optionally, the calculation formula of the Levy flight step length is:

[0025] ;

[0026] in, represents the Levy flight step length; 、 Represents a random number; Parameter representing the adjustment step size.

[0027] Optionally, the flight position update formula of the discoverer sparrow is:

[0028] ;

[0029] in, 、 They represent the flight position of the i'th discoverer sparrow in the j-th dimension at the t+1-th iteration and the flight position of the i'th discoverer sparrow in the j-th dimension at the t-th iteration respectively; represents the random decay factor; Indicates the maximum number of iterations; represents the step size factor; Represents a random number; represents the Levy flight step length; represents the global optimal flight position of the tth iteration; represents the exponential function; Indicates the warning value; Indicates the safety threshold.

[0030] Optionally, the flight position update formula of the joining sparrow is:

[0031] ;

[0032] in, 、 They represent the flight position of the i''th joiner sparrow in the j-th dimension at the t+1-th iteration and the flight position of the i''th joiner sparrow in the j-th dimension at the t-th iteration respectively; Represents a random number; represents the worst flight position at the tth iteration; represents the step size factor; represents the flight position of the discoverer sparrow followed in the t+1th iteration; represents a random vector that controls the search direction; represents the Levy flight step length; Represents the number of sparrows in a sparrow population.

[0033] Optionally, the flight position update formula of the sentinel sparrow is:

[0034] ;

[0035] in, 、 They represent the flight position of the i'''th alert sparrow in the j-th dimension at the t+1-th iteration and the flight position of the i'''th alert sparrow in the j-th dimension at the t-th iteration respectively; represents the global optimal flight position of the tth iteration; Parameter representing the adjustment step size; represents a constant; represents the step size factor; represents the Levy flight step length; Indicates direction control parameters; represents the worst flight position at the tth iteration; Indicates the fitness value of the i'''th alert sparrow; 、 They represent the optimal fitness value and the worst fitness value of the current sparrow respectively.

[0036] In a second aspect, the present invention provides a rolling bearing fault diagnosis system, comprising:

[0037] A signal acquisition module is used to: acquire a vibration signal of a rolling bearing to be diagnosed;

[0038] A fault diagnosis module is used to: input the rolling bearing vibration signal to be diagnosed into a pre-trained fault diagnosis model, and output a rolling bearing vibration fault diagnosis result;

[0039] The model training module is used for: wherein the fault diagnosis model is trained using the improved sparrow search algorithm, including:

[0040] Obtain the initial flight position of each sparrow in the sparrow population and calculate the initial fitness value of each sparrow;

[0041] The sparrow with the best initial fitness value is used as the discoverer sparrow for global search, and the remaining sparrows are used as joiner sparrows to follow the discoverer sparrow for local search. Some sparrows are randomly selected as sentinel sparrows to avoid predators.

[0042] In the preset search space, the flight positions of the discoverer sparrow, joiner sparrow, and guard sparrow are updated according to the Levy flight step length, and the new fitness value of each sparrow after the flight position update is calculated;

[0043] The initial fitness value is compared with the new fitness value, and the flight position corresponding to the sparrow with the best fitness value is used as the parameter of the fault diagnosis model to obtain a trained fault diagnosis model.

[0044] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the method described in the first aspect.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] The present invention uses a flying strategy to enhance the algorithm's global search capability and the ability to escape from local optimal solutions. Based on the improved sparrow search algorithm, the optimal parameters of the fault diagnosis model network structure are quickly found, and the model after parameter optimization can have a better fault extraction feature effect, so that the network iterative training can obtain a higher diagnostic accuracy and a lower function loss value, thereby improving the accuracy of rolling bearing fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 FIG. 1 is a flow chart of a rolling bearing fault diagnosis method according to an embodiment of the present invention;

[0048] Figure 2 FIG2 is a schematic diagram of a training process of a fault diagnosis model according to an embodiment of the present invention;

[0049] Figure 3 FIG2 is a schematic diagram showing an iterative comparison of the optimal fitness values ​​of the algorithm of the present invention and the existing algorithm in one embodiment;

[0050] Figure 4 Shown is a schematic diagram of a confusion matrix of a rolling bearing vibration fault diagnosis result in one embodiment of the present invention. DETAILED DESCRIPTION

[0051] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0052] The term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone, A and B exist simultaneously, or B exists alone. Additionally, the character " / " generally indicates an "or" relationship between the related objects.

[0053] Example 1

[0054] like Figure 1 As shown, this embodiment introduces a rolling bearing fault diagnosis method, which specifically includes the following steps:

[0055] Step 1: Obtain the vibration signal of the rolling bearing to be diagnosed and pre-process the vibration signal of the rolling bearing to be diagnosed, specifically:

[0056] Set the parameters of the variational mode decomposition (VMD) algorithm, including the number of modes and penalty factor;

[0057] Using the VMD algorithm, the vibration signal of the rolling bearing to be diagnosed is decomposed to obtain the intrinsic mode function (IMF) components. The IMF components are the inherent mode function components that represent the characteristics of different frequency scales.

[0058] The time-frequency domain features of the IMF components are extracted to obtain the preprocessed rolling bearing vibration signal, which is then used for fault diagnosis.

[0059] Step 2: Input the pre-processed rolling bearing vibration signal into the pre-trained fault diagnosis model and output the rolling bearing vibration fault diagnosis results, specifically:

[0060] In this embodiment, the fault diagnosis model adopts a Convolutional Neural Networks-Transformer (CNN-Transformer) model. The CNN-Transformer model is trained using an improved Tent Levy Sparrow Search Algorithm (TLSSA). The training steps are as follows:

[0061] Initialize the sparrow population size pop size and the maximum number of iterations , dimension dim, lower bound of sparrow population , the upper boundary of the sparrow population , safety threshold ST, control parameters of Tent chaos map , chaotic variables generated by Tent chaos map , the number of particles in the chaotic variable , and Levy flight function parameters , Levy flight random attenuation factor , Levy flight adjustment step parameters .

[0062] The improved sparrow search algorithm is a swarm intelligence optimization algorithm inspired by the foraging and anti-predation behavior of sparrows. It mainly improves the sparrow search algorithm by combining the Tent chaos map and the Levy flight strategy. Figure 2As shown in the figure, the fault diagnosis model parameters confirmed by the improved sparrow search algorithm are as follows:

[0063] The tent chaos map is used to obtain the initial flight position of each sparrow in the sparrow population. The mathematical expression of the tent chaos map is:

[0064] ;

[0065] The initial flight position of each sparrow in the sparrow population under the Tent chaos map is expressed as:

[0066] ;

[0067] in, 、 They represent the chaotic variables generated by the n+1th Tent chaotic mapping and the chaotic variables generated by the nth Tent chaotic mapping, respectively. ; represents the control parameters of the Tent chaotic map; represents the number of particles in the chaotic variable; express A random number between represents the initial flight position of each sparrow in the sparrow population; 、 They respectively represent the lower and upper bounds of the sparrow population.

[0068] The performance of the improved sparrow search algorithm is evaluated by calculating the initial fitness value of each sparrow. The initial fitness value of each sparrow is expressed as:

[0069] ;

[0070] in, represents the fitness value of the i-th sparrow; Represents the objective function value of the i-th sparrow.

[0071] The sparrow with the best initial fitness value is used as the discoverer sparrow for global search, the remaining sparrows are used as joiner sparrows to follow the discoverer sparrow for local search, and some sparrows are randomly selected as guard sparrows to avoid predators.

[0072] The algorithm begins its iteration process, which is repeated until the maximum number of iterations is met. In each iteration, the Levy flight step size is used to modify the sparrow's position update rule to enhance the algorithm's global search capability. That is, within the preset search space, the flight positions of the discoverer sparrow, the joiner sparrow, and the guard sparrow are updated according to the Levy flight step size.

[0073] The calculation formula of Levy flight step length is:

[0074] ;

[0075] in, represents the Levy flight step length; 、 represents a random number, From the normal distribution Generated random numbers, is from the standard normal distribution Generate a random number, Levy flight function parameters; Parameter representing the adjustment step size.

[0076] In the improved sparrow search algorithm, the position update of each sparrow can be summarized by the following formula:

[0077] ;

[0078] in, Updated flight positions of sparrows; Indicates the current flight position of the sparrow; Indicates the current optimal flight position.

[0079] The flight position update formula of the discoverer sparrow is:

[0080] ;

[0081] in, 、 They represent the flight position of the i'th discoverer sparrow in the j-th dimension at the t+1-th iteration and the flight position of the i'th discoverer sparrow in the j-th dimension at the t-th iteration respectively; represents the random decay factor; Indicates the maximum number of iterations; represents the step size factor; Represents a random number; represents the Levy flight step length; represents the global optimal flight position of the tth iteration; represents the exponential function; Indicates the warning value; Indicates the safety threshold.

[0082] The flight position update formula of the joining sparrow is:

[0083] ;

[0084] in, 、 They represent the flight position of the i''th joiner sparrow in the j-th dimension at the t+1-th iteration and the flight position of the i''th joiner sparrow in the j-th dimension at the t-th iteration respectively; represents the worst flight position at the tth iteration; represents the flight position of the discoverer sparrow followed in the t+1th iteration; The random vector that controls the search direction is a vector with a random value of 1 or -1; Represents the number of sparrows in a sparrow population.

[0085] The flight position update formula of the Vigilant Sparrow is:

[0086] ;

[0087] in, 、 They represent the flight position of the i'''th alert sparrow in the j-th dimension at the t+1-th iteration and the flight position of the i'''th alert sparrow in the j-th dimension at the t-th iteration respectively; represents a constant, which is a minimum constant set to avoid the denominator being zero; represents a random direction control parameter in [-1,1]; Indicates the fitness value of the i'''th alert sparrow; 、 They represent the optimal fitness value and the worst fitness value of the current sparrow respectively.

[0088] Calculate the new fitness value of each sparrow after the flight position is updated. The calculation formula is as above.

[0089] Compare the initial fitness value with the new fitness value. If the new fitness value is better than the initial fitness value, update the best sparrow. If the fitness value has not improved, do not update the best sparrow.

[0090] Check whether the maximum number of iterations has been reached. If so, output the best sparrow, that is, use the flight position corresponding to the sparrow with the best fitness value as the parameter of the fault diagnosis model to obtain a trained fault diagnosis model; otherwise, return to the starting step of the iteration.

[0091] The parameters of the optimized fault diagnosis model are the number of CNN layers, the number of channels per layer, the Transformer attention dimension, the number of encoder layers, and the number of multi-head attention heads.

[0092] In this embodiment, the global search capability of the algorithm and the ability to jump out of the local optimal solution are enhanced through the flight strategy. The optimal parameters of the fault diagnosis model network structure are quickly found based on the improved sparrow search algorithm. The model after parameter optimization can have a better fault extraction feature effect, so that the network iterative training can obtain a higher diagnostic accuracy and a lower function loss value, thereby improving the accuracy of rolling bearing fault diagnosis.

[0093] Example 2

[0094] Based on Example 1, this example introduces an experimental example of a rolling bearing fault diagnosis method:

[0095] The test data comes from the CRUW rolling bearing experiment. The experimental sample information is shown in Table 1.

[0096] Table 1 Experimental data information table

[0097]

[0098] The vibration signal of the rolling bearing to be diagnosed is obtained. In this embodiment, the number of training set samples is 70% of the total number of all samples, the number of validation set samples is 20% of the total number, and the number of test set samples is 10% of the total number.

[0099] The VMD algorithm is used to preprocess the vibration signal of the rolling bearing to be diagnosed.

[0100] Build a CNN-Transformer model.

[0101] Set the parameters required by the TLSSA algorithm, where pop size=10, =50, dim=10, ub=[1,10,32,1,2], lb=[3,32,128,3,2], ST=0.6, =0.5, 、 =10, =0.1, =0.2, =0.5.

[0102] The structural parameters of the CNN-Transformer model are optimized by improving the sparrow search algorithm. The detailed flowchart is as follows: Figure 2 As shown:

[0103] The vibration signal of the rolling bearing to be diagnosed is preprocessed and input into the trained CNN-Transformer model to output the rolling bearing vibration fault diagnosis result.

[0104] To verify the advantages of this embodiment, the Sparrow Search Algorithm (SSA), the Sparrow Search Algorithm (TSSA) with the Tent Chaotic Mapping method, the Sparrow Search Algorithm (LSSA) with the Levy Flight Strategy, and the TLSSA algorithm in this embodiment, which combines the two methods, were compared in iterative training of the optimal fitness value. The comparison results are shown in Figure 2. Figure 3 As shown in Table 2, it can be seen that the TLSSA algorithm of this embodiment has the best iterative effect. In addition, after the VMD algorithm preprocesses the bearing vibration signal data, multiple fault diagnoses are performed using CNN, CNN-Transformer, and SSA-CNN-Transformer diagnostic methods, and compared with the method of this embodiment. The results are shown in Table 2, which shows that the average accuracy of the method of this embodiment is the highest. Figure 4 As shown in the figure, except for the main diagonal, all the values ​​are 0, that is, the number of incorrectly predicted classification samples is 0, which fully demonstrates that the fault diagnosis model under the TLSSA algorithm is accurate in the classification and identification of bearing fault diagnosis.

[0105] Table 2 Average accuracy of fault diagnosis models under various algorithms

[0106]

[0107] Example 3

[0108] Based on embodiment 1 or 2, this embodiment introduces a rolling bearing fault diagnosis system, including:

[0109] A signal acquisition module is used to: acquire a vibration signal of a rolling bearing to be diagnosed;

[0110] A fault diagnosis module is used to: input the rolling bearing vibration signal to be diagnosed into a pre-trained fault diagnosis model, and output a rolling bearing vibration fault diagnosis result;

[0111] The model training module is used for: wherein the fault diagnosis model is trained using the improved sparrow search algorithm, including:

[0112] Obtain the initial flight position of each sparrow in the sparrow population and calculate the initial fitness value of each sparrow;

[0113] The sparrow with the best initial fitness value is used as the discoverer sparrow for global search, and the remaining sparrows are used as joiner sparrows to follow the discoverer sparrow for local search. Some sparrows are randomly selected as sentinel sparrows to avoid predators.

[0114] In the preset search space, the flight positions of the discoverer sparrow, joiner sparrow, and guard sparrow are updated according to the Levy flight step length, and the new fitness value of each sparrow after the flight position update is calculated;

[0115] The initial fitness value is compared with the new fitness value, and the flight position corresponding to the sparrow with the best fitness value is used as the parameter of the fault diagnosis model to obtain a trained fault diagnosis model.

[0116] The specific functional implementation of each of the above modules can be found in the relevant content of the method in Example 1 and will not be elaborated on here.

[0117] Example 4

[0118] This embodiment introduces a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the method described in embodiment 1 or 2 are implemented.

[0119] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0121] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0122] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0123] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which are all protected by the present invention.

Claims

1. A rolling bearing fault diagnosis method, characterized in that: include: Acquire vibration signals of rolling bearings to be diagnosed; Inputting the rolling bearing vibration signal to be diagnosed into a pre-trained fault diagnosis model, and outputting a rolling bearing vibration fault diagnosis result; The fault diagnosis model is trained using an improved sparrow search algorithm, including: Obtain the initial flight position of each sparrow in the sparrow population and calculate the initial fitness value of each sparrow; The sparrow with the best initial fitness value is used as the discoverer sparrow for global search, and the remaining sparrows are used as joiner sparrows to follow the discoverer sparrow for local search. Some sparrows are randomly selected as sentinel sparrows to avoid predators. In the preset search space, the flight positions of the discoverer sparrow, joiner sparrow, and guard sparrow are updated according to the Levy flight step length, and the new fitness value of each sparrow after the flight position update is calculated; The initial fitness value is compared with the new fitness value, and the flight position corresponding to the sparrow with the best fitness value is used as the parameter of the fault diagnosis model to obtain a trained fault diagnosis model.

2. The rolling bearing fault diagnosis method according to claim 1, characterized in that: Before inputting the rolling bearing vibration signal to be diagnosed into the pre-trained fault diagnosis model, the method further includes: Decomposing the vibration signal of the rolling bearing to be diagnosed by using a variational mode decomposition algorithm to obtain an IMF component; The time-frequency domain features of the IMF components are extracted to obtain a preprocessed rolling bearing vibration signal.

3. The rolling bearing fault diagnosis method according to claim 1, characterized in that: The acquisition of the initial flight position of each sparrow in the sparrow population includes: ； ； in, 、 They represent the chaotic variables generated by the n+1th Tent chaotic mapping and the nth Tent chaotic mapping respectively; represents the control parameters of the Tent chaotic map; represents the number of particles in the chaotic variable; Represents a random number; represents the initial flight position of each sparrow in the sparrow population; 、 They respectively represent the lower and upper bounds of the sparrow population.

4. The rolling bearing fault diagnosis method according to claim 1, characterized in that: The calculation formula for the sparrow's fitness value is: ； in, represents the fitness value of the i-th sparrow; Represents the objective function value of the i-th sparrow.

5. The rolling bearing fault diagnosis method according to claim 1, characterized in that: The calculation formula of the Levy flight step length is: ； in, represents the Levy flight step length; 、 Represents a random number; Parameter representing the adjustment step size.

6. The rolling bearing fault diagnosis method according to claim 1, characterized in that: The flight position update formula of the discoverer sparrow is: ； in, 、 They represent the flight position of the i'th discoverer sparrow in the j-th dimension at the t+1-th iteration and the flight position of the i'th discoverer sparrow in the j-th dimension at the t-th iteration respectively; represents the random decay factor; Indicates the maximum number of iterations; represents the step size factor; Represents a random number; represents the Levy flight step length; represents the global optimal flight position of the tth iteration; represents the exponential function; Indicates the warning value; Indicates the safety threshold.

7. The rolling bearing fault diagnosis method according to claim 1, characterized in that: The flight position update formula of the joining sparrow is: ； in, 、 They represent the flight position of the i''th joiner sparrow in the j-th dimension at the t+1-th iteration and the flight position of the i''th joiner sparrow in the j-th dimension at the t-th iteration respectively; Represents a random number; represents the worst flight position at the tth iteration; represents the step size factor; represents the flight position of the discoverer sparrow followed in the t+1th iteration; represents a random vector that controls the search direction; represents the Levy flight step length; Represents the number of sparrows in a sparrow population.

8. The rolling bearing fault diagnosis method according to claim 1, characterized in that: The flight position update formula of the sentinel sparrow is: ； in, 、 They represent the flight position of the i'''th alert sparrow in the j-th dimension at the t+1-th iteration and the flight position of the i'''th alert sparrow in the j-th dimension at the t-th iteration respectively; represents the global optimal flight position of the tth iteration; Parameter representing the adjustment step size; represents a constant; represents the step size factor; represents the Levy flight step length; Indicates direction control parameters; represents the worst flight position at the tth iteration; Indicates the fitness value of the i'''th alert sparrow; 、 They represent the optimal fitness value and the worst fitness value of the current sparrow respectively.

9. A rolling bearing fault diagnosis system, characterized in that: include: A signal acquisition module is used to: acquire a vibration signal of a rolling bearing to be diagnosed; A fault diagnosis module is used to: input the rolling bearing vibration signal to be diagnosed into a pre-trained fault diagnosis model, and output a rolling bearing vibration fault diagnosis result; The model training module is used for: wherein the fault diagnosis model is trained using the improved sparrow search algorithm, including: Obtain the initial flight position of each sparrow in the sparrow population and calculate the initial fitness value of each sparrow; The sparrow with the best initial fitness value is used as the discoverer sparrow for global search, and the remaining sparrows are used as joiner sparrows to follow the discoverer sparrow for local search. Some sparrows are randomly selected as sentinel sparrows to avoid predators. In the preset search space, the flight positions of the discoverer sparrow, joiner sparrow, and guard sparrow are updated according to the Levy flight step length, and the new fitness value of each sparrow after the flight position update is calculated; The initial fitness value is compared with the new fitness value, and the flight position corresponding to the sparrow with the best fitness value is used as the parameter of the fault diagnosis model to obtain a trained fault diagnosis model.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images