Gear fault intelligent diagnosis method and system, storage medium and electronic equipment

Abstract

The invention relates to a gear fault intelligent diagnosis method and system, a storage medium and electronic equipment. The method comprises the following steps: acquiring a gear vibration signal; decomposing the gear vibration signal by adopting a variational mode decomposition method to obtain a plurality of intrinsic mode functions; performing calculation and feature extraction on each intrinsic mode function to obtain an energy entropy feature vector; combining the energy entropy feature vector with a signal label to obtain a feature-label pair; the feature-tag pair is input into the bidirectional long-short-term memory network model for training, and a diagnosis result of the gear fault is obtained; and outputting the diagnosis result of the gear fault. Based on the variational mode decomposition, the energy entropy and the bidirectional long and short time memory network, the method can achieve the efficient feature extraction and accurate fault diagnosis of the gear vibration signal, and remarkably improves the diagnosis precision and reliability.

Description

Technical Field

[0001] The present invention relates to the technical field of gear fault diagnosis, and more specifically, to an intelligent gear fault diagnosis method, system, storage medium, and electronic device. Background Art

[0002] In a mechanical transmission system, gears, as key transmission components, are widely used in various industrial equipment, such as wind turbines, automotive gearboxes, aero-engines, etc. However, due to the influence of load, speed, vibration, and environmental factors during long-term operation, gears are prone to failures such as wear, fatigue cracks, and tooth breakage. If these failures are not detected and eliminated in time, it may lead to equipment shutdown, reduced production efficiency, and even serious safety accidents. Therefore, gear fault diagnosis technology is becoming increasingly important in the industrial field. Traditional gear fault diagnosis methods mainly rely on vibration signal analysis, and judge the operating state of gears by observing the time domain, frequency domain, or time-frequency domain. However, traditional methods have certain limitations in dealing with non-linear and non-stationary signals under complex working conditions, and the diagnostic accuracy is insufficient. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an intelligent gear fault diagnosis method, system, storage medium, and electronic device for the problems existing in the prior art.

[0004] The technical solution adopted by the present invention to solve its technical problems is: construct an intelligent gear fault diagnosis method, including the following steps:

[0005] Obtain the gear vibration signal;

[0006] Decompose the gear vibration signal by using the variational mode decomposition method to obtain a number of intrinsic mode functions;

[0007] Calculate and extract features for each of the intrinsic mode functions to obtain an energy entropy feature vector;

[0008] Combine the energy entropy feature vector with the signal label to obtain a feature-label pair;

[0009] Input the feature-label pair into a bidirectional long short-term memory network model for training to obtain a diagnosis result of gear faults;

[0010] Output the diagnosis result of gear faults.

[0011] In the intelligent gear fault diagnosis method of the present invention, the step of decomposing the gear vibration signal by using the variational mode decomposition method to obtain a number of intrinsic mode functions includes:

[0012] Construct a variational constraint problem based on the gear vibration signal;

[0013] Transform the variational constrained problem into a variational unconstrained problem;

[0014] Perform iterative update operations to obtain the spectra of several intrinsic mode functions;

[0015] Transform the spectra of the several intrinsic mode functions to obtain the several intrinsic mode functions.

[0016] In the gear fault intelligent diagnosis method of the present invention, the construction of the variational constrained problem based on the gear vibration signal includes:

[0017] Decompose the gear vibration signal to obtain multiple components;

[0018] Construct a constrained variational model based on the constraint conditions to complete the construction of the variational constrained problem.

[0019] In the gear fault intelligent diagnosis method of the present invention, the constraint condition is: the sum of the multiple components is equal to the gear vibration signal.

[0020] In the gear fault intelligent diagnosis method of the present invention, the transformation of the variational constrained problem into a variational unconstrained problem includes:

[0021] Use the Lagrange multiplier and the quadratic penalty factor to transform the variational constrained problem to obtain the variational unconstrained problem.

[0022] In the gear fault intelligent diagnosis method of the present invention, the performance of iterative update operations to obtain the spectra of several intrinsic mode functions includes:

[0023] Use the multiplicative operator alternating direction method to perform iterative update operations to obtain the spectra of the several intrinsic mode functions;

[0024] The transformation of the spectra of the several intrinsic mode functions to obtain the several intrinsic mode functions includes:

[0025] Through the inverse Fourier transform, transform the spectra of the several intrinsic mode functions from the frequency domain to the time domain to obtain the several intrinsic mode functions.

[0026] In the gear fault intelligent diagnosis method of the present invention, the calculation and feature extraction of each intrinsic mode function to obtain the energy entropy feature vector includes:

[0027] Calculate the energy of each intrinsic mode function;

[0028] Calculate the energy proportion of each intrinsic mode function;

[0029] According to the Shannon entropy principle, calculate the energy entropy of each of the intrinsic mode functions;

[0030] Construct the energy entropy feature vector based on the energy entropy.

[0031] The present invention also provides an intelligent gear fault diagnosis system, including:

[0032] A signal acquisition unit for acquiring gear vibration signals;

[0033] A signal decomposition unit for decomposing the gear vibration signals by using the variational mode decomposition method to obtain a plurality of intrinsic mode functions;

[0034] A feature extraction unit for calculating and extracting features from each of the intrinsic mode functions to obtain an energy entropy feature vector;

[0035] A label combination unit for combining the energy entropy feature vector with a signal label to obtain a feature-label pair;

[0036] A fault diagnosis unit for inputting the feature-label pair into a bidirectional long short-term memory network model for training to obtain a diagnosis result of gear faults;

[0037] A diagnosis result output unit for outputting the diagnosis result of the gear faults.

[0038] The present invention also provides a storage medium storing a computer program, which is adapted to be loaded by a processor to execute the steps of the above-mentioned intelligent gear fault diagnosis method.

[0039] The present invention also provides an electronic device including a memory and a processor, where the memory stores a computer program, and the processor executes the steps of the above-mentioned intelligent gear fault diagnosis method by calling the computer program stored in the memory.

[0040] Implementing the intelligent gear fault diagnosis method, system, storage medium and electronic device of the present invention has the following beneficial effects: including the following steps: acquiring gear vibration signals; decomposing the gear vibration signals by using the variational mode decomposition method to obtain a plurality of intrinsic mode functions; calculating and extracting features from each intrinsic mode function to obtain an energy entropy feature vector; combining the energy entropy feature vector with a signal label to obtain a feature-label pair; inputting the feature-label pair into a bidirectional long short-term memory network model for training to obtain a diagnosis result of gear faults; outputting the diagnosis result of gear faults. Based on variational mode decomposition, energy entropy and bidirectional long short-term memory network, the present invention can achieve efficient feature extraction and accurate fault diagnosis of gear vibration signals, and significantly improve the diagnosis accuracy and reliability. Description of the Drawings

[0041] The present invention will be further described below in conjunction with the accompanying drawings and embodiments. In the accompanying drawings:

[0042] Figure 1 is a schematic flow chart of the gear fault intelligent diagnosis method provided by the present invention;

[0043] Figure 2 is a gear fault diagnosis flow chart of VMD-EE-BiLSTM provided by the present invention;

[0044] Figure 3 is a decomposition result diagram of the vibration signal of a broken tooth fault after VMD provided by the present invention;

[0045] Figure 4 is the training result of the BiLSTM model provided by the present invention;

[0046] Figure 5 is a gearbox fault diagnosis result diagram of VMD-EE-BiLSTM provided by the present invention;

[0047] Figure 6 is a logic block diagram of the gear fault intelligent diagnosis system provided by the present invention. Specific embodiments

[0048] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0049] Compared with the traditional gear vibration signal processing methods that mainly rely on time-frequency analysis and empirical judgment, the present invention introduces the Variational Mode Decomposition (VMD) technology to decompose the gear vibration signal and reduce the complexity of the signal. The energy entropy (EE) index is combined to extract the energy characteristics of the Intrinsic Mode Function (IMF), and an Energy Entropy Feature Vector (EEFV) is constructed to characterize the signal energy distribution. The Bi-directional Long-short Term Memory Network (BiLSTM) is used to capture the dependency relationship of the information before and after the EEFV to achieve accurate diagnosis of gear faults.

[0050] Refer to Figure 1 , Figure 1This is a flowchart of a preferred embodiment of the gear fault intelligent diagnosis method provided by the present invention.

[0051] Specifically, as Figure 1 shown, the gear fault intelligent diagnosis method includes the following steps:

[0052] Step S101: Obtain the gear vibration signal.

[0053] Step S102: Decompose the gear vibration signal by using the variational mode decomposition method to obtain a number of intrinsic mode functions.

[0054] Optionally, in the embodiment of the present invention, decomposing the gear vibration signal by using the variational mode decomposition method to obtain a number of intrinsic mode functions includes: constructing a variational constraint problem based on the gear vibration signal; transforming the variational constraint problem into a variational unconstrained problem; performing iterative update operations to obtain the spectra of a number of intrinsic mode functions; and transforming the spectra of a number of intrinsic mode functions to obtain a number of intrinsic mode functions.

[0055] Among them, constructing a variational constraint problem based on the gear vibration signal includes: decomposing the gear vibration signal to obtain multiple components; constructing a constrained variational model based on the constraint conditions to complete the construction of the variational constraint problem. The constraint condition is that the sum of the multiple components is equal to the gear vibration signal. Transforming the variational constraint problem into a variational unconstrained problem includes: using the Lagrange multiplier and the quadratic penalty factor to transform the variational constraint problem to obtain a variational unconstrained problem. Performing iterative update operations to obtain the spectra of a number of intrinsic mode functions includes: using the multiplicative operator alternating direction method to perform iterative update operations to obtain the spectra of a number of intrinsic mode functions; and transforming the spectra of a number of intrinsic mode functions to obtain a number of intrinsic mode functions includes: through the inverse Fourier transform, transforming the spectra of a number of intrinsic mode functions from the frequency domain to the time domain to obtain a number of intrinsic mode functions.

[0056] Specifically, in the embodiment of the present invention, the VMD method is used to decompose the gear vibration signal (denoted as V(t)) to obtain a number of intrinsic mode functions IMF k , where k = 1, 2, 3, …, K, and K is the total number of IMFs. As Figure 2 shown, the decomposition of the gear vibration signal can be achieved through the following steps:

[0057] Sub-step A1: Construct a variational constraint problem. Assume that the original signal V(t) is decomposed into K components u, ensuring that the decomposition sequence is u with a center frequency and a finite bandwidth, and at the same time, the sum of the estimated bandwidths of each u is the smallest. The constraint condition is that the sum of all u is equal to the original signal. Then the VMD constrained variational model is as follows:

[0058]

[0059] In equation (1), u k ={u1, u2, ..., u K} represents each intrinsic mode function, i.e., IMF k ; ω k ={ω1, ω2, ..., ω K} represents the central frequency of each intrinsic mode component; represents taking the partial derivative; δ(t) represents the Dirac distribution function; * represents the convolution operation; V(t) is the gear vibration signal (i.e., the original signal). Among them, the formula in the first line of equation (1) represents a variational constraint problem, and "s.t." represents the constraint conditions of the problem, that is, the sum of all u is equal to the original signal.

[0060] Sub-step 2: By using the Lagrange multiplier λ and the quadratic penalty factor α, the variational constraint problem is transformed into a variational unconstrained problem, and the specific formula is as follows:

[0061]

[0062] In equation (2), L({u k}, {ω k}, λ) is the augmented Lagrangian formula.

[0063] Sub-step 3: By using the alternating direction multiplier, optimize and solve the central frequencies of each component and the response, and the specific formula is as follows:

[0064]

[0065] In equations (3) and (4), and are the Fourier transform forms of V(t), u(t), and λ(t) respectively; n is the number of iterations, ω represents the frequency, and the Fourier transform transforms the signal from the time domain to the frequency domain, so the independent variable of the signal changes from time "t" to frequency "ω".

[0066] When u reaches the given solution discrimination accuracy ε, stop the iteration, and the specific conditions are as follows:

[0067]

[0068] Sub-step 4: Through the inverse Fourier transform, transform the obtained from the frequency domain to the time domain u k (t), and u k (t) corresponds to the aforementioned IMF k .

[0069] Step S103: Calculate and extract features for each intrinsic mode function to obtain the energy entropy feature vector.

[0070] Optionally, in the embodiments of the present invention, calculating and feature extracting each intrinsic mode function to obtain an energy entropy feature vector includes: calculating the energy of each intrinsic mode function; calculating the energy proportion of each intrinsic mode function; calculating the energy entropy of each intrinsic mode function according to the Shannon entropy principle; constructing an energy entropy feature vector based on the energy entropy.

[0071] Specifically, calculate the EE of each IMF respectively, and construct an energy entropy feature vector (EEFV) based on the EE. Specifically, it may include the following steps:

[0072] Sub-step B1: Calculate the IMF k energy E(k), and the specific formula is as follows:

[0073]

[0074] In formula (6), u k (l) represents the l-th value in u k , and N is the number of data points.

[0075] Sub-step B2: Calculate the energy proportion p k of the IMF k , and the specific formula is as follows:

[0076]

[0077] In formula (7), is the total energy of all IMFs after a signal is decomposed.

[0078] Sub-step B3: Calculate the energy entropy EE k of the IMF according to the Shannon entropy principle k , and the specific formula is as follows:

[0079] EE k = -p k lgp k (8).

[0080] Sub-step B4: Construct the EEFV, and the specific expression is as follows:

[0081] EEFV = [EE1, EE2, EE3,..., EE K (9).

[0082] Step S104: Combine the energy entropy feature vector with the signal label to obtain a feature-label pair.

[0083] Specifically, the signal label is used to characterize whether the gear vibration signal is a fault signal. For example, the number "0" can be used to represent a normal signal, and the number "1" represents a fault signal. The feature vectors are placed in the first M + 1 columns of the table, and the signal labels are placed in the (M + 2)-th column of the table. For the feature vectors calculated from normal signals, the signal label is "0"; for the feature vectors calculated from fault signals, the label is "1".

[0084] Step S105: Input the feature-label pairs into a bidirectional long short-term memory network model for training to obtain the diagnosis result of gear faults.

[0085] Specifically, as Figure 2 shown, input the feature-label pairs of each signal into the BiLSTM model for training to obtain the diagnosis result of gear faults.

[0086] Step S106: Output the diagnosis result of gear faults.

[0087] Refer to Figure 6 , Figure 6 which is the logic block diagram of the gear fault intelligent diagnosis system provided by the present invention.

[0088] As Figure 6 shown, the gear fault intelligent diagnosis system includes:

[0089] A signal acquisition unit 601, configured to acquire gear vibration signals.

[0090] A signal decomposition unit 602, configured to decompose the gear vibration signals by using the variational mode decomposition method to obtain a plurality of intrinsic mode functions.

[0091] A feature extraction unit 603, configured to calculate and extract features from each intrinsic mode function to obtain an energy entropy feature vector.

[0092] A label combination unit 604, configured to combine the energy entropy feature vector with the signal label to obtain a feature-label pair.

[0093] A fault diagnosis unit 605, configured to input the feature-label pairs into a bidirectional long short-term memory network model for training to obtain the diagnosis result of gear faults.

[0094] A diagnosis result output unit 606, configured to output the diagnosis result of gear faults.

[0095] Specifically, the specific cooperation operation process among the units in the gear fault intelligent diagnosis system here can specifically refer to the above-mentioned gear fault intelligent diagnosis method, which will not be elaborated here.

[0096] The following takes gear fault data as an example and will be described in detail with reference to the accompanying drawings. It should be emphasized that the following description is merely exemplary and not intended to limit the scope of the present invention and its applications. As Figure 2 shown, the gear fault intelligent diagnosis method provided by the present invention is a gear intelligent diagnosis method based on VMD, PE, and BiLSTM. First, the gear vibration signal is decomposed into several components using VMD. Secondly, the EE of each component is calculated, and the EEFV is constructed based on this as the feature of the signal. Then, the EEFV and the gear state label corresponding to this signal are combined to form a "feature-label" pair. Finally, the "feature-label" pair is input into the BiLSTM model for training to obtain a gear fault diagnosis model.

[0097] Specifically, a certain gear fault data set is used for experimental verification. 500 pieces of data are extracted as the data set. Among them, there are 100 pieces for each of the five states of normal, defective, single tooth, crack, and wear. The sampling frequency is 5120 Hz, and each piece of data has 5120 data points.

[0098] Step 1: Use the VMD method to decompose the gear vibration signal V(t) to obtain several IMFs.

[0099] For this data set, the decomposition number of VMD is set to 8, α is set to 2000, and ε is set to 10 -6 . Here, a gear tooth breakage fault signal is taken as an example for display. The time domain diagrams and frequency domain diagrams of each IMF are as Figure 3 shown.

[0100] Step 2: Calculate the EE of each IMF respectively and construct the EEFV based on the EE.

[0101] The EE calculation results of the above gear tooth breakage fault signal are shown in Table 1, then EEFV = [0.352772, 0.224541, 0.275360, 0.208545, 0.188354, 0.135113, 0.045445, 0.198308].

[0102] Table 1 EE calculation results of each IMF of a certain tooth breakage fault signal

[0103]

[0104] Step 3: Combine the EEFV of the signal with the signal label to form a "feature-label" pair.

[0105] Here, a typical signal is selected from each of the five state signals in the dataset to display the "feature - label" pairs. The specific "feature - label" pairs are shown in Table 2. The label "0" represents the normal state; the label "1" represents the defect fault; the label "2" represents the broken tooth fault; the label "3" represents the crack fault; the label "4" represents the wear fault.

[0106] Table 2 Display of Typical "Feature - Label" Pairs

[0107]

[0108]

[0109] Step 4: Input the "feature - label" pairs of each signal into the BiLSTM model for training to obtain the gear fault intelligent diagnosis model.

[0110] For this dataset, the parameter settings of the BiLSTM model in the present invention are shown in Table 3.

[0111] Table 3 Model Parameter Settings

[0112]

[0113] Among the 500 data in the dataset, 350 data are randomly selected as the training set and 150 data are used as the test set. Ensure that the data volume ratio of the five states in the training set and the test set is 1:1:1:1:1.

[0114] Typical training results are as Figure 4 shown. Typical diagnosis results are as Figure 5 shown.

[0115] Here, 10 tests are carried out according to the above operations, and the average accuracy rate is 95.94%. The accuracy rate of each time is shown in Table 4.

[0116] Based on methods such as Variational Mode Decomposition (VMD), Energy Entropy (EE), and Bi - directional Long - short Term Memory Network (BiLSTM), the present invention proposes a new intelligent fault diagnosis method, aiming to achieve efficient feature extraction of gear vibration signals and accurate fault diagnosis.

[0117] The present invention introduces the VMD technology to decompose the gear vibration signal and reduce the signal complexity. Combines the EE index to extract the energy features of the IMF, and constructs the EEFV to characterize the signal energy distribution. Uses BiLSTM to capture the dependency relationship of the information before and after the EEFV to achieve accurate diagnosis of gear faults.

[0118] In addition, an electronic device according to the present invention includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the gear fault intelligent diagnosis method as described in any one of the above. Specifically, according to an embodiment of the present invention, the process described with reference to the flowchart above can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the method shown in the flowchart. In such an embodiment, when the computer program is downloaded and installed by the electronic device and executed, it executes the above functions defined in the method of the embodiment of the present invention. The electronic device in the present invention can be a terminal such as a notebook, a desktop computer, a tablet computer, a smart phone, etc., or a server.

[0119] In addition, a storage medium according to the present invention has a computer program stored thereon, and when the computer program is executed by a processor, it implements the gear fault intelligent diagnosis method as described in any one of the above. Specifically, it should be noted that the storage medium of the present invention above can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium can include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program codes. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, and the computer-readable signal medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program codes contained on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0120] The above computer-readable medium may be included in the above electronic device; or it may exist separately without being assembled into the electronic device.

[0121] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the various embodiments, reference can be made to each other. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple. For the relevant parts, reference can be made to the description in the method section.

[0122] Those skilled in the art can further realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0123] The steps of the methods or algorithms described in combination with the embodiments disclosed in this article can be directly implemented by hardware, software modules executed by a processor, or a combination of the two. The software modules can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0124] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They do not limit the protection scope of the present invention. Any equivalent changes and modifications made to the scope of the claims of the present invention shall fall within the scope covered by the claims of the present invention.

Claims

1. A gear fault intelligent diagnosis method, characterized in that: The following steps are involved: Get gear vibration signal; Decomposing the gear vibration signal by using a variational mode decomposition method to obtain several eigenmode functions; Calculating and extracting features of each of the intrinsic mode functions to obtain an energy entropy feature vector; Combining the energy entropy feature vector with the signal label to obtain a feature-label pair; Input the feature-label pair into a bidirectional long short-term memory network model for training to obtain a gear fault diagnosis result; The diagnosis result of the gear fault is output.

2. The gear fault intelligent diagnosis method according to claim 1, characterized in that: The gear vibration signal is decomposed by using a variational mode decomposition method to obtain several eigenmode functions including: Constructing a variational constraint problem based on the gear vibration signal; Converting the variational constrained problem into a variational unconstrained problem; Perform iterative updating operations to obtain frequency spectra of several intrinsic mode functions; The frequency spectra of the plurality of intrinsic mode functions are transformed to obtain the plurality of intrinsic mode functions.

3. The gear fault intelligent diagnosis method according to claim 2, characterized in that: The constructing of the variational constraint problem based on the gear vibration signal includes: Decomposing the gear vibration signal to obtain multiple components; Construct a constrained variational model based on the constraint conditions to complete the construction of the variational constraint problem.

4. The gear fault intelligent diagnosis method according to claim 3 is characterized in that: The constraint condition is that the sum of the multiple components is equal to the gear vibration signal.

5. The gear fault intelligent diagnosis method according to claim 2, characterized in that: The converting the variational constrained problem into a variational unconstrained problem comprises: The variational constrained problem is transformed using Lagrange multipliers and quadratic penalty factors to obtain the variational unconstrained problem.

6. The gear fault intelligent diagnosis method according to claim 2, characterized in that: The iterative update operation to obtain the frequency spectra of several intrinsic mode functions includes: Using a multiplication operator alternating direction method to perform iterative update operations to obtain frequency spectra of the plurality of intrinsic mode functions; The transforming the frequency spectra of the plurality of intrinsic mode functions to obtain the plurality of intrinsic mode functions comprises: The frequency spectra of the plurality of intrinsic mode functions are converted from the frequency domain to the time domain through inverse Fourier transformation to obtain the plurality of intrinsic mode functions.

7. The gear fault intelligent diagnosis method according to claim 1, characterized in that: The step of calculating and extracting features of each of the intrinsic mode functions to obtain the energy entropy feature vector comprises: Calculating the energy of each of the eigenmode functions; Calculating the energy proportion of each of the intrinsic mode functions; According to the Shannon entropy principle, the energy entropy of each of the eigenmode functions is calculated; The energy entropy feature vector is constructed based on the energy entropy.

8. A gear fault intelligent diagnosis system, characterized in that: include: A signal acquisition unit, used for acquiring a gear vibration signal; A signal decomposition unit, used for decomposing the gear vibration signal by using a variational mode decomposition method to obtain a number of intrinsic mode functions; A feature extraction unit, used for calculating and extracting features of each of the intrinsic mode functions to obtain an energy entropy feature vector; A label combination unit, used to combine the energy entropy feature vector with the signal label to obtain a feature-label pair; A fault diagnosis unit, used for inputting the feature-label pair into a bidirectional long short-term memory network model for training to obtain a diagnosis result of a gear fault; The diagnosis result output unit is used to output the diagnosis result of the gear fault.

9. A storage medium, characterized in that: The storage medium stores a computer program, and the computer program is suitable for being loaded by a processor to execute the steps of the gear fault intelligent diagnosis method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps of the gear fault intelligent diagnosis method as claimed in any one of claims 1 to 7 by calling the computer program stored in the memory.

Images