Rolling bearing fault diagnosis method, device and electronic equipment
Through adaptive determination of the optimal variational modal decomposition parameters and envelope demodulation analysis, the problem of difficulty in separation of fault information in vibration signals in the prior art is solved, and accurate fault diagnosis and early fault identification of rolling bearings are achieved.
Patent Information
- Application Number
- CN202210591010.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-27
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-05-27
AI Technical Summary
The prior art cannot adaptively, accurately and efficiently separate the fault information of rolling bearings from the vibration signal, resulting in a low accuracy of the fault detection result.
By obtaining the vibration signal that characterizes the state of the rolling bearing, the optimal variational modal decomposition parameters are adaptively determined, and variational modal decomposition is performed. Combined with envelope demodulation analysis and fast Fourier transform, the optimal eigenmodal function is obtained. Envelope spectrum and determine the characteristic frequency based on the envelope spectrum for troubleshooting.
It realizes accurate diagnosis of rolling bearing status, improves the accuracy and efficiency of fault detection, and can adaptively optimize the variable modal decomposition parameters, highlight the impact characteristics caused by the fault, and accurately extract the early feature frequency.
Smart Images

Figure CN115031967B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of rolling bearings, and in particular to a rolling bearing fault diagnosis method, device and electronic equipment. Background Art
[0002] Rolling bearings are a key core component of mechanical transmission systems and are widely used in many industrial scenarios. The health of rolling bearings plays a very important role in the safe and stable operation of equipment. In the early stages of bearing failure, the fault signal is often mixed with other strong interference noises, resulting in weak fault features and difficulty in extraction, making it impossible to accurately diagnose the fault. Therefore, accurately extracting fault features from early bearing fault signals is the key to accurately identifying early bearing faults.
[0003] In the related art, manual methods are mainly used to set the optimal variational mode decomposition parameters based on experience. However, the above method of obtaining the optimal variational mode decomposition parameters requires continuous human attempts and cannot guarantee the optimal parameters, resulting in the inability to extract effective characteristic frequencies when the signal contains more noise components.
[0004] Therefore, how to adaptively, accurately and efficiently determine the optimal variational mode decomposition parameters and then fully separate the fault information from the signal to achieve accurate diagnosis of the bearing condition has become an urgent problem to be solved. Summary of the invention
[0005] The present application aims to solve one of the technical problems in the related art at least to some extent.
[0006] To this end, the first purpose of the present application is to propose a rolling bearing fault diagnosis method to solve the technical problems existing in the prior art that fault information cannot be adaptively, accurately and efficiently separated from vibration signals, signal feature extraction is inaccurate, and thus the accuracy of rolling bearing fault detection results is low.
[0007] In order to achieve the above-mentioned objectives, the first aspect of the present application provides a rolling bearing fault diagnosis method, which includes: obtaining a vibration signal characterizing the rolling bearing state information; adaptively determining the optimal variational modal decomposition parameters for fully separating the fault information from the vibration signal; performing variational modal decomposition on the vibration signal according to the optimal variational modal decomposition parameters to obtain at least two intrinsic mode function components; selecting the optimal intrinsic mode function from the intrinsic mode function components, and combining envelope demodulation analysis and fast Fourier transform to obtain the envelope spectrum of the optimal intrinsic mode function; according to the envelope spectrum, obtaining the characteristic frequency of the vibration signal, and performing fault diagnosis on the rolling bearing according to the characteristic frequency.
[0008] In addition, the rolling bearing fault diagnosis method according to the above embodiment of the present application may also have the following additional technical features:
[0009] According to one embodiment of the present application, the adaptive determination of the optimal variational modal decomposition parameters for fully separating fault information from the vibration signal includes: obtaining the number of candidate intrinsic modal functions, candidate penalty factors and search steps, and searching according to the number of candidate intrinsic modal functions, the candidate penalty factors and the search step to obtain candidate variational modal decomposition parameters; performing variational modal decomposition on the vibration signal according to the candidate variational modal decomposition parameters, and obtaining the kurtosis of each candidate intrinsic modal function; obtaining the optimal intrinsic modal function according to the kurtosis of each candidate intrinsic modal function and based on the maximum kurtosis principle; obtaining the optimal variational modal decomposition parameters according to the optimal intrinsic modal function.
[0010] According to one embodiment of the present application, the searching according to the number of candidate intrinsic mode functions, the candidate penalty factors and the search step size to obtain candidate variational mode decomposition parameters includes: arranging the number of candidate intrinsic mode functions in ascending order to obtain a ranking result of the number of candidate intrinsic mode functions; searching the number of candidate intrinsic mode functions starting from the number of candidate intrinsic mode functions that is at the first place in the ranking result, and traversing the candidate penalty factors according to the search step size to obtain the candidate variational mode decomposition parameters.
[0011] According to an embodiment of the present application, starting from the number of candidate intrinsic mode functions that is at the first place in the sorting result, searching the number of candidate intrinsic mode functions, and traversing the candidate penalty factors according to the search step to obtain the candidate variational mode decomposition parameters, includes: obtaining the arrangement order of the candidate penalty factors; starting from the number of candidate intrinsic mode functions that is at the first place in the sorting result, traversing all the candidate penalty factors in the arrangement order according to the number of candidate intrinsic mode functions in the current rank and the search step to obtain at least one group of candidate variational mode decomposition parameters corresponding to the number of candidate intrinsic mode functions in the current rank, wherein the candidate variational mode decomposition parameters include the number of candidate intrinsic mode functions and any of the candidate penalty factors.
[0012] According to an embodiment of the present application, the obtaining of the optimal intrinsic mode function based on the maximum kurtosis principle includes: obtaining the maximum kurtosis from all the kurtosis; and taking the candidate intrinsic mode function corresponding to the maximum kurtosis as the optimal intrinsic mode function.
[0013] According to one embodiment of the present application, the method further includes: obtaining a first sparse measurement index of the optimal intrinsic mode function corresponding to the number of the candidate intrinsic mode functions in the current order; judging whether to stop variational mode decomposition based on the first sparse measurement index, and obtaining the optimal variational mode decomposition parameters when it is determined to stop variational mode decomposition.
[0014] According to an embodiment of the present application, judging whether to stop variational mode decomposition based on the first sparse measurement index, and obtaining the optimal variational mode decomposition parameters when it is determined to stop variational mode decomposition, includes: judging whether the first sparse measurement index is the current minimum sparse measurement index; when it is determined that the first sparse measurement index is the current minimum sparse measurement index, obtaining the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of at least two subsequent orders; when it is determined that each of the second sparse measurement indicators is greater than the first sparse measurement indicator, stopping variational mode decomposition, and using the candidate variational mode decomposition parameters corresponding to the first sparse measurement indicator as the optimal variational mode decomposition parameters.
[0015] According to one embodiment of the present application, obtaining the characteristic frequency of the vibration signal according to the envelope spectrum includes: drawing a corresponding waveform curve according to the envelope spectrum; obtaining the peak of the waveform curve, and obtaining the characteristic frequency of the vibration signal according to the peak.
[0016] In order to achieve the above-mentioned purpose, the second aspect of the present application provides a rolling bearing fault diagnosis device, which includes: a first acquisition module, used to acquire a vibration signal characterizing the rolling bearing state information; a second acquisition module, used to adaptively determine the optimal variational modal decomposition parameters for fully separating the fault information from the vibration signal; a third acquisition module, used to perform variational modal decomposition on the vibration signal according to the optimal variational modal decomposition parameters to obtain at least two intrinsic modal function components; a fourth acquisition module, used to select the optimal intrinsic modal function from the intrinsic modal function components, and obtain the envelope spectrum of the optimal intrinsic modal function in combination with envelope demodulation analysis and fast Fourier transform; a fault diagnosis module, used to acquire the characteristic frequency of the vibration signal according to the envelope spectrum, and perform fault diagnosis on the rolling bearing according to the characteristic frequency.
[0017] In addition, the rolling bearing fault diagnosis device according to the above embodiment of the present application may also have the following additional technical features:
[0018] According to one embodiment of the present application, the second acquisition module is further used to: obtain the number of candidate intrinsic mode functions, candidate penalty factors and search steps, and search according to the number of candidate intrinsic mode functions, the candidate penalty factors and the search step to obtain candidate variational modal decomposition parameters; perform variational modal decomposition on the vibration signal according to the candidate variational modal decomposition parameters, and obtain the kurtosis of each candidate intrinsic mode function; obtain the optimal intrinsic mode function according to the kurtosis of each candidate intrinsic mode function and based on the maximum kurtosis principle; obtain the optimal variational modal decomposition parameters according to the optimal intrinsic mode function.
[0019] According to one embodiment of the present application, the second acquisition module is further used to: arrange the number of candidate intrinsic mode functions in ascending order to obtain a ranking result of the number of candidate intrinsic mode functions; search the number of candidate intrinsic mode functions starting from the number of candidate intrinsic mode functions that is at the first place in the ranking result, and traverse the candidate penalty factors according to the search step size to obtain the candidate variational mode decomposition parameters.
[0020] According to one embodiment of the present application, the second acquisition module is also used to: obtain the arrangement order of the candidate penalty factors; starting from the number of candidate intrinsic mode functions that is at the first place in the sorting result, according to the number of candidate intrinsic mode functions in the current rank and the search step, traverse all the candidate penalty factors in the arrangement order to obtain at least one group of candidate variational mode decomposition parameters corresponding to the number of candidate intrinsic mode functions in the current rank, wherein the candidate variational mode decomposition parameters include the number of candidate intrinsic mode functions and any candidate penalty factor.
[0021] According to an embodiment of the present application, the second acquisition module is further used to: obtain the maximum kurtosis from all the kurtosis; and use the candidate intrinsic mode function corresponding to the maximum kurtosis as the optimal intrinsic mode function.
[0022] According to one embodiment of the present application, the second acquisition module is also used to: obtain the first sparse measurement index of the optimal intrinsic mode function corresponding to the number of the candidate intrinsic mode functions in the current order; determine whether to stop variational mode decomposition based on the first sparse measurement index, and when it is determined to stop variational mode decomposition, obtain the optimal variational mode decomposition parameters.
[0023] According to one embodiment of the present application, the third acquisition module is also used to: determine whether the first sparse measurement index is the current minimum sparse measurement index; when it is determined that the first sparse measurement index is the current minimum sparse measurement index, obtain the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of at least two subsequent orders; when it is determined that each of the second sparse measurement indicators is greater than the first sparse measurement indicator, stop variational mode decomposition, and use the candidate variational mode decomposition parameters corresponding to the first sparse measurement indicator as the optimal variational mode decomposition parameters.
[0024] According to one embodiment of the present application, the fault diagnosis module is further used to: draw a corresponding waveform curve according to the envelope spectrum; obtain the peak of the waveform curve, and obtain the characteristic frequency of the vibration signal according to the peak.
[0025] In order to achieve the above-mentioned purpose, the third aspect embodiment of the present application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the rolling bearing fault diagnosis method as described in any one of the embodiments of the first aspect of the present application.
[0026] In order to achieve the above-mentioned objectives, the fourth aspect embodiment of the present application provides a non-transitory computer-readable storage medium storing computer instructions, which are used to implement the rolling bearing fault diagnosis method as described in any one of the first aspect embodiments of the present application when executed by a computer. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 A schematic diagram of a rolling bearing fault diagnosis method disclosed in an embodiment of the present application.
[0028] Figure 2 A schematic diagram of a rolling bearing fault diagnosis method disclosed in another embodiment of the present application.
[0029] Figure 3 A schematic diagram of a rolling bearing fault diagnosis method disclosed in another embodiment of the present application.
[0030] Figure 4 A schematic diagram of a rolling bearing fault diagnosis method disclosed in another embodiment of the present application.
[0031] Figure 5 A schematic diagram of a rolling bearing fault diagnosis method disclosed in another embodiment of the present application.
[0032] Figure 6A schematic diagram of a rolling bearing fault diagnosis method disclosed in another embodiment of the present application.
[0033] Figure 7 A schematic diagram of a rolling bearing fault diagnosis method disclosed in another embodiment of the present application.
[0034] Figure 8 A schematic diagram of a rolling bearing fault diagnosis method disclosed in another embodiment of the present application.
[0035] Fig. 9 A schematic diagram of an optimization process disclosed in one embodiment of the present application.
[0036] Fig.10 A schematic diagram of a rolling bearing fault diagnosis method disclosed in another embodiment of the present application.
[0037] Fig.11 A schematic flow chart of a rolling bearing fault diagnosis method disclosed in another embodiment of the present application.
[0038] Fig.12 A schematic diagram of the structure of a rolling bearing fault diagnosis device disclosed in one embodiment of the present application.
[0039] Fig.13 A schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application. DETAILED DESCRIPTION
[0040] In order to better understand the above technical solution, exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.
[0041] A rolling bearing fault diagnosis method and device according to an embodiment of the present application will be described below with reference to the accompanying drawings.
[0042] Figure 1 It is a flow chart of a rolling bearing fault diagnosis method according to an embodiment disclosed in the present application.
[0043] like Figure 1 As shown, the rolling bearing fault diagnosis method proposed in the embodiment of the present application specifically includes the following steps:
[0044] S101, obtaining a vibration signal representing state information of a rolling bearing.
[0045] Among them, rolling bearing refers to a precision mechanical component that converts the sliding friction between the rotating shaft and the shaft seat into rolling friction, reducing friction loss.
[0046] It should be noted that the present application does not limit the specific method for obtaining the vibration sampling signal of the rolling bearing, which can be set according to actual conditions.
[0047] As a possible implementation manner, optionally, a vibration acceleration sensor may be radially mounted on the rolling bearing, and a vibration signal representing the state information of the rolling bearing may be obtained by using the vibration acceleration sensor.
[0048] Advantageously, after the vibration signal representing the rolling bearing state information is obtained, the vibration signal may be processed to obtain the vibration signal.
[0049] It should be noted that the present application does not limit the specific method for obtaining the vibration signal of the rolling bearing, which can be set according to actual conditions.
[0050] As a possible implementation manner, optionally, the vibration signal may be subjected to windowing processing, and a vibration sampling signal with a fixed number of sampling points may be intercepted to obtain the vibration signal of the rolling bearing.
[0051] The windowing process refers to truncating the original signal X(t) at the sampling time (T), that is, limiting the original signal X(t) to XT(t) and then further processing it. This process is called windowing process.
[0052] S102: Adaptively determine optimal variational mode decomposition parameters for fully separating fault information from the vibration signal.
[0053] Among them, variational mode decomposition (VMD) can determine the frequency center and bandwidth of each component by iteratively searching for the optimal solution of the variational model in the process of obtaining the decomposed components.
[0054] The variable mode decomposition parameters may include but are not limited to the number of intrinsic mode functions k and the penalty factor α.
[0055] In the embodiment of the present application, after the vibration signal is acquired, the optimal variational mode decomposition parameters for fully separating the fault information from the vibration signal can be adaptively determined.
[0056] S103. Perform variational modal decomposition on the vibration signal according to the optimal variational modal decomposition parameters to obtain at least two intrinsic mode function components.
[0057] In the embodiment of the present application, after the optimal variational modal decomposition parameters are obtained, variational modal decomposition can be performed on the vibration signal according to the optimal variational modal decomposition parameters to obtain at least two intrinsic mode function components.
[0058] S104, selecting an optimal intrinsic mode function from the intrinsic mode function components, and combining envelope demodulation analysis and fast Fourier transform to obtain an envelope spectrum of the optimal intrinsic mode function.
[0059] It should be noted that the specific method of selecting the optimal intrinsic mode function from the intrinsic mode function components is not limited in the present application and can be selected according to actual conditions.
[0060] As a possible implementation manner, optionally, the optimal IMF may be selected from the IMF components based on a maximum kurtosis criterion.
[0061] In the embodiment of the present application, after the optimal intrinsic mode function is obtained, envelope demodulation analysis can be performed on the optimal intrinsic mode function, and a fast Fourier transform can be performed to obtain the envelope spectrum of the optimal intrinsic mode function.
[0062] S105. Acquire a characteristic frequency of the vibration signal according to the envelope spectrum, so as to perform fault diagnosis on the rolling bearing according to the characteristic frequency.
[0063] It should be noted that after the envelope spectrum is obtained, the characteristic frequency of the vibration signal can be obtained according to the spectrum peak of the envelope spectrum.
[0064] Furthermore, after the characteristic frequency of the vibration signal is acquired, the rolling bearing fault diagnosis can be performed according to the characteristic frequency.
[0065] As a possible implementation method, optionally, the attribute parameters of the rolling bearing can be obtained, and the inherent fault characteristic frequency of the rolling bearing can be extracted from the attribute parameters, and the rolling bearing fault diagnosis can be performed based on the characteristic frequency and the inherent fault characteristic frequency.
[0066] It should be noted that the specific method for rolling bearing fault diagnosis is not limited in the present application and can be selected according to actual conditions.
[0067] As a possible implementation manner, optionally, when the characteristic frequency is consistent with the inherent fault characteristic frequency, it indicates that there is a fault in the rolling bearing.
[0068] As a possible implementation manner, optionally, an inherent fault characteristic frequency range may be determined based on the inherent fault characteristic frequency. When the characteristic frequency is within the inherent fault characteristic frequency range, it indicates that a rolling bearing fault exists.
[0069] It should be noted that the above description of rolling bearing fault diagnosis based on characteristic frequency is only a possible example, and in practical applications, it can be set according to actual conditions.
[0070] For example, optionally, the attribute parameters of the rolling bearing may be acquired, and the inherent normal operating characteristic frequency of the rolling bearing may be extracted from the attribute parameters, and the rolling bearing may be fault diagnosed based on the characteristic frequency and the inherent normal operating characteristic frequency.
[0071] The rolling bearing fault diagnosis method provided by the present application obtains a vibration signal that characterizes the state information of the rolling bearing, and adaptively determines the optimal variational modal decomposition parameters that fully separate the fault information from the vibration signal, and according to the optimal variational modal decomposition parameters, performs variational modal decomposition on the vibration signal to obtain at least two intrinsic modal function components, selects the optimal intrinsic modal function from the intrinsic modal function components, and combines envelope demodulation analysis and fast Fourier transform to obtain the envelope spectrum of the optimal intrinsic modal function, and according to the envelope spectrum, obtains the characteristic frequency of the vibration signal, and performs fault diagnosis on the rolling bearing according to the characteristic frequency. Thus, the present application provides a rolling bearing fault diagnosis method, which can adaptively, accurately and efficiently determine the optimal variational modal decomposition parameters, and then fully separate the fault information from the original vibration signal to achieve accurate diagnosis of the bearing state characteristic frequency, the method has the advantages of good adaptability, simple application operation, high fault diagnosis accuracy, good fault information separation effect, etc., and can accurately diagnose the early faults of the rolling bearing.
[0072] It should be noted that in the present application, when obtaining the optimal variational mode decomposition parameters, the number k of candidate intrinsic mode functions and all possible values of the candidate penalty factor α can be gridded, and the number k of candidate intrinsic mode functions can be searched from small to large, and each candidate penalty factor α value corresponding to the specific parameter k is preferentially traversed, and the parameter combination [k, α] corresponding to the grid point is selected in turn, and then the optimal variational mode decomposition parameters are selected.
[0073] As a possible implementation, Figure 2 As shown, based on the above embodiment, the specific process of adaptively determining the optimal variational mode decomposition parameters for fully separating fault information from the vibration signal in the above step S102 includes the following steps:
[0074] S201, obtaining the number of candidate intrinsic mode functions, candidate penalty factors, and search steps, and performing a search based on the number of candidate intrinsic mode functions, candidate penalty factors, and search steps to obtain candidate variational mode decomposition parameters.
[0075] It should be noted that, in the present application, when adaptively determining the optimal variational mode decomposition parameters for fully separating fault information from the vibration signal, the number of candidate intrinsic mode functions, the candidate penalty factor α and the search step size may be first obtained.
[0076] It should be noted that the specific value of the number of candidate intrinsic mode functions is not limited in the present application and can be selected according to actual conditions.
[0077] For example, the number of candidate IMFs may be 2, 5, or 14. Preferably, the number of candidate IMFs is at least 2.
[0078] It should be noted that the setting of the search step length is not limited in the present application and can be selected according to actual conditions.
[0079] Optionally, the search step size can be set to 1, 2, 5, etc.
[0080] Among them, the penalty factor can be used to weigh the weight of loss and classification interval.
[0081] It should be noted that the setting of the penalty factor is not limited in this application and can be selected according to actual conditions.
[0082] Optionally, the penalty factor may be set to an integer from 100 to 4000.
[0083] Further, after obtaining the number of candidate intrinsic mode functions, candidate penalty factors, and search step length, a search may be performed based on the number of candidate intrinsic mode functions, candidate penalty factors, and search step length to obtain candidate variational mode decomposition parameters.
[0084] As a possible implementation, Figure 3 As shown, based on the above embodiment, the specific process of searching according to the number of candidate intrinsic mode functions, candidate penalty factors and search steps in the above step S201 to obtain candidate variational mode decomposition parameters includes the following steps:
[0085] S301 , arranging the number of candidate intrinsic mode functions in ascending order to obtain a ranking result of the number of candidate intrinsic mode functions.
[0086] In the embodiment of the present application, after the number of candidate intrinsic mode functions is obtained, the number can be arranged in ascending order to obtain a ranking result for each number.
[0087] For example, if the number k of candidate intrinsic mode functions is 2, the number k of candidate intrinsic mode functions is 3, and if the number k of candidate intrinsic mode functions is 4, in this case, the numbers are arranged in ascending order, and the sorting results of each number can be obtained as follows: number k is 2, number k is 3, and number k is 4.
[0088] S302 , starting from the candidate intrinsic mode function number that ranks first in the sorting result, searching the candidate intrinsic mode function number, and traversing the candidate penalty factors according to the search step size to obtain the candidate variational mode decomposition parameters.
[0089] As a possible implementation, Figure 4 As shown, based on the above embodiment, the specific process of searching the number of candidate intrinsic mode functions starting from the number of candidate intrinsic mode functions that is ranked first in the sorting result in the above step S302, and traversing the candidate penalty factors according to the search step length to obtain the candidate variational mode decomposition parameters includes the following steps:
[0090] S401: Obtain the ranking order of candidate penalty factors.
[0091] It should be noted that the specific method for obtaining the arrangement order of the candidate penalty factors is not limited in the present application and can be selected according to actual conditions.
[0092] Optionally, the candidate penalty factors α may be divided into grids to obtain the arrangement order of the candidate penalty factors.
[0093] For example, if the penalty factors are 100, 101, ... 4000, in this case, the arrangement in positive order is [100, 101, ... 4000].
[0094] S402. Starting from the number of candidate intrinsic mode functions that ranks first in the sorting result, according to the number of candidate intrinsic mode functions in the current rank and the search step, all candidate penalty factors are traversed in the order of arrangement to obtain at least one set of candidate variational mode decomposition parameters corresponding to the number of candidate intrinsic mode functions in the current rank, wherein the candidate variational mode decomposition parameters include the number of candidate intrinsic mode functions and any candidate penalty factor.
[0095] It should be noted that in the present application, the search step size corresponding to the number k of candidate intrinsic mode functions and the search step size corresponding to the candidate penalty factor α can be set separately, that is, the search step size corresponding to the number k of candidate intrinsic mode functions and the search step size corresponding to the candidate penalty factor α can be consistent or inconsistent.
[0096] For example, the search step length corresponding to the number of candidate intrinsic mode functions k can be set to 1, and the search step length corresponding to the candidate penalty factor α can be set to 1; the search step length corresponding to the number of candidate intrinsic mode functions k can be set to 1, and the search step length corresponding to the candidate penalty factor α can be set to 5.
[0097] Furthermore, according to the number of candidate intrinsic mode functions of the current rank and the search step, all candidate penalty factors are traversed in order of arrangement to obtain at least one set of candidate variational mode decomposition parameters corresponding to the number of the current rank.
[0098] The candidate variational mode decomposition parameters include the number of candidate intrinsic mode functions k and any candidate penalty factor α.
[0099] For example, if the number of candidate intrinsic mode functions k is 2, the candidate penalty factor α is 100 to 4000, and the search step is 1, the candidate variational mode decomposition parameters corresponding to the number of candidate intrinsic mode functions in the current order can be obtained: [2, 100], [2, 101], [2, 102]...[2, 4000].
[0100] S202. Perform variational modal decomposition on the vibration signal according to the candidate variational modal decomposition parameters, and obtain the kurtosis of each candidate intrinsic mode function.
[0101] For example, if the candidate variational modal decomposition parameters are [2, 100], [2, 101], [2, 102] ... [2, 4000], the vibration signal can be subjected to variational modal decomposition according to the candidate variational modal decomposition parameters to obtain the kurtosis of each candidate intrinsic mode function.
[0102] Among them, kurtosis refers to a characteristic parameter used to measure the outlier degree of outlier data.
[0103] It should be noted that when trying to obtain the kurtosis of the candidate intrinsic mode function, the following formula can be used to obtain it.
[0104]
[0105] Among them, Kurt is the kurtosis value of the candidate intrinsic mode function, N is the number of data points of the candidate intrinsic mode function, x i is the i-th data point of the candidate intrinsic mode function, μ is the mean of the candidate intrinsic mode function sequence, and σ is the variance of the candidate intrinsic mode function sequence.
[0106] S203 . According to the kurtosis of each candidate eigenmode function and based on the kurtosis maximum principle, an optimal eigenmode function is obtained.
[0107] Among them, the maximum kurtosis principle refers to the principle of obtaining the optimal intrinsic mode function on the basis of maximum kurtosis.
[0108] As a possible implementation, Figure 5 As shown, based on the above embodiment, the specific process of obtaining the optimal intrinsic mode function according to the kurtosis of each candidate intrinsic mode function and based on the kurtosis maximum principle in the above step S203 includes the following steps:
[0109] S501. Obtain the maximum kurtosis from all kurtosis.
[0110] It should be noted that since the process of variational mode decomposition is a layer-by-layer decomposition process, the maximum number of kurtosis can only be 1, and all kurtosis are not equal.
[0111] S502: taking the candidate eigenmode function corresponding to the maximum kurtosis as the optimal eigenmode function.
[0112] For example, if the kurtosis of candidate eigenmode function 1 is 3, the kurtosis of candidate eigenmode function 2 is 4, and the kurtosis of candidate eigenmode function 3 is 5, in this case, the maximum kurtosis is 5, and the optimal eigenmode function is candidate eigenmode function 3.
[0113] S204. Obtain optimal variational mode decomposition parameters according to the optimal eigenmode function.
[0114] The rolling bearing fault diagnosis method provided by the present application can obtain the first sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of the current order, and judge whether to stop variational mode decomposition according to the first sparse measurement index, and obtain the optimal variational mode decomposition parameters when it is determined to stop variational mode decomposition. Therefore, the present application can select the number k and penalty factor α of the optimal intrinsic mode by obtaining the optimal variational mode decomposition parameters, can retain important information and enhance the vibration signal denoising ability, and prevent the vibration signal from being over-decomposed, and realize the adaptive optimization selection of variational mode decomposition parameters, improve the accuracy of rolling bearing fault information decomposition, highlight the impact characteristics caused by rolling bearing faults, and can accurately extract the early characteristic frequencies of vibration signals.
[0115] Figure 6 It is a flow chart of a rolling bearing fault diagnosis method according to an embodiment disclosed in the present application.
[0116] like Figure 6 As shown, the rolling bearing fault diagnosis method proposed in the embodiment of the present application specifically includes the following steps:
[0117] S601, obtaining a vibration signal representing state information of a rolling bearing.
[0118] S602: Arrange the number of candidate intrinsic mode functions in ascending order to obtain a ranking result of the number of candidate intrinsic mode functions.
[0119] S603: Obtain the ranking order of candidate penalty factors.
[0120] S604. Starting from the number of candidate intrinsic mode functions that ranks first in the sorting result, according to the number of candidate intrinsic mode functions in the current rank and the search step, all candidate penalty factors are traversed in the order of arrangement to obtain at least one set of candidate variational mode decomposition parameters corresponding to the number of candidate intrinsic mode functions in the current rank, wherein the candidate variational mode decomposition parameters include the number of candidate intrinsic mode functions and any candidate penalty factor.
[0121] S605 . Perform variational modal decomposition on the vibration signal according to the candidate variational modal decomposition parameters, and obtain the kurtosis of each candidate intrinsic mode function.
[0122] S606. Obtain the maximum kurtosis from all kurtosis.
[0123] S607: taking the candidate eigenmode function corresponding to the maximum kurtosis as the optimal eigenmode function.
[0124] S608. Obtain optimal variational mode decomposition parameters according to the optimal eigenmode function.
[0125] Furthermore, in the present application, after obtaining the optimal intrinsic mode function, it can be determined whether to stop the variational mode decomposition.
[0126] As a possible implementation, Figure 7 As shown, based on the above embodiment, the following steps are specifically included:
[0127] S701. Obtain a first sparse measurement index of an optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of the current rank.
[0128] The sparsity index may be a power spectrum entropy, that is, the first sparsity measurement index may be a first power spectrum entropy.
[0129] It should be noted that, in the embodiment of the present application, when trying to obtain the first sparse measurement index, it can be obtained according to the following formula:
[0130]
[0131]
[0132] Among them, U IMF (f) is the spectrum obtained after Fourier transform of the intrinsic mode function, S IMF(f) is the power spectrum data sequence, q i is the percentage of the ith power spectrum value in the entire power spectrum sequence, H f It is the first sparse measure.
[0133] S702. Determine whether to stop variational mode decomposition according to the first sparse measurement index, and obtain optimal variational mode decomposition parameters when it is determined to stop variational mode decomposition.
[0134] As a possible implementation, Figure 8 As shown, based on the above embodiment, in the above step S702, according to the first sparse measurement index, it is determined whether to stop the variational mode decomposition, and when it is determined to stop the variational mode decomposition, the specific process of obtaining the optimal variational mode decomposition parameters includes the following steps:
[0135] S801. Determine whether the first sparse measurement index is the current minimum sparse measurement index.
[0136] In an embodiment of the present application, after obtaining the first sparse measurement index, it can be determined whether the current first sparse measurement index is the current minimum sparse measurement index after the current round of decomposition during the variational mode decomposition process.
[0137] S802: When it is determined that the first sparse measure index is the current minimum sparse measure index, obtain second sparse measure indexes of the optimal intrinsic mode functions corresponding to the numbers of candidate intrinsic mode functions of at least two subsequent orders.
[0138] For example, if the number of candidate intrinsic mode functions in the current order is k=5, it is determined that the first sparse measurement index corresponding to k=5 is 10, and it is the current minimum sparse measurement index. In this case, the decomposition continues. At the same time, in order to avoid over-decomposition, the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions in at least two subsequent orders can be obtained. For example, the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions in the subsequent two orders can be obtained. The second sparse measurement index corresponding to the number of candidate intrinsic mode functions k=6 in the next order is 11, and the second sparse measurement index corresponding to the number of candidate intrinsic mode functions k=7 in the next two orders is 12.
[0139] It should be noted that the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of at least two subsequent ranks proposed in the present application is only a possible example. In actual applications, the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of any subsequent consecutive ranks can be obtained. For example, the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of any subsequent ranks, such as at least three ranks, four ranks, etc., can be obtained.
[0140] S803. When it is determined that each second sparse measurement index is greater than the first sparse measurement index, stop the variational mode decomposition, and use the candidate variational mode decomposition parameters corresponding to the first sparse measurement index as the optimal variational mode decomposition parameters.
[0141] For example, if the number of candidate intrinsic mode functions in the current order is k=5, the first sparse measurement index corresponding to k=5 is determined to be 10, and it is the current minimum sparse measurement index, the second sparse measurement index corresponding to the number of candidate intrinsic mode functions in the next order k=6 is 11, and the second sparse measurement index corresponding to the number of candidate intrinsic mode functions in the next two orders k=7 is 12, that is, when each second sparse measurement index is greater than the first sparse measurement index, in this case, the variational mode decomposition can be stopped, and the candidate variational mode decomposition parameters corresponding to the first sparse measurement index are used as the optimal variational mode decomposition parameters.
[0142] It should be noted that in order to more clearly demonstrate that the present application proposes an early termination criterion and related processing procedures based on a sparse measurement indicator, the following explanation is given by taking the sparse measurement indicator as power spectrum entropy as an example.
[0143] For example, the number of IMF components k is searched between 2 and 15 with a step size of 1, and the penalty factor α is searched between 100 and 4000 with a step size of 100. The grid search method is used to optimize the parameters. The optimization process is as follows: Fig. 9 As shown, the optimal parameter combination is finally determined to be [5,3400].
[0144] The power spectrum entropy value in the figure refers to the minimum power spectrum entropy. There are 40 sets of α parameters for one k. The sign change of one grid point represents a set of parameter combinations. When the minimum power spectrum entropy is determined to be unchanged when k=6 and k=7 during 240 calculations, the optimal parameter combination is determined to be [5,3400].
[0145] The rolling bearing fault diagnosis method provided by the present application can determine whether the first sparse measurement index is the current minimum sparse measurement index, and when it is determined that the first sparse measurement index is the current minimum sparse measurement index, obtain the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of at least two subsequent orders, and then stop the variational mode decomposition when it is determined that each second sparse measurement index is greater than the first sparse measurement index, and use the candidate variational mode decomposition parameters corresponding to the first sparse measurement index as the optimal variational mode decomposition parameters. Therefore, the present application can effectively prevent excessive decomposition of the signal by proposing an early termination criterion based on the sparse measurement index, that is, the minimum sparse measurement index terminates the calculation when there is no change in all α values corresponding to the continuous traversal of the two parameter k values.
[0146] It should be noted that, in the present application, after obtaining the envelope spectrum, a corresponding waveform curve can be drawn, and the characteristic frequency of the vibration signal can be obtained according to the peak of the waveform curve.
[0147] As a possible implementation, Fig.10 As shown, based on the above embodiment, the specific process of obtaining the characteristic frequency of the vibration signal according to the envelope spectrum in the above step S105 includes the following steps:
[0148] S1001. Draw a corresponding waveform curve according to the envelope spectrum.
[0149] In the embodiment of the present application, after the envelope spectrum is obtained, a corresponding waveform curve can be drawn according to the envelope spectrum, wherein the abscissa of the waveform curve is the frequency of the vibration signal, and the ordinate is the amplitude of the vibration signal.
[0150] S1002: Obtain a peak of the waveform curve, and obtain a characteristic frequency of the vibration signal according to the peak.
[0151] In an embodiment of the present application, the peak of the waveform curve can be obtained, and the frequency corresponding to the waveform curve can be used as the characteristic frequency of the vibration signal. The characteristic frequency of the vibration signal is compared with the natural frequency of the rolling bearing, and then the rolling bearing fault is diagnosed based on the comparison result.
[0152] It should be noted that the present application does not limit the specific method of comparing the wave peak with the natural frequency of the rolling bearing and diagnosing the rolling bearing fault based on the comparison result, and does not limit the specific setting of the natural frequency, which can be selected according to actual conditions.
[0153] As a possible implementation method, the inherent frequency can be set as the inherent fault characteristic frequency. Optionally, when the identification characteristic frequency and the inherent fault characteristic frequency are consistent, it can be determined that there is a fault in the rolling bearing; optionally, the inherent fault characteristic frequency range can be determined based on the inherent fault characteristic frequency, and then when the identification characteristic frequency is within the inherent fault characteristic frequency range, it can be determined that there is a fault in the rolling bearing.
[0154] As another possible implementation method, the natural frequency can be set as the inherent normal operating characteristic frequency. Optionally, when the identification characteristic frequency and the inherent normal operating characteristic frequency are inconsistent, it can be determined that there is a fault in the rolling bearing; optionally, the inherent normal operating characteristic frequency range can be determined based on the inherent normal operating characteristic frequency, and then when the identification characteristic frequency is not within the inherent normal operating characteristic frequency range, it can be determined that there is a fault in the rolling bearing.
[0155] The rolling bearing fault diagnosis method provided by the present application can draw a corresponding waveform curve according to the envelope spectrum, and obtain the peak of the waveform curve, obtain the characteristic frequency of the vibration signal according to the peak, and then perform fault diagnosis on the rolling bearing according to the characteristic frequency. The present application provides a rolling bearing fault diagnosis method, which has the advantages of good adaptability, simple application operation, high recognition accuracy, etc., can accurately diagnose early faults of rolling bearings, and can be applied to fault diagnosis of other rotating machinery.
[0156] In summary, if Fig.11 As shown, the rolling bearing fault diagnosis method proposed in the present application can first obtain the vibration sampling signal, determine the variational modal decomposition parameter combination, and perform variational modal decomposition, obtain the optimal intrinsic mode function based on the maximum kurtosis criterion, and then calculate and update the minimum power spectrum entropy value, by judging whether it meets the early termination criterion or traversing all variational modal decomposition parameter combinations. If not, the variational modal decomposition parameter combination is re-determined. If it meets, variational modal decomposition is performed based on the optimized parameters, and based on the maximum kurtosis criterion, the optimal intrinsic mode function is obtained, and envelope demodulation and fast Fourier transform are performed to obtain the envelope spectrum, and then the characteristic frequency is obtained according to the envelope spectrum, and compared with the fixed characteristic frequency to perform fault diagnosis on the rolling bearing.
[0157] The present application utilizes variational modal decomposition to decompose the vibration signal of a rolling bearing, accurately selects the optimal intrinsic mode function based on the maximum kurtosis, takes the minimum power spectrum entropy of the optimal intrinsic mode function as the goal, and sets an early termination criterion to achieve adaptive optimization selection of variational modal decomposition parameters. At the same time, grid division is performed according to all possible values of the number of intrinsic mode functions k and the penalty factor α, the number of intrinsic mode functions k is searched from small to large, and each α value corresponding to the specific number of intrinsic mode functions k is traversed preferentially, and the parameter combination [k, α] corresponding to the grid point is selected in turn, and the rolling bearing vibration signal is subjected to variational modal decomposition solution, and the minimum power spectrum entropy value is used as the early termination criterion when all α values corresponding to the two parameter k values are continuously traversed without change.
[0158] Therefore, the present application accurately selects the optimal intrinsic mode function based on the maximum kurtosis criterion, takes the minimum power spectrum entropy of the optimal intrinsic mode function as the goal, sets the early termination criterion, realizes the adaptive optimization selection of variational mode decomposition parameters, improves the accuracy of fault information decomposition, highlights the impact characteristics caused by the fault, can accurately extract early characteristic frequencies, and at the same time sets the variational mode decomposition parameter optimization rules and early termination criteria, and can select the appropriate number of intrinsic mode components k and penalty factor α to retain important information and enhance signal denoising capabilities, while preventing excessive signal decomposition. The rolling bearing fault diagnosis method proposed in the present application has the advantages of good adaptability, simple application operation, high fault diagnosis accuracy, and good fault information separation effect. It can better solve the problem of early bearing fault diagnosis and can also be applied to fault feature extraction and diagnosis of other rotating machinery.
[0159] Fig.12 It is a structural schematic diagram of a rolling bearing fault diagnosis device according to an embodiment disclosed in the present application.
[0160] like Fig.12 As shown, the rolling bearing fault diagnosis device 100 includes: a first acquisition module 11, a second acquisition module 12, a third acquisition module 13, a fourth acquisition module 14 and a fault diagnosis module 15.
[0161] A first acquisition module 11 is used to acquire a vibration signal representing the state information of the rolling bearing;
[0162] The second acquisition module 12 is used to adaptively determine the optimal variational mode decomposition parameters for fully separating the fault information from the vibration signal;
[0163] A third acquisition module 13 is used to perform variational modal decomposition on the vibration signal according to the optimal variational modal decomposition parameter to obtain at least two intrinsic mode function components;
[0164] A fourth acquisition module 14 is used to select an optimal intrinsic mode function from the intrinsic mode function components, and obtain an envelope spectrum of the optimal intrinsic mode function by combining envelope demodulation analysis and fast Fourier transform;
[0165] The fault diagnosis module 15 is used to obtain the characteristic frequency of the vibration signal according to the envelope spectrum, and perform fault diagnosis on the rolling bearing according to the characteristic frequency.
[0166] According to one embodiment of the present application, the second acquisition module 12 is also used to: obtain the number of candidate intrinsic mode functions, candidate penalty factors and search steps, and search according to the number of candidate intrinsic mode functions, the candidate penalty factors and the search step length to obtain candidate variational modal decomposition parameters; perform variational modal decomposition on the vibration signal according to the candidate variational modal decomposition parameters, and obtain the kurtosis of each candidate intrinsic mode function; obtain the optimal intrinsic mode function according to the kurtosis of each candidate intrinsic mode function and based on the maximum kurtosis principle; obtain the optimal variational modal decomposition parameters according to the optimal intrinsic mode function.
[0167] According to one embodiment of the present application, the second acquisition module 12 is further used to: arrange the number of candidate intrinsic mode functions in ascending order to obtain a ranking result of the number of candidate intrinsic mode functions; search the number of candidate intrinsic mode functions starting from the number of candidate intrinsic mode functions that is at the first place in the ranking result, and traverse the candidate penalty factors according to the search step size to obtain the candidate variational mode decomposition parameters.
[0168] According to one embodiment of the present application, the second acquisition module 12 is also used to: obtain the arrangement order of the candidate penalty factors; starting from the number of candidate intrinsic mode functions that is at the first place in the sorting result, according to the number of candidate intrinsic mode functions in the current rank and the search step, traverse all the candidate penalty factors in the arrangement order to obtain at least one group of candidate variational mode decomposition parameters corresponding to the number of candidate intrinsic mode functions in the current rank, wherein the candidate variational mode decomposition parameters include the number of candidate intrinsic mode functions and any candidate penalty factor.
[0169] According to an embodiment of the present application, the second acquisition module 12 is further used to: acquire the maximum kurtosis from all the kurtosis; and use the candidate intrinsic mode function corresponding to the maximum kurtosis as the optimal intrinsic mode function.
[0170] According to one embodiment of the present application, the second acquisition module 12 is also used to: obtain a first sparse measurement index of the optimal intrinsic mode function corresponding to the number of the candidate intrinsic mode functions of the current order; determine whether to stop variational mode decomposition based on the first sparse measurement index, and when it is determined to stop variational mode decomposition, obtain the optimal variational mode decomposition parameters.
[0171] According to one embodiment of the present application, the third acquisition module 13 is also used to: determine whether the first sparse measurement index is the current minimum sparse measurement index; when it is determined that the first sparse measurement index is the current minimum sparse measurement index, obtain the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of at least two subsequent orders; when it is determined that each of the second sparse measurement indicators is greater than the first sparse measurement indicator, stop the variational mode decomposition, and use the candidate variational mode decomposition parameters corresponding to the first sparse measurement indicator as the optimal variational mode decomposition parameters.
[0172] According to an embodiment of the present application, the fault diagnosis module 15 is further used to: draw a corresponding waveform curve according to the envelope spectrum; obtain the peak of the waveform curve, and obtain the characteristic frequency of the vibration signal according to the peak.
[0173] The embodiment of the present application provides a rolling bearing fault diagnosis device, which obtains a vibration signal representing the state information of the rolling bearing, and adaptively determines the optimal variational modal decomposition parameters for fully separating the fault information from the vibration signal, performs variational modal decomposition on the vibration signal according to the optimal variational modal decomposition parameters, so as to obtain at least two intrinsic modal function components, selects the optimal intrinsic modal function from the intrinsic modal function components, and combines envelope demodulation analysis and fast Fourier transform to obtain the envelope spectrum of the optimal intrinsic modal function, and obtains the characteristic frequency of the vibration signal according to the envelope spectrum, and performs fault diagnosis on the rolling bearing according to the characteristic frequency. Therefore, the present application provides a rolling bearing fault diagnosis method, which can adaptively, accurately and efficiently determine the optimal variational modal decomposition parameters, and then fully separate the fault information from the original vibration signal to achieve accurate diagnosis of the bearing state characteristic frequency, the method has the advantages of good adaptability, simple application operation, high fault diagnosis accuracy, good fault information separation effect, etc., and can accurately diagnose the early faults of the rolling bearing.
[0174] In order to implement the above embodiment, the present application also proposes an electronic device 2000, such as Fig.13 As shown, it includes a memory 210, a processor 220, and a computer program stored in the memory 210 and executable on the processor 220. When the processor executes the program, the aforementioned rolling bearing fault diagnosis method is implemented.
[0175] In order to implement the above-mentioned embodiment, the present application also proposes a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to implement the above-mentioned rolling bearing fault diagnosis method when executed by a computer.
[0176] In the description of the present application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.
[0177] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.
[0178] In this application, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.
[0179] In the present application, unless otherwise clearly specified and limited, a first feature being “above” or “below” a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being “above”, “above”, and “above” a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being “below”, “below”, and “below” a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.
[0180] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0181] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.
Claims
1. A rolling bearing fault diagnosis method, It is characterized in that include: Acquire vibration signals representing rolling bearing status information; Adaptively determine the optimal variational mode decomposition parameters to fully separate fault information from vibration signals; According to the optimal variational modal decomposition parameters, performing variational modal decomposition on the vibration signal to obtain at least two intrinsic mode function components; Selecting an optimal intrinsic mode function from the intrinsic mode function components, and combining envelope demodulation analysis and fast Fourier transform to obtain an envelope spectrum of the optimal intrinsic mode function; According to the envelope spectrum, acquiring the characteristic frequency of the vibration signal, and performing fault diagnosis on the rolling bearing according to the characteristic frequency; The adaptive determination of optimal variational mode decomposition parameters for fully separating fault information from the vibration signal includes: Obtaining the number of candidate intrinsic mode functions, the candidate penalty factor, and the search step, and performing a search based on the number of candidate intrinsic mode functions, the candidate penalty factor, and the search step to obtain candidate variational mode decomposition parameters; According to the candidate variational modal decomposition parameters, performing variational modal decomposition on the vibration signal, and obtaining the kurtosis of each candidate intrinsic mode function; According to the kurtosis of each candidate intrinsic mode function and based on the kurtosis maximum principle, an optimal intrinsic mode function is obtained; According to the optimal intrinsic mode function, obtaining the optimal variational mode decomposition parameter; The searching according to the number of candidate intrinsic mode functions, the candidate penalty factors and the search step length to obtain candidate variational mode decomposition parameters includes: Arranging the number of candidate intrinsic mode functions in ascending order to obtain a ranking result of the number of candidate intrinsic mode functions; Starting from the number of candidate intrinsic mode functions that is at the first place in the sorting result, searching the number of candidate intrinsic mode functions, and traversing the candidate penalty factors according to the search step size to obtain the candidate variational mode decomposition parameters; The step of searching the candidate intrinsic mode function quantity starting from the candidate intrinsic mode function quantity that is at the first place in the sorting result, and traversing the candidate penalty factors according to the search step length to obtain the candidate variational mode decomposition parameters includes: Obtaining the arrangement order of the candidate penalty factors; Starting from the number of candidate intrinsic mode functions that is at the first place in the sorting result, according to the number of candidate intrinsic mode functions at the current rank and the search step, all the candidate penalty factors are traversed in the arrangement order to obtain at least one set of candidate variational mode decomposition parameters corresponding to the number of candidate intrinsic mode functions at the current rank, wherein the candidate variational mode decomposition parameters include the number of candidate intrinsic mode functions and any candidate penalty factor; Also includes: Obtaining a first sparse measurement index of the optimal intrinsic mode function corresponding to the number of the candidate intrinsic mode functions of the current rank; According to the first sparse measurement index, determine whether to stop variational mode decomposition, and when it is determined to stop variational mode decomposition, obtain the optimal variational mode decomposition parameters; The determining whether to stop variational mode decomposition according to the first sparse measurement index, and obtaining the optimal variational mode decomposition parameters when it is determined to stop variational mode decomposition, includes: Determine whether the first sparse measurement index is the current minimum sparse measurement index; When it is determined that the first sparse measure index is the current minimum sparse measure index, obtaining second sparse measure indexes of the optimal intrinsic mode function corresponding to the numbers of the candidate intrinsic mode functions of at least two subsequent orders respectively; When it is determined that each of the second sparse measurement indicators is greater than the first sparse measurement indicator, the variational mode decomposition is stopped, and the candidate variational mode decomposition parameters corresponding to the first sparse measurement indicator are used as the optimal variational mode decomposition parameters.
2. The rolling bearing fault diagnosis method according to claim 1, It is characterized in that The step of obtaining the optimal intrinsic mode function according to the kurtosis of each candidate intrinsic mode function and based on the maximum kurtosis principle includes: Obtaining the maximum kurtosis from all of said kurtosis; The candidate eigenmode function corresponding to the maximum kurtosis is used as the optimal eigenmode function.
3. The rolling bearing fault diagnosis method according to claim 1, It is characterized in that The step of acquiring the characteristic frequency of the vibration signal according to the envelope spectrum includes: According to the envelope spectrum, drawing a corresponding waveform curve; The peak of the waveform curve is obtained, and the characteristic frequency of the vibration signal is obtained according to the peak.
4. A rolling bearing fault diagnosis device, It is characterized in that include: A first acquisition module is used to acquire a vibration signal representing the state information of the rolling bearing; The second acquisition module is used to adaptively determine the optimal variational mode decomposition parameters for fully separating the fault information from the vibration signal; A third acquisition module, configured to perform variational modal decomposition on the vibration signal according to the optimal variational modal decomposition parameter to obtain at least two intrinsic mode function components; A fourth acquisition module is used to select an optimal intrinsic mode function from the intrinsic mode function components, and obtain an envelope spectrum of the optimal intrinsic mode function by combining envelope demodulation analysis and fast Fourier transform; A fault diagnosis module, used for acquiring a characteristic frequency of the vibration signal according to the envelope spectrum, and performing fault diagnosis on the rolling bearing according to the characteristic frequency; The second acquisition module is further used to: obtain the number of candidate intrinsic mode functions, candidate penalty factors and search steps, and search according to the number of candidate intrinsic mode functions, the candidate penalty factors and the search step to obtain candidate variational mode decomposition parameters; According to the candidate variational modal decomposition parameters, the vibration signal is subjected to variational modal decomposition, and the kurtosis of each candidate intrinsic modal function is obtained; according to the kurtosis of each candidate intrinsic modal function and based on the maximum kurtosis principle, an optimal intrinsic modal function is obtained; according to the optimal intrinsic modal function, the optimal variational modal decomposition parameters are obtained; The second acquisition module is further used to: arrange the number of candidate intrinsic mode functions in ascending order to obtain a ranking result of the number of candidate intrinsic mode functions; search the number of candidate intrinsic mode functions starting from the number of candidate intrinsic mode functions that is at the top of the ranking result, and traverse the candidate penalty factors according to the search step size to obtain the candidate variational mode decomposition parameters; The second acquisition module is further used to: obtain the arrangement order of the candidate penalty factors; starting from the number of candidate intrinsic mode functions that is at the first place in the sorting result, according to the number of candidate intrinsic mode functions at the current rank and the search step, traverse all the candidate penalty factors in the arrangement order to obtain at least one set of candidate variational mode decomposition parameters corresponding to the number of candidate intrinsic mode functions at the current rank, wherein the candidate variational mode decomposition parameters include the number of candidate intrinsic mode functions and any candidate penalty factor; The second acquisition module is further used to: obtain a first sparse measure index of the optimal intrinsic mode function corresponding to the number of the candidate intrinsic mode functions of the current order; determine whether to stop variational mode decomposition according to the first sparse measure index, and when it is determined to stop variational mode decomposition, obtain the optimal variational mode decomposition parameters; The second acquisition module is also used to: determine whether the first sparse measurement index is the current minimum sparse measurement index; when it is determined that the first sparse measurement index is the current minimum sparse measurement index, obtain the second sparse measurement index of the optimal intrinsic mode function corresponding to the number of candidate intrinsic mode functions of at least two subsequent orders; when it is determined that each of the second sparse measurement indicators is greater than the first sparse measurement indicator, stop variational mode decomposition, and use the candidate variational mode decomposition parameters corresponding to the first sparse measurement indicator as the optimal variational mode decomposition parameters.
5. An electronic device, include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 3.
