Vibration sensor fault detection method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of vibration sensor fault detection, in particular to a vibration sensor fault detection method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a vibration signal of a vibration sensor, determining a delay parameter and an embedded dimension of a phase space matrix based on the vibration signal, constructing the phase space matrix based on the delay parameter and the embedded dimension, performing singular decomposition on the phase space matrix to obtain a singular value sequence, calculating a mean value of a singular energy difference spectrum based on the singular value sequence; detecting singular value serial numbers of all abrupt change extreme points greater than a mean value in the singular energy difference spectrum, taking the maximum serial number as a noise reduction order, replacing all singular values of which the serial numbers are greater than the noise reduction order with zero, completing noise reduction of the singular energy difference spectrum, calculating a similar entropy between a vibration signal after noise reduction and a standard vibration signal, and calculating the similarity entropy between the vibration signal after noise reduction and the standard vibration signal. And judging whether the vibration sensor has a fault based on the similar entropy. According to the invention, the accuracy of vibration sensor fault detection can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vibration sensor fault detection, and in particular to a vibration sensor fault detection method, device, electronic equipment and storage medium. Background Art

[0002] In industrial production, vibration sensors are widely used to monitor the status of various mechanical equipment, such as engines, pumps, fans, centrifuges, etc.

[0003] Existing methods for detecting vibration sensor faults primarily include impedance analysis, response verification, and sensitivity assessment. Impedance analysis utilizes an impedance analyzer to comprehensively test the sensor's resistance, inductance, and capacitance within its operating frequency band. Response verification utilizes a signal generator to generate a signal of a specific frequency. After passing through the sensor, the output voltage is observed using an oscilloscope to verify consistency between the output and input signals. However, these methods often suffer from low accuracy and efficiency in detecting vibration sensor faults.

[0004] Therefore, a vibration sensor fault detection method is urgently needed to solve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide a vibration sensor fault detection method, device, electronic device and storage medium to solve the technical problems of low detection accuracy and low efficiency in existing solutions for vibration sensor fault detection.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] In one aspect, a vibration sensor fault detection method is provided, characterized in that the method comprises:

[0008] Step 1: Acquire a vibration signal from a vibration sensor, determine a delay parameter and an embedding dimension of a phase space matrix based on the vibration signal, construct a phase space matrix based on the delay parameter and the embedding dimension, perform a singular decomposition on the phase space matrix to obtain a singular value sequence, and calculate a mean of a singular energy difference spectrum based on the singular value sequence;

[0009] Step 2: Detect the singular value numbers of all mutation extreme points in the singular energy difference spectrum that are greater than the mean, use the largest number as the denoising order, and replace all singular values ​​with numbers greater than the denoising order with zero;

[0010] Step 3: Perform the inverse singular value decomposition process and inversely reconstruct the denoised signal from the phase space to complete the first denoising. The denoised signal is used as a new signal to replace the original vibration signal. Steps 1 to 3 are repeated until the termination condition of the singular energy difference spectrum denoising is met.

[0011] Step 4: Calculate the similarity entropy between the noise-reduced vibration signal and the standard vibration signal, and determine whether the vibration sensor is faulty based on the similarity entropy.

[0012] Furthermore, determining the delay parameter and embedding dimension of the phase space matrix based on the vibration signal, and constructing the phase space matrix based on the delay parameter and embedding dimension includes the following process:

[0013] Set the vibration signal to x=(x1, x2, ..., x N ), x i represents the data points contained in the vibration signal, where N is the number of data points, i = 1, 2, ..., N;

[0014]

[0015] Among them, I(τ) represents the function expression containing the delay parameter τ, p(x i , x i+τ ) represents the data point x i , x i+τ The probability of points appearing simultaneously, p(x i ) represents the data point x i The probability of occurrence, p(x i+τ ) represents the data point x i+τ Probability of occurrence;

[0016] The value corresponding to the first minimum value of I(τ) is recorded as the delay parameter τ;

[0017]

[0018] Where E(d) represents a function expression with embedding dimension d, X i (d) represents the i-th reconstruction vector when the embedding dimension is d, X i，d (d) represents X i (d)’s nearest neighbor, X i (d+1) represents the i-th reconstruction vector when the embedding dimension is d+1, X i，d (d+1) represents X i The nearest neighbor of (d+1), |||| represents the modulo value calculation;

[0019] The d value corresponding to the fluctuation range of E(d) / E(d+1) within 0.05 is the embedding dimension;

[0020] Construct the phase space matrix A based on the delay parameter and embedding dimension:

[0021]

[0022] Wherein, m=N-(d-1)τ.

[0023] Furthermore, performing singular decomposition on the phase space matrix to obtain a singular value sequence, and calculating the mean of the singular energy difference spectrum based on the singular value sequence includes the following process:

[0024] For the phase space matrix A, we get a pair of orthogonal matrices U = (u1, u2, ..., x 1+(d-1)τ ) and V=(v1,v2,...,v m ), and based on U, V and A, we get the following relationship:

[0025] A=USV T ;

[0026] Among them, V T represents the transpose of V, S is the singular value sequence of the phase space matrix, S = (λ1, λ2, ..., λ h ), where h = 1 + (d - 1) τ × m;

[0027] The singular energy E is calculated based on the singular value sequence k :

[0028] E k =λ k 2 ; k = 1, 2, ..., h;

[0029] Singular energy difference spectrum b k Expressed as:

[0030] b k =E k -E k+1 ;

[0031] Calculate the mean avg of the singular energy difference spectrum:

[0032]

[0033] Furthermore, calculating the similarity entropy between the noise-reduced vibration signal and the standard vibration signal includes the following process:

[0034] The time domain signal corresponding to the noise-reduced vibration signal is set to Y=(y1, y2, ..., y G ), y g represents the data points contained in the time domain signal, where G is the number of data points, g = 1, 2, ..., G;

[0035] Calculate the eigenvector F based on the time domain signal g :

[0036] F g =(y g ,y g+1 ,...,yg+B-1 )-y0;g=1,2,...,G-B+1;

[0037] Among them, B is the length of the vector, y0 is the average value of each vector,

[0038] Calculate the eigenvector F g and the standard vibration signal eigenvector F j Euclidean distance

[0039] Based on Euclidean distance Calculate the similarity expression function R between the denoised vibration signal and the standard vibration signal B :

[0040]

[0041] Based on the expression function R B Calculate the similarity entropy H RB :

[0042] H RB =ln(R B / R B+1 ).

[0043] Furthermore, judging whether the vibration sensor is faulty based on similar entropy includes the following process:

[0044] The similar entropy threshold is loaded, and it is determined whether the similar entropy exceeds the similar entropy threshold. If so, it is determined that the vibration sensor is faulty; if not, it is determined that the vibration sensor is not faulty.

[0045] Furthermore, the termination condition of the singular energy difference spectrum denoising includes that, among the singular value numbers of all detected mutation extreme value points greater than the mean, the maximum number is less than a preset value.

[0046] On the other hand, a gyroscope monitoring device is applicable to the vibration sensor fault detection method according to any one of claims 1 to 6, characterized in that the device comprises:

[0047] The vibration signal denoising module obtains the vibration signal of the vibration sensor, determines the delay parameter and embedding dimension of the phase space matrix based on the vibration signal, constructs the phase space matrix based on the delay parameter and embedding dimension, performs singular decomposition on the phase space matrix to obtain a singular value sequence, and calculates the mean of the singular energy difference spectrum based on the singular value sequence; detects the singular value sequence numbers of all mutation extreme points in the singular energy difference spectrum that are greater than the mean, takes the largest sequence number as the denoising order, and replaces all singular values ​​with sequence numbers greater than the denoising order with zero; performs the inverse singular value decomposition process, and inversely reconstructs the denoised signal from the phase space to complete a denoising operation, uses the signal that has completed the denoising operation as a new signal to replace the original vibration signal, and repeats steps one to three until the termination condition of the singular energy difference spectrum denoising is met;

[0048] The fault detection module calculates the similarity entropy between the noise-reduced vibration signal and the standard vibration signal, and determines whether the vibration sensor is faulty based on the similarity entropy.

[0049] On the other hand, a device includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform any one of the vibration sensor fault detection methods described above.

[0050] Finally, a storage medium stores computer instructions, wherein the computer instructions enable a computer to execute any one of the vibration sensor fault detection methods described above.

[0051] Compared with the existing solutions, the present invention achieves the following beneficial effects:

[0052] The present invention obtains a vibration signal from a vibration sensor and performs noise reduction processing on the vibration signal, calculates the similarity entropy between the noise-reduced vibration signal and a standard vibration signal, and determines whether the vibration sensor has a fault based on the similarity entropy, thereby improving the accuracy and efficiency of vibration sensor fault detection.

[0053] The present invention reduces the probability of failure and maintenance costs, and improves production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0055] Figure 1 is a flow chart of a vibration sensor fault detection method according to an embodiment of the present invention;

[0056] Figure 2This is a structural block diagram of a vibration sensor fault detection device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] In addition, the described features, structures or characteristics can be combined in any suitable manner in one or more example embodiments. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure can be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. can be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0059] This embodiment provides a vibration sensor fault detection method. Figure 1 FIG. 1 is a flow chart of a vibration sensor fault detection method according to an embodiment of the present invention. Figure 1 As shown, the method includes the following steps:

[0060] Step 1: Acquire a vibration signal from a vibration sensor, determine a delay parameter and an embedding dimension of a phase space matrix based on the vibration signal, construct a phase space matrix based on the delay parameter and the embedding dimension, perform a singular decomposition on the phase space matrix to obtain a singular value sequence, and calculate a mean of a singular energy difference spectrum based on the singular value sequence;

[0061] Step 2: Detect the singular value numbers of all mutation extreme points in the singular energy difference spectrum that are greater than the mean, use the largest number as the denoising order, and replace all singular values ​​with numbers greater than the denoising order with zero;

[0062] Step 3: Perform the inverse singular value decomposition process and inversely reconstruct the denoised signal from the phase space to complete the first denoising. The denoised signal is used as a new signal to replace the original vibration signal. Steps 1 to 3 are repeated until the termination condition of the singular energy difference spectrum denoising is met.

[0063] Step 4: Calculate the similarity entropy between the noise-reduced vibration signal and the standard vibration signal, and determine whether the vibration sensor is faulty based on the similarity entropy.

[0064] In summary, the present invention obtains the vibration signal of the vibration sensor and performs noise reduction processing on the vibration signal, calculates the similarity entropy between the noise-reduced vibration signal and the standard vibration signal, and judges whether the vibration sensor has a fault based on the similarity entropy, which can improve the accuracy and efficiency of vibration sensor fault detection.

[0065] In some embodiments, determining a delay parameter and an embedding dimension of a phase space matrix based on the vibration signal, and constructing the phase space matrix based on the delay parameter and the embedding dimension includes the following process:

[0066] Set the vibration signal to x=(x1, x2, ..., x N ), x i represents the data points contained in the vibration signal, where N is the number of data points, i = 1, 2, ..., N;

[0067]

[0068] Among them, I(τ) represents the function expression containing the delay parameter τ, p(x i , x i+τ ) represents the data point x i , x i+τ The probability of points appearing simultaneously, p(x i ) represents the data point x i The probability of occurrence, p(x i+τ ) represents the data point x i+τ Probability of occurrence;

[0069] The value corresponding to the first minimum value of I(τ) is recorded as the delay parameter τ;

[0070]

[0071] Where E(d) represents a function expression with embedding dimension d, X i (d) represents the i-th reconstruction vector when the embedding dimension is d, X i，d (d) represents X i (d)’s nearest neighbor, X i (d+1) represents the i-th reconstruction vector when the embedding dimension is d+1, X i，d (d+1) represents X i The nearest neighbor of (d+1), |||| represents the modulo value calculation;

[0072] The d value corresponding to the fluctuation range of E(d) / E(d+1) within 0.05 is the embedding dimension;

[0073] Construct the phase space matrix A based on the delay parameter and embedding dimension:

[0074]

[0075] Wherein, m=N-(d-1)τ.

[0076] In some embodiments, performing singular decomposition on the phase space matrix to obtain a singular value sequence, and calculating the mean of the singular energy difference spectrum based on the singular value sequence includes the following process:

[0077] For the phase space matrix A, we get a pair of orthogonal matrices U = (u1, u2, ..., x 1+(d-1)τ ) and V=(v1,v2,...,v m ), and based on U, V and A, we get the following relationship:

[0078] A=USV T ;

[0079] Among them, V T represents the transpose of V, S is the singular value sequence of the phase space matrix, S=(λ1,λ2,...,λ h ), where h = 1 + (d - 1) τ × m;

[0080] The singular energy E is calculated based on the singular value sequence k :

[0081] E k =λ k 2 ; k = 1, 2, ..., h;

[0082] Singular energy difference spectrum b k Expressed as:

[0083] b k =E k -E k+1 ;

[0084] Calculate the mean avg of the singular energy difference spectrum:

[0085]

[0086] In some embodiments, calculating the similarity entropy between the noise-reduced vibration signal and the standard vibration signal includes the following process:

[0087] The time domain signal corresponding to the noise-reduced vibration signal is set to Y=(y1, y2, ..., y G ), y g represents the data points contained in the time domain signal, where G is the number of data points, g = 1, 2, ..., G;

[0088] Calculate the eigenvector F based on the time domain signal g :

[0089] F g =(y g ,y g+1 ,...,y g+B-1 )-y0;g=1,2,...,G-B+1;

[0090] Among them, B is the length of the vector, y0 is the average value of each vector,

[0091] Calculate the eigenvector F g and the standard vibration signal eigenvector F j Euclidean distance

[0092] Based on Euclidean distance Calculate the similarity expression function R between the denoised vibration signal and the standard vibration signal B :

[0093]

[0094] Based on the expression function R B Calculate the similarity entropy H RB :

[0095] H RB =ln(R B / R B+1 ).

[0096] Furthermore, judging whether the vibration sensor is faulty based on similar entropy includes the following process:

[0097] The similar entropy threshold is loaded, and it is determined whether the similar entropy exceeds the similar entropy threshold. If so, it is determined that the vibration sensor is faulty; if not, it is determined that the vibration sensor is not faulty.

[0098] It is worth noting that the termination condition of the singular energy difference spectrum denoising includes that among the singular value numbers of all detected mutation extreme points greater than the mean, the maximum number is less than a preset value.

[0099] on the other hand, Figure 2 This is a structural block diagram of a vibration sensor fault detection device according to an embodiment of the present invention. Figure 2 As shown, the device includes:

[0100] The vibration signal denoising module obtains the vibration signal of the vibration sensor, determines the delay parameter and embedding dimension of the phase space matrix based on the vibration signal, constructs the phase space matrix based on the delay parameter and embedding dimension, performs singular decomposition on the phase space matrix to obtain a singular value sequence, and calculates the mean of the singular energy difference spectrum based on the singular value sequence; detects the singular value sequence numbers of all mutation extreme points in the singular energy difference spectrum that are greater than the mean, takes the largest sequence number as the denoising order, and replaces all singular values ​​with sequence numbers greater than the denoising order with zero; performs the inverse singular value decomposition process, and inversely reconstructs the denoised signal from the phase space to complete a denoising operation, uses the signal that has completed the denoising operation as a new signal to replace the original vibration signal, and repeats steps one to three until the termination condition of the singular energy difference spectrum denoising is met;

[0101] The fault detection module calculates the similarity entropy between the noise-reduced vibration signal and the standard vibration signal, and determines whether the vibration sensor is faulty based on the similarity entropy.

[0102] On the other hand, the present invention further provides a device, comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute any vibration sensor fault detection method.

[0103] Finally, the present invention also provides a storage medium, which stores computer instructions, and the computer instructions enable a computer to execute any vibration sensor fault detection method.

[0104] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0105] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0106] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0107] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is only for some logical functions. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0108] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0109] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A vibration sensor fault detection method, characterized in that: Methods include: Step 1: Obtain a vibration signal from a vibration sensor, determine a delay parameter and an embedding dimension of a phase space matrix based on the vibration signal, construct a phase space matrix based on the delay parameter and the embedding dimension, perform singular decomposition on the phase space matrix to obtain a singular value sequence, and calculate a mean of a singular energy difference spectrum based on the singular value sequence; Step 2: Detect the singular value numbers of all mutation extreme points in the singular energy difference spectrum that are greater than the mean, use the largest number as the denoising order, and replace all singular values ​​whose numbers are greater than the denoising order with zero; Step 3: perform the inverse process of singular value decomposition, and inversely reconstruct the denoised signal from the phase space to complete the denoising. The signal that has completed the denoising is used as a new signal to replace the original vibration signal, and steps 1 to 3 are repeated until the termination condition of the singular energy difference spectrum denoising is met; Step 4: Calculate the similarity entropy between the vibration signal after noise reduction and the standard vibration signal, and determine whether the vibration sensor is faulty based on the similarity entropy.

2. The vibration sensor fault detection method according to claim 1, characterized in that: Determining the delay parameter and the embedding dimension of the phase space matrix based on the vibration signal, and constructing the phase space matrix based on the delay parameter and the embedding dimension includes the following processes: Set the vibration signal to x = (x1, x2, ..., x N ), x i represents the data points contained in the vibration signal, where N is the number of data points, i=1, 2, ..., N; Among them, I(τ) represents the function expression containing the delay parameter τ, p(x i , x i+τ ) represents the data point x i , x i+τ The probability of points appearing simultaneously, p(x i ) represents the data point x i The probability of occurrence, p(x i+τ ) represents the data point x i+τ Probability of occurrence; The value corresponding to the first minimum value of I(τ) is recorded as the delay parameter τ; Where E(d) represents a function expression with embedding dimension d, X i (d) represents the i-th reconstruction vector reconstructed when the embedding dimension is d, X i，d (d) indicates X i (d)’s nearest neighbor, X i (d+1) represents the i-th reconstruction vector reconstructed when the embedding dimension is d+1, X i，d (d+1) represents X i The nearest neighbor of (d+1), |||| represents the modulus value calculation; The d value corresponding to the fluctuation range of E(d) / E(d+1) within 0.05 is the embedding dimension; Construct the phase space matrix A based on the delay parameter and the embedding dimension: Wherein, m=N-(d-1)τ.

3. The vibration sensor fault detection method according to claim 2, characterized in that: The phase space matrix is ​​singularly decomposed to obtain a singular value sequence, and the mean of the singular energy difference spectrum is calculated based on the singular value sequence. The process includes: For the phase space matrix A, we get a pair of orthogonal matrices U = (u1, u2, ..., x 1+(d-1)τ ) and V = (v1, v2, ..., v m ), and based on U, V and A, we get the following relationship: A=USV T ; Among them, V T represents the transpose of V, S is the singular value sequence of the phase space matrix, S = (λ1, λ2, ..., λ h ), where h = 1 + (d-1) τ × m; The singular energy E is calculated based on the singular value sequence k : E k =λ k 2 ;k=1,2,...,h; Singular energy difference spectrum b k It is expressed as: b k =E k -E k+1 ; Calculate the mean avg of the singular energy difference spectrum:

4. The vibration sensor fault detection method according to claim 1, characterized in that: Calculating the similarity entropy between the noise-reduced vibration signal and the standard vibration signal includes the following process: The time domain signal corresponding to the vibration signal after noise reduction is set to Y = (y1, y2, ..., y G ), y g represents the data points contained in the time domain signal, where G is the number of data points, g = 1, 2, ..., G; Calculate the eigenvector F based on the time domain signal g : F g =(y g ,y g+1 ,...,y g+B-1 )-y0;g=1,2,...,G-B+1; Among them, B is the length of the vector, y0 is the average value of each vector, Calculate the eigenvector F g Compared with the standard vibration signal eigenvector F j The Euclidean distance Based on Euclidean distance Calculate the similarity expression function R between the noise-reduced vibration signal and the standard vibration signal 1 : Based on the expression function R B Calculate the similarity entropy H RB : H RB =ln(R B / R B+1 )。 5. The vibration sensor fault detection method according to claim 4, characterized in that: Judging whether the vibration sensor is faulty based on similar entropy includes the following processes: The similar entropy threshold is loaded to determine whether the similar entropy exceeds the similar entropy threshold. If so, it is determined that the vibration sensor is faulty; if not, it is determined that the vibration sensor is not faulty.

6. The vibration sensor fault detection method according to claim 1, characterized in that: The termination condition of the singular energy difference spectrum denoising includes that among the singular value numbers of all detected mutation extreme value points greater than the mean, the maximum number is less than a preset value.

7. A vibration sensor fault detection device, applicable to the vibration sensor fault detection method according to any one of claims 1 to 6, characterized in that: The device includes: The vibration signal denoising module obtains the vibration signal of the vibration sensor, determines the delay parameter and embedding dimension of the phase space matrix based on the vibration signal, constructs the phase space matrix based on the delay parameter and the embedding dimension, performs singular decomposition on the phase space matrix to obtain a singular value sequence, and calculates the mean of the singular energy difference spectrum based on the singular value sequence; detects the singular value sequence numbers of all mutation extreme points in the singular energy difference spectrum that are greater than the mean, takes the maximum sequence number as the denoising order, and replaces all singular values ​​whose sequence numbers are greater than the denoising order with zero; performs the inverse process of singular value decomposition, and inversely reconstructs the denoised signal from the phase space to complete a denoising, uses the signal that has completed a denoising as a new signal to replace the original vibration signal, and repeats steps one to three until the termination condition of the singular energy difference spectrum denoising is met; The fault detection module calculates the similarity entropy between the noise-reduced vibration signal and the standard vibration signal, and determines whether the vibration sensor has a fault based on the similarity entropy.

8. A device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to execute the vibration sensor fault detection method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium stores computer instructions, which enable a computer to execute the vibration sensor fault detection method according to any one of claims 1 to 6.

Images