Variable speed bearing fault diagnosis method and system based on SVMD and rotating speed estimation

Abstract

The invention provides a variable-speed bearing fault diagnosis method and system based on SVMD and rotating speed estimation, and the method comprises the steps: obtaining an accurate fault feature value of a bearing through estimating the rotating speed of the bearing, employing SVMD to decompose a vibration signal, carrying out the denoising reconstruction, carrying out the envelope demodulation and order analysis of a reconstructed signal, comparing the feature frequencies of different parts of equipment with an envelope threshold value according to the rotating speed, and carrying out the fault diagnosis of the bearing. Therefore, the fault part is judged, the function of accurately identifying the fault part of the bearing is realized, and the bearing fault can be effectively diagnosed. The rotating speed of the variable-speed equipment is obtained without installing a rotating speed sensor, so that feature extraction and fault diagnosis are carried out on the bearing, and the method is particularly suitable for the working condition of power frequency change caused by rotating speed change of the variable-speed equipment. According to the method, aiming at noise interference and power frequency change in the vibration signals of the variable-speed bearing, all intrinsic mode function components (IMF) are extracted in a continuous mode by adopting a sequential variational mode decomposition algorithm (SVMD), so that the noise interference in the signals is effectively removed.

Description

Technical Field

[0001] The present invention belongs to the technical field of variable-speed bearing fault diagnosis, and particularly relates to a variable-speed bearing fault diagnosis method and system based on SVMD and speed estimation. Background Technique

[0002] During the operation of equipment, it often causes equipment damage due to the failure to detect faults in time, resulting in huge economic losses. The existing solution is to conduct fault diagnosis on the equipment to distinguish different faults of the equipment. The key to fault diagnosis is the recognition of fault signals. Since current equipment is becoming more and more complex, fault signals are often submerged, so it is necessary to denoise the fault signals and extract the effective components.

[0003] In the existing fault diagnosis technology, the variational mode decomposition algorithm VMD often reduces the performance of VMD due to inaccurate decomposition mode numbers. The sequential variational mode decomposition algorithm SVMD extracts all intrinsic mode function components IMF in a continuous manner; compared with VMD, SVMD does not need to know the number of modes and has lower computational complexity, and can effectively remove the noise interference in the signal.

[0004] On the other hand, variable-speed equipment often causes inaccurate extraction of bearing fault characteristic values due to the change of power frequency caused by the speed change, and some equipment cannot obtain accurate speed due to the difficulty in installing speed sensors. Summary of the Invention

[0005] The technical problem to be solved by the present invention is: to provide a variable-speed bearing fault diagnosis method and system based on SVMD and speed estimation for accurately identifying the fault location of the bearing.

[0006] The technical solution adopted by the present invention to solve the above technical problem is: a variable-speed bearing fault diagnosis method based on SVMD and speed estimation, including the following steps: S1: Obtain the vibration signals of the equipment at different speeds and normalize them, and estimate the speeds corresponding to the vibration signals; S2: Perform SVMD decomposition on the vibration signals to obtain basic mode components, and form a reconstructed signal according to the correlation with the vibration signals; S3: Perform envelope demodulation and order analysis on the reconstructed signal, compare the characteristic frequencies of different parts of the equipment with the envelope threshold according to the speed, and judge the fault location.

[0007] According to the above solution, in the step S1, the specific steps for estimating the speeds corresponding to the vibration signals are: S11: Discretize and perform time-frequency transformation on the vibration signals, and take the first ten-order data; S12: Establish a large loop for the first ten-order data before assignment. In each loop, obtain the amplitudes corresponding to 1 times, 0.5 times, 2 times, 3 times, and 4 times (loop count + 5). S13: Establish a small loop for screening (loop count + 5) under the large loop, where the loop count of the small loop is less than that of the large loop. S14: Compose the screened (loop count + 5) into the first vector M; traverse the first vector and assign values to the first vector according to conditions to obtain the fourth vector. S15: If the length of the fourth vector is greater than 0, use the mode of the fourth vector as the rotational speed; if the length of the fourth vector is less than or equal to 0, set the rotational speed equal to 25.

[0008] Further, in the step S11, the specific steps are as follows: S111: Filter the vibration signal. S112: Perform FFT transformation on the filtered data and take the first half of the data. S113: Arrange the data after FFT transformation from largest to smallest and take the first ten-order data after arrangement.

[0009] Further, in the step S12, the specific steps of the large loop are as follows: If the ratio of the first-order data to the second-order data is greater than 5, set the values of the second-order to tenth-order data equal to the first-order data. Or if the ratios of the first-order data, the second-order data, and the third-order data are all greater than 5, find the index positions of the first-order to tenth-order data in the data after FFT transformation: If the number of indices of the first-order data that can be divisible by the indices of the third-order to tenth-order data is greater than the number of indices of the second-order data that can be divisible by the indices of the third-order to tenth-order data, set the values of the second-order to tenth-order data equal to the first-order data. If the number of indices of the first-order data that can be divisible by the indices of the third-order to tenth-order data is less than the number of indices of the second-order data that can be divisible by the indices of the third-order to tenth-order data, set the values of the first-order, third-order to tenth-order data equal to the second-order data.

[0010] Further, in the step S13, the specific steps of the small loop are as follows: If the amplitude corresponding to 3 times (loop count + 5) is equal to the current first ten-order data, and the ratio of the amplitude corresponding to 3 times (loop count + 5) to the amplitude corresponding to 1 time (loop count + 5) is less than 5, and the amplitude corresponding to 1 time (loop count + 5) is greater than 0.2, record (loop count + 5). If the amplitude corresponding to 1 times (cycle number + 5) is equal to the current top ten order data, and the amplitude corresponding to 1 times (cycle number + 5) is greater than 0.2, and the amplitudes corresponding to 0.5 times (cycle number + 5) and 2 times (cycle number + 5) are within the range of the top ten order data, record (cycle number + 5); If the amplitude corresponding to 1 times (cycle number + 5) is equal to the current top ten order data, and the amplitude corresponding to 1 times (cycle number + 5) is greater than 0.2, and the amplitudes corresponding to 0.5 times (cycle number + 5) and 4 times (cycle number + 5) are not within the range of the top ten order data, record (cycle number + 5).

[0011] Further, in the step S14, the specific steps are as follows: S141: Traverse the first vector. If the absolute difference between 4 times of a certain data in the large loop and a certain data in the small loop is less than 2, then create another small loop: Traverse the first vector. If the absolute difference between 2 times of a certain data in the large loop and another data in the small loop is less than 2, let the other data in the small loop be equal to the certain data in the large loop; obtain the second vector; S142: Traverse the second vector. If the absolute difference between 3 times of a certain data in the large loop and a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; If the absolute difference between 4 times of a certain data in the large loop and a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; If the absolute difference between 5 times of a certain data in the large loop and a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; obtain the third vector; S143: Traverse the third vector. If the absolute difference between a certain data in the large loop and a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; If a certain data in the large loop is divisible by a certain data in the small loop, let the certain data in the small loop be equal to the certain data in the large loop; If the absolute difference between a certain data in the large loop and 0.5 times of a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; obtain the fourth vector.

[0012] According to the above scheme, in the step S2, the specific steps are as follows: S21: Perform SVMD decomposition on the vibration signal to obtain the basic mode components; S22: Normalize each basic mode component; S23: Calculate the correlation coefficients between each normalized vibration signal and each normalized basic mode component, and add up the basic mode components with correlation coefficients greater than the preset value to obtain the reconstructed signal.

[0013] According to the above solution, in step S3, the specific steps are as follows: S31: Perform envelope demodulation on the reconstructed signal and draw the order envelope spectrum at different speeds. S32: Calculate the order of the characteristic frequency of the device according to the speed. S33: Compare the amplitude of the characteristic frequency of the device in the order envelope spectrum with the envelope threshold: If the amplitude of a certain characteristic frequency exceeds the envelope threshold, the device component corresponding to this characteristic frequency fails.

[0014] A variable-speed bearing fault diagnosis system based on SVMD and speed estimation A speed estimation sub-module, used to obtain the vibration signals of the device at different speeds and normalize them, and estimate the speeds corresponding to the vibration signals. A reconstruction sub-module, used to perform SVMD decomposition on the vibration signal to obtain basic mode components, and form a reconstructed signal according to the correlation with the vibration signal. A fault judgment sub-module, used to perform envelope demodulation and order analysis on the reconstructed signal, compare the characteristic frequencies of different parts of the device with the envelope threshold according to the speed, and judge the fault location.

[0015] A computer memory, which stores a computer program executable by a computer processor, and the computer program executes a variable-speed bearing fault diagnosis method based on SVMD and speed estimation.

[0016] The beneficial effects of the present invention are as follows: 1. The variable-speed bearing fault diagnosis method and system based on SVMD and speed estimation of the present invention estimate the bearing speed to obtain accurate fault characteristic values of the bearing, use SVMD to decompose and denoise and reconstruct the vibration signal, perform envelope demodulation and order analysis on the reconstructed signal, compare the characteristic frequencies of different parts of the device with the envelope threshold according to the speed, thereby judging the fault location, realizing the function of accurately identifying the fault location of the bearing, and being able to effectively diagnose bearing faults.

[0017] 2. The present invention estimates the bearing speed through a speed estimation algorithm to obtain accurate fault characteristic values of the bearing, without installing a speed sensor to obtain the speed of a variable-speed device, thereby performing feature extraction and fault diagnosis on the bearing, and is particularly suitable for working conditions where the power frequency changes due to speed changes in variable-speed devices.

[0018] 3. In view of the noise interference and power frequency variation in the vibration signals of variable-speed bearings, the present invention adopts the sequential variational mode decomposition algorithm (SVMD) to extract all intrinsic mode function components (IMFs) in a continuous manner, effectively removing the noise interference in the signals.

[0019] Of course, it is not necessary for any product implementing the present invention to achieve all the above-mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following briefly introduces the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 It is a flowchart of an embodiment of the present invention.

[0022] Figure 2 It is a flowchart of the analysis algorithm of an embodiment of the present invention.

[0023] Figure 3 It is a flowchart of the rotational speed estimation algorithm of an embodiment of the present invention.

[0024] Figure 4 It is the order envelope spectrum diagram at a rotational speed of 900 r / min in an embodiment of the present invention.

[0025] Figure 5 It is the order envelope spectrum diagram at a rotational speed of 1800 r / min in an embodiment of the present invention.

[0026] Figure 6 It is the order envelope spectrum diagram at a rotational speed of 3600 r / min in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the following further elaborates on the present invention in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0028] Embodiment 1 Refer to Figure 1 , the specific steps of a variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation are as follows: S1: Obtain the vibration signals of the device at different rotational speeds and normalize them, and estimate the rotational speeds corresponding to the vibration signals; S2: Perform SVMD decomposition on the vibration signals to obtain basic modal components, and form a reconstructed signal according to the correlation with the vibration signals; S3: Perform envelope demodulation and order analysis on the reconstructed signal, compare the characteristic frequencies of different parts of the device with the envelope threshold according to the rotational speed, and determine the fault location.

[0029] Further, in step S1, the specific steps for estimating the rotational speed corresponding to the vibration signal are as follows: S11: Discretize and perform time-frequency transformation on the vibration signal, and take the first ten-order data; S12: Establish a large loop for assigning the first ten-order data. In each loop, obtain the amplitudes corresponding to 1 times, 0.5 times, 2 times, 3 times, and 4 times (loop count + 5); S13: Establish a small loop for screening (loop count + 5) under the large loop. The loop count of the small loop is less than that of the large loop; S14: Compose the screened (loop count + 5) into the first vector M; traverse the first vector and assign values to the first vector according to conditions to obtain the fourth vector; S15: If the length of the fourth vector is greater than 0, use the mode of the fourth vector as the rotational speed; if the length of the fourth vector is less than or equal to 0, set the rotational speed equal to 25.

[0030] Further, in step S11, the specific steps are as follows: S111: Filter the vibration signal; S112: Perform FFT transformation on the filtered data and take the first half of the data; S113: Arrange the data after FFT transformation from largest to smallest and take the first ten-order data after arrangement.

[0031] Further, in step S12, the specific steps of the large loop are as follows: If the ratio of the first-order data to the second-order data is greater than 5, set the values of the second-order to tenth-order data equal to the first-order data; Or if the ratios of the first-order data, the second-order data, and the third-order data are all greater than 5, find the index positions of the first-order to tenth-order data in the data after FFT transformation: If the number of indices of the first-order data that can be divisible by the indices of the third-order to tenth-order data is greater than the number of indices of the second-order data that can be divisible by the indices of the third-order to tenth-order data, set the values of the second-order to tenth-order data equal to the first-order data; If the number of indices of the first-order data that can be divisible by the indices of the third-order to tenth-order data is less than the number of indices of the second-order data that can be divisible by the indices of the third-order to tenth-order data, set the values of the first-order, third-order to tenth-order data equal to the second-order data.

[0032] Further, in step S13, the specific steps of the small loop are as follows: If the amplitude corresponding to 3 times (number of cycles + 5) is equal to the current top ten-order data, and the ratio of the amplitude corresponding to 3 times (number of cycles + 5) to the amplitude corresponding to 1 time (number of cycles + 5) is less than 5, and the amplitude corresponding to 1 time (number of cycles + 5) is greater than 0.2, record (number of cycles + 5); If the amplitude corresponding to 1 time (number of cycles + 5) is equal to the current top ten-order data, and the amplitude corresponding to 1 time (number of cycles + 5) is greater than 0.2, and the amplitudes corresponding to 0.5 times (number of cycles + 5) and 2 times (number of cycles + 5) are within the range of the top ten-order data, record (number of cycles + 5); If the amplitude corresponding to 1 time (number of cycles + 5) is equal to the current top ten-order data, and the amplitude corresponding to 1 time (number of cycles + 5) is greater than 0.2, and the amplitudes corresponding to 0.5 times (number of cycles + 5) and 4 times (number of cycles + 5) are not within the range of the top ten-order data, record (number of cycles + 5).

[0033] Further, in step S14, the specific steps are as follows: S141: Traverse the first vector. If the absolute difference between 4 times a certain data in the large loop and a certain data in the small loop is less than 2, then create another small loop: Traverse the first vector. If the absolute difference between 2 times a certain data in the large loop and another data in the small loop is less than 2, let the other data in the small loop be equal to the certain data in the large loop; obtain the second vector; S142: Traverse the second vector. If the absolute difference between 3 times a certain data in the large loop and a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; If the absolute difference between 4 times a certain data in the large loop and a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; If the absolute difference between 5 times a certain data in the large loop and a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; obtain the third vector; S143: Traverse the third vector. If the absolute difference between a certain data in the large loop and a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; If a certain data in the large loop is divisible by a certain data in the small loop, let the certain data in the small loop be equal to the certain data in the large loop; If the absolute difference between a certain data in the large loop and 0.5 times a certain data in the small loop is less than 2, let the certain data in the small loop be equal to the certain data in the large loop; obtain the fourth vector.

[0034] In step S2, the specific steps are as follows: S21: Perform SVMD decomposition on the vibration signal to obtain the basic mode components; S22: Normalize each basic modal component; S23: Calculate the correlation coefficient between each vibration signal after normalization and each basic modal component after normalization, and add up the basic modal components with correlation coefficients greater than the preset value to obtain the reconstructed signal.

[0035] In step S3, the specific steps are as follows: S31: Perform envelope demodulation on the reconstructed signal and draw the order envelope spectrum at different rotational speeds; S32: Calculate the order of the characteristic frequency of the device according to the rotational speed; S33: Compare the amplitude of the characteristic frequency of the device in the order envelope spectrum with the envelope threshold: If the amplitude of a certain characteristic frequency exceeds the envelope threshold, the device component corresponding to this characteristic frequency fails.

[0036] In this embodiment, the accurate fault characteristic value of the bearing is obtained by estimating the bearing rotational speed. The SVMD decomposition is used to denoise and reconstruct the vibration signal. Envelope demodulation and order analysis are performed on the reconstructed signal. The characteristic frequencies of different parts of the device are compared with the envelope threshold according to the rotational speed, so as to judge the fault location, realizing the function of accurately identifying the fault location of the bearing and effectively diagnosing the bearing fault.

[0037] Embodiment 2 The steps of this embodiment are the same as those of Embodiment 1, except that each step is applied to a specific example. Figure 2 The flowchart of the analysis algorithm for this embodiment is as follows. Specifically, it includes the following steps: S1: Obtain the original acceleration vibration signal data x1, x2,..., xn of the variable-speed device at different rotational speeds; normalize the vibration signal X composed of the original acceleration vibration signal data to obtain X_rec; estimate the rotational speed rpm of the vibration signal X by using the rotational speed estimation algorithm; Figure 3 The flowchart of the rotational speed estimation algorithm for this embodiment. The specific steps for estimating the rotational speed corresponding to the vibration signal are as follows: S11: Discretize and perform time-frequency transformation on the vibration signal X, and take the first ten-order data X(x1, x2,..., x 10 ); The specific steps are as follows: S111: Filter the original acceleration vibration signal data per second, and the filtering frequency is [5, 215]; S112: Perform FFT transformation on the filtered data and take the first half of the data; S113: Arrange the data after FFT transformation from large to small; take the first ten-order data X(x1, x2,..., x 10 ); S12: Establish a large loop for the first ten-order data before assignment, with the loop count i ranging from 0 to 65; in each loop, obtain the amplitudes A, 0.5A, 2A, 3A, and 4A corresponding to 5 + i, 0.5*(5 + i), 2*(5 + i), 3*(5 + i), and 4*(5 + i) on the spectrogram; the specific steps of the large loop are as follows: If the ratio of x1 to x2 is greater than 5, then set the values of x2, …, x 10 to be equal to x1. Or if the ratios of x1, x2 to x3 are all greater than 5, find the index positions of x1, x2, …, x 10 in the data after FFT transformation. If the number of indices of x1 that can be divisible by the indices of x3, x4, …, x 10 is greater than the number of indices of x2 that can be divisible by the indices of x3, x4, …, x 10 then set the values of x2, …, x 10 to be equal to x1; conversely, if the number of indices of x1 that can be divisible by the indices of x3, x4, …, x 10 is less than the number of indices of x2 that can be divisible by the indices of x3, x4, …, x 10 then set the values of x1, …, x 10 to be equal to x2. S13: Establish a small loop for screening (5 + i) under the large loop, with the loop count j ranging from 1 to 10; the specific steps of the small loop are as follows: If 3A is equal to X(j), and the ratio of 3A to A is less than 5 and A is greater than 0.2, record 5 + i. If A is equal to X(j), and A is greater than 0.2 and 0.5A, 2A are within the range of X, record 5 + i. If A is equal to X(j), and A is greater than 0.2 and 0.5A, 4A are not within the range of X, record 5 + i. S14: Compose the (5 + i) recorded in step S13 into vector M; traverse the first vector and assign values to the first vector according to the conditions to obtain the fourth vector; the specific steps are as follows: S141: Traverse vector M. If the absolute difference between 4 times one data M(k) under the large loop and the next data M(l) under the small loop is less than 2, then establish another small loop, traverse vector M. If the absolute difference between 2 times M(k) and the next data M(h) under the small loop is less than 2, set M(h) equal to M(k); obtain the new vector M1. S142: Traverse vector M1. If the absolute difference between three times of one data M(kk) in the outer loop and the next data M(ll) in the inner loop is less than 2, let M(ll) be equal to M(kk); if the absolute difference between four times of one data M(kk) in the outer loop and the next data M(ll) in the inner loop is less than 2, let M(ll) be equal to M(kk); if the absolute difference between five times of one data M(kk) in the outer loop and the next data M(ll) in the inner loop is less than 2, let M(ll) be equal to M(kk); obtain the new vector M2; S143: Traverse vector M2. If the absolute difference between one data M(kkk) in the outer loop and the next data M(lll) in the inner loop is less than 2, let M(lll) be equal to M(kkk); if one data M(kkk) in the outer loop is divisible by the next data M(lll) in the inner loop, let M(lll) be equal to M(kkk); if the absolute difference between one data M(kkk) in the outer loop and 0.5 times of the next data M(lll) in the inner loop is less than 2, let M(lll) be equal to M(kkk); thus obtain the new vector N; S15: If the length of vector N is greater than 0, find the mode of vector N to obtain r; otherwise r is equal to 25; the rotational speed is 60 * r.

[0038] S2: Perform SVMD decomposition on the vibration signal X to obtain the basic mode components IMF, and form the reconstructed signal Y according to the correlation with the vibration signal X; the specific steps are as follows: S21: Perform SVMD decomposition on the vibration signal X to obtain the basic mode components IMF; S22: Normalize each basic mode component to obtain IMF_rec; S23: Calculate the correlation coefficient between X_rec and each normalized basic mode component IMF_rec, and add the basic mode components IMF with a correlation coefficient greater than 0.3 to obtain the reconstructed signal Y; S3: Perform envelope demodulation and order analysis on the reconstructed signal Y, compare the characteristic frequencies of different parts of the device with the envelope threshold according to the rotational speed, and judge the fault location; the specific steps are as follows: S31: Perform envelope demodulation on the reconstructed signal Y, and then draw the order envelope spectrum at different rotational speeds according to the rotational speed; S32: Calculate the order of the bearing characteristic frequency according to the rotational speed rpm; S33: Check the amplitude size of the bearing characteristic frequency in the order envelope spectrum. If the amplitude of a certain characteristic frequency exceeds the envelope threshold, it is judged that the bearing component corresponding to this characteristic frequency has a fault.

[0039] Example 3 The steps of this embodiment are the same as those of Embodiment 1, except that each step is applied to a specific example: According to steps S11 - S15, the rotational speed of the variable - speed equipment acceleration signal is estimated, and rotational speeds of 900 r / min, 1800 r / min, and 3600 r / min are obtained.

[0040] Normalize the original signal according to step S1.

[0041] Set the maximum balance parameter , the double - rising step size is 0, and the tolerance of the convergence criterion is 10 -6 , decompose the acceleration signal by SVMD according to step S21 to obtain a series of intrinsic mode functions IMF.

[0042] Normalize the intrinsic mode functions IMF according to step S22.

[0043] Calculate the correlation coefficient between the normalized acceleration signal and the normalized intrinsic mode functions. Add the intrinsic mode functions with a correlation coefficient greater than 0.3 to obtain a reconstructed signal.

[0044] Perform envelope demodulation on the reconstructed signal according to step S31, and draw the order envelope spectra at different rotational speeds, as Figure 4 , Figure 5 and Figure 6 shown.

[0045] Calculate the orders of the bearing characteristic frequencies at different rotational speeds according to step S32, as shown in Table 1.

[0046] The selected bearing model is 6314 deep - groove ball bearing, and the bearing parameters are: the number of rolling elements , the diameter of the rolling element , the pitch diameter of the bearing , the contact angle .

[0047] Table 1 Bearing characteristic frequency orders

[0048] As Figure 4 , Figure 5 and Figure 6 shown, according to step S33, substitute the bearing characteristic frequencies into the envelope spectrum to find the amplitudes of the corresponding frequencies. It is found by experience that the amplitude of the 4.9 order is relatively prominent, and thus it can be preliminarily judged that there is a fault in the inner ring of the bearing.

[0049] The verification results show that this embodiment can accurately identify the fault location of the bearing, and thus can effectively diagnose the bearing fault.

[0050] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not imply the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0051] Embodiment 4 This embodiment is used to implement the principle of the above method embodiment to construct a variable-speed bearing fault diagnosis system based on SVMD and speed estimation, including a speed estimation sub-module, a reconstruction sub-module, and a fault judgment sub-module.

[0052] The speed estimation sub-module is used to obtain the vibration signals of the device at different speeds and normalize them, and estimate the speeds corresponding to the vibration signals; The reconstruction sub-module is used to perform SVMD decomposition on the vibration signals to obtain the basic mode components, and form a reconstructed signal according to the correlation with the vibration signals; The fault judgment sub-module is used to perform envelope demodulation and order analysis on the reconstructed signal, and compare the characteristic frequencies of different parts of the device with the envelope threshold according to the speed to judge the fault location.

[0053] Each sub-module is mainly used to implement each step of the method embodiment, which will not be elaborated here.

[0054] It should be noted that according to the needs of implementation, each step / component described in the present application can be split into more steps / components, or two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0055] This embodiment further includes a processor, a communication interface, a memory, and a communication bus; wherein the processor, the communication interface, and the memory complete communication with each other through the communication bus; a computer program is stored in the memory, and when the program is executed by the processor, the processor executes the steps of a variable-speed bearing fault diagnosis method based on SVMD and speed estimation.

[0056] This embodiment further provides a computer-readable storage medium, on which executable instructions are stored, and when the instructions are executed by the processor, the processor implements a variable-speed bearing fault diagnosis method based on SVMD and speed estimation.

[0057] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects.

[0058] Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0059] The present application is described with reference to the flowchart of the method and computer program product according to Embodiment 1 of the present application and the block diagram of the device (system) according to Embodiment 3. It should be understood that each process or block in the flowchart or block diagram, as well as the combination of processes or blocks in the flowchart or block diagram, can be implemented by computer program instructions.

[0060] These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices produce a variable-speed bearing fault diagnosis system based on SVMD and speed estimation for implementing the functions specified in one Figure 1 one process or multiple processes or blocks Figure 1 one block or multiple blocks.

[0061] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured article including an instruction device, and the instruction device implements the functions specified in one Figure 1 one process or multiple processes or blocks Figure 1 one block or multiple blocks.

[0062] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, so that a series of operation steps are executed on the computer or other programmable devices to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable devices provide the steps of a variable-speed bearing fault diagnosis method based on SVMD and speed estimation for implementing the functions specified in one Figure 1 one process or multiple processes or blocks Figure 1 one block or multiple blocks.

[0063] The above embodiments are only used to illustrate the design concept and characteristics of the present invention, and the purpose is to enable those skilled in the art to understand the content of the present invention and implement it accordingly. The protection scope of the present invention is not limited to the above embodiments. Therefore, all equivalent changes or modifications made according to the principles and design concepts disclosed by the present invention are within the protection scope of the present invention.

Claims

1. A variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation, characterized in that: It includes the following steps: S1: Obtain the vibration signals of the device at different rotational speeds and normalize them, and estimate the rotational speed corresponding to the vibration signals; S2: Perform SVMD decomposition on the vibration signals to obtain basic mode components, and form a reconstructed signal according to the correlation with the vibration signals; S3: Perform envelope demodulation and order analysis on the reconstructed signal, compare the characteristic frequencies of different parts of the device with the envelope threshold according to the rotational speed, and judge the fault location.

2. The variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation according to claim 1, characterized in that: In the step S1, the specific steps for estimating the rotational speed corresponding to the vibration signals are as follows: S11: Discretize and perform time-frequency transformation on the vibration signals, and take the first ten-order data; S12: Establish a large loop for assigning values to the first ten-order data. Each time the loop is executed, obtain the amplitudes corresponding to 1 times, 0.5 times, 2 times, 3 times, and 4 times (loop count + 5); S13: Establish a small loop for screening (loop count + 5) under the large loop. The loop count of the small loop is less than that of the large loop; S14: Compose the screened (loop count + 5) into the first vector M; traverse the first vector and assign values to the first vector according to the conditions to obtain the fourth vector; S15: If the length of the fourth vector is greater than 0, use the mode of the fourth vector as the rotational speed; if the length of the fourth vector is less than or equal to 0, set the rotational speed equal to 25.

3. A variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation according to claim 2, characterized in that: In the step S11, the specific steps are as follows: S111: Filter the vibration signals; S112: Perform FFT transformation on the filtered data and take the first half of the data; S113: Arrange the data after FFT transformation from large to small and take the first ten-order data after arrangement.

4. A variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation according to claim 2, characterized in that: In the step S12, the specific steps of the large loop are as follows: If the ratio of the first-order data to the second-order data is greater than 5, set the values of the second-order to tenth-order data equal to the first-order data; Or if the ratios of the first-order data, the second-order data, and the third-order data are all greater than 5, find the index positions of the first-order to tenth-order data in the data after FFT transformation: If the number of the first-order data that can be divisible by the indices of the third-order to tenth-order data is greater than the number of the second-order data that can be divisible by the indices of the third-order to tenth-order data, set the values of the second-order to tenth-order data equal to the first-order data; If the number of the first-order data that can be divisible by the indices of the third-order to tenth-order data is less than the number of the second-order data that can be divisible by the indices of the third-order to tenth-order data, set the values of the first-order, third-order to tenth-order data equal to the second-order data.

5. A variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation according to claim 2, characterized in that: In the step S13, the specific steps of the small loop are as follows: If the amplitude corresponding to 3 times (loop count + 5) is equal to the current first ten-order data, and the ratio of the amplitude corresponding to 3 times (loop count + 5) to the amplitude corresponding to 1 time (loop count + 5) is less than 5, and the amplitude corresponding to 1 time (loop count + 5) is greater than 0.2, record (loop count + 5); If the amplitude corresponding to 1 time (loop count + 5) is equal to the current first ten-order data, and the amplitude corresponding to 1 time (loop count + 5) is greater than 0.2, and the amplitudes corresponding to 0.5 times (loop count + 5) and 2 times (loop count + 5) are within the range of the first ten-order data, record (loop count + 5); If the amplitude corresponding to 1 times (cycle number + 5) is equal to the current top ten order data, and the amplitude corresponding to 1 times (cycle number + 5) is greater than 0.2, and the amplitudes corresponding to 0.5 times (cycle number + 5) and 4 times (cycle number + 5) are not within the range of the top ten order data, record (cycle number + 5).

6. The variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation according to claim 2, wherein: In the step S14 described above, the specific steps are as follows: S141: Traverse the first vector. If the absolute difference between 4 times a certain data in the large loop and a certain data in the small loop is less than 2, then create another small loop: Traverse the first vector. If the absolute difference between 2 times a certain data in the large loop and another data in the small loop is less than 2, make the other data in the small loop equal to the certain data in the large loop; obtain the second vector; S142: Traverse the second vector. If the absolute difference between 3 times a certain data in the large loop and a certain data in the small loop is less than 2, make the certain data in the small loop equal to the certain data in the large loop; If the absolute difference between 4 times a certain data in the large loop and a certain data in the small loop is less than 2, make the certain data in the small loop equal to the certain data in the large loop; If the absolute difference between 5 times a certain data in the large loop and a certain data in the small loop is less than 2, make the certain data in the small loop equal to the certain data in the large loop; obtain the third vector; S143: Traverse the third vector. If the absolute difference between a certain data in the large loop and a certain data in the small loop is less than 2, make the certain data in the small loop equal to the certain data in the large loop; If a certain data in the large loop is divisible by a certain data in the small loop, make the certain data in the small loop equal to the certain data in the large loop; If the absolute difference between a certain data in the large loop and 0.5 times a certain data in the small loop is less than 2, make the certain data in the small loop equal to the certain data in the large loop; obtain the fourth vector.

7. A variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation according to claim 1, characterized in that: In the step S2 described above, the specific steps are as follows: S21: Perform SVMD decomposition on the vibration signal to obtain the basic mode components; S22: Normalize each basic mode component; S23: Calculate the correlation coefficient between each normalized vibration signal and each normalized basic mode component, and add the basic mode components with the correlation coefficient greater than the preset value to obtain the reconstructed signal.

8. A variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation according to claim 1, characterized in that: In the step S3 described above, the specific steps are as follows: S31: Perform envelope demodulation on the reconstructed signal and draw the order envelope spectrum at different rotational speeds; S32: Calculate the order of the characteristic frequency of the equipment according to the rotational speed; S33: Compare the amplitude of the characteristic frequency of the equipment in the order envelope spectrum with the envelope threshold: If the amplitude of a certain characteristic frequency exceeds the envelope threshold, the equipment component corresponding to this characteristic frequency has a fault.

9. A variable-speed bearing fault diagnosis system based on SVMD and rotational speed estimation, characterized in that: A rotational speed estimation sub-module, configured to obtain the vibration signals of the equipment at different rotational speeds and normalize them, and estimate the rotational speeds corresponding to the vibration signals; A reconstruction sub-module, configured to perform SVMD decomposition on the vibration signals to obtain the basic mode components, and form a reconstructed signal according to the correlation with the vibration signals; A fault judgment sub-module, configured to perform envelope demodulation and order analysis on the reconstructed signal, and compare the characteristic frequencies of different parts of the equipment with the envelope threshold according to the rotational speed to judge the fault location.

10. A computer memory, characterized in that: It stores a computer program that can be executed by a computer processor, and the computer program executes a variable-speed bearing fault diagnosis method based on SVMD and rotational speed estimation as described in any one of claims 1 to 8.

Images