Sliding bearing oil film vortex motion fault feature extraction method based on synchronous energy entropy

Abstract

The invention discloses a sliding bearing oil film vortex motion fault feature extraction method based on synchronous energy entropy, and the method comprises the steps: (1), selecting the type and installation position of a sensor, and determining related parameters; (2) acquiring synchronous vibration data; (3) carrying out standardization processing on the synchronous vibration data; step (4), reconstructing the synchronous vibration data after standardization processing into multi-dimensional synchronous vibration data; step (5), centralized processing of each dimension of synchronous vibration data is carried out; (6) calculating a first synchronous energy entropy; (7) performing subtraction processing on two adjacent columns of data of the multi-dimensional synchronous vibration data; (8) calculating a second synchronous energy entropy; and (9) taking the two synchronous energy entropies as two fault features. By extracting the two synchronous energy entropies, the sliding bearing oil film vortex motion fault can be reliably depicted, and the method is low in calculation cost and can be used for real-time fault detection and diagnosis.

Description

Technical Field

[0001] The present invention belongs to the technical field of rotating mechanical system fault detection, and in particular relates to a sliding bearing fault feature extraction method based on synchronous energy entropy (SEE). Background Art

[0002] Sliding bearings are key components in rotating machinery systems, such as engines and generators. Their health directly impacts the performance of these systems and can even cause equipment downtime. To ensure the safe operation of rotating machinery systems using sliding bearings, real-time monitoring of their health is essential.

[0003] During the operation of sliding bearings, a highly harmful self-excited vibration often occurs, namely oil film vortex. The vibration frequency of oil film vortex is about half the rotational frequency of the shaft, so oil film vortex is also called half-speed vortex.

[0004] Currently, researchers have proposed various entropy concepts, such as approximate entropy, sample entropy, fuzzy entropy, permutation entropy, and energy entropy. Based on these entropy methods, various rotating machinery fault feature extraction and fault detection methods have been proposed. Patent No. 2023103945416 discloses a fault feature extraction method based on multi-scale permutation entropy and a kurtosis value fusion factor. While this method can extract rolling bearing fault information, it may not be able to extract oil film vortex fault information in sliding bearings. Patent No. 2021110169321 discloses a rolling bearing operating condition assessment method based on EEMD energy entropy. This method uses EEMD energy entropy as a fault signature to characterize rolling bearing faults. While this method can identify bearing fault states, its computational cost is relatively high, making it unsuitable for real-time detection of high-speed bearing faults. Patent No. 2024110253664 discloses a bearing fault diagnosis method based on wavelet packet energy entropy, which is used to extract bearing fault features. The calculation of wavelet packet energy entropy requires signal processing using both ensemble empirical mode decomposition (EEMD) and wavelet packet decomposition. Wavelet packet decomposition requires the selection of appropriate wavelet basis functions. Inappropriate wavelet basis functions not only reduce the fault diagnosis accuracy of this method but can also lead to erroneous diagnostic results. Furthermore, this method is computationally expensive, making it unsuitable for real-time detection of high-speed bearing faults.

[0005] Considering that the frequency of oil film vortex faults is generally half the shaft rotational frequency, this patent proposes a method for extracting the characteristics of oil film vortex faults in sliding bearings based on synchronous energy entropy. This method not only reliably extracts the characteristics of oil film vortex faults in sliding bearings, but also has low computational cost and can be used for real-time fault detection and diagnosis. Summary of the Invention

[0006] To address the above problems, the present invention proposes a method for extracting the characteristics of sliding bearing oil film vortex faults based on synchronous energy entropy. By extracting two synchronous energy entropies, this method can reliably characterize the sliding bearing oil film vortex fault. This method has low computational cost and can be used for real-time fault detection and diagnosis.

[0007] The technical solution adopted by the present invention is: a method for extracting the characteristics of sliding bearing oil film vortex fault based on synchronous energy entropy, comprising:

[0008] Step (1): Set the number of synchronous sampling points M and the number of shaft revolutions N per shaft revolution;

[0009] Step (2): using a vibration sensor to obtain synchronous vibration data X;

[0010] Step (3): standardize the synchronous vibration data X to obtain the synchronous vibration data Y;

[0011] Step (4): reconstructing the standardized synchronous vibration data Y into M-dimensional synchronous vibration data Z according to the number of synchronous sampling points M per rotation of the shaft;

[0012] Step (5): using a zero-centering method to perform zero-centering on each N-dimensional vector in the M-dimensional synchronous vibration data Z, and obtaining the zero-centering M-dimensional synchronous data Z1;

[0013] Step (6): Calculate the energy of each row of data in the M-dimensional synchronous vibration data Z1 to obtain energy data E and relative energy data According to the relative energy data Calculate the first synchronization energy entropy SEE1;

[0014] Step (7): For the M-dimensional synchronous vibration data Z, subtract the data in the i-th column from the data in the i+1th column or subtract the data in the i-th column from the data in the i+1th column to obtain data D;

[0015] Step (8): Calculate the energy of each row of data in data D and obtain energy data S and relative energy data According to the relative energy data Calculate the second synchronization energy entropy SEE2;

[0016] Step (9): Consider the two synchronous energy entropies as two fault signatures.

[0017] Furthermore, in step (1), the sensors are a vibration sensor and a speed sensor; the vibration sensor can be a vibration displacement sensor, a vibration velocity sensor or a vibration acceleration sensor; the vibration sensor is installed near the bearing to measure the vibration of the bearing; the speed sensor is installed near the shaft to measure the shaft speed; the relevant parameters include: the number of synchronous sampling points M per shaft revolution and the number of shaft revolutions N.

[0018] Furthermore, the implementation process of step (2) is as follows: using the shaft speed signal as a reference signal, according to the number of synchronous sampling points M and the shaft rotation number N set in step (1), a vibration sensor is used to obtain synchronous vibration data X,

[0019] X={x(1),x(2),x(3)...x(MN)}.

[0020] Furthermore, the implementation process of step (3) is as follows:

[0021] The Z-score standardization method is used to standardize the synchronous vibration data X to obtain the standardized synchronous vibration data Y.

[0022]

[0023] Where μ represents the mean of the synchronous vibration data X, and σ represents the standard deviation of the synchronous vibration data X.

[0024] Furthermore, the implementation process of step (4) is as follows:

[0025] According to the number of synchronous sampling points M per rotation of the shaft, the standardized synchronous vibration data Y is reconstructed into M-dimensional synchronous vibration data Z. The reconstructed M-dimensional synchronous data Z is shown as follows:

[0026]

[0027] To simplify writing, Then the M-dimensional synchronous vibration data Z can be expressed as follows:

[0028]

[0029] Furthermore, the implementation process of step (5) is as follows:

[0030] The zero-centering method is used to perform zero-centering on each N-dimensional vector in the M-dimensional synchronous vibration data Z. The M-dimensional synchronous data Z1 after centering is as follows:

[0031]

[0032] Among them, μ i(i=1,2,3,…,M) is the average value of the i-th row data in the M-dimensional synchronous vibration data Z1,

[0033]

[0034] Furthermore, the implementation process of step (6) is as follows:

[0035] Calculate the energy of each row of data in the M-dimensional synchronous vibration data Z1 and obtain the energy data E and relative energy data

[0036] E={e1 e2 e3…e i e M},

[0037]

[0038] Among them, e i Represents the energy of the i-th row of data in the M-dimensional synchronous vibration data Z1, and also represents the mean square value of the i-th row of data in the M-dimensional synchronous vibration data Z1; represents the relative energy of the i-th row of data in the M-dimensional synchronous vibration data Z1,

[0039]

[0040] According to the relative energy data Calculate the first synchronous energy entropy SEE1,

[0041]

[0042] Furthermore, the implementation process of step (7) is as follows:

[0043] For the M-dimensional synchronous vibration data Z, the data in the i+1th column is subtracted from the data in the ith column to obtain the data D.

[0044]

[0045] In the process of calculating the data D, the data in the i+1th column may be subtracted from the data in the i-th column.

[0046] Furthermore, the implementation process of step (8) is as follows:

[0047] Calculate the energy of each row of data in data D respectively to obtain energy data S and relative energy data

[0048] S={s1 s2 s3…s i s M},

[0049]

[0050] Among them, s i Represents the energy of the i-th row of data in data D, and also represents the mean square value of the i-th row of data in data D; Represents the relative energy of the i-th row of data in data D,

[0051]

[0052] According to the relative energy data Calculate the second synchronization energy entropy SEE2,

[0053]

[0054] Furthermore, the implementation process of step (9) is as follows:

[0055] The two calculated synchronous energy entropies SEE1 and SEE2 are regarded as two fault characteristics of the sliding bearing.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] (1) The extraction method of the present invention can extract two fault features, and the extracted fault features can reliably characterize the oil film vortex fault of the sliding bearing.

[0058] (2) The extraction method of the present invention has low computational cost and can be applied to real-time detection of high-speed sliding bearing faults.

[0059] (3) The extraction method of the present invention requires little expert knowledge and only requires two sampling parameters to be set in advance.

[0060] (4) The extraction method of the present invention is not affected by changes in rotational speed and can be applied to sliding bearing fault detection at various rotational speeds. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a flow chart of the method for extracting the characteristics of sliding bearing oil film vortex fault based on synchronous energy entropy disclosed by the present invention;

[0062] Figure 2 This is the synchronous vibration data of the sliding bearing when it is healthy in an embodiment of the present invention;

[0063] Figure 3 The synchronous vibration data of the sliding bearing oil film vortex fault in the embodiment of the present invention;

[0064] Figure 4 This is the standardized synchronous vibration data of the sliding bearing when it is healthy in the embodiment of the present invention;

[0065] Figure 5This is the standardized synchronous vibration data of the sliding bearing oil film vortex fault in the embodiment of the present invention;

[0066] Figure 6 This is the first fault feature SEE1 graph extracted by the method of the present invention;

[0067] Figure 7 This is the second fault feature SEE2 graph extracted by the method of the present invention;

[0068] Figure 8 The SEE1 and SEE2 distribution diagrams of the sliding bearing when healthy and faulty are extracted by the method of the present invention. DETAILED DESCRIPTION

[0069] For a better understanding of the present invention, the present invention is further described below with reference to the accompanying drawings in the examples of the present invention, but is not intended to limit the present invention. Various modifications and improvements made to the technical solution of the present invention by ordinary persons in the art without departing from the design concept of the present invention should fall within the scope of protection of the present invention.

[0070] A method for extracting the characteristics of oil film vortex fault in sliding bearings based on synchronous energy entropy, the process is as follows Figure 1 As shown, the specific steps include:

[0071] Step (1): Select the sensor type and installation location, and determine the relevant parameters;

[0072] Step (2): Acquire synchronous vibration data;

[0073] Step (3): standardize the synchronous vibration data;

[0074] Step (4): reconstructing the standardized synchronous vibration data into multi-dimensional synchronous vibration data;

[0075] Step (5): separately centralize and process the synchronous vibration data of each dimension;

[0076] Step (6): Calculate the first synchronization energy entropy;

[0077] Step (7): performing subtraction processing on two adjacent columns of multi-dimensional synchronous vibration data;

[0078] Step (8): Calculate the second synchronization energy entropy;

[0079] Step (9): Consider the two synchronous energy entropies as two fault signatures.

[0080] Furthermore, in step (1), the sensors are a vibration sensor and a speed sensor; the vibration sensor may be a vibration displacement sensor, a vibration velocity sensor, or a vibration acceleration sensor; the vibration sensor is used to measure the vibration of the bearing; the speed sensor is used to measure the speed of the rotating shaft, such as the installation position and method disclosed in CN 110186510 B; the relevant parameters include: the number of synchronous sampling points M per rotation of the rotating shaft and the number of rotations N of the rotating shaft.

[0081] In this embodiment, the number M of synchronous sampling points per rotation of the rotating shaft is set to 100, and the number N of rotations of the rotating shaft is set to 10.

[0082] Furthermore, the implementation process of step (2) is as follows: using the shaft speed signal as a reference signal, according to the number of synchronous sampling points M and the shaft rotation number N set in step (1), a vibration sensor is used to obtain synchronous vibration data X,

[0083] X={x(1),x(2),x(3)...x(MN)}.

[0084] In this embodiment, the synchronous vibration data obtained is as follows: Figure 2 and 3 shown. Figure 2 shows the synchronous vibration data of the plain bearing when it is healthy, and Figure 3 The synchronous vibration data of a sliding bearing with oil film whirl failure is shown. It is difficult to distinguish the healthy and faulty states of the bearing from the synchronous vibration data.

[0085] Furthermore, the implementation process of step (3) is as follows:

[0086] The Z-score standardization method is used to standardize the synchronous vibration data X to obtain the standardized synchronous vibration data Y.

[0087]

[0088] Where μ represents the mean of the synchronous vibration data X, and σ represents the standard deviation of the synchronous vibration data X.

[0089] The synchronous vibration data after normalization are as follows: Figure 4 and 5 As shown in the figure, we can see that the two sets of data are unified to the same scale and have a mean of zero.

[0090] Furthermore, the implementation process of step (4) is as follows:

[0091] According to the number of synchronous sampling points M per rotation of the shaft, the standardized synchronous vibration data Y is reconstructed into M-dimensional synchronous vibration data Z. The reconstructed M-dimensional synchronous vibration data Z is shown as follows:

[0092]

[0093] To simplify writing, Then the M-dimensional synchronous vibration data Z can be expressed as follows:

[0094]

[0095] Furthermore, the implementation process of step (5) is as follows:

[0096] The zero-centering method is used to perform zero-centering on each N-dimensional vector in the M-dimensional synchronous vibration data Z. The M-dimensional synchronous data Z1 after zero-centering is as follows:

[0097]

[0098] Among them, μ i (i=1,2,3,…,M) is the average value of the i-th row data in the M-dimensional synchronous vibration data Z1,

[0099]

[0100] Furthermore, the implementation process of step (6) is as follows:

[0101] Calculate the energy of each row of data in the M-dimensional synchronous vibration data Z1 and obtain the energy data E and relative energy data

[0102] E={e1 e2 e3…e i …e M},

[0103]

[0104] Among them, e i Represents the energy of the i-th row of data in the M-dimensional synchronous vibration data Z1, and also represents the mean square value of the i-th row of data in the M-dimensional synchronous vibration data Z1; represents the relative energy of the i-th row of data in the M-dimensional synchronous vibration data Z1,

[0105]

[0106] According to the relative energy data Calculate the first synchronous energy entropy SEE1,

[0107]

[0108] Furthermore, the implementation process of step (7) is as follows:

[0109] For the M-dimensional synchronous vibration data Z, the data in the i+1th column is subtracted from the data in the ith column to obtain the data D.

[0110]

[0111] In the process of calculating the data D, the data in the i+1th column may be subtracted from the data in the i-th column.

[0112] Furthermore, the implementation process of step (8) is as follows:

[0113] Calculate the energy of each row of data in data D respectively to obtain energy data S and relative energy data

[0114] S={s1 s2 s3…s i …s M},

[0115]

[0116] Among them, s i Represents the energy of the i-th row of data in data D, and also represents the mean square value of the i-th row of data in data D; Represents the relative energy of the i-th row of data in data D,

[0117]

[0118] According to the relative energy data Calculate the second synchronization energy entropy SEE2,

[0119]

[0120] Furthermore, the implementation process of step (9) is as follows:

[0121] The two calculated synchronous energy entropies SEE1 and SEE2 are regarded as two fault characteristics of the sliding bearing.

[0122] When applied, the two fault characteristics are compared with the thresholds of the fault state and the healthy state to determine whether it is in a healthy or faulty state.

[0123] In the embodiment, the two fault features SEE1 and SEE2 extracted by the method of the present invention are respectively as follows: Figure 6 and Figure 7 As shown in the figure, it can be seen that both synchronous energy entropies SEE1 and SEE2 can well distinguish the healthy and faulty states of the sliding bearing. Figure 8The SEE1 and SEE2 distributions for healthy and faulty sliding bearings, extracted using the method of the present invention, are shown. As can be seen from the figures, healthy and faulty sliding bearing states are easily distinguishable. The two synchronized energy entropies, SEE1 and SEE2, extracted using the method of the present invention, can effectively characterize oil film vortex faults in sliding bearings.

[0124] The method for extracting sliding bearing oil film vortex fault features based on synchronized energy entropy, as disclosed in this invention, can reliably extract sliding bearing oil film vortex fault features. Furthermore, we tested the computational cost of this method on a desktop computer (Intel Core i7-9700 3.00GHz CPU), and the time required to extract two fault features was approximately 1.16 milliseconds. This method has low computational cost and can be applied to real-time fault detection.

[0125] The above embodiments are intended only to illustrate the technical concepts and features of the present invention. Their purpose is to enable those skilled in the art to understand the contents of the present invention and implement them accordingly. They are not intended to limit the scope of protection of the present invention. Any equivalent changes or modifications made in accordance with the spirit of the present invention shall be covered by the present invention.

Claims

1. A method for extracting the characteristics of oil film vortex fault of sliding bearings based on synchronous energy entropy, characterized in that: include: Step (1): Set the number of synchronous sampling points M and the number of shaft revolutions N per shaft revolution; Step (2): using a vibration sensor to obtain synchronous vibration data X; Step (3): standardize the synchronous vibration data X to obtain the synchronous vibration data Y; Step (4): reconstructing the standardized synchronous vibration data Y into M-dimensional synchronous vibration data Z according to the number of synchronous sampling points M per rotation of the shaft; Step (5): using a zero-centering method to perform zero-centering on each N-dimensional vector in the M-dimensional synchronous vibration data Z, and obtaining the zero-centering M-dimensional synchronous data Z1; Step (6): Calculate the energy of each row of data in the M-dimensional synchronous vibration data Z1 to obtain energy data E and relative energy data According to the relative energy data Calculate the first synchronization energy entropy SEE1; Step (7): For the M-dimensional synchronous vibration data Z, subtract the data in the i-th column from the data in the i+1th column or subtract the data in the i-th column from the data in the i+1th column to obtain data D; Step (8): Calculate the energy of each row of data in data D and obtain energy data S and relative energy data According to the relative energy data Calculate the second synchronization energy entropy SEE2; Step (9): Consider the two synchronous energy entropies as two fault signatures.

2. The method for extracting the characteristics of sliding bearing oil film vortex fault based on synchronous energy entropy according to claim 1 is characterized in that: In step (2), synchronize the vibration data X: X={x(1),x(2),x(3)...x(MN)}.

3. The method for extracting the characteristics of sliding bearing oil film vortex fault based on synchronous energy entropy according to claim 1, characterized in that: Synchronous vibration data Y after normalization: Where μ represents the mean of the synchronous vibration data X, and σ represents the standard deviation of the synchronous vibration data X.

4. The method for extracting the characteristics of sliding bearing oil film vortex fault based on synchronous energy entropy according to claim 1, characterized in that: The reconstructed M-dimensional synchronous vibration data Z: make Then the M-dimensional synchronous vibration data Z can be expressed as follows:

5. The method for extracting the characteristics of sliding bearing oil film vortex fault based on synchronous energy entropy according to claim 1, characterized in that: M-dimensional synchronized data Z1 after centralization: Among them, μ i (i=1,2,3,…,M) is the average value of the i-th row data in the M-dimensional synchronous vibration data Z1, 6. The method for extracting the characteristics of sliding bearing oil film vortex fault based on synchronous energy entropy according to claim 1, characterized in that: Energy data E and relative energy data And={e1 e2 e3…and i and M }, Among them, e i Represents the energy of the i-th row of data in the M-dimensional synchronous vibration data Z1, and also represents the mean square value of the i-th row of data in the M-dimensional synchronous vibration data Z1; represents the relative energy of the i-th row of data in the M-dimensional synchronous vibration data Z1, According to the relative energy data Calculate the first synchronous energy entropy SEE1, 7. The method for extracting the characteristics of sliding bearing oil film vortex fault based on synchronous energy entropy according to claim 1, characterized in that: Data D:

8. The method for extracting the characteristics of sliding bearing oil film vortex fault based on synchronous energy entropy according to claim 1, characterized in that: Energy data S and relative energy data S={s1 s2 s3…s i s M }, Among them, s i Represents the energy of the i-th row of data in data D, and also represents the mean square value of the i-th row of data in data D; Represents the relative energy of the i-th row of data in data D, According to the relative energy data Calculate the second synchronization energy entropy SEE2,

Images