Rolling bearing lubrication state detection method
By collecting and processing vibration acceleration signals at different speeds of bearing seats and calculating lubrication status characterization indicators, the problem of poor lubrication status detection effect in the prior art is solved, efficient and accurate lubrication status judgment is achieved, and maintenance time is reduced.
Patent Information
- Application Number
- CN202411057711.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-22
AI Technical Summary
The existing bearing lubrication status detection methods have problems such as waste of lubricant, poor economic and timeliness, or excessive time-consuming and limited accuracy.
By collecting the vibration acceleration signals of bearing seats at different speeds, performing discrete Fourier transform and center frequency calculations, calculate bearing lubrication status characterization indicators, judge the lubrication status, and reduce maintenance time.
Accurate and efficient judgment of bearing lubrication status, guide reasonable lubricant filling, and improve the economic and efficiency of inspection.
Smart Images

Figure CN119000072A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of rolling bearing maintenance, and in particular relates to a rolling bearing lubrication state detection method. Background Art
[0002] Currently, there are generally two ways to detect the lubrication status of bearings: the first method requires regularly releasing the lubricating oil in the bearing seat and determining the status of the lubricating oil by performing quality inspection on the lubricating oil. When the viscosity of the lubricating oil decreases or there are too many impurities, the bearing needs to be refilled with lubricating oil; the second method uses a temperature sensor to collect the temperature signal of the bearing seat and uses the temperature rise rate of the bearing seat to determine the lubrication status of the bearing. If the temperature rise of the equipment is too fast, it means that the bearing is poorly lubricated and the bearing needs to be refilled with lubricant.
[0003] Although regular status detection of the lubricating oil in the bearing seat can accurately determine the lubrication status of the bearing, this method has a certain amount of lubricating oil waste. At the same time, the spectrum or iron spectrum analysis of the lubricating oil is expensive and the detection time is long, so the economy and timeliness of this method are poor. Using a temperature sensor to detect the temperature rise of the bearing seat can obtain the lubrication status of the bearing. When the temperature rises too fast, it means that the lubrication quality of the bearing has decreased. However, this method takes too long, and the temperature rise detection value is easily affected by other forced heat dissipation equipment, so the accuracy and efficiency of this method are limited. Summary of the invention
[0004] The technical problem solved by the present invention is the poor effect of detecting the state of the lubricating oil in the bearing seat. The present invention provides a rolling bearing lubrication state detection method. The present invention can calculate the bearing lubrication state characterization index, and can accurately and efficiently realize the rolling bearing lubrication state judgment, so as to guide the bearing maintenance personnel to reasonably perform the lubricating oil filling operation and reduce the bearing maintenance time.
[0005] In order to achieve the above object, the present invention is implemented by the following technical solutions:
[0006] A method for detecting the lubrication state of a rolling bearing comprises the following steps:
[0007] Step 1: collecting bearing seat vibration acceleration signals at at least two different rotational speeds; wherein the bearing seat vibration acceleration signals are periodically collected at each rotational speed;
[0008] Step 2: Use a high-pass digital filter to extract the bearing seat vibration acceleration signal at each speed;
[0009] Step 3: Perform discrete Fourier transform on the vibration acceleration signals of the bearing seats at each speed to obtain the signal spectrum at each speed;
[0010] Step 4: Calculate the center frequency of the signal spectrum at each speed using the center frequency calculation method;
[0011] Step 5: Calculate the difference between the center frequencies of the signal spectra at each speed, and divide the difference between the center frequencies of the signal spectra by the speed difference between the speeds to obtain the bearing lubrication state characterization index;
[0012] Step 6: Obtain multiple bearing lubrication status characterization indicators in the same manner as in step 5, calculate the mean of the multiple bearing lubrication status characterization indicators, and determine whether the mean of the bearing lubrication status indicator is less than a threshold value. If so, an alarm is issued.
[0013] Optionally, in step one, the vibration acceleration sensor is installed on the surface of the bearing seat by bolting or gluing, and a signal collector is used to collect the vibration acceleration signals of the bearing seat at each rotation speed. The vibration frequency of the vibration acceleration signal collected at each rotation speed is not less than 10000 Hz, and the measuring range is not less than 100g.
[0014] Furthermore, the acceleration signal of the bearing seat at each speed is collected at each speed, and the vibration acceleration signal of the bearing seat can be collected at each speed. The signal collection time is T seconds, and the signal collection time is not less than 10 seconds.
[0015] Optionally, in step 2, the digital high-pass filter is used to filter the vibration acceleration signal of the bearing seat at each speed, and the frequency extracted by the digital high-pass filter for the vibration acceleration signal of the bearing seat at each speed is a vibration frequency higher than 2000 Hz. The cutoff frequency of the digital high-pass filter is 2000 Hz, and the passband ripple is 1 dB.
[0016] Optionally, in step three, the signal spectrum X(k) of the discrete Fourier transform of the vibration acceleration signal at each operating speed is calculated as formula (2):
[0017]
[0018] In formula (1), x is the discrete signal of the vibration acceleration signal of the bearing seat collected, x(n) represents the vibration acceleration signal of the bearing seat collected at the n×Ts second, j represents an imaginary number, N represents the total number of signal points of the vibration acceleration signal of the bearing seat collected, n represents the nth sampling point in the signal, k represents the kth spectrum line in the frequency, is the rotation factor of the signal in the fast Fourier transform of the signal, e represents the natural base, X′(k) or It represents the Fast Fourier calculation result of the signal, X(k) represents the signal spectrum, and the value of X(k) is equal to the modulus of X′(k).
[0019] In formula (2), conj(X′(k)) represents taking the conjugate of the signal fast Fourier transform result X′(k). After running the conj(X′(k)) operation, the conj(X′(k)) operation result output is that the real part of X′(k) remains unchanged, and the imaginary part becomes the opposite number.
[0020] Optionally, in step 4, the calculation of the signal spectrum center frequency of each vibration acceleration signal is formula (3):
[0021]
[0022] Wherein, X(k) is the signal spectrum of the discrete Fourier transform of the vibration acceleration signal, k represents an integer between 0 and N, and N is the total number of vibration acceleration signals of the bearing seat collected; There are N spectral lines in the signal spectrum. Multiply the amplitude of the kth spectral line by k, where k is an integer between 0 and N. Then sum the results of multiplying the amplitudes of the N spectral lines by k. It means that there are N spectral lines in the signal spectrum. The amplitudes of the N spectral lines are summed and then the sum is multiplied by the signal acquisition time T. f represents the center frequency of the signal spectrum.
[0023] Optionally, in step 5, the bearing lubrication state characterization index is I, the speed difference = high speed - low speed, the speed difference unit is revolutions per second, and the calculation formula is formula (4):
[0024]
[0025] Where f is the center frequency of the signal spectrum, f i represents the center frequency of the signal spectrum at the i-th speed, n represents the speed of the shaft supported by the bearing, and n i Indicates the speed under the i-th working condition, I i It represents the i-th bearing lubrication state characterization index calculated by the center frequency of the i+1-th signal spectrum and the center frequency of the i-th signal spectrum.
[0026] Optionally, in step six, the average of the bearing lubrication status characterization indexes under multiple different speed differences is calculated. If the average is greater than 20, the bearing lubrication status is good, otherwise the bearing is refilled with lubricating oil.
[0027] Furthermore, the multiple bearing lubrication state characterization indicators I are divided by the number of bearing lubrication state characterization indicators I to obtain the average value of the bearing lubrication state characterization indicators.
[0028] Beneficial effects of the present invention:
[0029] The present invention performs discrete Fourier transform on the bearing seat vibration acceleration signal at each rotational speed to obtain the signal spectrum at each rotational speed, calculates the center frequency of the signal spectrum at each rotational speed through each signal spectrum, calculates the difference between the center frequencies of the signal spectrum at each rotational speed, and obtains the bearing lubrication state characterization index by dividing the difference between the center frequencies of the signal spectrum by the speed difference between the rotational speeds. The multiple bearing lubrication state characterization indicators obtained are averaged, and the lubrication state of the bearing is determined by judging the average of the bearing lubrication state indicators. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0031] Figure 1 It is a schematic diagram of the workflow of the present invention;
[0032] Figure 2 It is the signal spectrum of the bearing seat vibration acceleration signal at low speed under the condition of good lubrication condition;
[0033] Figure 3 It is the signal spectrum of the bearing seat vibration acceleration signal when the bearing seat is at medium speed under the condition of good lubrication condition;
[0034] Figure 4 It is the signal spectrum of the bearing seat vibration acceleration signal when the bearing seat is at high speed under the condition of good lubrication condition;
[0035] Figure 5 It is the signal spectrum of the bearing seat vibration acceleration signal at low speed under the condition of poor lubrication;
[0036] Figure 6 It is the signal spectrum of the bearing seat vibration acceleration signal at medium speed under the condition of poor lubrication;
[0037] Figure 7 It is the signal spectrum of the bearing seat vibration acceleration signal when the bearing seat is at high speed under the condition of poor lubrication;
[0038] Figure 8 It is the center frequency of the signal spectrum of the filtered bearing seat vibration acceleration signal under good lubrication conditions;
[0039] Fig. 9It is the center frequency of the signal spectrum of the filtered bearing seat vibration acceleration signal under the condition of poor lubrication;
[0040] Fig.10 It is an indicator of the lubrication status of the bearing under good lubrication conditions;
[0041] Fig.11 It is an indicator of the lubrication status of the bearing under poor lubrication conditions. DETAILED DESCRIPTION
[0042] The embodiments of the present application are described in detail below with reference to the accompanying drawings.
[0043] Example 1
[0044] like Figure 1 As shown, this embodiment provides a rolling bearing lubrication state detection method, comprising the following steps:
[0045] Step 1: collecting bearing seat vibration acceleration signals at at least two different rotational speeds; wherein the bearing seat vibration acceleration signals are periodically collected at each rotational speed;
[0046] Step 2: Use a high-pass digital filter to extract the bearing seat vibration acceleration signal at each speed;
[0047] Step 3: Perform discrete Fourier transform on the vibration acceleration signals of the bearing seats at each speed to obtain the signal spectrum at each speed;
[0048] Step 4: Calculate the center frequency of the signal spectrum at each speed using the center frequency calculation method;
[0049] Step 5: Calculate the difference between the center frequencies of the signal spectra at each speed, and divide the difference between the center frequencies of the signal spectra by the speed difference between the speeds to obtain the bearing lubrication state characterization index;
[0050] Step 6: Obtain multiple bearing lubrication status characterization indicators in the same manner as in step 5, calculate the mean of the multiple bearing lubrication status characterization indicators, and determine whether the mean of the bearing lubrication status indicator is less than a threshold value. If so, an alarm is issued.
[0051] Specifically, in step one, the vibration acceleration sensor is installed on the surface of the bearing seat by bolting or gluing, and the vibration acceleration signal of the bearing seat at each speed is collected by a signal collector. The vibration frequency of the vibration acceleration signal collected at each speed is not less than 10000 Hz, and the range is not less than 100g.
[0052] Furthermore, the acceleration signal of the bearing seat at each speed is collected at each speed, and the vibration acceleration signal of the bearing seat can be collected at each speed. The signal collection time is T seconds, and the signal collection time is not less than 10 seconds.
[0053] In step two, a digital high-pass filter is used to filter the vibration acceleration signal of each bearing seat at each speed. The digital high-pass filter extracts the vibration acceleration signal of each bearing seat at each speed with a vibration frequency above 2000 Hz. The cutoff frequency of the digital high-pass filter is 2000 Hz, and the passband ripple is 1 dB.
[0054] In step three, the discrete Fourier transform signal spectrum X(k) of the vibration acceleration signal at each operating speed is calculated, which is formula (2):
[0055]
[0056] In formula (1), x is the discrete signal of the vibration acceleration signal of the bearing seat collected, x(n) represents the vibration acceleration signal of the bearing seat collected at the n×Ts second, j represents an imaginary number, N represents the total number of signal points of the vibration acceleration signal of the bearing seat collected, n represents the nth sampling point in the signal, k represents the kth spectrum line in the frequency, is the rotation factor of the signal in the fast Fourier transform of the signal, e represents the natural base, X′(k) or It represents the Fast Fourier calculation result of the signal, X(k) represents the signal spectrum, and the value of X(k) is equal to the modulus of X′(k).
[0057] In formula (2), conj(X′(k)) represents taking the conjugate of the signal fast Fourier transform result X′(k). After running the conj(X′(k)) operation, the conj(X′(k)) operation result output is that the real part of X′(k) remains unchanged, and the imaginary part becomes the opposite number.
[0058] In step 4, the calculation of the signal spectrum center frequency of each vibration acceleration signal is as follows:
[0059]
[0060] Wherein, X(k) is the signal spectrum of the discrete Fourier transform of the vibration acceleration signal, k represents an integer between 0 and N, and N is the total number of vibration acceleration signals of the bearing seat collected; There are N spectral lines in the signal spectrum. Multiply the amplitude of the kth spectral line by k, where k is an integer between 0 and N. Then sum the results of multiplying the amplitudes of the N spectral lines by k. It means that there are N spectral lines in the signal spectrum. The amplitudes of the N spectral lines are summed and then the sum is multiplied by the signal acquisition time T. f represents the center frequency of the signal spectrum.
[0061] In step 5, the bearing lubrication state characterization index is I, the speed difference = high speed - low speed, the speed difference unit is revolutions per second, and the calculation formula is formula (4):
[0062]
[0063] Where f is the center frequency of the signal spectrum, f i represents the center frequency of the signal spectrum at the i-th speed, n represents the speed of the shaft supported by the bearing (in revolutions per second), n i Indicates the speed under the i-th working condition, I i It represents the i-th bearing lubrication state characterization index calculated by the center frequency of the i+1-th signal spectrum and the center frequency of the i-th signal spectrum.
[0064] In step six, the mean of the bearing lubrication state characterization index under multiple different speed differences is calculated. If the mean is greater than 20, the bearing lubrication state is good, otherwise the bearing is refilled with lubricating oil. Among them, the mean of the bearing lubrication state characterization index is obtained by dividing the multiple bearing lubrication state characterization indexes I by the number of bearing lubrication state characterization indexes I.
[0065] This embodiment allows a rotating shaft to rotate in a bearing. The rotation of the rotating shaft causes the bearing to vibrate. The rotating shaft has different rotation speeds. Two different rotation speeds of the rotating shaft are randomly selected. The rotating shaft rotates at different rotation speeds. The bearing seat will have different vibrations at different rotating shaft rotation speeds. The bearing seat generates different bearing seat vibration acceleration signals at different rotation speeds, that is, the two different rotation speeds each generate a bearing seat vibration acceleration signal of a different frequency.
[0066] In this embodiment, the vibration acceleration signals of the bearing seat with different frequencies are subjected to discrete Fourier transform to obtain signal spectra with different rotation speeds.
[0067] The bearing seat vibration acceleration signals of different frequencies have different signal spectrum center frequencies.
[0068] Of course, in this embodiment, three or more different rotation speeds of the rotating shaft may be randomly selected.
[0069] Example 2
[0070] Based on Example 1, Steps 5 and 6 can select two or more different rotational speeds multiple times and randomly. Therefore, the rotational speeds randomly selected each time are different, and the signal spectra at the rotational speeds randomly selected each time are different, and the difference between the center frequencies of the signal spectra is also different. The rotational speed difference between the rotational speeds is different each time the rotational speed is randomly selected, and there are multiple bearing lubrication state characterization indicators obtained by dividing the difference between the center frequencies of the signal spectra by the rotational speed difference between the rotational speeds;
[0071] The mean of multiple bearing lubrication state characterization indicators is calculated to determine whether the mean of the bearing lubrication state indicator is less than a threshold value, and if so, an alarm is issued. In this way, the mean of the bearing lubrication state characterization indicator is obtained by dividing the multiple bearing lubrication state characterization indicators I by the number of bearing lubrication state characterization indicators I.
[0072] Example 3
[0073] Based on Examples 1-2, the present invention proposes a bearing lubrication state detection method based on the difference in bearing vibration frequency. When the shaft rotates, when the rolling bearing is in good lubrication state, the contact state between the rolling element (ball) and the raceway of the bearing is an elastohydrodynamic lubrication contact state, and an oil film is formed between the rolling element and the raceway. The stiffness of the oil film will be affected by the shaft rotation speed. The faster the speed, the greater the stiffness of the oil film. When the impurities in the lubricating oil increase or the viscosity decreases, the contact state between the rolling element and the raceway of the bearing changes from an elastohydrodynamic lubrication contact state to a dry contact state. The contact stiffness of the rolling element and the raceway in the dry contact state is not affected by the shaft rotation speed. Since the resonant frequency of the bearing is proportional to the contact stiffness between the rolling element and the raceway, the lubrication state of the bearing can be determined by collecting the vibration signal of the bearing seat and detecting the frequency change of the bearing seat vibration signal at different rotation speeds.
[0074] The present invention detects the bearing lubrication state by collecting the vibration signal of the bearing seat. It does not need to judge the bearing lubrication state based on the detection result of the lubricating oil inside the bearing seat, nor does it judge the bearing lubrication state based on the temperature rise of the bearing seat. Instead, it uses the frequency difference within the vibration frequency band of the bearing under different lubrication conditions to realize the bearing lubrication state detection. Since the frequency difference within the vibration frequency band of the bearing under different lubrication conditions is obvious, and the bearing vibration can be quickly measured after the equipment speed is stable, the method has the characteristics of high detection accuracy, good economy, and high efficiency.
[0075] Example 4
[0076] Based on all the above embodiments, Figure 2-Figure 4 It is the signal spectrum of the bearing seat vibration acceleration signal at low, medium and high speeds under good lubrication conditions. The specific speed depends on the actual working conditions.
[0077] Figure 5-Figure 7 It is the signal spectrum of the bearing seat vibration acceleration signal at low, medium and high speeds under poor lubrication conditions.
[0078] Figure 8 It is the center frequency of the signal spectrum of the filtered bearing seat vibration acceleration signal under good lubrication conditions.
[0079] Fig. 9 It is the center frequency of the signal spectrum of the filtered bearing seat vibration acceleration signal under poor lubrication conditions.
[0080] Fig.10 It is an indicator of the lubrication status of the bearing under good lubrication conditions.
[0081] Fig.11 It is an indicator of the lubrication status of the bearing under poor lubrication conditions.
[0082] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope of the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A method for detecting the lubrication state of a rolling bearing, characterized in that: The steps include: Step 1: collecting bearing seat vibration acceleration signals at at least two different rotational speeds; wherein, the bearing seat vibration acceleration signals are periodically collected at each rotational speed; Step 2: Use a high-pass digital filter to extract the bearing seat vibration acceleration signal at each speed; Step 3: Perform discrete Fourier transform on the vibration acceleration signals of the bearing seats at each speed to obtain the signal spectrum at each speed; Step 4: Calculate the center frequency of the signal spectrum at each speed using the center frequency calculation method; Step 5: Calculate the difference between the center frequencies of the signal spectra at each speed, and divide the difference between the center frequencies of the signal spectra by the speed difference between the speeds to obtain the bearing lubrication state characterization index; Step 6: Obtain multiple bearing lubrication status characterization indicators in the same manner as in step 5, calculate the mean of the multiple bearing lubrication status characterization indicators, and determine whether the mean of the bearing lubrication status indicator is less than a threshold value. If so, an alarm is issued.
2. A rolling bearing lubrication status detection method according to claim 1, characterized in that: In step one, the vibration acceleration sensor is installed on the surface of the bearing seat by bolting or gluing, and the vibration acceleration signal of the bearing seat at each rotation speed is collected by a signal collector. The vibration frequency of the vibration acceleration signal collected at each rotation speed is not less than 10000 Hz, and the measuring range is not less than 100g.
3. A rolling bearing lubrication status detection method according to claim 2, characterized in that: The acceleration signal of the bearing seat at each speed is collected at each speed, and the vibration acceleration signal of the bearing seat can be collected at each speed. The signal collection time is T seconds, and the signal collection time is not less than 10 seconds.
4. A rolling bearing lubrication status detection method according to claim 1, characterized in that: In the step 2, the digital high-pass filter is used to filter the vibration acceleration signal of the bearing seat at each rotational speed. The frequency extracted by the digital high-pass filter from the vibration acceleration signal of the bearing seat at each rotational speed is a vibration frequency higher than 2000 Hz. The cutoff frequency of the digital high-pass filter is 2000 Hz, and the passband ripple is 0-1dB.
5. A rolling bearing lubrication status detection method according to claim 1, characterized in that: In step three, the signal spectrum X(k) of the discrete Fourier transform of the vibration acceleration signal at each operating speed is calculated as formula (2): In formula (1), x is the discrete signal of the vibration acceleration signal of the bearing seat collected, x(n) represents the vibration acceleration signal of the bearing seat collected at the n×Ts second, j represents an imaginary number, N represents the total number of signal points of the vibration acceleration signal of the bearing seat collected, n represents the nth sampling point in the signal, k represents the kth spectrum line in the frequency, is the rotation factor of the signal in the fast Fourier transform of the signal, e represents the natural base, X′(k) or It represents the Fast Fourier calculation result of the signal, X(k) represents the signal spectrum, and the value of X(k) is equal to the modulus of X′(k). In formula (2), conj(X′(k)) represents taking the conjugate of the signal fast Fourier transform result X′(k). After running the conj(X′(k)) operation, the conj(X′(k)) operation result output is that the real part of X′(k) remains unchanged, and the imaginary part becomes the opposite number.
6. A rolling bearing lubrication status detection method according to claim 1, characterized in that: In step 4, the calculation of the signal spectrum center frequency of each vibration acceleration signal is formula (3): Wherein, X(k) is the signal spectrum of the discrete Fourier transform of the vibration acceleration signal, k represents an integer between 0 and N, and N is the total number of vibration acceleration signals of the bearing seat collected; There are N spectral lines in the signal spectrum. Multiply the amplitude of the kth spectral line by k, where k is an integer between 0 and N. Then sum the results of multiplying the amplitudes of the N spectral lines by k. It means that there are N spectral lines in the signal spectrum, the amplitudes of the N spectral lines are summed, and then the summation result is multiplied by the signal acquisition time T; f represents the center frequency of the signal spectrum.
7. A rolling bearing lubrication status detection method according to claim 1, characterized in that: In step 5, the bearing lubrication state characterization index is I, the speed difference unit is revolutions per second, and the calculation formula is formula (4): Where f is the center frequency of the signal spectrum, f i represents the center frequency of the signal spectrum at the i-th speed, n represents the speed of the shaft supported by the bearing, and n i Indicates the speed under the i-th working condition, I i It represents the i-th bearing lubrication state characterization index calculated by the center frequency of the i+1-th signal spectrum and the center frequency of the i-th signal spectrum.
8. A rolling bearing lubrication status detection method according to claim 1, characterized in that: In step six, the average of the bearing lubrication status characterization indexes under multiple different speed differences is calculated. If the average is greater than 20, the bearing lubrication status is good, otherwise the bearing is refilled with lubricating oil.
9. A rolling bearing lubrication status detection method according to claim 8, characterized in that: The mean value of the bearing lubrication state characterization index I is obtained by dividing the multiple bearing lubrication state characterization indexes I by the number of bearing lubrication state characterization indexes I.
