Quantitative calculation method for damage size of rolling bearing raceway

Abstract

The invention provides a method for quantitatively calculating the damage size of a rolling bearing raceway. According to the method, a time-frequency distribution diagram of bearing fault signal pulses is extracted through an instantaneous energy analysis method, a vibration energy curve is obtained, on this basis, the time interval when a rolling body enters and leaves the raceway local damage is obtained, and finally, the size of the damage is obtained through calculation. Through different damage size fault experiments and decomposition inspection verification of the rolling bearing raceway, the result shows that quantitative estimation of different damage sizes of the bearing at different rotating speeds is accurate and effective, and a new reliable way can be provided for real-time quantitative estimation of the local damage size of the rolling bearing.

Description

Technical Field

[0001] The present application belongs to the field of rotating machinery fault diagnosis, and specifically relates to a quantitative calculation method for the size of local damage to a rolling bearing raceway. Background Art

[0002] Rolling bearings are one of the most common and core components in rotating machinery and are widely used in aerospace, railway, construction, steel, papermaking, textile, mining and other industries. Their operating status is directly related to the safety of equipment. In recent years, with the advancement of technology, early fault diagnosis technology for rolling bearings has become increasingly mature, such as envelope analysis methods, various wavelet analysis methods, blind deconvolution methods, signal decomposition methods, machine learning methods, etc. However, a complete prediction and health management system for rolling bearings should also include a quantitative estimation of the degree of damage; but at present, a large amount of research on rolling bearings is focused on fault diagnosis. The assessment of the degree of bearing damage can be mainly divided into two categories: (1) using statistical methods to form trend degradation indicators to assess the degree of damage; (2) directly quantitatively estimating the damage size of the bearing through damage impact characteristics or physical models. In recent years, although there have been many studies on bearing life prediction, most of them are evaluated through trend degradation indicators, and most of the work is concentrated on the former. Quantitative estimation of bearing damage size has clear physical meaning, but it is more challenging.

[0003] Therefore, this application addresses the difficulty in quantitatively calculating the size of local damage in bearing raceways and proposes a fast calculation method based on instantaneous vibration energy, which can provide an important technical approach for online monitoring of the evolution and severity of rolling bearing faults. Summary of the Invention

[0004] The purpose of this application is to provide a method for quantitatively calculating the size of local damage in rolling bearings based on instantaneous vibration energy. This method can quickly and conveniently estimate the size of local damage in rolling bearings.

[0005] The technical solution of this application is:

[0006] A method for quantitatively calculating the size of local damage on inner and outer raceways of a rolling bearing, characterized by comprising:

[0007] Step 1: Collect vibration signals containing different local damage sizes on the inner and outer raceways of the rolling bearing. According to the sampling frequency and sampling time of the bearing signal, select a single pulse signal from the periodic impact waveform of the time domain signal; the single pulse signal can be selected arbitrarily without considering factors such as amplitude.

[0008] Step 2: Perform time-frequency calculation on a single fault pulse (different time-frequency analysis methods can be used) to obtain the time-frequency distribution of the single fault pulse time domain waveform. This application uses the smoothed pseudo-Wigner-Ville distribution with high time-frequency analysis accuracy for verification, but is not limited to other time-frequency analysis methods, such as time-frequency analysis methods based on wavelet analysis, etc.

[0009] Step 3: Extract the instantaneous energy from the time-frequency distribution diagram in step 2 to obtain the time interval ΔT between the rolling element entering and leaving the damaged area. The formula used to calculate the instantaneous energy of the time-frequency distribution is:

[0010]

[0011] Where |E(t)| 2 is the instantaneous energy of the vibration signal; f1 and f2 are the frequency ranges of a single fault pulse; WVD SP (t,f) represents the result of the smoothed pseudo-Wigner-Ville distribution performed on a single fault pulse, and df represents the integral of the frequency. The time nodes when the rolling element enters and leaves the damage are extracted by the change in instantaneous energy. At the moment the rolling element enters the damage, the instantaneous energy curve will change from a smooth state to a small peak fluctuation, which is recorded as point A here; when the rolling element enters the damage area and collides with the damage edge, a huge collision energy peak will be generated and oscillate, and the maximum peak is recorded as point B here. When the rolling element completely leaves the damage area, the re-stress process will produce a smaller peak again, which is recorded as point C here; the time difference ΔT between AC is the time it takes for the rolling element to pass through the local damage area; the so-called "re-stress process" is a common concept in this field. When the rolling element enters and leaves the damage area, it will experience a process in which the stress between the rolling element and the bearing disappears and then reappears, which is called the "re-stress process".

[0012] Step 4: Substitute ΔT into the quantitative calculation formula for the local damage size of the inner and outer rings of the rolling bearing to obtain the estimated result. For the local damage on the outer raceway of the bearing, the damage size L out The relationship with ΔT is:

[0013]

[0014] Where D m is the bearing pitch diameter, d is the rolling element diameter, α is the contact angle, f n For the local damage on the inner raceway, the damage size L in The relationship with ΔT is:

[0015]

[0016] When the rolling bearing is a deep groove ball bearing and its contact angle is 0 degrees, the local damage size L of the inner raceway and outer raceway can be further simplified as:

[0017]

[0018] In at least one embodiment of the present invention, the present invention is applied to calculate the size of local damage on the inner and outer raceways of a rolling bearing, and its effectiveness and accuracy are verified. The invention has at least the following beneficial technical effects:

[0019] The present invention provides a quantitative calculation method for the local damage size of rolling bearings based on the instantaneous energy of vibration. It can quickly, effectively and conveniently calculate the local damage size on the rolling bearing raceway, and can provide an important technical approach for realizing online monitoring of the evolution and severity of rolling bearing faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more apparent.

[0021] Figure 1 It is a flow chart of the present invention for quantitatively calculating the local damage size of a rolling bearing.

[0022] Figure 2 This is a schematic diagram of a faulty bearing with different sizes of local damage on the inner and outer raceways.

[0023] Figure 3 It is a schematic diagram of the test collection device.

[0024] Figure 4a This is a schematic diagram of the instantaneous energy change and calculation results of different local damage sizes on the outer raceway at a speed of 1200 r / min (1.0 mm, ΔT = 0.000699 s, L = 0.97 mm).

[0025] Figure 4b This is a schematic diagram of the instantaneous energy change and calculation results of different local damage sizes on the outer raceway at a speed of 1200 r / min (1.4 mm, ΔT = 0.001060 s, L = 1.47 mm).

[0026] Figure 4c This is a schematic diagram of the instantaneous energy change and calculation results of different local damage sizes on the outer raceway at a speed of 1200 r / min (1.8 mm, ΔT = 0.00132 s, L = 1.83 mm).

[0027] Figure 4dThis is a schematic diagram of the instantaneous energy change and calculation results of different local damage sizes on the outer raceway at a speed of 1200 r / min (2.2 mm, ΔT = 0.001571 s, L = 2.17 mm).

[0028] Figure 5a This is a schematic diagram of the instantaneous energy change and calculation results of different local damage sizes on the inner raceway at a speed of 1200 r / min (1.0 mm, ΔT = 0.000694 s, L = 0.96 mm).

[0029] Figure 5b This is a schematic diagram of the instantaneous energy change and calculation results of different local damage sizes on the inner raceway at a speed of 1200 r / min (1.4 mm, ΔT = 0.001068 s, L = 1.48 mm).

[0030] Figure 5c This is a schematic diagram of the instantaneous energy change and calculation results of different local damage sizes on the inner raceway at a speed of 1200 r / min (1.8 mm, ΔT = 0.00134 s, L = 1.85 mm).

[0031] Figure 5d This is a schematic diagram of the instantaneous energy change and calculation results of different local damage sizes on the inner raceway at a speed of 1200 r / min (2.2 mm, ΔT = 0.00162 s, L = 2.24 mm). DETAILED DESCRIPTION

[0032] like Figure 1 As shown, the present invention provides a method for quantitatively calculating the damage size of a rolling bearing raceway, comprising the following steps:

[0033] Step 1: Collect vibration signals with different local damage sizes on the inner and outer raceways of the rolling bearing. Select a single pulse signal from the periodic impact waveform of the time domain signal based on the sampling frequency and sampling time of the bearing signal.

[0034] Step 2: Perform time-frequency calculation on a single fault pulse (different time-frequency analysis methods can be used) to obtain the time-frequency distribution of the single fault pulse time domain waveform. This application uses the smoothed pseudo-Wigner-Ville distribution with high time-frequency analysis accuracy for verification, but is not limited to other time-frequency analysis methods, such as time-frequency analysis methods based on wavelet analysis, etc.

[0035] Step 3: Extract the instantaneous energy from the time-frequency distribution diagram in step 2 to obtain the time interval ΔT between the rolling element entering and leaving the damaged area. The formula used to calculate the instantaneous energy of the time-frequency distribution is:

[0036]

[0037] Where |E(t)| 2 is the instantaneous energy of the vibration signal; f1 and f2 are the frequency ranges of a single fault pulse; SPWVD x (t,f) represents the result of a smoothed pseudo-Wigner-Ville distribution applied to a single fault pulse, and df represents the integral over frequency. Integrating the frequency range in the time-frequency distribution plot yields the instantaneous vibration energy. The changes in instantaneous energy are used to extract the time points when the rolling element enters and exits the damage zone. At the moment the rolling element enters the damage zone, the instantaneous energy curve changes from a smooth state to a small peak fluctuation, noted here as point A. When the rolling element enters the damaged area and collides with the damaged edge, a large collision energy peak is generated and oscillates, with the maximum peak value noted here as point B. When the rolling element completely exits the damaged area, the re-stressing process produces another smaller peak, noted here as point C. The time difference ΔT between AC represents the time it takes for the rolling element to pass through the localized damage zone.

[0038] Step 4: Substitute ΔT into the quantitative calculation formula for the local damage size of the inner and outer rings of the rolling bearing to obtain the estimated result. For the local damage on the outer raceway of the bearing, the damage size L out The relationship with ΔT is:

[0039]

[0040] Where D m is the bearing pitch diameter, d is the rolling element diameter, α is the contact angle, f n For the local damage on the inner raceway, the damage size L in The relationship with ΔT is:

[0041]

[0042] When the rolling bearing is a deep groove ball bearing and its contact angle is 0 degrees, the local damage size L of the inner raceway and outer raceway can be further simplified as:

[0043]

[0044] The present invention uses local damage tests on the inner and outer raceways of bearings of different sizes for verification. The fault processing locations include the inner and outer rings. Different fault sizes are processed by manual wire cutting. The groove widths include 1.0mm, 1.4mm, 1.8mm and 2.2mm, and the groove depth is 1.0mm. The 6206 deep groove ball bearing is the test object. Figure 2 As shown. The ABLT-1A bearing fatigue testing machine is used. This testing machine can perform rolling bearing failure simulation tests, such as Figure 3The faulty test bearing was placed in the bearing seat, the test sampling frequency was 128kHz, the data logger was NI-USB 9234, and the vibration sensor model was AI002. The speeds were 1200r / min, 1500r / min, 1800r / min, and 2400r / min. The specific parameters of the 6206 rolling bearing are shown in Table 1:

[0045] Table 1

[0046]

[0047] Taking the speed of 1200r / min as an example, for local damage on the outer ring, we first perform SPWVD time-frequency analysis on the single impact pulse in the vibration signal collected at different damage sizes and extract its instantaneous energy. Then, we extract the time interval ΔT when the rolling element passes through the damaged area and substitute it into the formula for calculation. Figure 4a 、 Figure 4b 、 Figure 4c 、 Figure 4d The calculation results of four different damage sizes on the outer raceway are shown in Figure 2. From the results, it can be seen that the change pattern of instantaneous energy is the same for the four different damage sizes.

[0048] When the rolling element is in the entry phase, the transient energy will fluctuate upward due to the low-frequency response. During the impact phase, the high-frequency response causes the transient energy to increase sharply and oscillate. Then, the rolling element leaves the damaged area and re-contacts the raceway. The low-frequency response generated by this process will again cause the transient energy to peak. After that, the energy oscillation gradually decreases to zero. It is worth noting that the transient energy change caused by the low-frequency response is smaller than the energy change caused by the high-frequency response, and the change caused by the re-stressing process is greater than the change caused by the stress relief process. In addition, in other stages, the transient energy is almost zero, with no significant change. Therefore, based on the time interval ΔT between the two low-frequency responses, the corresponding local damage size of the outer ring is calculated, and the result is generally consistent with the actual size.

[0049] For the quantitative calculation of local damage of the inner raceway, the instantaneous energy changes of the four damage sizes at 1200r / min are as follows: Figure 5a 、 Figure 5b 、 Figure 5c 、 Figure 5d As shown, the instantaneous energy variation pattern is similar to that of the outer ring. The time interval between the entry and exit phases can be clearly extracted from the figure, resulting in a good estimation result. To ensure the accuracy of the estimated size, five pulses were selected for each damage size, and the average values ​​and errors are given in Tables 2 and 3, respectively. The tables show that the quantitative estimation results of the proposed method are very accurate, with an error of less than 5%, for both outer and inner ring damage.

[0050] Table 2

[0051]

[0052] Table 3

[0053]

[0054]

[0055] To verify the results at different speeds, Tables 4 and 5 present the quantitative calculations of different damage sizes on the outer and inner raceways at four speeds, compared with the actual damage sizes. As can be seen from the tables, the calculated results of the method provided by the present invention are all less than 10%, meeting the requirements of engineering applications.

[0056] Table 4

[0057]

[0058] Table 5

[0059]

[0060] The present invention provides a method for accurately and quantitatively calculating the damage dimensions of the inner and outer raceways of rolling bearings. There are numerous methods and approaches for implementing this technical solution. The foregoing is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, such as employing different time-frequency analysis methods. These improvements and modifications are also within the scope of protection of the present invention. Any components not specified in this embodiment may be implemented using existing technologies.

Claims

1. A method for quantitatively calculating the damage size of the inner and outer raceways of rolling bearings, characterized in that: include: Step 1: Collect vibration signals with different local damage sizes on the inner and outer raceways of the rolling bearing. Select a single fault pulse signal from the periodic impulse waveform of the time domain signal based on the sampling frequency and sampling time of the bearing signal. Step 2: Perform time-frequency calculation on a single fault pulse to obtain a time-frequency distribution diagram of the single fault pulse time domain waveform; Step 3: Calculate the vibration energy of the time-frequency distribution in step 2 to obtain the time interval ΔT between the rolling element entering and leaving the damaged area; Step 4: Substitute ΔT into the quantitative calculation formula for the local damage size of the inner and outer rings of the rolling bearing to obtain the estimated results of the damage size of the inner and outer raceways of the rolling bearing; Step 5: Repeat steps 2 to 4 above in a loop to calculate the estimated values ​​of multiple fault pulses and obtain the average result.

2. A method for quantitatively calculating the damage size of inner and outer raceways of rolling bearings according to claim 1, characterized in that: The step 2 adopts a smoothed pseudo Wigner-Ville time-frequency distribution method.

3. A method for quantitatively calculating the damage size of inner and outer raceways of rolling bearings according to claim 1, characterized in that: In step 2, the time-frequency calculation of a single fault pulse is performed using a time-frequency analysis method based on wavelet analysis.

4. The method according to claim 1, wherein In step 3, the vibration energy is extracted from the time-frequency distribution diagram of step 2 using the following formula: Where |E(t)| 2 is the instantaneous energy of the vibration signal; f1 and f2 are the frequency ranges of a single fault pulse; SPWVD x (t,f) represents the result of the smoothed pseudo-Wigner-Ville time-frequency distribution method performed on a single fault pulse, df represents the frequency integration, t represents time, and f represents frequency. The frequency range in the time-frequency distribution diagram is integrated to obtain the instantaneous energy of vibration, and the time nodes when the rolling body enters and leaves the damage are extracted through the change in instantaneous energy. At the moment the rolling body enters the damage, the instantaneous energy curve will change from a smooth state to a small peak fluctuation, which is recorded as point A here. When the rolling body enters the damage area and collides with the damage edge, a huge collision energy peak is generated and oscillates. The maximum peak is recorded as point B here. When the rolling body completely leaves the damage area, the re-stress process will produce a smaller peak again, which is recorded as point C here. The time difference ΔT between AC is the time it takes for the rolling body to pass through the local damage area.

5. A method for quantitatively calculating the damage size of inner and outer raceways of rolling bearings according to claim 1, characterized in that: In step 4, for the local damage on the outer raceway of the bearing, the damage size L out The relationship with ΔT is: Where D m is the bearing pitch diameter, d is the rolling element diameter, α is the contact angle, f n is the speed frequency; For local damage on the inner raceway, the damage size L in The relationship with ΔT is: When the rolling bearing is a deep groove ball bearing, its contact angle is 0 degrees. At this time, the calculation formula for the local damage size L of the inner raceway and the outer raceway will be the same, that is:

6. A method for quantitatively calculating the damage size of inner and outer raceways of rolling bearings according to claim 1, characterized in that: In the step 5, the calculation of estimated values ​​of multiple fault pulses is to calculate estimated values ​​of more than 5 fault pulses.

Images