An experimental method for quantitatively judging blade rub

Abstract

The application discloses an experimental method for quantitatively judging blade rub-impact, which comprises a motor, a rotating shaft, a blade, a casing and a bearing seat, the motor and the rotating shaft are both supported by the bearing seat, the output shaft of the motor is connected with the rotating shaft through a shaft coupling, the blade is fixedly arranged on the rotating shaft, the casing is arranged on the periphery of the blade, the gap between the casing and the blade is adjusted through a three-direction micro-motion platform, and the method comprises the following steps: strain gauges are arranged on each blade, and an eddy current sensor is arranged on the rotating shaft; strain signals of each blade at different times are measured i and time t i , vibration signals of each blade are processed, evaluation index values of strain conditions of rub-impact blades and non-rub-impact blades are obtained according to the obtained Δε and Δt, and the strain ratio r of rub-impact data and non-rub-impact data of the same blade at the same rotating speed is calculated. The application aims to provide an experimental method for quantitatively judging blade rub-impact and an evaluation index, and has important significance for experimental research on rub-impact failure mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of rubbing experiments, and in particular to an experimental method for quantitatively judging blade rubbing. Background Art

[0002] Rotor-stator rubbing is a common fault type in rotating machinery. Blade-casing rubbing can cause complex rotor vibrations, degrading system performance, shortening blade life, and even leading to serious failures such as blade loss. Experimental investigation of the nonlinear dynamic characteristics of rubbing rotor systems is crucial for aircraft engine design and development. Current fault experimental studies are mostly based on rotor displacement or casing acceleration signals, with limited consideration of blade vibration signals. Blade-casing rubbing faults occur at the blade tip. Impact and friction forces cause localized deformation of the blade. The resulting stress waves are then transmitted through the blade body to the shaft and casing. Therefore, blade vibrations contain stronger fault signals, making them valuable for blade fault diagnosis. Blade-casing rubbing is complex and can occur in complex forms, such as single-blade and multi-blade rubbing. Currently, qualitative methods are often used to assess blade rubbing, and quantitative evaluation indicators are lacking to aid fault diagnosis. Therefore, designing an experimental method and evaluation indicator for quantitatively assessing blade rubbing is crucial for experimental research into rubbing fault mechanisms. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies in the above-mentioned prior art and to provide an experimental method and evaluation index for quantitatively judging blade rub, which is of great significance for experimental research on the mechanism of rub failure.

[0004] To achieve the above object, the present invention is implemented through the following technical solutions:

[0005] An experimental method for quantitatively judging blade rubbing includes a motor, a rotating shaft, a blade, a casing, and a bearing seat. The motor and the rotating shaft are both mounted through the bearing seat. The output shaft of the motor is connected to the rotating shaft through a coupling. The blade is fixed to the rotating shaft through a disc. A casing is provided on the periphery of the blade. The gap between the casing and the blade is adjusted by a three-directional micro-motion platform. A high-speed slip ring is provided at one end of the rotating shaft connected to the bearing seat. The method includes the following steps:

[0006] (1) Install a strain gauge on each blade and an eddy current sensor on the shaft. Connect the strain gauge and eddy current sensor to a data acquisition instrument to collect vibration signals of the blade and shaft.

[0007] (2) Measure the strain signal ε of each blade at different times i and time t i , the vibration signals of each blade are processed, and the formula is as follows:

[0008]

[0009] t i is the time point, ε i is the strain of the blade at time t i , Δε is the strain difference between different time points, and Δt is the time difference between different time points;

[0010] (3) According to the obtained Δε and Δt, an evaluation index value V ε reflecting the strain condition of the rubbing blade and an evaluation index v ε reflecting the strain condition of the non-rubbing blade, are obtained, and the formula is as follows:

[0011]

[0012] V ε is the maximum strain change rate of the blade when rubbing, and v ε is the maximum strain change rate of the blade when not rubbing;

[0013] (4) The rubbing data and non-rubbing data of the same blade at the same rotating speed are selected to calculate the strain ratio r, and the formula is as follows:

[0014] r=V ε / v ε (1-3)

[0015] If r≤1, the blade does not rub with the casing, if r>1 and , the blade rubs with the casing, and are the maximum strain change rates of the rubbing blade at different rubbing rotating speeds of the rotating speed ω1 and the rotating speed ω2, respectively.

[0016] Further, the strain gauges in step (1) are arranged at the same position of the root of each blade, and the bridge type of the strain gauges is selected as a half-bridge mode.

[0017] Further, the eddy current sensors in step (1) are arranged at two positions on the rotating shaft, and the installation angle between the two eddy current sensors is 90°.

[0018] Further, before collecting the vibration signals of the blade and the rotating shaft in step (1), the rotating-static gap between the casing and the blade is adjusted by using a three-direction micro-motion platform, when collecting the vibration signals in the non-rubbing fault, the rotating-static gap is adjusted to ensure that the blade and the casing do not rub, the motor is started, the rotating speed is increased, and multiple sets of stable vibration signals in the non-rubbing fault are collected, when collecting the vibration signals in the rubbing fault, the rotating-static gap is adjusted to ensure that the blade and the casing do not rub at the initial rotating speed, the motor is started, the rotating speed is increased to simulate the rubbing fault of the blade and the casing, and multiple sets of vibration signals in the rubbing fault are collected.

[0019] Furthermore, when calculating the strain ratio r in step (4), the rubbing data V of the same blade at the same speed should be selected. ε and friction-free data v ε .

[0020] Furthermore, when the strain ratio r calculated in step (4) is greater than 1, if at this time This indicates that the blade strain ratio r>1 is due to vibration transmission and no friction occurs.

[0021] Furthermore, in the rubbing data of step (4), the strain evaluation index of the same blade at different rubbing speeds is determined. When ω1<ω2, the speed ω1 and the speed ω2 are adjacent experimental speeds and ω2-ω1≤10rad / s.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] The present invention defines the evaluation index value V of the blade strain condition ε , the evaluation index v of the strain condition of the blade without friction ε The strain ratio r can be used to intuitively determine whether blade rubbing is present and the extent of blade rubbing. This method directly measures the blade strain signal using strain gauges and uses high-speed slip rings to output the signal. This method is more accurate and can identify strain ratios r greater than 1 due to vibration transmission when no rubbing occurs. This makes it more suitable for diagnosing blade rubbing in weak rubbing faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a schematic diagram of the experimental platform layout structure of the present invention;

[0025] Figure 2 Schematic diagram of the arrangement of the connection between the strain gauge and the high-speed slip ring conductor in the present invention;

[0026] Figure 3 This is a waterfall diagram of the rotor without friction in this embodiment;

[0027] Figure 4 : This is the frequency spectrum of the rotor at the rubbing speed of 2 in this embodiment;

[0028] Figure 5 : is the axis trajectory diagram of the rotor at the rubbing speed of 2 in this embodiment;

[0029] Figure 6 is the strain signal of the blade at the rubbing speed of 2 in this embodiment;

[0030] Figure 7This is the frequency spectrum of the rotor at a friction speed of 4 in this embodiment;

[0031] Figure 8 This is a diagram of the rotor axis trajectory at a friction speed of 4 in this embodiment.

[0032] Reference numerals:

[0033] 1-motor, 2-rotating shaft, 3-blade, 4-casing, 5-bearing seat, 6-coupling, 7-three-direction micro-motion platform, 8-high-speed slip ring, 9-strain gauge, 10-eddy current sensor. DETAILED DESCRIPTION

[0034] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0035] like Figure 1 and Figure 2 As shown, an experimental method for quantitatively judging blade rubbing includes a motor 1, a rotating shaft 2, a blade 3, a casing 4 and a bearing seat 5. The motor 1 and the rotating shaft 2 are both installed through the bearing seat 5. The output shaft of the motor 1 is connected to the rotating shaft 2 through a coupling 6. The blade 3 is fixed on the rotating shaft 2 through a disc. The casing 4 is arranged on the periphery of the blade 3. The gap between the casing 4 and the blade 3 is adjusted by a three-directional micro-motion platform 7. A high-speed slip ring 8 is provided at one end of the rotating shaft 2 connected to the bearing seat 5, including the following steps:

[0036] (1) Strain gauges 9 are installed on each blade 3 at the same position at the root of each blade 3. The strain gauge 9 conductors are connected to a data acquisition instrument through a strain adapter to collect strain signals from each blade. The strain gauge 9 uses a half-bridge bridge type, which is suitable for measuring simple tensile, compressive, or bending strains in harsh environments. Other bridge types can also be selected based on actual working conditions and research focus. Eddy current sensors 10 are installed on the rotating shaft 2. Two eddy current sensors 10 are installed on the rotating shaft 2, and the two eddy current sensors 10 are installed at an angle of 90° to collect the time domain signal of the vibration of the rotating shaft 2.

[0037] Before collecting vibration signals from blades 3 and shaft 2, the three-directional micro-motion platform 7 is used to adjust the rotor-static clearance between casing 4 and blades 3. When collecting vibration signals without a rubbing fault, the rotor-static clearance is adjusted to ensure that blades 3 and casing 4 do not rub against each other. Then, motor 1 is started and the speed is increased to collect multiple sets of stable vibration signals without a rubbing fault. When collecting vibration signals with a rubbing fault, the rotor-static clearance is adjusted to ensure that blades 3 and casing 4 do not rub against each other at the initial speed. Then, motor 1 is started and the speed is increased to simulate a rubbing fault between blades 3 and casing 4, and multiple sets of vibration signals with a rubbing fault are collected.

[0038] (2) Measure the strain signal ε of each blade 3 at different times iand time t i The vibration signal of each blade 3 is processed as follows:

[0039]

[0040] t i is the time point, and ε i is the blade strain at time t i , Δε is the strain difference between different times, and Δt is the time difference between different times.

[0041] (3) According to the obtained Δε and Δt, the evaluation index value V ε reflecting the strain condition of the rubbing blade 3 and the evaluation index v ε of the strain condition of the non-rubbing blade 3 can be obtained, and the formula is as follows:

[0042]

[0043] V ε is the maximum strain change rate of the blade 3 during rubbing, v ε is the maximum strain change rate of the blade 3 during non-rubbing, V ε and v ε have the physical meaning of the maximum value of the square of the strain change rate of the blade 3 in a stable motion cycle.

[0044] (4) The rubbing data and non-rubbing data of the same blade 3 at the same speed are selected to calculate the strain ratio r. When calculating the strain ratio r, the rubbing data V ε and the non-rubbing data v ε of the same blade at the same speed are selected, and the formula is as follows:

[0045] r = V ε / v ε (1-3)

[0046] If r≤1, it means that the blade 3 does not rub with the casing 4.

[0047] Since the vibration signal is transmitted through the blade 3 and the disc, when a certain blade 3 rubs, the strain ratio r of the blade 3 is obviously greater than 1, but the strain ratio r of other blades 3 will also change. If the strain ratio r>1, it is still necessary to compare the strain signals of the rubbing blades at different speeds.

[0048] The strain of the blade at speed ω1 is The strain of the blade at speed ω2 is At a lower experimental speed, the influence of centrifugal effect and aerodynamic force on the bending strain of the blade is small. If At this time it means that the strain ratio r>1 of the blade is due to the vibration transmission effect and does not rub.

[0049] If the blades rub at a speed of ω1, the strain ratio r will be significantly greater than 1, and increasing the speed will intensify the rubbing, resulting in Then the blade rubs against the casing 4. and The maximum strain change rate of the rubbing blade at different rubbing speeds of ω1 and ω2 is used to determine the strain evaluation index of the same blade at different rubbing speeds. When ω1<ω2, the speed ω1 and the speed ω2 are adjacent experimental speeds and ω2-ω1≤10rad / s.

[0050] In order to better understand the technical solution of the present invention, a single-axis experimental platform with four blades is taken as an example. The sampling frequency f in the experiment s The experimental speed is controlled at 2kHz. Considering the safety of the experiment, the experimental speed is controlled at Ω=27rad / s~120rad / s. First, the strain signals of the four blades and the vibration displacement signal of the shaft 2 are collected at 11 groups of speeds without friction. Figure 3 This is the waterfall diagram of the lateral vibration of the shaft 2 when there is no friction. It can be concluded that the frequency characteristics of the shaft 2 are mainly based on the rotation frequency. Since the experimental platform may have slight misalignment and other problems, there are also small amplitudes at the harmonics in the waterfall diagram.

[0051] In order to facilitate data analysis, we take κ = 10 11 ,satisfy analyze Values ​​are sufficient. Tables 1 and 2 show the experimental data of the blades at different speeds in the speed-up experiment. Four groups of rubbing speeds were selected: rubbing speed 1 (ω = 37.68 rad / s), rubbing speed 2 (ω = 55.26 rad / s), rubbing speed 3 (ω = 67.20 rad / s), and rubbing speed 4 (ω = 107.40 rad / s); and four groups of non-rubbing speeds: non-rubbing speed 1 (ω = 39.88 rad / s), non-rubbing speed 2 (ω = 57.78 rad / s), non-rubbing speed 3 (ω = 69 rad / s), and non-rubbing speed 4 (ω = 108 rad / s).

[0052] Table 1 shows the strain of the blade at different speeds

[0053]

[0054] Table 2 shows the strain of the blade under conditions with and without friction

[0055]

[0056] Table 1 is the strain data of the blades in the experiment of the no-rubbing condition at two similar rotating speeds, it can be observed that the strain ratio of the four blades is less than 1 or fluctuates around 1 at the rubbing rotating speed 1, i.e. ω = 37.68 rad / s, which indicates that the blades are in the no-rubbing state at this rotating speed.

[0057] By comparing the strain signals of the blades at the rubbing rotating speed 2 and the strain signals of the blades at the rubbing rotating speed 3 in Table 2, it can be observed that the strain values of the blade 1 and the blade 2 have little change after the rotating speed is increased, which indicates that the blade 1 and the blade 2 do not have the rubbing fault at the two rotating speeds. At the rubbing rotating speed 3, the strain ratio of the blade 2 is 1.63, which indicates that the vibration transmission causes the change of the strain ratio of the blade. At the rubbing rotating speed 2, the strain ratio of the blade 4 is 1.81, although the strain ratio of the blade 2 is 1.4, the blade 2 does not have the rubbing fault. Therefore, considering the vibration transmission of the blade, the case that the increase of the strain value is less than 1.8 times is regarded as the no-rubbing fault. It is judged that the blade 4 has the rubbing fault at the rubbing rotating speed 2 and the rubbing rotating speed 3.

[0058] At the rubbing rotating speed 2, the inner concave phenomenon appears in the shaft center locus diagram of the rotor, and the frequency spectrum diagram has the frequency component of the multiple frequency of the rotating frequency, see Figure 4 and Figure 5 , which further verifies the above result. Figure 6 Therefore, the strain signals of the four blades can be found, and it is difficult to judge whether the blade has the rubbing only by whether the obvious pulse signal appears, and only the preliminary qualitative judgment can be made, therefore, the quantitative evaluation index and the blade rubbing judgment method have important significance for the rubbing fault experiment research.

[0059] Table 3 is the strain of the blades in the no-rubbing condition

[0060]

[0061] The rubbing condition of the rubbing rotating speed 4 in Table 3 is judged. It can be obtained from the strain ratio of the blades that the four blades all participate in the rubbing at this time, and the blade 4 has the most serious rubbing. Figure 7 and Figure 8 are the frequency spectrum diagram and the shaft center locus diagram of the rotor at this rotating speed, the frequency components in the frequency spectrum diagram are rich, and the shaft center locus is more complex.

[0062] The above only describes the preferred embodiments of the present application and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An experimental method for quantitatively judging blade rubbing, comprising a motor (1), a rotating shaft (2), a blade (3), a casing (4) and a bearing seat (5), wherein the motor (1) and the rotating shaft (2) are both mounted via the bearing seat (5), the output shaft of the motor (1) is connected to the rotating shaft (2) via a coupling (6), the blade (3) is fixedly arranged on the rotating shaft (2) via a disc, a casing (4) is arranged on the periphery of the blade (3), the casing (4) is adjusted to a clearance between the blade (3) and the casing (4) via a three-directional micro-motion platform (7), a high-speed slip ring (8) is arranged at one end of the rotating shaft (2) connected to the bearing seat (5), and the method is characterized in that: The steps include: (1) a strain gauge (9) is provided on each blade (3), an eddy current sensor (10) is provided on the rotating shaft (2), and the strain gauge (9) and the eddy current sensor (10) are connected to a data acquisition instrument to collect vibration signals of the blade (3) and the rotating shaft (2); (2) Measure the strain signal ε of each blade (3) at different times i and time t i , the vibration signals of each blade (3) are processed, and the formula is as follows: t i is the time point, ε i t i The bending strain of the blade (3) at the moment, Δε is the strain difference at different moments, and Δt is the time difference at different moments; (3) Based on the obtained Δε and Δt, the evaluation index value V that can reflect the strain of the rubbing blade ε , Evaluation index of strain condition of blade without friction v ε , the formula is as follows: V ε is the maximum strain change rate of the blade during rubbing, v ε is the maximum strain change rate of the blade when there is no friction; (4) The rubbing data and non-rubbing data of the same blade (3) at the same speed are selected to calculate the strain ratio r, and the formula is as follows: r=V ε / v ε (1-3) If r≤1, the blade (3) has no friction with the casing. If r>1 and Then the blade (3) rubs against the casing. and They are the maximum strain change rates of the rubbing blades at different rubbing speeds of ω1 and ω2 respectively.

2. The experimental method for quantitatively judging blade rub according to claim 1, characterized in that: The strain gauge (9) in step (1) is set at the same position at the root of each blade (3), and the bridge type of the strain gauge (9) is a half-bridge method.

3. The experimental method for quantitatively judging blade rub according to claim 1, characterized in that: Two eddy current sensors (10) in step (1) are provided on the rotating shaft (2), and the installation angle between the two eddy current sensors (10) is 90°.

4. The experimental method for quantitatively judging blade rub according to claim 1, characterized in that: Before collecting the vibration signals of the blade (3) and the rotating shaft (2) in step (1), the three-directional micro-motion platform (7) is first used to adjust the rotation-static clearance between the casing (4) and the blade (3); when collecting the vibration signals when there is no rubbing fault, the rotation-static clearance is adjusted to ensure that the blade (3) and the casing (4) do not rub against each other; the motor (1) is started, the speed is increased, and multiple groups of stable vibration signals when there is no rubbing fault are collected; when collecting the vibration signals when there is a rubbing fault, the rotation-static clearance is adjusted to ensure that the blade (3) and the casing (4) do not rub against each other at the initial speed; the motor (1) is started, the speed is increased to simulate the rubbing fault between the blade (3) and the casing (4), and multiple groups of vibration signals of the rubbing fault are collected.

5. The experimental method for quantitatively judging blade rub according to claim 1, characterized in that: When calculating the strain ratio r in step (4), the rubbing data V of the same blade (3) at the same speed should be selected. ε and friction-free data v ε .

6. The experimental method for quantitatively judging blade rub according to claim 1, characterized in that: When the strain ratio r calculated in step (4) is greater than 1, if at this time This indicates that the blade strain ratio r>1 is due to vibration transmission and no friction occurs.

7. The experimental method for quantitatively judging blade rub according to claim 1, characterized in that: In the rubbing data of step (4), the strain evaluation index of the same blade (3) at different rubbing speeds is determined. When ω1<ω2, the speed ω1 and the speed ω2 are adjacent experimental speeds and ω2-ω1≤10rad / s.

Images