A radial and axial blade tip timing sensor measurement device and method

Abstract

The present disclosure discloses a radial and axial blade tip timing sensor measurement device, comprising: a drive assembly, a blade assembly and an excitation assembly, wherein the drive assembly is used to drive the blade assembly to rotate; the excitation assembly is used to excite the blade assembly during the rotation process; the device also includes a signal acquisition assembly for collecting the speed signal of the blade assembly during the rotation process and the vibration signal generated during the excitation process.

Description

Technical Field

[0001] The present disclosure belongs to the technical field of non-destructive testing of rotating machinery rotor blades, and particularly relates to a radial and axial blade tip timing sensor measurement device and method. Background Art

[0002] As an important component of an aero-engine, blades are exposed to high temperatures, high pressures, and high speeds during operation, which can easily cause blade vibrations, leading to fatigue fractures and serious accidents. Therefore, it is of great significance to conduct research on the condition monitoring and fault diagnosis of aero-engine blades. At present, there are two main methods for blade monitoring. The first is contact measurement methods such as gluing strain gauges. This method is time-consuming and labor-intensive to install strain gauges, and has a short actual working time. In addition, it is difficult to monitor the vibration of all blades in an aero-engine, making it difficult to achieve large-scale application. The second is non-contact measurement methods represented by blade tip timing. By installing a few sensors in the engine casing, the vibration status of all blades in the first stage of an aero-engine can be monitored.

[0003] However, the existing blade monitoring methods based on blade tip timing all install the blade tip timing sensor in the radial direction of the blade disk, and there are few precedents for the application of axially arranged sensors. However, in engineering practice, radially installed blade tip timing sensors mean that holes must be drilled in the engine casing, which is not allowed for casings with coatings. Moreover, the vibration generated by the blades during high-speed rotation is not necessarily along the radial direction. Its vibration components also include components along the axial direction of the blade disk and the torsion of the blade itself. This causes the measuring point of the radially arranged blade tip sensor to change. For the axially installed blade tip timing sensor, its measuring point is less affected by the blade vibration. Therefore, in theory, the vibration signal accuracy obtained by the axially installed sensor should be higher than that of the radially installed blade tip timing sensor. Summary of the Invention

[0004] In response to the deficiencies in the prior art, the purpose of the present disclosure is to provide a radial and axial blade tip timing sensor measurement device. This device arranges two radial and axial sensors at the blade end, and simultaneously measures the vibration data of the blade in the radial and axial directions of the blade disk. The vibration data measured in the two directions are compared and verified to obtain a more accurate blade vibration shape.

[0005] To achieve the above objectives, the present disclosure provides the following technical solutions:

[0006] A radial and axial blade tip timing sensor measuring device includes: a drive assembly, a blade assembly and an excitation assembly, wherein:

[0007] The driving assembly is used to drive the blade assembly to rotate;

[0008] The excitation component is used to excite the blade component during the rotation process;

[0009] The device also includes a signal acquisition component for acquiring a rotation speed signal of the blade component during rotation and a vibration signal generated during excitation.

[0010] Preferably, the drive assembly comprises a motor connected to the blade assembly via a coupling and a main shaft.

[0011] Preferably, the blade assembly includes a blade disk, and a plurality of blades are circumferentially arranged on the blade disk.

[0012] Preferably, an organic casing is provided outside the blade disk.

[0013] Preferably, the excitation component includes a nozzle bracket, a plurality of nozzles are provided on the nozzle bracket, and each nozzle is externally connected to a high-pressure air pump.

[0014] Preferably, the signal acquisition component includes a radial blade tip timing sensor, an axial blade tip timing sensor and a rotational speed sensor.

[0015] The present disclosure also provides a radial and axial blade tip timing sensor measurement method, comprising the following steps:

[0016] S100: Simplify the circumferential vibration of the blade into multi-frequency vibration and construct the steering vector;

[0017] S200: Install radial and axial sensors at the same location and collect blade tip timing signals from both sensors simultaneously;

[0018] S300: estimating the autocorrelation matrix of the leaf-end timing signal through an iterative update algorithm;

[0019] S400: Calculating the minimum variance power spectrum of the leaf-end timing signal according to the autocorrelation matrix;

[0020] S500: Compare the results of the radial sensor and the axial sensor based on the minimum variance power spectrum.

[0021] Preferably, in step S100, the circumferential vibration of the blade is simplified to multi-frequency vibration and expressed as:

[0022]

[0023] Where i represents the loop variable, which refers to all integers from 1 to K, j represents the imaginary unit, and k represents the kth component of the blade circumferential vibration. represents the vibration phase, b k represents the vibration amplitude, f k represents the vibration frequency, Indicates frequency fk The corresponding array steering vector, M represents the steering vector length.

[0024] Preferably, in step S300, the autocorrelation matrix of the leaf-end timing signal is estimated by the following formula:

[0025]

[0026]

[0027] in, represents the autocorrelation matrix obtained at the i+1th iteration, A(f G , -t M-1 ) represents the steering matrix and the steering matrix is ​​composed of the steering vector a(f k , t M-1 ) constitutes, f G Indicates that the frequencies that make up the steering matrix range from f1 to f G , G represents the number of frequencies that make up the steering matrix, t M-1 Indicates the time from t0 to t that constitutes the steering matrix M-1 , represents the estimation of power spectrum density, y represents blade vibration displacement, x represents blade vibration signal, diag(·) represents taking the main diagonal elements of the matrix, (·) H represents the conjugate transpose, (·) -1 Indicates matrix inversion, A(f G , -t M-1 )=[a(f1,t M-1 ) a(f2,t M-1 )…a(f G , t M-1 )] represents the steering matrix, and the dimension of the matrix is ​​G×M.

[0028] Preferably, in step S400, the minimum variance power is calculated by the following formula:

[0029]

[0030] in, represents the inverse matrix of the leaf-end timing signal autocorrelation matrix, a(f, t M-1 ) represents the steering vector of frequency f, which is used to calculate the minimum variance power value at frequency f. When f is taken from 1 to the required maximum frequency value, the minimum variance power spectrum of the signal to be calculated is obtained; x represents the blade vibration signal, (·) H represents the conjugate transpose, represents the array steering vector corresponding to frequency f, and M represents the length of the steering vector.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present disclosure can more accurately obtain the vibration mode of the blade by arranging radial sensors and axial sensors at the same angle, thereby overcoming the disadvantage that the sensor can only be installed near the axial position of the blade in engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A front view of a radial and axial blade tip timing sensor measurement device provided in accordance with one embodiment of the present disclosure;

[0034] Figure 2 A rear view of a radial and axial blade tip timing sensor measurement device provided in accordance with one embodiment of the present disclosure;

[0035] Figure 3 for Figure 1 A schematic structural diagram of the middle bearing assembly in the device shown;

[0036] Figure 4 for Figure 1 a structural diagram of the blade assembly in the device shown;

[0037] Figure 5(a) is the vibration displacement diagram of blade No. 1 obtained by the radial blade tip timing sensor;

[0038] Figure 5(b) is the vibration displacement diagram of blade No. 1 obtained by the axial blade tip timing sensor;

[0039] Figure 6(a) is the vibration displacement diagram of blade No. 2 obtained by the radial blade tip timing sensor;

[0040] Figure 6(b) is the vibration displacement diagram of blade No. 2 obtained by the axial blade tip timing sensor;

[0041] Figure 7 This is a partial enlarged view of the vibration displacement of blade No. 1 obtained by the radial and axial blade tip timing sensors;

[0042] Figure 8 This is a partial enlarged view of the vibration displacement of blade No. 2 obtained by the radial and axial blade tip timing sensors;

[0043] Figure 9(a) is the pseudo-amplitude spectrum of the radial vibration displacement data of blade No. 1;

[0044] Figure 9(b) is the pseudo-amplitude spectrum of the axial vibration displacement data of blade No. 1;

[0045] Figure 10(a) is the pseudo-amplitude spectrum of the radial vibration displacement data of blade No. 2;

[0046] Figure 10(b) is the pseudo-amplitude spectrum of the axial vibration displacement data of blade No. 2;

[0047] The following are the descriptions of the reference numerals:

[0048] 1. Base plate; 2. Motor support; 3. Motor; 4. Coupling; 5. Bearing seat; 6. Bearing seat base plate; 7. Blade; 8. Casing; 9. Sensor support; 10. Radial blade tip timing sensor; 11. 61906 bearing; 12. Main shaft; 13. 6006 bearing; 14. Flat key; 15. Fastening nut; 16. Speed ​​sensor; 17. Nozzle bracket; 18. Nozzle hose; 19. Axial blade tip timing sensor; 20. Nozzle; 21. Nozzle bracket column; 22. Blade end cover; 23. Spacer; 24. Blade. DETAILED DESCRIPTION

[0049] The following will refer to the attached Figures 1 to 10(b) Specific embodiments of the present disclosure are described in detail. Although specific embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. Instead, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0050] It should be noted that certain words are used in the specification and claims to refer to specific components. Those skilled in the art should understand that technicians may use different nouns to refer to the same component. This specification and claims do not use the difference in nouns as a way to distinguish components, but use the difference in the functions of the components as the criterion for distinction. As mentioned throughout the specification and claims, "including" or "comprising" is an open term, so it should be interpreted as "including but not limited to". The subsequent description of the specification is a preferred embodiment of the present disclosure, but the description is based on the general principles of the specification and is not used to limit the scope of the present disclosure. The scope of protection of the present disclosure shall be as defined by the attached claims.

[0051] To facilitate understanding of the embodiments of the present disclosure, further explanation will be given below using specific embodiments as examples in conjunction with the accompanying drawings, and the accompanying drawings do not constitute a limitation on the embodiments of the present disclosure.

[0052] In one embodiment, if Figure 1 As shown, the present disclosure provides a radial and axial blade tip timing sensor measurement device, comprising:

[0053] Bottom plate 1,

[0054] A driving assembly and a blade assembly are provided on the bottom plate 1, wherein the driving assembly is used to drive the blade assembly to rotate;

[0055] An excitation component is also provided on the bottom plate to excite the blade component during the rotation process;

[0056] A signal acquisition component is also provided on the bottom plate for collecting the rotation speed signal of the blade component during the rotation process and the vibration signal generated during the excitation process.

[0057] In another embodiment, the drive assembly includes a motor support 2, which is mounted on the base plate 1 via hexagon socket bolts, a motor 3 is mounted on the motor support 2 via hexagon socket bolts, and the motor 3 is connected to the blade assembly via a coupling 4. Furthermore, the drive assembly also includes a bearing seat 5, which is connected to the base plate 1 via a bearing seat base plate 6 to fix the coupling 4. Further, as Figure 3 As shown, a concave platform is provided on the bearing seat 5, and a 61906 bearing 11 and a 6006 bearing 13 are respectively provided on both sides of the concave platform. One side of the 61906 bearing is connected to the coupling, and the other side is connected to the main shaft 12. The other side of the main shaft is connected to the 6006 bearing, and the other side of the 6006 bearing is connected to the blade assembly. By arranging the 61906 bearing and the 6006 bearing, the main shaft can be supported.

[0058] In another embodiment, Figure 4 As shown, the blade assembly includes a blisk 7, which is axially fixed to a main shaft 12 via a flat key 14. The drive assembly drives the blisk 7 to rotate via the main shaft 12. A fastening nut 15 is provided on the main shaft 12 for axially securing the blisk. The blisk 7 has multiple circumferential grooves, each of which houses a blade 24. Spacers 23 are provided between adjacent blades 24. A blisk end cap 22 is also provided on the outside of the blisk 7 to protect it.

[0059] In another embodiment, Figure 2 As shown, the excitation assembly includes a nozzle holder 17, which is secured to a nozzle holder column 21 via hexagon socket bolts. Multiple nozzles 20 are secured to the nozzle holder 17 via fastening bolts. Each nozzle is connected to an external high-pressure air pump via a hose. When the blisk rotates at high speed under the drive assembly, the high-pressure air pump sprays high-pressure gas from the nozzles to the blades through the hose, thereby exciting the blades to vibrate.

[0060] In another embodiment, the signal acquisition component includes a radial blade tip timing sensor 10 , an axial blade tip timing sensor 19 and a rotational speed sensor 16 .

[0061] In this embodiment, a speed sensor 16 is mounted on the bearing housing 5 to collect speed signals from the blades during rotation. A casing 8 is secured to the baseplate via hexagonal bolts. Circumferentially arranged on the casing 8 are rectangular slots within which are mounted sensor holders 9. Multiple radial tip timing sensors 10 (i.e., positioned radially along the blade disc) are threadedly mounted on these sensor holders 9. Furthermore, the nozzle holder also has circumferentially arranged rectangular slots within which are mounted multiple axial tip timing sensors 19 (i.e., positioned axially along the blade disc). By arranging radial blade tip timing sensors and axial blade tip timing sensors to simultaneously collect vibration data in different directions of the blade and perform comparative verification, the blade vibration shape can be obtained more accurately (because the vibration frequency of the blade is usually 10 times or even higher than the blade rotation frequency, and the sampling frequency of the sensor is much lower than the vibration frequency of the blade. If the blade tip timing sensor is only installed in the radial direction of the blade disk according to the existing blade monitoring method, it will inevitably affect the sampling accuracy of the vibration signal. Therefore, in order to overcome this problem, this embodiment arranges sensors in both radial and axial directions, and calculates the blade vibration displacement and vibration frequency through a suitable algorithm, so as to obtain a higher equivalent sampling rate, thereby improving the sampling accuracy of the vibration signal).

[0062] In another embodiment, the present disclosure further provides a radial and axial blade tip timing sensor measurement method, comprising the following steps:

[0063] S100: Simplify the circumferential vibration of the blade into multi-frequency vibration and construct the steering vector;

[0064] In this step, the circumferential vibration of the blade is simplified to the superposition of sinusoidal signals, so the circumferential vibration of the blade can be expressed as: Where i represents the loop variable, which refers to all integers from 1 to K, j represents the imaginary unit, and k represents the kth component of the blade circumferential vibration. represents the vibration phase, b k represents the vibration amplitude, f k represents the vibration frequency, Indicates frequency f k The corresponding array steering vector, M is the steering vector length.

[0065] S200: Install radial and axial sensors at the same location and collect blade tip timing signals from both sensors simultaneously;

[0066] In this step, blade tip timing sensors are installed at the same angle in the radial and axial directions around the blade disk. To ensure the reliability of the results, the angular installation error between the radial and axial sensors should be less than 2°. Data from the radial and axial sensors are then collected simultaneously, and the collection process ends at the same time, resulting in blade tip timing data for both directions.

[0067] S300: estimating the autocorrelation matrix of the leaf-end timing signal through an iterative update algorithm;

[0068] In this step, since the blade-end timing sensors are arranged in a non-uniform manner, the signal autocorrelation matrix is ​​not directly calculated by taking the inner product of the blade-end timing signal. Initialize the power spectral density function of the blade-end timing signal Iteratively update the autocorrelation matrix and power spectral density function The specific steps are as follows:

[0069]

[0070]

[0071] in, represents the autocorrelation matrix obtained at the i+1th iteration, A(f G , -t M-1 ) represents the steering matrix and the steering matrix is ​​composed of the steering vector a(f k , t M-1 ) constitutes, f G Indicates that the frequencies that make up the steering matrix range from f1 to f G , G represents the number of frequencies that make up the steering matrix, t M-1 Indicates the time from t0 to t that constitutes the steering matrix M-1 , represents the estimation of power spectrum density, y represents blade vibration displacement, x represents blade vibration signal, diag(·) represents taking the main diagonal elements of the matrix, (·) H represents the conjugate transpose, (·) -1 Indicates matrix inversion, A(f G , -t M-1 )=[a(f1,t M-1 ) a(f2,t M-1 )…a(f G , t M-1 ] represents the steering matrix, and the dimension of the matrix is ​​G×M.

[0072] S400: Calculating the minimum variance power spectrum of the leaf-end timing signal according to the autocorrelation matrix;

[0073] In this step, the minimum variance power is calculated as follows:

[0074]

[0075] in, represents the inverse matrix of the leaf-end timing signal autocorrelation matrix, a(f, t M-1) represents the steering vector of frequency f, which is used to calculate the minimum variance power value at frequency f. When f is taken from 1 to the required maximum frequency value, the minimum variance power spectrum of the signal to be calculated is obtained; x represents the blade vibration signal, (·) H represents the conjugate transpose, represents the array steering vector corresponding to frequency f, and M represents the length of the steering vector.

[0076] S500: Compare the results of the radial sensor and the axial sensor based on the minimum variance power spectrum.

[0077] In this step, the natural frequencies of the finite element simulation are compared with the experimental results. The blade tip vibration displacements collected by the radial sensor and the axial sensor are compared in the time domain, and the Pearson correlation coefficient between the two signals is calculated using the following formula:

[0078]

[0079] Among them, x1 and x2 represent radial vibration displacement and axial vibration displacement respectively, and σ x Denote the mean and standard deviation of the signal respectively, and cov(·) denotes the covariance. Based on the pseudo-spectra of the blade tip timing signals in the two directions obtained in step S400, the waveforms and amplitudes of the spectra are compared to obtain the magnitude of the blade's first-order natural frequency.

[0080] In this exemplary embodiment, the signals collected by the radial tip timing sensor and the axial tip timing sensor are analyzed separately. The radial tip timing sensor is arranged on the casing 8, with the angles between the sensors being [91°, 150°, 193°, 210°, 255°]. The axial tip timing sensor is arranged on the nozzle bracket 17, with the angles between the sensors being the same as the angles of the radially mounted sensors [91°, 150°, 193°, 210°, 255°]. The arrival times of the rotating blades are respectively obtained using the radial tip timing sensor 10 and the axial tip timing sensor 19. The blade disk is a split disk with 12 blades, made of 45# steel, with a blade radius of R = 68 mm, a blade thickness of d = 1 mm, and a blade width of w = 20 mm. Two cases are considered: with and without jet excitation. Jet excitation is achieved by spraying 0.3 MPa high-pressure gas through four jet excitation nozzles 20 mounted on the nozzle sensor bracket.

[0081] Figure 5(a) and Figure 5(b) are the vibration displacement diagrams of blade No. 1 measured by the radial and axial blade tip timing sensors, respectively. Figure 7 Figure 6(a) and Figure 6(b) are the vibration displacement diagrams of blade No. 2 measured by the radial and axial blade tip timing sensors, respectively. Figure 8Figure 6(a) and Figure 6(b) show enlarged views of the areas corresponding to the vertical lines. Due to the different radial and axial measurement points, the blade vibration amplitudes measured by the two sensors differ somewhat, but their transformation trends are very similar, as can be seen from the local enlarged view. The correlation coefficients between the vibration displacements measured by the radially and axially mounted sensors were calculated. The Pearson correlation coefficient for the vibration displacement of blade 1 was 0.9623≈1.0, and the Pearson correlation coefficient for the vibration displacement of blade 2 was 0.9834≈1.0, indicating that the two vibration displacements are very similar. This preliminarily proves the equivalence between the radially mounted and axially mounted sensors in the time domain.

[0082] According to the blade dynamics model, finite element analysis using ANSYS shows that the first-order natural frequency of the blade is around 400Hz. Therefore, the frequency range of the vibration signal to be identified is determined to be 1Hz to 1000Hz, and the frequency identification resolution Δf = 1Hz. Since the method described here requires that the rotation speed of the blade disk should be approximately uniform, the original radial vibration displacement signal and axial vibration displacement signal are intercepted, and the resonance region data with a data range of about 20s is selected. The data length is set to 256, and the maximum number of autocorrelation matrix iterations is set to 15. The autocorrelation matrices of the signals of blades 1 and 2 under the two sensor installation methods are estimated respectively.

[0083] Construct peak search vector a(f, t M-1 ), and according to the peak search vector a(f,t M-1 ) Calculate the pseudo-amplitude spectrum s of the signal matrix in the frequency range [1, 1000] Hz f , Figure 9 (a) and Figure 9 (b) are the pseudo-amplitude spectra of radial and axial vibration displacement data of the resonance zone of blade No. 1. Figure 10 (a) and Figure 10 (b) are the pseudo-amplitude spectra of radial and axial vibration displacement data of the resonance zone of blade No. 2. f The frequencies corresponding to the corresponding peaks are the frequency components contained in the blade vibration. The peaks in Figures 9(a), 9(b) and 10(a), 10(b) are mainly the blade rotation frequency and its multiple frequencies. eo , n∈[1, 2, …], the remaining frequency components are mainly the first-order natural frequency f1 of the blade vibration, among which the natural frequency of blade No. 1 at the corresponding speed is 391 Hz, and the first-order natural frequency of blade No. 2 is 388 Hz, which is consistent with the analysis results of the blade finite element. At the same time, the pseudo-spectrum estimation results of the radial sensor and the axial sensor have certain differences in amplitude due to the different measurement point positions, but the estimation results of each frequency component including the natural frequency are the same, which further verifies the equivalence of the radially installed sensor and the axially installed sensor.

[0084] Although embodiments of the present invention have been described above with reference to the accompanying drawings, the present invention is not limited to the specific embodiments and applications described above. The verification targets are not limited to specific sensor placement angles or split-blade configurations. The specific embodiments described above are merely illustrative and instructive, and are not restrictive. Persons skilled in the art, informed by this specification and without departing from the scope of the claims, may devise numerous other variations, all of which fall within the scope of protection of the present invention.

Claims

1. A method for measuring radial and axial blade tip timing sensors, comprising the following steps: S100: Simplify the circumferential vibration of the blade into multi-frequency vibration and construct the steering vector; S200: Install radial and axial sensors at the same location and collect blade tip timing signals from both sensors simultaneously; S300: estimating the autocorrelation matrix of the leaf-end timing signal through an iterative update algorithm; S400: Calculating the minimum variance power spectrum of the leaf-end timing signal according to the autocorrelation matrix; S500: Compare the results of the radial sensor and the axial sensor based on the minimum variance power spectrum; in, In step S100, the circumferential vibration of the blade is simplified to multi-frequency vibration and expressed as: ， in, i represents the loop variable, which refers to All integers from K; j represents the imaginary unit; k represents the kth component of the blade circumferential vibration, represents the vibration phase, represents the vibration amplitude, represents the vibration frequency, Indicates frequency The corresponding array steering vector, represents the length of the steering vector; In step S300, the autocorrelation matrix of the leaf-end timing signal is estimated by the following formula: ， ， in, Indicates the The autocorrelation matrix obtained by the iteration is represents the steering matrix and the steering matrix is ​​composed of the steering vector constitute, Indicates the frequencies that make up the steering matrix from arrive , G represents the number of frequencies that make up the steering matrix, Indicates the time of forming the steering matrix from arrive , represents an estimate of the power spectral density, represents the blade vibration displacement, It means taking the main diagonal elements of the matrix, represents the conjugate transpose, represents matrix inversion, represents the steering matrix, the dimension of the matrix is ​​G×M; In step S400, the minimum variance power is calculated by the following formula: ， in, represents the inverse matrix of the leaf-end timing signal autocorrelation matrix, Indicates frequency The steering vector is used to calculate the frequency The minimum variance power value when From 1 to the required maximum frequency value, the minimum variance power spectrum of the signal to be determined is obtained; represents the blade vibration signal, represents the conjugate transpose, Indicates frequency The corresponding array steering vector, M represents the steering vector length.

2. The method according to claim 1, wherein The method is performed by measuring the radial and axial blade tip timing sensors. The device comprises: a driving component, a blade component and an excitation component, The driving assembly is used to drive the blade assembly to rotate; The excitation component is used to excite the blade component during the rotation process; The device also includes a signal acquisition component for acquiring a rotation speed signal of the blade component during rotation and a vibration signal generated during excitation.

3. The method according to claim 2, wherein: The drive assembly includes a motor connected to the blade assembly via a coupling and a main shaft.

4. The method according to claim 2, wherein: The blade assembly includes a blade disk, and a plurality of blades are circumferentially arranged on the blade disk.

5. The method according to claim 4, wherein An organic casing is arranged outside the blade disc.

6. The method according to claim 2, wherein: The excitation component includes a nozzle bracket, on which a plurality of nozzles are arranged, and each nozzle is externally connected to a high-pressure air pump.

7. The method according to claim 2, wherein: The signal acquisition component includes a radial blade tip timing sensor, an axial blade tip timing sensor and a rotation speed sensor.

Images