A method for identifying the dynamic vibration mode of aero-engine blades based on inner product correlation

Abstract

The present application provides a method for identifying the dynamic vibration mode of an aero-engine blade based on the inner product correlation, including: arranging a plurality of stress measurement points on the blade to obtain the resonance speed, resonance frequency and measured stress value of the blade, and expressing each measured stress value as a measured stress vector; carrying out dynamic frequency analysis under different speed conditions to obtain the frequency values and nodal stresses of each order under each speed condition, and establishing a dynamic frequency database; selecting the dynamic frequency speed closest to the resonance speed from the dynamic frequency database, and selecting a plurality of frequencies within a predetermined range from the resonance frequency as the suspected vibration mode frequencies; respectively selecting the calculated stress values of a plurality of stress measurement points in the suspected vibration mode, and expressing them as the calculated stress vectors of each suspected vibration mode; calculating the inner product correlation between the measured stress vector and each calculated stress vector; determining the most relevant inner product correlation between the calculated stress vector and the measured stress vector according to the magnitude, and the suspected vibration mode frequency corresponding to the most relevant inner product correlation is the vibration mode frequency corresponding to the resonance vibration mode.

Description

Technical Field

[0001] This application belongs to the technical field of aeroengines, and particularly relates to a method for identifying the dynamic vibration mode of aeroengine blades based on inner product correlation. Background Technique

[0002] Due to the influence of factors such as centrifugal force, temperature, and aerodynamic force, the compressor and turbine blades in an aeroengine will bear large loads under the actual working conditions of the engine, and forced resonance and blade vibration induced by air flow may occur, and even destructive vibration may be generated, endangering the safety and service life of the engine. Therefore, in the work of engine development and fault troubleshooting, etc., it is necessary to carry out the dynamic stress measurement of the blades.

[0003] The dynamic stress measurement test of engine blades can obtain the resonance frequency of the blades and the dynamic stress values at the strain gauge positions under different working conditions. Due to the limited number of strain gauges, it is difficult to comprehensively monitor the maximum vibration stress of each resonance vibration mode. Therefore, it is necessary to judge the resonance vibration mode of the blades and calculate the maximum stress based on limited data information, and then judge whether the maximum vibration stress points under each resonance vibration mode meet the requirements of high-cycle fatigue life to ensure the safe operation of the engine.

[0004] In the current dynamic stress measurement (referred to as dynamic measurement) test of engine blades, the judgment process of the blade resonance vibration mode is as follows:

[0005] 1) Select the strain gauges on the same blade, and record the measured resonance frequency of the blade and the measured vibration stress of different strain gauges at a certain resonance moment;

[0006] 2) Conduct dynamic frequency analysis on the blade at corresponding or similar rotational speeds to obtain the calculated resonance frequency of the blade considering the influence of centrifugal force and temperature, as well as the calculated relative vibration stress at the corresponding strain gauge position and strain gauge direction;

[0007] 3) Compare the measured resonance frequency with the theoretically calculated frequency. If the frequency difference is within an acceptable range, it is determined as a suspected vibration mode;

[0008] 4) Determine the suspected vibration mode according to the frequency, and compare the ratio of the measured and calculated vibration stresses at different measurement points. If the ratio is not reversed and the difference in the ratio is within an acceptable range, it is determined as this order of vibration mode.

[0009] However, the above method or process for judging the blade resonance vibration mode has the following disadvantages:

[0010] a) Judging the blade vibration mode according to the ratio of the vibration stresses at each measurement point is cumbersome and has a certain subjectivity;

[0011] b) During multiple rounds of dynamic measurement, affected by factors such as working line adjustment, air flow attack angle change, and inlet distortion, the resonance frequencies are very rich, and it is necessary to repeatedly compare different vibration modes, resulting in low work efficiency;

[0012] c) For high - order vibration modes, the large stress points of vibration are mostly concentrated at the corner positions, and the stress at the positions where there is no obvious vibration is too small. The stress differences at each measurement point are extremely large, making it difficult to form a unified identification standard for high - order vibration modes. Moreover, for the too - small measured stress values, it is difficult to exclude the test system error factors, and this error will have a great impact on the stress ratio, increasing the difficulty of high - order vibration mode identification. Summary of the Invention

[0013] The purpose of this application is to provide a dynamic vibration mode identification method for aero - engine blades based on inner - product correlation to solve or mitigate at least one problem in the background technology.

[0014] The technical solution of this application is: a dynamic vibration mode identification method for aero - engine blades based on inner - product correlation, and the method includes:

[0015] Arrange a plurality of strain gauges on the blade to form stress measurement points and conduct a dynamic stress measurement test on the blade to obtain the resonance speed, resonance frequency of the blade, and the measured stress values of each measurement point. Represent the measured stress values of each measurement point in sequence as a measured stress vector.

[0016] Conduct dynamic frequency analysis under different speed conditions to obtain the frequency values of each order and the nodal stress at the positions of each strain gauge under each speed condition, and establish a dynamic frequency database for this blade and the corresponding strain gauge arrangement scheme.

[0017] Select the dynamic frequency speed closest to the resonance speed from the dynamic frequency database, and then select multiple frequencies within a predetermined range from the dynamic frequency database that are close to the resonance frequency as the suspected vibration mode frequencies.

[0018] Respectively select the calculated stress values of multiple stress measurement points in the suspected vibration mode, and represent them in sequence as the calculated stress vectors of each suspected vibration mode.

[0019] Calculate the inner - product correlation between the measured stress vector and each calculated stress vector.

[0020] Compare the relative magnitudes of each inner - product correlation, determine the inner - product correlation that is most relevant between the calculated stress vector and the measured stress vector. Then select the suspected vibration mode frequency corresponding to the most relevant inner - product correlation in the dynamic frequency database as the vibration mode frequency corresponding to the resonance vibration mode.

[0021] Furthermore, the setting positions of the strain gauges are selected according to the previous dynamic stress measurement results or the Campbell diagram analysis results at the positions where the vibration is more obvious to be close to the maximum stress point where the resonance vibration mode may occur.

[0022] Furthermore, the positions where the vibration is more obvious include the root position applicable to the first - order vibration mode and the tip position applicable to the high - order vibration mode.

[0023] Furthermore, the node stress direction of each strain gauge is consistent with the pasting direction of the strain gauge.

[0024] Furthermore, when multiple frequencies are selected from the dynamic frequency database and determined as suspected mode frequencies, the range difference from the resonance frequency is 5%.

[0025] Furthermore, the calculation method of the inner product correlation degree between the measured stress vector and each calculated stress vector is as follows:

[0026]

[0027] In the formula, C n is the inner product correlation degree;

[0028] is the measured stress vector;

[0029] is the calculated stress vector under the nth-order mode;

[0030] σ A σ B σ C σ Z are the measured stress values of the measuring points A, B, C... Z;

[0031] σ An σ Bn σ Cn σ Zn are the corresponding calculated stress values of the measuring points A, B, C... Z under the nth-order mode.

[0032] Furthermore, the inner product correlation degree C max of the identified mode = max{C1, C2,..., C n}.

[0033] The engine blade dynamic mode shape identification method based on the calculation of the inner product correlation degree proposed in this application quickly sorts the correlation degrees between the measured vibration stress and the calculated vibration stress of adjacent orders, provides an exact basis for mode shape identification, and significantly improves work efficiency; moreover, the method of this application has good applicability to high-order modes and more reasonably considers the influence of too small measured stress values during the mode shape identification process. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions provided in this application, the drawings will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application.

[0035] Figure 1 is a schematic flow chart of the dynamic mode shape identification method of this application.

[0036] Figure 2 Schematic diagram of the position where strain gauges are pasted on the blade in the embodiment of the present application. Detailed implementation manners

[0037] To make the purpose, technical solutions and advantages of the implementation of the present application clearer, the technical solutions in the embodiments of the present application will be described in more detail below with reference to the accompanying drawings in the embodiments of the present application.

[0038] Based on the dynamic stress measurement and analysis results of engine blades, the present application establishes a relative vibration stress database for each node of the blade under different vibration modes, and establishes a dynamic vibration mode recognition method for aero-engine blades based on the inner product correlation degree by introducing the inner product correlation degree calculation method.

[0039] As Figure 1 shown, the dynamic vibration mode recognition method for aero-engine blades provided by the present application includes the following steps:

[0040] 1) Set strain gauges on the blade according to the patch plan. The patch position should be selected at a position with obvious vibration in combination with previous dynamic measurement results or Campbell diagram analysis results, close to the maximum stress point where resonance vibration modes may occur, so as to improve the accuracy of vibration mode recognition. Among them, the first-order vibration mode is generally located at the blade root position, and the high-order vibration modes are generally located at the blade tip position.

[0041] For example, in this embodiment, there are four measurement points A, B, C, and D on a certain blade. The positions of the four measurement points formed by pasting four strain gauges on the blade are shown in Figure 2 shown. Two strain gauges A and B are arranged at the axial front and rear edges of the blade tip, one strain gauge D is arranged in the middle of the blade root, and one strain gauge C is arranged at the front edge of the middle part of the blade in the radial direction. In the following text, the description will be based on the blade having four measurement points A, B, C, and D.

[0042] Conduct dynamic measurement tests and record the resonance speed N t , resonance frequency F t and the measured stress values σ A , σ B , σ C , σ D at each measurement point. These stress values are jointly represented as the measured stress vector in order

[0043] For example, the resonance speed N t obtained through dynamic measurement tests is 7000 r / min, the corresponding resonance frequency F t is 3000 Hz, and the measured stress values σ A , σ B , σ C , σ D are 40, 60, 80, and 120. Therefore, the measured stress vector That is, (40, 60, 80, 120).

[0044] 2) Conduct dynamic frequency analysis under different rotational speeds for use in querying the corresponding dynamic measurement resonance rotational speed; output the frequency values of each order and the nodal stresses at each patch position under each rotational speed condition. It should be noted that the direction of the nodal force should be consistent with the patch direction of the strain gauge; establish a dynamic frequency database for this blade and the corresponding patch scheme.

[0045] 3) Select from the dynamic frequency database the dynamic frequency rotational speed N t closest to the resonance rotational speed N f , and then select from it the frequencies within a certain range (usually 5%) of the resonance frequency F t as the suspected mode frequencies, which can be marked as F f1 , F f2 , F f3 … F fn .

[0046] For example, according to the analysis results, the dynamic frequency rotational speed N t closest to the resonance rotational speed N f is 7200 r / min, and the range of the corresponding suspected mode frequency band is 2850 - 3150 Hz. Except for the starting frequency and the ending frequency of this frequency band, taking 100 Hz as the step size, five suspected mode frequencies are obtained, that is, the determined suspected mode frequencies are F f1 = 2850, F f2 = 2900, F f3 = 3000, F f4 = 3100, F f5 = 3150 Hz.

[0047] 4) Respectively extract the calculated stress values of the four measuring points A, B, C, and D in the suspected mode, and jointly represent them in order as the calculated stress vectors of each suspected mode:

[0048] For example, the calculated stress vector in this embodiment is:

[0049] 5) Calculate the inner product correlation degrees C1, C2, C3, and C4 between the measured stress vector and each calculated stress vector and respectively according to the formula.

[0050] In this embodiment, the measured stress vector obtained through the above formula and the calculated stress vector The inner product correlations are C1 = 0.9959, C2 = 0.9335, C3 = 0.9998, C4 = 0.8908, and C5 = 0.5734 respectively;

[0051] 7) Compare the relative magnitudes of the inner product correlations C1 to C4. Determine the inner product correlation C max that is most relevant to the calculated stress vector and the measured stress vector through the formula C max = max{C1, C2, C3, C4}. Then, select the suspected modal frequency F max corresponding to the most relevant inner product correlation C fn as the modal frequency corresponding to the resonant mode.

[0052] In this embodiment, C3 is determined as the most relevant inner product correlation through C max = max{C1, C2, C3, C4}. Then, select the suspected modal frequency F f3 = 3000 Hz as the modal frequency corresponding to the resonant mode.

[0053] Based on the determined modal frequency, the maximum stress conversion work can be carried out according to the relevant data in the dynamic frequency database associated with the above modal frequency, and then it can be judged whether the maximum vibration stress point under the resonant mode meets the high-cycle fatigue life requirements. The above content will not be elaborated here.

[0054] The method of this application has the following advantages:

[0055] a) Represent the measured vibration stress and the calculated vibration stress as a vector (n measuring points) respectively. In the n-dimensional Euclidean space R n , use the included angle of the vectors to describe the correlation between the test mode and the calculated mode. The workload is small and the idea is clear, and objective mode judgment can be achieved;

[0056] b) For different resonant frequencies, by using the previously established relative vibration stress databases of each order and combining the inner product correlation calculation method, the correlations between the measured vibration stress and the calculated vibration stress of similar orders can be quickly sorted, providing a definite basis for mode identification and significantly improving the work efficiency;

[0057] c) For high-order modes, even if the stress ratio differences at each measuring point are large, the inner product correlation model is still applicable, and the calculated value is mainly dominated by the relatively large stress measured at the high-order measuring points, and the relatively small stress measured at the low-order measuring points has little influence on the calculated value, which is consistent with the actual mode judgment method.

[0058] The engine blade dynamic vibration mode identification method based on inner product correlation calculation proposed in this application quickly sorts the correlation between the measured vibration stress and the calculated vibration stress of adjacent orders, provides an exact basis for vibration mode identification, and significantly improves work efficiency. Moreover, the method of this application has good applicability to high-order vibration modes and more reasonably considers the influence brought by too small measured stress values during the vibration mode identification process.

[0059] As described above, it is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in this application should be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. A method for identifying the dynamic vibration mode of aero-engine blades based on inner product correlation, characterized in that, The method includes: Arranging a plurality of strain gauges on the blade to form stress measurement points and conducting a dynamic stress measurement test on the blade to obtain the resonance speed, resonance frequency of the blade, and the measured stress values of each measurement point, and jointly representing the measured stress values of each measurement point in order as a measured stress vector; Conducting dynamic frequency analysis under different speed conditions to obtain the frequency values of each order and the nodal stresses at the positions of each strain gauge under each speed condition, and establishing a dynamic frequency database for the blade and the corresponding strain gauge arrangement scheme; Selecting the dynamic frequency speed closest to the resonance speed from the dynamic frequency database, and then selecting multiple frequencies within a predetermined range from the resonance frequency in the dynamic frequency database as the suspected mode frequencies; Respectively selecting the calculated stress values of multiple stress measurement points in the suspected mode and jointly representing them in order as the calculated stress vectors of each suspected mode; Calculating the inner product correlation degree between the measured stress vector and each calculated stress vector; Comparing the relative magnitudes of the inner product correlation degrees, determining the inner product correlation degree with the highest correlation between the calculated stress vector and the measured stress vector, and then selecting the suspected mode frequency corresponding to the most relevant inner product correlation degree in the dynamic frequency database as the mode frequency corresponding to the resonance mode.

2. The method for identifying the dynamic vibration mode of an aeroengine blade based on the inner product correlation degree according to claim 1, wherein, The installation position of the strain gauge is selected according to the previous dynamic stress measurement results or the Campbell diagram analysis results at the position where the vibration is more obvious to be close to the maximum stress point where the resonance mode may occur.

3. The method for identifying the dynamic vibration mode of an aeroengine blade based on inner product correlation as described in claim 2, characterized in that, The position where the vibration is more obvious includes the root position applicable to the first-order mode and the tip position applicable to the high-order mode.

4. The method for identifying the dynamic vibration mode of an aero-engine blade based on the inner product correlation degree according to claim 1, wherein The nodal stress direction of each strain gauge is consistent with the bonding direction of the strain gauge.

5. The method for identifying the dynamic vibration mode of an aero-engine blade based on the inner product correlation degree according to claim 1, wherein When selecting multiple frequencies in the dynamic frequency database as the suspected mode frequencies, the range difference from the resonance frequency is 5%.

6. The method for identifying the dynamic vibration mode of an aeroengine blade based on the inner product correlation degree according to claim 1, wherein The calculation method of the inner product correlation degree between the measured stress vector and each calculated stress vector is: where C n is the inner product relevance; is the measured stress vector; Calculate the stress vector for the n-th vibration mode; σ A , σ B , σ C , σ Z are the measured stress values at measuring points A, B, C... Z; σ An 、 σ Bn 、 σ Cn 、 σ Zn are the corresponding calculated stress values of measuring points A, B, C... Z under the nth order vibration mode.

7. The method for identifying the dynamic vibration mode of an aero-engine blade based on inner product correlation as claimed in claim 6, wherein Inner product correlation degree C of the identified vibration modes max = max{C1, C2,..., C n}.

Images