Aero-engine high-speed test bed main shaft fatigue crack propagation state monitoring method

Abstract

The invention relates to the technical field of dynamic monitoring of test spindles, and discloses a fatigue crack propagation monitoring method for a spindle of a high-speed test bed of an aero-engine. A guide rail rotating device is coaxially arranged in the circumferential direction of a preset monitoring point of a central hole of the test spindle and keeps fixed lift-off from the central hole of the test spindle; a sliding rotation detection box in which a profiling array eddy current detection sensor, a signal processing unit and a wireless communication module are integrated is designed and packaged, and the sliding rotation detection box is controlled by a driving motor to rotate at a high speed along a guide rail; the profiling array eddy current detection sensor and the test main shaft form asynchronous relative movement which is simultaneous and same in direction and keeps a stable relative speed, and the relative speed is the limit detection speed of the profiling array eddy current detection sensor, so that the eddy current detection sensor can effectively obtain a detection signal; effective dynamic monitoring of crack propagation conditions of the test spindle in high-speed operation is realized, and defect change signal capture and analysis during high-speed operation of a real aero-engine are simulated.

Description

Technical Field

[0001] The invention relates to the technical field of nondestructive testing, and in particular to a method for monitoring the fatigue crack growth state of a main shaft of an aero-engine high-speed test bench. Background Art

[0002] The speed of aircraft engines is usually above 10,000 rpm. In the process of research and development, in order to obtain first-hand information to promote subsequent optimization and improvement, it is necessary to obtain it through test bench tests. Through long-term research, it is found that the original static detection method cannot effectively monitor the fatigue crack propagation trend. In the prior art, the patent with application number CN202411092811.9 discloses a real-time monitoring method for fatigue cracks in the main shaft of an engine test bench. It adopts eddy current detection technology and speed reduction test method. By prefabricating fatigue sources, it realizes the dynamic monitoring of fatigue source crack propagation of the test shaft on the test bench under different speeds, loads and other parameters, providing a basis for dynamic monitoring research. On this basis, how to realize the capture and analysis of defect change signals that meet the requirements of real high-speed operation of aircraft engines has become a new technical problem in this research work. Summary of the invention

[0003] In order to solve the above problems, the present invention provides a method for monitoring the fatigue crack growth state of a main shaft of an aircraft engine high-speed test bench, which is implemented as follows:

[0004] A method for monitoring the fatigue crack growth state of a main shaft of an aero-engine high-speed test bench comprises the following steps:

[0005] Step 1: Install a guide rail rotating device coaxially in the circumferential direction of the preset monitoring point of the center hole of the test spindle, the guide rail rotating device maintains a fixed distance from the inner surface of the center hole of the test spindle, and the guide rail rotating device includes a guide rail, a sliding rotation detection box, and a driving motor;

[0006] The sliding and rotating detection box is encapsulated with a contour array eddy current detection sensor, a signal processing unit and a wireless communication module, and the detection surface of the contour array eddy current detection sensor faces the inner surface of the central hole of the test spindle;

[0007] The driving motor is used to drive the sliding and rotating detection box to rotate at high speed along the guide rail;

[0008] Step 2: setting the rotation speed of the sliding rotation detection box to lag behind the rotation speed of the test spindle, so that the contoured array eddy current detection sensor and the inner surface of the center hole of the test spindle form asynchronous relative motion;

[0009] Step 3: Obtain the eddy current detection signal diagram of the real-time monitoring array for fatigue crack propagation at the preset monitoring point of the test spindle center hole.

[0010] Furthermore, in step 2, the asynchronous relative motion is the relative rotational motion of the contoured array eddy current detection sensor and the test spindle in the same direction but at different speeds at the same time; and when the test spindle is rotating at a variable speed, the rotation speed of the sliding rotation detection box is set to change synchronously with the test spindle, and the two maintain a constant relative speed.

[0011] Furthermore, the relative speed of rotation of the test spindle and the sliding rotation detection box is the limit detection speed of the contoured array eddy current detection sensor.

[0012] Furthermore, in step three, the acquired detection data is processed, and a dynamic compensation algorithm model is pre-established. According to the real-time rotation speed and dynamic characteristics of the test spindle and the sliding rotation detection box, the baseline and amplitude of the detection signal are adjusted to dynamically supplement the detection data and overcome interference.

[0013] Furthermore, the construction of the dynamic supplement algorithm model includes:

[0014] (1) Acquisition of the motion path of the test spindle and the sliding rotating detection box: When loading at different speeds, the test spindle and the sliding rotating detection box are subjected to forces that produce vibration, strong centrifugal force, and structural thermal expansion. The motion paths of the monitoring points and the motion paths of the detection sensors in the sliding rotating detection box under the influence of multiple factors are simulated and drawn by computer;

[0015] (2) Signal pre-compensation: Generate a simulated relative lift-off path map based on the simulated motion path of the monitoring point and the motion path of the detection sensor, and generate the first dynamic signal compensation data link based on the simulated relative lift-off path map;

[0016] Furthermore, the first dynamic signal compensation data chain is converted into sensor dynamic parameter adjustment data, and the sensor dynamic parameter adjustment data is loaded in real time during the monitoring process, so as to ensure that the detection sensor overcomes the jitter lift-off during the monitoring process and obtains an effective monitoring signal;

[0017] (3) Monitoring signal correction: The acquired monitoring signal is converted through the pre-established eddy current detection signal-lift-off value conversion model to obtain the actual relative lift-off route map. During the data processing, the simulated relative lift-off route map and the actual relative lift-off route map are compared, and then a second dynamic signal data link is generated. The second dynamic signal compensation is performed on the actually acquired monitoring signal, and the baseline and amplitude of the monitoring signal are adjusted accordingly to achieve signal correction.

[0018] Furthermore, the eddy current detection signal-lift-off value conversion model is established by: using numerical simulation technology to simulate the lift-off value caused by the vibration of the contoured array eddy current detection sensor relative to the test spindle and the corresponding eddy current detection signal value under the lift-off, and constructing the eddy current detection signal-lift-off value conversion model of the contoured array eddy current detection sensor relative to the test spindle during the entire monitoring process.

[0019] Furthermore, the monitoring method also includes establishing an intelligent identification model for the crack propagation stage;

[0020] The crack recognition model is trained and constructed by collecting the frequency domain characteristics and time domain characteristics of cracks of different types, sizes and depths as crack samples;

[0021] The corrected monitoring signal is introduced into the intelligent recognition model of the crack propagation stage to automatically identify the crack propagation stage and provide rapid feedback of the monitoring results.

[0022] Compared with the existing ones, this application can obtain the following technical effects:

[0023] The present invention sets a guide rail rotating device in the center hole of the test spindle to keep fixed and lifted from it, and designs a sliding rotating detection box encapsulating a profile array eddy current detection sensor, a signal processing unit and a wireless communication module to rotate at high speed along the guide rail, so that the profile array eddy current detection sensor and the test spindle form an asynchronous relative motion, the relative speed of the asynchronous relative motion is stable, and the relative speed is the limit detection speed of the profile array eddy current detection sensor, so that effective dynamic monitoring of crack propagation under high-speed operation of the test spindle is achieved, and the defect change signal capture and analysis when simulating the high-speed operation of a real aircraft engine is simulated. At the same time, due to the stable relative speed of the asynchronous relative motion, the signal-to-noise ratio is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the examples of the present invention or the technical solutions in the prior art or the drawings required for use in the description of the prior art, a brief introduction is given. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0025] Figure 1 Schematic diagram of the asynchronous rotation of the center hole of the aircraft engine test spindle and the eddy current detection sensor.

[0026] Figure 2 This is a simple structural diagram of the test spindle and guide rail rotating device.

[0027] In the figure:

[0028] 10-test spindle; 20-guide rail; 30-sliding rotation detection box, 31-contact array eddy current detection sensor. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but only represents selected embodiments of the present invention.

[0030] The engine test bench can evaluate various performances, perform fatigue life tests, perform fault analysis, and perform design verification by simulating the operation of the engine under actual working conditions. The most common test is the simulated rotation test, and various parameters are detected or monitored during the test. The rotation speed of an aircraft engine is usually above 10,000 revolutions. Since the rotation speed is too fast, taking a test shaft with a diameter of 0.5 meters as an example, its linear speed reaches hundreds of meters per second. So far, there is no method that can reliably detect the expansion of surface defects at such a high speed. Therefore, the speed can only be reduced during testing, such as to 500 to 1000 revolutions. This cannot satisfy the defect change signal capture and analysis when the aircraft engine is actually running at high speed. The present invention discloses a method for monitoring the fatigue crack expansion state of the main shaft of an aircraft engine high-speed test bench, comprising the following steps:

[0031] Step 1: A guide rail rotating device is coaxially installed in the circumferential direction of the preset monitoring point of the center hole of the test spindle 10, and the guide rail rotating device maintains a fixed distance with the inner surface of the center hole of the test spindle. The guide rail rotating device includes a guide rail 20, a sliding rotation detection box 30, and a driving motor;

[0032] The sliding rotation detection box 30 is encapsulated with a contour array eddy current detection sensor 31, a signal processing unit and a wireless communication module, and the detection surface of the contour array eddy current detection sensor 31 faces the inner surface of the center hole of the test spindle 10;

[0033] The driving motor is used to drive the sliding rotation detection box 30 to rotate at high speed along the guide rail 20;

[0034] By selecting a suitable profiling array eddy current detection sensor 31, it is ensured that it can work normally in a rotating state, and the guide rail rotation device is optimized so that the profiling array eddy current detection sensor 31 can remain stable during the rotation process; in this embodiment, the sliding rotation detection box 30 can be set to one or more, refer to the attached Figure 2If a plurality of sliding rotation detection boxes 30 are provided, the plurality of sliding rotation detection boxes 30 are equidistantly arranged in the guide rail 20 and rotate synchronously for detection.

[0035] Step 2: setting the rotation speed of the sliding rotation detection box 30 to lag behind the rotation speed of the test spindle 10, so that the contoured array eddy current detection sensor 31 and the inner surface of the center hole of the test spindle 10 form asynchronous relative motion;

[0036] Step 3: Obtain the eddy current detection signal diagram of the real-time monitoring array for fatigue crack propagation at the preset monitoring point of the test spindle center hole.

[0037] Furthermore, in step 2, the asynchronous relative motion is the relative rotational motion of the contoured array eddy current detection sensor and the test spindle in the same direction but at different speeds at the same time; and when the test spindle is rotating at a variable speed, the rotation speed of the sliding rotation detection box is set to change synchronously with the test spindle, and the two maintain a constant relative speed.

[0038] A linked motor control system is used to control the rotation speed of the contour array eddy current detection sensor 31 and the test spindle 10, and the two simultaneously rotate relative to each other in the same direction but at different speeds, ensuring that the contour array eddy current detection sensor can effectively capture the eddy current signal during the detection process.

[0039] The contour array eddy current detection sensor adopts a high-frequency data acquisition system to ensure that the changes in eddy current signals can be stimulated quickly and accurately, and the collected eddy current signals are processed, such as filtering and amplifying, to improve the signal-to-noise ratio of the signal.

[0040] Furthermore, the relative speed of rotation of the test spindle and the sliding rotation detection box is the limit detection speed of the contoured array eddy current detection sensor.

[0041] Different eddy current detection sensors have different limit detection speeds. The external loading speed of the sliding rotation detection box is adjusted in real time according to the limit detection speed of the selected eddy current detection sensor and the speed change of the test spindle.

[0042] That is, the linear speed of the engine test spindle is V, the limit detection speed of the eddy current detection sensor is V(sensor), then the external loading speed of the sliding rotation detection box is:

[0043] V(load)=VV(sensor)

[0044] Furthermore, the loading speed of the eddy current detection sensor in asynchronous monitoring can be calculated through the conversion formula of the rotational speed N, the angular velocity ω, and the linear velocity V.

[0045] Take the engine test spindle speed N = 10000 (unit: revolutions per minute, RPM) and the test spindle center hole radius R = 0.25 (unit: meter, m) as an example; the limit detection speed of the eddy current detection sensor is N (sensor) = 1000 (unit: revolutions per minute, RPM), and the eddy current detection sensor rotation radius r = 0.25 (unit: meter, m) (in actual monitoring, the eddy current detection sensor and the test spindle center hole should be kept appropriately lifted);

[0046] Angular velocity (unit: radians per second, rad / s):

[0047] Linear speed (unit: meter / second, m / s): v = r·ω

[0048] but:

[0049]

[0050] Vload=V-Vsensor=261.8m / s-26.18m / s=235.62m / s

[0051] The external loading speed V(load) of the eddy current sensor can be converted back to its rotational speed (RPM):

[0052]

[0053] The external loading speed of the sliding rotating detection box or the rotation speed of the sliding rotating detection box can be obtained.

[0054] In order to improve the accuracy and reliability of the detection signal, further, in step three, the acquired detection data is processed, and a dynamic compensation algorithm model is pre-established. According to the real-time rotation speed and dynamic characteristics of the test spindle and the sliding rotation detection box, the baseline and amplitude of the detection signal are adjusted to dynamically supplement the detection data and overcome interference.

[0055] Furthermore, the construction of the dynamic supplement algorithm model includes:

[0056] (1) Acquisition of the motion path of the test spindle and the sliding rotating detection box: When loading at different speeds, the test spindle and the sliding rotating detection box are subjected to forces that produce vibration, strong centrifugal force, and structural thermal expansion. The motion paths of the monitoring points and the motion paths of the detection sensors in the sliding rotating detection box under the influence of multiple factors are simulated and drawn by computer;

[0057] (2) Signal pre-compensation: Generate a simulated relative lift-off path map based on the simulated motion path of the monitoring point and the motion path of the detection sensor, and generate the first dynamic signal compensation data link based on the simulated relative lift-off path map;

[0058] Furthermore, the first dynamic signal compensation data chain is converted into sensor dynamic parameter adjustment data, and the sensor dynamic parameter adjustment data is loaded in real time during the monitoring process, so as to ensure that the detection sensor overcomes the jitter lift-off during the monitoring process and obtains an effective monitoring signal;

[0059] (3) Monitoring signal correction: The acquired monitoring signal is converted through the pre-established eddy current detection signal-lift-off value conversion model to obtain the actual relative lift-off route map. During the data processing, the simulated relative lift-off route map and the actual relative lift-off route map are compared, and then a second dynamic signal data link is generated. The second dynamic signal compensation is performed on the actually acquired monitoring signal, and the baseline and amplitude of the monitoring signal are adjusted accordingly to achieve signal correction.

[0060] Furthermore, the eddy current detection signal-lift-off value conversion model is established by: using numerical simulation technology to simulate the lift-off value caused by the vibration of the contoured array eddy current detection sensor relative to the test spindle in the step two and the corresponding eddy current detection signal value under the lift-off, and constructing the eddy current detection signal-lift-off value conversion model of the contoured array eddy current detection sensor relative to the test spindle during the entire monitoring process.

[0061] Furthermore, the monitoring method also includes establishing an intelligent identification model for the crack propagation stage;

[0062] The crack recognition model is trained and constructed by collecting the frequency domain characteristics and time domain characteristics of cracks of different types, sizes and depths as crack samples;

[0063] The corrected monitoring signal is imported into the intelligent recognition model of crack extension stage to automatically identify the crack extension stage and provide rapid feedback of the monitoring results. The recognition results can be displayed through a graphical interface, which makes it easier for test personnel to quickly and intuitively understand the relationship between crack extension and rotation speed at different stages.

[0064] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for monitoring the fatigue crack growth state of a main shaft of an aircraft engine high-speed test bench, characterized in that: The steps include: Step 1: Install a guide rail rotating device coaxially in the circumferential direction of the preset monitoring point of the center hole of the test spindle, the guide rail rotating device maintains a fixed distance from the inner surface of the center hole of the test spindle, and the guide rail rotating device includes a guide rail, a sliding rotation detection box, and a driving motor; The sliding and rotating detection box is encapsulated with a contour array eddy current detection sensor, a signal processing unit and a wireless communication module, and the detection surface of the contour array eddy current detection sensor faces the inner surface of the central hole of the test spindle; The driving motor is used to drive the sliding and rotating detection box to rotate at high speed along the guide rail; Step 2: setting the rotation speed of the sliding rotation detection box to lag behind the rotation speed of the test spindle, so that the contoured array eddy current detection sensor and the inner surface of the center hole of the test spindle form asynchronous relative motion; Step 3: Obtain the eddy current detection signal diagram of the real-time monitoring array for fatigue crack propagation at the preset monitoring point of the test spindle center hole.

2. The method for monitoring fatigue crack growth state of a main shaft of an aircraft engine high-speed test bench according to claim 1, characterized in that: In step 2, the asynchronous relative motion is the relative rotational motion of the contoured array eddy current detection sensor and the test spindle in the same direction but at different speeds at the same time; and when the test spindle is rotating at a variable speed, the rotation speed of the sliding rotation detection box is set to change synchronously with the test spindle, and the two maintain a constant relative speed.

3. The method for monitoring fatigue crack growth state of a main shaft of an aircraft engine high-speed test bench according to claim 2, characterized in that: The relative speed of the test spindle and the sliding rotation detection box is set to the limit detection speed of the contour array eddy current detection sensor.

4. The method for monitoring fatigue crack growth state of a main shaft of an aircraft engine high-speed test bench according to claim 1, characterized in that: In step three, the acquired detection data is processed, and a dynamic compensation algorithm model is pre-established. According to the real-time rotation speed and dynamic characteristics of the test spindle and the sliding rotation detection box, the detection data is dynamically supplemented to overcome interference.

5. The method for monitoring fatigue crack growth state of a main shaft of an aircraft engine high-speed test bench according to claim 4, characterized in that: The construction of the dynamic compensation algorithm model includes: (1) Acquisition of the motion path of the test spindle and the sliding rotating detection box: When loading at different speeds, the test spindle and the sliding rotating detection box are subjected to forces that produce vibration, strong centrifugal force, and structural thermal expansion. The motion paths of the monitoring points and the motion paths of the detection sensors in the sliding rotating detection box under the influence of multiple factors are simulated and drawn by computer; (2) Signal pre-compensation: Generate a simulated relative lift-off path map based on the simulated motion path of the monitoring point and the motion path of the detection sensor, and generate the first dynamic signal compensation data link based on the simulated relative lift-off path map; Furthermore, the first dynamic signal compensation data chain is converted into sensor dynamic parameter adjustment data, and the sensor dynamic parameter adjustment data is loaded in real time during the monitoring process, so as to ensure that the detection sensor overcomes jitter lift-off and eliminates interference during the monitoring process, and obtains effective monitoring signals; (3) Monitoring signal correction: The acquired monitoring signal is converted through the pre-established eddy current detection signal-lift-off value conversion model to obtain the actual relative lift-off route map. During the data processing, the simulated relative lift-off route map and the actual relative lift-off route map are compared, and then a second dynamic signal data link is generated. The second dynamic signal compensation is performed on the actually acquired monitoring signal, and the baseline and amplitude of the monitoring signal are adjusted accordingly to achieve signal correction.

6. The method for monitoring fatigue crack growth state of a main shaft of an aircraft engine high-speed test bench according to claim 5, characterized in that: The eddy current detection signal-to-lift-off value conversion model is established by: using numerical simulation technology to simulate the lift-off value caused by the vibration of the contoured array eddy current detection sensor relative to the test spindle in the step two and the corresponding eddy current detection signal value under the lift-off, and constructing the eddy current detection signal-to-lift-off value conversion model of the contoured array eddy current detection sensor relative to the test spindle during the entire monitoring process.

7. The method for monitoring fatigue crack growth state of a main shaft of an aircraft engine high-speed test bench according to claim 5, characterized in that: The monitoring method also includes establishing an intelligent recognition model for crack growth stages; The crack recognition model is trained and constructed by collecting the frequency domain characteristics and time domain characteristics of cracks of different types, sizes and depths as crack samples; The corrected monitoring signal is introduced into the intelligent recognition model of the crack propagation stage to automatically identify the crack propagation stage and provide rapid feedback of the monitoring results.

Images