A method and device for diagnosing loosening fault of pipe clamp of aircraft engine accessories

Abstract

The present invention provides a method and device for diagnosing the loose clamp fault of an aero-engine accessory pipeline, belonging to the field of aero-engines; the method steps include evaluating the state of the aero-engine accessory pipeline, selecting the installation position of a speed sensor according to the evaluation result; collecting the signal of the aero-engine accessory pipeline through the speed sensor; processing the signal of the aero-engine accessory pipeline; calculating the fault characteristics of the aero-engine accessory pipeline according to the signal processing result; and diagnosing the fault of the aero-engine accessory pipeline according to the fault characteristics. The present invention considers the actual nonlinear boundaries and constraints of the aero-engine accessory pipeline, takes the local structure of the accessory pipeline as the object of analysis and diagnosis, and simplifies the difficulty of pipeline dynamics modeling, analysis and calculation.

Description

Technical Field

[0001] The invention belongs to the field of aero-engines, and in particular relates to a method and equipment for diagnosing a loosening fault of a pipeline clamp of an aero-engine accessory. Background Art

[0002] As an important part of the aircraft engine system, the aircraft engine accessory pipeline is mainly used to transport media such as air, lubricating oil and fuel, so it is called the "heart" of the aircraft engine. The relative fixation between the accessory pipelines and between the pipelines and the casing is mainly achieved through clamps with pads. Since the aircraft engine inevitably has dynamic vibration loads such as rotor vibration, fluid pulsation, oscillating combustion and heat flow circulation during operation, vibration failure problems such as loose clamps in the accessory pipelines have always been very prominent.

[0003] The clamp with pad is generally composed of a metal clamp and a rubber gasket. The suspended clamp is directly connected by bolts, and the grounding clamp is fixed to the engine casing by bolts. According to the different loosening parts and loosening mechanisms, the common clamp loosening failure modes are mainly the loosening of the clamp locking bolt and the loosening of the rubber gasket. During the installation of the clamp, the bolt is tightened to a predetermined torque, and the bending, torsion and tensile energy in the connector is balanced with the friction constraint between the thread mating surface and the load-bearing surface. When this balance is destroyed by the dynamic vibration load generated by the engine, the bolt preload gradually decreases and causes the clamp to have a bolt loosening failure. The rubber gasket can effectively avoid friction between the clamp and the pipeline, increase the wrapping of the pipeline, and prevent the clamp from wearing the pipeline. However, due to the poor high temperature resistance and aging resistance of rubber materials, the gasket material is prone to problems such as loose density and weakened elasticity after long-term use, causing the rubber gasket to loosen. The loose clamp failure will induce fatigue fracture failure of the pipeline, seriously damage the integrity of the accessory pipeline system, affect the reliability of the overall structure of the aircraft engine, and cause catastrophic accidents such as aircraft destruction and death. For example, the clamp gasket of a certain type of aircraft engine fuel main pipe cracked, causing the pipe joint to vibrate more and break; the fuel supply pipe of a certain type of aircraft engine broke due to the clamp falling off, causing the engine to leak fuel in the air and stop.

[0004] Whether the locking bolts are loose or the rubber gaskets are loose, it takes a certain amount of time and process for the clamp loosening fault to germinate, develop and eventually cause the pipeline to break and fail. Therefore, if we make full use of this "time difference" and "process difference" to timely judge and accurately locate the clamp loosening fault of the aircraft engine accessory pipeline, we can eliminate the safety hazards before the pipeline breaks and fails, and ensure the stable operation of the engine and the aircraft. At the same time, the diagnosis results of the clamp loosening fault are also of great reference and guidance significance for the structural design, installation layout and maintenance strategy of the aircraft engine accessory pipeline.

[0005] Considering the structural form, vibration characteristics and clamp position of the aircraft engine accessory pipeline, the dynamic model of the pipeline is often simplified to a beam-like structure with multiple points of support. Therefore, the diagnosis of the loose clamp fault in the pipeline is correspondingly transformed into the identification of the loose connection in the beam-like structure. The main basis of the loose fault diagnosis method based on mechanical model and vibration characteristics is that the fault will change the local geometric and physical properties of the structure, thereby affecting the vibration characteristics of the entire structure. Therefore, by monitoring and analyzing the changes in the structural vibration characteristics, it is possible to reversely judge whether the fault occurs (fault identification) and the location of the fault (fault location) and other information. Common vibration characteristics include the modal parameters of the structure (natural frequency and modal vibration shape, etc.), frequency response function, output spectrum and transfer function. Compared with other characteristics, the transfer function (generally defined as the ratio of the output spectrum at different positions of the structure) has obvious advantages. For example, the transfer function is directly calculated only from the output response, avoiding the structural modal analysis and input excitation measurement; the transfer function only depends on the structural zero point, is more sensitive to the changes in the local parameters of the structure caused by the fault, and is more suitable for the diagnosis of local faults.

[0006] In 2002, Johnson and Adams from Purdue University in the United States first defined the nonlinear transfer function and determined the location of loose connections in beam-like multi-degree-of-freedom structures by simulating loose faults as internal forces with cubic stiffness nonlinearity. Afterwards, Adams' team carried out experimental verification of loose fault location on aircraft skin and automobile suspension. In order to eliminate the interference of external excitation and noise on the sensitivity of fault identification factors, the output response weighting coefficient is used to increase the difference in nonlinear transfer functions between loose structures and healthy structures. Based on the Volterra series theory, the prior art proposes a transfer function based on the nonlinear output frequency response function (NOFRF) to diagnose multiple nonlinear loose units in a one-dimensional beam-like structure. In order to avoid the complex decomposition of the nonlinear output frequency response function, the prior art directly uses the high-order nonlinear output spectrum to define the transfer function, and proposes a diagnostic method that can distinguish the number of faults. In order to effectively improve the sensitivity of the fault identification factor and the accuracy of the location factor, the identification of bolt loosening faults in satellite structures is completed based on the transfer function of the second order output spectrum (SOOS). On this basis, the influence of inherent material and boundary nonlinearity in the structure is considered and eliminated, and the identification sensitivity of multiple bolt loosening faults in satellite structures is improved. Combining the transfer function based on the second-order output spectrum and the local adjustment parameter method, an indicator factor that can accurately identify bolt loosening faults in one-dimensional ring structures is proposed, providing more detailed loosening fault location information.

[0007] From the above research status, it can be seen that the diagnostic method based on nonlinear transfer function considers the loose structural mechanics model (piecewise linear or hysteresis nonlinear model) that is more in line with the actual situation. While continuously improving the sensitivity of loose fault identification, scholars strive to provide more accurate fault location information. However, there are still three difficult problems to be solved in the diagnosis of loose clamp faults in aircraft engine accessory pipelines.

[0008] (1) Most diagnostic methods ignore the nonlinear boundaries and constraints of accessory pipes, which results in large errors in the diagnosis of loose clamp faults in actual aircraft engine accessory pipes.

[0009] (2) Most diagnostic methods require the accessory pipeline as a whole to be analyzed and diagnosed, which increases the difficulty of accessory pipeline dynamics modeling, analysis, and calculation, and also increases the workload of fault diagnosis;

[0010] (3) Most diagnostic methods require the reference state of the accessory pipeline as a diagnostic benchmark, which severely limits the effective monitoring of some aircraft engine accessory pipelines where the reference state is unknown or difficult to measure. Summary of the invention

[0011] Technical issues to be solved:

[0012] In order to avoid the shortcomings of the prior art, the present invention provides a method for diagnosing loose clamp faults of aircraft engine accessory pipelines, taking into account the actual nonlinear boundaries and constraints of the aircraft engine accessory pipelines, and taking the local structure of the accessory pipelines as the object of analysis and diagnosis, thereby solving the problems faced by the current method based on nonlinear transfer function in the process of diagnosing loose clamp faults of aircraft engine accessory pipelines.

[0013] The technical solution of the present invention is: a method for diagnosing a loosening fault of an aero-engine accessory pipeline clamp, which is characterized by the following specific steps:

[0014] Step 1: Evaluate the status of the aircraft engine accessory pipelines and select the speed sensor installation location based on the evaluation results;

[0015] Step 2: Collecting signals from the aircraft engine accessory pipeline through the speed sensor;

[0016] Step 3: Process the signal of the aircraft engine accessory pipeline;

[0017] Step 4: Calculate the fault characteristics of the aircraft engine accessory pipeline according to the signal processing result of step 3;

[0018] Step 5: Diagnose the fault of aircraft engine accessory pipeline through fault characteristics.

[0019] A further technical solution of the present invention is: in the step 1, a state and environment assessment is performed according to the structural characteristics and installation method of a certain type of aircraft engine accessory pipeline to be diagnosed, and a speed sensor is installed at a position where a clamp loosening failure is likely to occur.

[0020] A further technical solution of the present invention is: in step 2, the velocity sensor is installed at a selected position and connected to a host computer through a signal conditioner; an excitation system is installed on the accessory casing, and the excitation system is connected to the host computer; and the pipeline vibration velocity time domain signal is collected through the host computer.

[0021] A further technical solution of the present invention is: in step 3, the vibration velocity time domain signal is subjected to noise reduction processing, the time domain signal is transformed into frequency domain data, and the output spectrum and transfer function are calculated.

[0022] A further technical solution of the present invention is: the output spectrum is The superscript n represents the nth set of measured data, n = 1, 2, 3, the subscript a represents different sensor channels, a = 1, 2, 3, j is the complex frequency domain parameter, ω = 64π is the complex frequency;

[0023] The transfer function is Subscripts a and b represent different sensor channels, a, b = 1, 2, 3, a ≠ b.

[0024] A further technical solution of the present invention is: in said step 4, the fault characteristics of the accessory pipeline are calculated according to the obtained transfer function;

[0025] Based on the transfer function calculated from the first and second groups of data, the fault characteristics are:

[0026]

[0027] Among them, the superscript 12 represents the first group of data and the second group of data;

[0028] According to the transfer function calculated from the second and third groups of data, the fault characteristics are:

[0029]

[0030] The superscript 23 represents the second and third groups of data.

[0031] A further technical solution of the present invention is: in step 5, according to the obtained fault characteristics, the diagnostic characteristics are calculated to determine the clamp loosening fault in the pipeline diagnosis area; the diagnostic characteristic calculation formula is as follows:

[0032]

[0033] When the diagnostic feature β(jω) is significantly greater than zero, there is a clamp loose fault in the diagnostic area.

[0034] A diagnostic device for loose clamp faults in an aircraft engine accessory pipeline, characterized in that it comprises a speed sensor, a signal conditioner, a host computer, and a vibration excitation system; the speed sensor is installed in the middle of the pipeline, corresponding to the place where the loose clamp fault is most likely to occur, and is connected to the host computer through the signal conditioner; the vibration excitation system is installed on the accessory casing, the vibration excitation system comprises an exciter and a signal amplifier, and the exciter is connected to the host computer through the signal amplifier;

[0035] The excitation frequency and amplitude, sensor sensitivity, signal sampling frequency and signal sampling time are set on the software side of the host computer to collect the pipeline vibration velocity time domain signal.

[0036] A further technical solution of the present invention is: three speed sensors are installed in the diagnostic area in the middle of the pipeline, the speed sensor model is B&K VS-080, and the sensitivity is 75mv / (mm / s); the signal conditioner has a synchronous sampling frequency of 204.8kHz.

[0037] A further technical solution of the present invention is: the excitation system is a B&K 4842 modal excitation system, the maximum output load is 100N, the frequency range is DC-5000Hz, and the peak-to-peak value of the stroke displacement is 1inch; the sampling frequency is set to 8192Hz, and the sampling time is 12s.

[0038] Beneficial Effects

[0039] The beneficial effects of the present invention are as follows: the present invention considers the actual nonlinear boundaries and constraints of the aircraft engine accessory pipeline, takes the local structure of the accessory pipeline as the object of analysis and diagnosis, and simplifies the difficulty of pipeline dynamics modeling, analysis and calculation. At the same time, the diagnostic method only considers the aircraft engine accessory pipeline to be diagnosed, without the need for reference state and reference characteristics, reduces the workload of clamp loose fault diagnosis, and improves the practicality and effectiveness of the clamp loose fault diagnosis method. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flow chart of the diagnostic method.

[0041] Figure 2 It is a schematic diagram of signal acquisition.

[0042] Figure 3 is a bar graph of diagnostic results.

[0043] Explanation of the accompanying drawings: 1. Accessory pipeline; 2. Threaded joint; 3. Metal clamp; 4. Speed ​​sensor; 5. Signal conditioner; 6. Computer; 7. Accessory receiver; 8. Diagnostic area; 9. Vibrator; 10. Signal amplifier; 11. Horizontal axis (diagnostic area); 12. Vertical axis (diagnostic characteristic value); 13. Coordinate origin; 14. Diagnostic area; 15. Graphic label. DETAILED DESCRIPTION

[0044] The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0045] This embodiment is a method for diagnosing a loosening fault of a certain type of aircraft engine accessory pipe clamp. Figure 1 As shown below:

[0046] Step 1, assessing the status of aircraft engine accessory pipelines;

[0047] According to the structural characteristics and installation method of a certain type of aircraft engine accessory pipeline to be diagnosed, the status and environment assessment are carried out, and the vibration velocity sensor is installed at the selected measuring point.

[0048] In this embodiment, a certain type of aircraft engine accessory pipeline to be diagnosed is a single-tube three-point support beam structure. Both ends of the pipeline are connected to the aircraft engine accessory casing through threaded joints, and the middle of the pipeline is fixed to the casing by a clamp. The speed sensor measuring point is selected at the middle position of the pipeline where the clamp loosening failure is most likely to occur.

[0049] Step 2, collecting signals from the aircraft engine accessory pipeline;

[0050] Install a vibration exciter on the accessory housing to provide a simulated vibration source, and connect the vibration exciter to the computer through a signal amplifier. Install a vibration velocity sensor in the diagnostic area, and connect the sensor to the computer through a signal conditioner. Set the excitation frequency and amplitude, sensor sensitivity, signal sampling frequency and signal sampling time on the computer software to collect the pipeline vibration velocity time domain signal.

[0051] In this implementation example, Figure 2 As shown, there are 3 speed sensor signals in the diagnostic area connected to the computer through a signal conditioner. The speed sensor model is B&K VS-080, with a sensitivity of 75mv / (mm / s). The signal conditioner is independently developed and has a synchronous sampling frequency of 204.8kHz. The signal acquisition software is independently developed and can simultaneously collect vibration time domain data of 8 channels. The excitation system (exciter and signal amplifier) ​​is a B&K 4842 modal excitation system with a maximum output load of 100N, a frequency range of DC-5000Hz, and a peak-to-peak stroke displacement of 1inch. The sampling frequency is set to 8192Hz and the sampling time is 12s.

[0052] For the first set of data, the excitation frequency and amplitude were set to 16 Hz and 6 V respectively, and the time domain signals of three speed sensors were collected.

[0053] For the second set of data, the excitation frequency and amplitude were set to 16 Hz and 8 V respectively, and the time domain signals of 3 speed sensors were collected.

[0054] For the third set of data, the excitation frequency and amplitude were set to 16 Hz and 20 V respectively, and the time domain signals of three speed sensors were collected.

[0055] Step 3, processing the signal of the aircraft engine accessory pipeline;

[0056] De-noise the vibration velocity time domain signal, transform the time domain signal into frequency domain data, and calculate the output spectrum and transfer function.

[0057] In this implementation example, the first set of data is divided into 6 segments of 2s data for each sensor channel, and then a 2s noise-reduced data segment is obtained by averaging in the time domain. The 2s data is transformed into frequency domain data through a self-written Fourier transform program, and the vibration amplitude at a frequency of 32Hz is selected to calculate the transfer function shown in the following formula.

[0058]

[0059] in, is the output spectrum, the superscript 1 represents the first set of data, the subscript a represents different sensor channels, j is the complex frequency domain parameter, and ω=64π is the complex frequency. is the transfer function, the superscript 1 represents the first set of data, and the subscripts a and b represent different sensor channels.

[0060] For the second set of data, the 12s duration data of each sensor channel is divided into 6 segments of 2s duration data, and then a 2s duration noise-reduced data is obtained by averaging in the time domain. The 2s duration data is transformed into frequency domain data through a self-written Fourier transform program, and the vibration amplitude at a frequency of 32Hz is selected to calculate the transfer function shown in the following formula.

[0061]

[0062] in, is the output spectrum, the superscript 2 represents the second set of data, the subscript a represents different sensor channels, j is the complex frequency domain parameter, and ω=64π is the complex frequency. is the transfer function, the superscript 2 represents the second set of data, and the subscripts a and b represent different sensor channels.

[0063] For the third set of data, the 12s duration data of each sensor channel is divided into 6 segments of 2s duration data, and then a 2s duration noise-reduced data is obtained by averaging in the time domain. The 2s duration data is transformed into frequency domain data through a self-written Fourier transform program, and the vibration amplitude at a frequency of 32Hz is selected to calculate the transfer function shown in the following formula.

[0064]

[0065] in, is the output spectrum, the superscript 3 represents the third set of data, the subscript a represents different sensor channels, j is the complex frequency domain parameter, and ω=64π is the complex frequency. is the transfer function, the superscript 3 represents the third set of data, and the subscripts a and b represent different sensor channels.

[0066] Step 4, calculating the characteristics of the aircraft engine accessory pipeline;

[0067] According to the obtained transfer function, the fault characteristics of the accessory pipeline are calculated.

[0068] In this implementation example, the fault characteristics shown in the following formula are calculated based on the transfer function calculated based on the first set of data and the second set of data.

[0069]

[0070] Among them, α 12 (jω) is the fault characteristic, and the superscript 12 represents the first set of data and the second set of data.

[0071] Based on the transfer function calculated from the second and third sets of data, the fault characteristics are calculated as shown in the following formula.

[0072]

[0073] Among them, α 23 (jω) is the fault characteristic, and the superscript 23 represents the second and third sets of data.

[0074] Step 5, diagnose the fault of the aircraft engine accessory pipeline.

[0075] According to the obtained fault characteristics, the diagnostic characteristics are calculated to determine the loose clamp fault in the pipeline diagnosis area.

[0076] In this implementation example, the diagnostic feature shown in the following formula is calculated. Whether there is a clamp loosening fault in the pipeline diagnosis area is determined based on whether the feature value approaches zero.

[0077]

[0078] Among them, β(jω) is the diagnostic feature.

[0079] like Figure 3 As shown in Figure 1, the diagnostic feature β(jω) is significantly greater than zero. Therefore, there is a clamp loosening fault in the diagnostic area.

[0080] At this point, the diagnosis of the loose clamp failure of a certain type of aircraft engine accessory pipe has been completed.

[0081] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention without departing from the principles and intent of the present invention.

Claims

1. A method for diagnosing a loose clamp fault in an aircraft engine accessory pipeline, characterized in that The specific steps are as follows: Step 1: Evaluate the status of the aircraft engine accessory pipeline, and select the speed sensor installation position based on the evaluation results; the aircraft engine accessory pipeline to be diagnosed is a single-tube three-point support beam structure, with both ends of the pipeline connected to the aircraft engine accessory casing through threaded joints, and the middle of the pipeline is fixed to the casing through a clamp. The speed sensor measurement point is selected at the middle position of the pipeline where the clamp loosening failure is most likely to occur; Step 2: Collect signals from the aircraft engine accessory pipeline through the speed sensor; install the speed sensor at a selected position and connect it to the host computer through a signal conditioner; install a vibration excitation system on the accessory casing, and connect the vibration excitation system to the host computer; collect pipeline vibration velocity time domain signals through the host computer; Step 3: Process the signal of the aircraft engine accessory pipeline; reduce the noise of the vibration velocity time domain signal, transform the time domain signal into frequency domain data, and calculate the output spectrum and transfer function; the output spectrum is , the superscript n represents the nth set of measured data, n=1, 2, 3, and the subscript a represents different sensor channels, a=1, 2, 3, is the complex frequency domain parameter, is the complex frequency; the transfer function is ,Subscripts a, b represent different sensor channels, a, b = 1, 2, 3, a ≠ b; Step 4: Calculate the fault characteristics of the accessory pipeline of the aircraft engine according to the signal processing result of step 3; calculate the fault characteristics of the accessory pipeline according to the obtained transfer function; Based on the transfer function calculated from the first and second groups of data, the fault characteristics are: Among them, the superscript 12 represents the first group of data and the second group of data; According to the transfer function calculated from the second and third groups of data, the fault characteristics are: Among them, the superscript 23 represents the second and third groups of data; Step 5: diagnose the fault of the aircraft engine accessory pipeline through the fault characteristics; calculate the diagnostic characteristics according to the obtained fault characteristics, and determine the clamp loosening fault in the pipeline diagnosis area; the diagnostic characteristic calculation formula is as follows: When diagnostic features If it is significantly greater than zero, there is a loose clamp fault in the diagnostic area.

2. A device for implementing the method for diagnosing the loosening fault of the pipe clamp of an aircraft engine accessory according to claim 1, characterized in that: It includes a speed sensor, a signal conditioner, a host computer, and an excitation system; the speed sensor is connected to the host computer through the signal conditioner; the excitation system is installed on the accessory casing, the excitation system includes an exciter and a signal amplifier, and the exciter is connected to the host computer through the signal amplifier; The excitation frequency and amplitude, sensor sensitivity, signal sampling frequency and signal sampling time are set on the software side of the host computer to collect the pipeline vibration velocity time domain signal.

3. The device according to claim 2, characterized in that: Three velocity sensors are installed in the diagnostic area in the middle of the pipeline. The velocity sensor model is B&K VS-080 with a sensitivity of 75mv / (mm / s); the signal conditioner has a synchronous sampling frequency of 204.8kHz.

4. The device according to claim 2, characterized in that: The excitation system is a B&K 4842 modal excitation system with a maximum output load of 100N, a frequency range of DC-5000Hz, and a peak-to-peak value of stroke displacement of 1inch; the sampling frequency is set to 8192Hz, and the sampling time is 12s.

Images