A method and system for diagnosing rotor system rub fault of an aero-engine

Abstract

The present application relates to a kind of aero-engine rotor system rub-impact fault diagnosis method and system, specific process is: state evaluation, signal acquisition, signal processing, feature calculation, fault diagnosis.The present application occurs in the rub-impact fault of specific area (wheel disc) of rotor system.Using finite element method, the dynamic model of specific area rotor structure is established, its dynamic characteristics are analyzed, and by the method of multiple excitation, the nonlinear dynamics response of specific area rotor structure and related transfer function and feature are obtained, whether specific area has rub-impact fault is identified and positioned.The present application method only needs the nonlinear dynamics response of specific area of the rotor system to be diagnosed and transfer function, without reference system and reference feature, more suitable for real-time monitoring and diagnosis of rotor system.

Description

Technical Field

[0001] This invention belongs to the field of aero-engines and relates to a diagnostic method and system for rotor-static contact rubbing faults in aero-engine rotor systems. Background Technology

[0002] Due to their complex structure and harsh operating conditions, vibration problems in aero engines have always been prominent and challenging. As engine structures become more flexible and bear more complex loads, vibration issues are bound to become even more pronounced, increasing the likelihood of vibration-related failures. Vibration failures can severely reduce the structural reliability of an engine, affect its performance stability, shorten its service life, and even lead to catastrophic accidents resulting in the destruction of the aircraft and loss of life.

[0003] Rotor-stationary rubbing is one of the major vibration faults in aero-engines. It manifests as a reduction in the clearance between the rotor and stator structures, leading to contact, collision, and friction between the rotating and stationary components. Rubbing introduces nonlinear factors, severely affecting the vibration characteristics of the rotor system and even causing serious problems such as rotor instability. Therefore, scholars both domestically and internationally have proposed various rubbing diagnosis methods based on nonlinear frequency domain vibration characteristics (frequency response function, output spectrum, and transfer function).

[0004] The frequency response function is defined as the ratio of the output spectrum to the input excitation, which contains multi-mode information of the rotor system over a wide frequency range. The simulated faulty rotor system is a volterra system. Tang et al. identified the imbalance and rubbing faults in the rotor system by changing the frequency response function and combining it with principal component analysis. See: H.Tang, YHLiao, JYCao, H.Xie. Fault diagnosis approach based on volterra models[J], Mechanical Systems and Signal Processing, 2010, 24(4):1099-1113. Using synchronous sampling of input and output signals to obtain a simplified model of the rubbing rotor system, Jiang Jing et al. proposed a nonlinear spectrum analysis method for rotor rubbing faults. See: Jiang Jing, Li Zhinong, Zeng Yudong, Yuan Xianfeng. Nonlinear spectrum analysis of rotor rubbing faults[J], Vibration and Shock, 2010, 29(5):1-2. Based on this, Li Zhinong et al. introduced the nonlinear output frequency response function into the rotor system with misalignment-rubbing coupling fault, analyzed the influence of the fault on the nonlinear output frequency response function of each order of the rotor system, and proposed a corresponding rotor system misalignment-rubbing coupling fault identification method. See the reference: Li Zhinong, Li Yunlong, Diao Haiyang. Research on rotor misalignment-rubbing coupling fault diagnosis method based on nonlinear output frequency response function [J], Journal of Mechanical Engineering, 2019, 55(19):84-91.In recent years, Professor Y. Liu's team at Northeastern University proposed a fault feature based on second-order weighted contribution ratio and nonlinear output frequency response function. They further improved the sensitivity and accuracy of the fault diagnosis features in rotor systems by combining algorithms such as Crenshaw-Curtis integral, stochastic resonance, minimum cross-entropy, and relative entropy. See the following references: Y. Liu, YL Zhao, JT Li, FG Xi, SH Yu, Y. Zhang. Research on fault feature extraction method based on NOFRFs and its application in rotor faults[J], Shock and Vibration, 2019, 3524948:1-12. and Y. Liu, JT Li, KPFeng, YL Zhao, XX Yan, H. Ma. A novel fault diagnosis method for rotor rub-impact based on nonlinear output frequency response functions and stochastic resonance[J], Journal of Sound and Vibration, 2020, 481(115421):1-17.

[0005] To address the challenge of determining and measuring unknown input excitations in engineering applications, the output spectrum is directly used for rubbing diagnosis of rotor systems. These methods primarily employ nonlinear, non-stationary signal decomposition algorithms to process the vibration signals of faulty rotor systems and extract rubbing features from the output spectrum to identify rubbing faults. Commonly used algorithms include: Fourier transform (see reference: Liu Lijuan, Chen Guo, Li Chenggang, Feng Guoquan, Wang Deyou. Research on intelligent identification technology of radial rubbing position of rotor and stationary parts in aero-engine[J], Vibration and Shock, 2013, 32(3):1-5.), wavelet transform (see reference: Liu Xiandong, Li Qihan. Application of wavelet transform in early rubbing fault diagnosis of rotor system dynamic and static parts[J], Acta Aeronautica Sinica, 1999, 20(3):220-223.), empirical mode decomposition (see reference: YGLei, ZJHe, YYZi. Application of the EEMD method to rotorfault diagnosis of rotating machinery[J], Mechanical Systems and Signal Processing, 2009, 23(4):1327-1338.), local mean decomposition (see reference: LFDeng, RZZhao. Fault feature extraction of a rotor system based on local mean decomposition and teager energy kurtosis[J], Journal of Mechanical Science and Technology, 2014, 28(4): 1161-1169. Variational mode decomposition, see reference: YXWang, R.Markert, JWXiang, WGZheng. Research on variational mode decomposition and its application indetecting rub-impact fault of the rotor system[J], Mechanical Systems and Signal Processing, 2015, 60(1): 243-251. Stochastic resonance theory, see reference: NQHu, M.Chen, XSWen.The application of stochastic resonance theory for early detectingrub-impact fault of rotor system[J],Mechanical Systems and Signal Processing, 2003, 17(4): 883-895. Cyclic stationary analysis, see reference: Chen Zhongsheng, Yang Yongmin, Hu Yingqing, Shen Guoji. Application of second-order cyclic stationary analysis in early rotor rubbing fault identification [J], Mechanical Science and Technology, 2004, 23(2): 221-223. Periodic sequence transformation, see reference: Li Yungong, Liu Jie, Zhang Jinping. Rotor rubbing fault feature extraction method based on measured impact response [J], Journal of Mechanical Engineering, 2007, 43(4): 224-228. Singular value decomposition, see reference: He Tian, ​​Liu Xiandong, Li Qihan. An improved method for diagnosing rotor-stationary rubbing faults in aero-engines [J], Journal of Aerospace Power, 2008, 23(6): 1093-1097. Matching decomposition transformation, see reference: SBWang, XFChen, GYLi, X.Li, ZJHe. Matching demodulation transform with application to feature Extraction of rotor rub-impact fault[J], IEEE Transactions on Instrumentation and Measurement, 2013, 63(5): 1372-1383., Morphological component analysis, see reference: Chen Xiangmin, Yu Dejie, Li Xing, Li Rong. Application of morphological component analysis in early rotor rub-impact fault diagnosis[J], Journal of Vibration Engineering, 2014, 27(3): 466-472., Normalized complex energy operator demodulation method, see reference: XLAn. Local rub-impact fault diagnosis of a rotor system based on adaptive local iterative filtering[J], Transactions of the Institute of Measurement and Control, 2017, 395(5): 748-753., Adaptive local iterative filtering, see reference: M. Zeng, Y. Yang, JD Zheng, JSCheng.Normalized complex teager energy operatordemodulation method and its application to fault diagnosis in a rubbing rotorsystem[J], Mechanical Systems and Signal Processing, 2015, 50(1): 380-399, etc. .

[0006] The transfer function is defined as the ratio of the nonlinear output spectrum at different locations in the rotor system, eliminating the influence of system poles on diagnostic features. Professor Wen Bangchun's team at Northeastern University analyzed the relationship between the nonlinear transfer function and the parameters in the dynamic stiffness matrix of the rubbing rotor system, and proposed a rubbing identification feature based on the nonlinear transfer function for rubbing single-disc rotor systems and rubbing multi-disc rotor systems. See the following references: Han Qingkai, Yang Ying, Lang Zhiqiang, Wen Bangchun. Research on the location method of rubbing fault in rotor system based on nonlinear output frequency response function [J], Science & Technology Review, 2009, 27(2):29-32. and HLYao, QKHan, LXLi, BCWen. Detection of rubbing location in rotor system by super-harmonic responses [J], Journal of Mechanical Science and Technology, 2012, 26(8):2431-2437.

[0007] Compared to other nonlinear frequency domain vibration characteristics, the transfer function is calculated solely from the rotor system's output spectrum, thus avoiding complex modal analysis and input measurements. Furthermore, the transfer function encompasses multi-mode parameters and spatial location information of the rotor system over a wide frequency range, making it more sensitive to changes in local rotor parameters caused by rubbing. However, to date, research on the application of nonlinear transfer functions in rotor system rotor-to-station rubbing fault diagnosis has only been addressed in a very limited number of publications. Therefore, research on the application of nonlinear transfer functions in the rotor system rotor-to-station rubbing diagnosis of aero-engines still has the following shortcomings:

[0008] (1) Existing methods mainly consider single rubbing fault rotor systems, but actual rotor systems generally have rubbing coupled faults, with faults affecting each other and the coupling mechanism being complex.

[0009] (2) Existing rubbing diagnosis mainly relies on the variation law of the nonlinear transfer function properties of the overall rotor system with the rubbing position. The accuracy of the diagnosis results is easily affected by the structural form of the rotor system.

[0010] (3) Existing rubbing diagnosis methods require the rotor system to be in good condition as a reference, and the location of rubbing is determined by obvious changes in characteristics, which severely limits the diagnosis of rotor systems with unknown reference conditions. Summary of the Invention

[0011] Technical problems to be solved

[0012] To overcome the shortcomings of existing technologies, this invention proposes a diagnostic method and system for rotor-to-stationary rubbing faults in aero-engine rotor systems, addressing the problems existing in current methods based on nonlinear transfer functions during the diagnosis of rotor-to-stationary rubbing faults in aero-engine rotor systems. For example, existing methods primarily consider individual rotor systems experiencing rubbing faults; existing rubbing diagnoses mainly rely on the variation of the nonlinear transfer function properties of the overall rotor system with the rubbing location; and existing rubbing diagnosis methods require a reference to the intact state of the rotor system, determining the rubbing location through significant changes in characteristics.

[0013] Technical solution

[0014] A diagnostic method for rotor-stationary rubbing faults in an aero-engine rotor system, characterized by the following steps:

[0015] Step 1: Assess the condition of the aero-engine rotor system: Based on the structural form and support method of the aero-engine rotor system to be diagnosed, install multiple vibration displacement sensors at the locations of the measurement points to be diagnosed.

[0016] Step 2: Acquire signals from the aero-engine rotor system: The motor drives the rotor system to run at a set rotor system speed. At a set signal sampling frequency and time, four different unbalance quantities are applied to the rotor disk of the aero-engine rotor system to obtain four sets of time-domain signals from the multi-position displacement sensor of the measurement point to be diagnosed.

[0017] Step 3: Take each set of time-domain signals from each sensor channel, average them in the time domain to obtain the noise-reduced data, and then transform the time-domain data into frequency-domain data. Choose the vibration amplitude at any frequency and calculate the transfer function as shown in the following formula:

[0018]

[0019] in: The transfer function is defined by j, which is a complex frequency domain parameter, and ω is a complex frequency. The superscript n represents the nth group of data, the subscript a represents different sensor channels of the measurement point to be diagnosed, and the subscript i represents the ith measurement point to be diagnosed.

[0020] Step 4: Calculate the characteristics of the aero-engine rotor system: Based on the transfer function obtained from the first, second, and third sets of data, calculate the fault characteristics as shown in the following formula:

[0021]

[0022] Based on the transfer function calculated using the second, third, and fourth sets of data, the fault characteristics shown in the following formula are calculated:

[0023]

[0024] Where: [IF(jω)] = [TF(jω)] -1 For matrix [TF] i 1,2,3 (jω)] and [TF i 2,3,4 The inverse matrix of [jω];

[0025] The

[0026] The

[0027] Step 5: Diagnose faults in the aircraft engine rotor system:

[0028] Calculate diagnostic features Λ i (jω)

[0029]

[0030] When the feature value Λ of the i-th diagnostic measurement point i If (jω) is greater than 5%, it is determined that there is a rubbing fault at the i-th test point to be diagnosed in the rotor system.

[0031] The signal sampling frequency and signal sampling time satisfy the sampling theorem for engineering vibration signals.

[0032] The magnitude of the imbalance must be such that the acquired signal remains stable.

[0033] The diagnostic test point can be a single or multiple diagnostic test points.

[0034] The diagnostic test point is the disc part of the rotor system.

[0035] A system for diagnosing rotor-stationary rubbing faults in an aero-engine rotor system, characterized in that it includes a displacement sensor, a computer, and a memory, wherein the computer executes a computer program stored in the memory to implement the steps of the data migration method.

[0036] A readable storage medium for storing a diagnostic method for rotor-stationary rubbing faults in an aero-engine rotor system, characterized in that the readable storage medium stores a computer program that, when executed by a processor, implements the steps of the data migration method.

[0037] Beneficial effects

[0038] This invention proposes a diagnostic method and system for rotor-to-static rubbing faults in aero-engine rotor systems. The specific process includes: state assessment, signal acquisition, signal processing, feature calculation, and fault diagnosis. This invention addresses rubbing faults occurring in a specific region of the rotor system (the rotor disk). A dynamic model of the rotor structure in this specific region is established using the finite element method, its dynamic characteristics are analyzed, and the nonlinear dynamic response and related transfer functions and characteristics of the rotor structure in this specific region are obtained through multiple excitation methods, thus identifying and locating whether a rubbing fault exists in the specific region. The main innovations and contributions of this invention are as follows:

[0039] 1. By focusing on specific areas of the rotor system rather than the entire rotor system as the object of analysis and diagnosis, the difficulty of modeling and analysis is simplified, while also facilitating the accurate location of faults;

[0040] 2. Although other types of faults may affect the response of the rotor structure in a specific area, the faults do not directly affect the specific area, that is, they will not affect the diagnosis of rubbing faults in the specific area.

[0041] 3. The method of the present invention only requires the nonlinear dynamic response and transfer function of a specific region of the rotor system to be diagnosed, without the need for a reference system and reference features, making it more suitable for real-time monitoring and diagnosis of rotor systems. Attached Figure Description

[0042] Figure 1 Method Flowchart

[0043] Figure 2 Schematic diagram of signal acquisition method in the embodiment

[0044] In the diagram: 1. Left support; 2. Left bearing housing; 3. Wheel A (rubbing area A); 4. Wheel B (rubbing area B); 5. Shaft; 6. Right bearing housing; 7. Right support; 8. Signal conditioner; 9. Computer; 10. First displacement sensor 1 for wheel A; 11. Second displacement sensor 2 for wheel A; 12. Third displacement sensor 3 for wheel A; 13. Fourth displacement sensor 4 for wheel A; 14. First displacement sensor 8 for wheel B; 15. Second displacement sensor 7 for wheel B; 16. Third displacement sensor 6 for wheel B; 17. Fourth displacement sensor 5 for wheel B.

[0045] Figure 3 Diagnostic results bar chart Detailed Implementation

[0046] The present invention will now be further described in conjunction with the embodiments and accompanying drawings:

[0047] This implementation example demonstrates a method for diagnosing rotor-stationary rubbing faults in a certain type of aircraft engine rotor system. The specific operational procedures are based on... Figure 1 As shown below:

[0048] Step 1: Assess the condition of the aircraft engine rotor system;

[0049] Based on the structural form and support method of the rotor system of a certain type of aero-engine to be diagnosed, condition and environmental assessments are conducted, and vibration displacement sensors are installed at the selected measurement points.

[0050] In this implementation example, the rotor system of a certain type of aero-engine to be diagnosed is a single-shaft double-disc structure. The two ends of the shaft are connected to the bearing housing through bearings and fixed to the support. The vibration displacement sensor measuring point is selected at the disc where rotational-static collision is likely to occur.

[0051] Step 2: Acquire signals from the aero-engine rotor system;

[0052] The rotor system is driven by a motor. Vibration displacement sensors are installed in the diagnostic area, and the sensors are connected to a computer via a signal conditioner. The rotor system speed, sensor sensitivity, signal sampling frequency, and signal sampling time are set on the computer software to acquire the time-domain signal of the rotor system's vibration displacement.

[0053] In this implementation example, such as Figure 2 As shown, the eight vibration displacement sensor signals in diagnostic areas A and B are connected to the computer via a signal conditioner. The displacement sensor model is B&K IN-081, with a sensitivity of 8mV / um. The signal conditioner is independently developed, featuring a synchronous sampling frequency of 204.8kHz and 24-bit resolution. The signal acquisition software is also independently developed, capable of simultaneously acquiring vibration time-domain data from eight channels. The rotor system speed is 1200r / min, the sampling frequency is 8192Hz, and the sampling time is 12s.

[0054] The first set of data, with imbalance values ​​of 0.5g / 0°, 0.5g / 10°, 0.5g / 20°, and 0.5g / 30°, is applied to wheel A, and time-domain signals from 8 displacement sensors are collected.

[0055] The second set of data, with imbalance values ​​of 0.5g / 0°, 0.5g / 10°, 0.5g / 20°, and 0.5g / 30°, was applied to wheel A, and with imbalance values ​​of 0.5g / 0° and 0.5g / 10°, it was applied to wheel B. Time-domain signals from 8 displacement sensors were collected.

[0056] The third set of data, with imbalance values ​​of 0.5g / 0°, 0.5g / 10°, 0.5g / 20°, and 0.5g / 30°, is applied to wheel A, and with imbalance values ​​of 0.5g / 0°, 0.5g / 10°, 0.5g / 20°, and 0.5g / 30°, it is applied to wheel B. Time-domain signals from 8 displacement sensors are collected.

[0057] The fourth set of data, with imbalance values ​​of 0.5g / 0°, 0.5g / 10°, 0.5g / 20°, and 0.5g / 30°, was applied to wheel A. The imbalance values ​​of 0.5g / -10°, 0.5g / -20°, 0.5g / 0°, 0.5g / 10°, 0.5g / 20°, and 0.5g / 30° were applied to wheel B. Time-domain signals from 8 displacement sensors were collected.

[0058] Step 3: Process the signals from the aircraft engine rotor system;

[0059] The vibration displacement time-domain signal is denoised by transforming the time-domain signal into frequency-domain data and calculating the output spectrum and transfer function.

[0060] In this implementation example

[0061] The first set of data was divided into six 2-second segments from each sensor channel's 12-second data period. A 2-second segment of noise-reduced data was obtained through time-domain averaging. The 2-second data was then transformed into frequency-domain data using a self-developed Fourier transform program. The vibration amplitude at 40Hz was selected, and the transfer function was calculated as shown in the following formula.

[0062]

[0063]

[0064] in, The output spectrum is defined by the superscript 1, which represents the first set of data, the subscript a, which represents different sensor channels, j, which is the complex frequency domain parameter, and ω = 80π, which is the complex frequency. For the transfer function, the superscript 1 represents the first set of data, and the subscripts a and b represent different sensor channels.

[0065] The second set of data was divided into six 2-second segments from the 12-second data of each sensor channel. A time-domain average was used to obtain a 2-second segment of noise-reduced data. A self-developed Fourier transform program was used to transform the 2-second data into frequency domain data. The vibration amplitude at 40Hz was selected, and the transfer function was calculated as shown in the following formula.

[0066]

[0067]

[0068] in, The output spectrum is represented by the superscript 2, which indicates the second set of data, the subscript a, which indicates different sensor channels, j, which is the complex frequency domain parameter, and ω = 80π, which is the complex frequency. For the transfer function, the superscript 2 represents the second set of data, and the subscripts a and b represent different sensor channels.

[0069] The third set of data was divided into six 2-second segments from each sensor channel's 12-second data period. A 2-second segment of noise-reduced data was obtained through time-domain averaging. The 2-second data was then transformed into frequency-domain data using a self-developed Fourier transform program. The vibration amplitude at 40Hz was selected, and the transfer function was calculated as shown in the following formula.

[0070]

[0071]

[0072] in, The output spectrum is defined by the superscript 3, which represents the third set of data, the subscript a, which represents different sensor channels, j, which is a complex frequency domain parameter, and ω = 80π, which is a complex frequency. For the transfer function, the superscript 3 represents the third set of data, and the subscripts a and b represent different sensor channels.

[0073] The fourth set of data was divided into six 2-second segments from the 12-second data of each sensor channel. A time-domain average was used to obtain a 2-second segment of noise-reduced data. A self-developed Fourier transform program was used to transform the 2-second data into frequency domain data. The vibration amplitude at 40Hz was selected, and the transfer function was calculated as shown in the following formula.

[0074]

[0075]

[0076] in, For the output spectrum, the superscript 4 represents the fourth group of data, the subscript a represents different sensor channels, j is the complex frequency domain parameter, and ω=80π is the complex frequency. For the transfer function, the superscript 4 represents the fourth set of data, and the subscripts a and b represent different sensor channels.

[0077] Step 4: Calculate the characteristics of the aero-engine rotor system;

[0078] Based on the obtained transfer function, calculate the fault characteristics of the rotor system.

[0079] In this implementation example

[0080] Based on the transfer function calculated from the first set of data, the second set of data, and the third set of data, the fault characteristics shown in the following formula are calculated.

[0081]

[0082] in, The fault characteristics are indicated by the superscript 123 representing the first, second, and third sets of data, and the subscript c representing different diagnostic areas.

[0083] Based on the transfer function calculated from the second, third, and fourth sets of data, the fault characteristics shown in the following formula are calculated.

[0084]

[0085] in, The fault characteristics are indicated by the superscript 234, which represents the first, second, and third sets of data, and the subscript c, which represents different diagnostic areas.

[0086] Where: [IF(jω)] = [TF(jω)] -1 For matrix [TF] i 1,2,3 (jω)] and [TF i 2,3,4 The inverse matrix of [jω];

[0087] The

[0088] The

[0089] Step 5: Diagnose faults in the aircraft engine rotor system.

[0090] Based on the obtained fault characteristics, diagnostic characteristics are calculated to determine the rubbing faults in each diagnostic area of ​​the rotor system.

[0091] In this implementation example, the diagnostic characteristics shown in the following formula are calculated. Whether the characteristic value approaches zero determines whether a rubbing fault exists in each diagnostic region of the rotor system.

[0092]

[0093] Among them, Λ i (jvω)(c=A,B) represents diagnostic features, with the subscript c representing different diagnostic regions.

[0094] like Figure 3 As shown, because of diagnostic feature Λ A (jvω) is significantly greater than zero, which is a diagnostic feature Λ B (jvω) clearly approaches zero. The rubbing fault only exists in the diagnostic region A.

[0095] Thus, the diagnosis of the rotor-static contact rubbing fault of a certain type of aero-engine rotor system was completed.

[0096] This implementation example considers the actual nonlinear boundaries and constraints of the aero-engine rotor system, using the local rotor disk as the object of analysis and diagnosis, thus simplifying the difficulty of dynamic modeling, analysis, and calculation of the faulty rotor system. Simultaneously, the diagnostic method only considers the aero-engine rotor system to be diagnosed, eliminating the need for reference states and characteristics, reducing the workload of rotor-to-static contact rubbing diagnosis, and improving the practicality of the diagnostic method.

Claims

1. A diagnostic method for rotor-stationary rubbing faults in an aero-engine rotor system, characterized in that... The steps are as follows: Step 1: Assess the condition of the aero-engine rotor system: Based on the structural form and support method of the aero-engine rotor system to be diagnosed, install multiple vibration displacement sensors at the locations of the measurement points to be diagnosed. Step 2: Acquire signals from the aero-engine rotor system: The motor drives the rotor system to run at a set rotor system speed. At a set signal sampling frequency and time, four different unbalance quantities are applied to the rotor disk of the aero-engine rotor system to obtain four sets of time-domain signals from multiple displacement sensors at the measurement points to be diagnosed. Step 3: Take each set of time-domain signals from each sensor channel, average them in the time domain to obtain the noise-reduced data, and then transform the time-domain data into frequency-domain data. , ; Select the vibration amplitude at any frequency and calculate the transfer function as shown in the following formula: in: For transfer functions, For complex frequency domain parameters, For complex frequencies, superscript n Representing the n Group data, where subscript 'a' represents different sensor channels of the measurement point to be diagnosed, and subscript 'i' represents the first... i One diagnostic point; Step 4: Calculate the characteristics of the aero-engine rotor system: Based on the transfer function obtained from the first, second, and third sets of data, calculate the fault characteristics as shown in the following formula: Based on the transfer function calculated using the second, third, and fourth sets of data, the fault characteristics shown in the following formula are calculated: in: For matrix and The inverse matrix; The The Step 5: Diagnose faults in the aircraft engine rotor system: Calculate diagnostic features When the i Feature values ​​of each diagnostic measurement point If the value is greater than 5%, determine the rotor system's first... i One of the test points to be diagnosed has a friction fault.

2. The diagnostic method for rotor-stationary rubbing faults in an aero-engine rotor system according to claim 1, characterized in that: The signal sampling frequency and signal sampling time satisfy the sampling theorem for engineering vibration signals.

3. The diagnostic method for rotor-stationary rubbing faults in an aero-engine rotor system according to claim 1, characterized in that: The magnitude of the imbalance must be such that the acquired signal remains stable.

4. The diagnostic method for rotor-stationary rubbing faults in an aero-engine rotor system according to claim 1, characterized in that: The diagnostic test point can be a single or multiple diagnostic test points.

5. The diagnostic method for rotor-stationary rubbing faults in an aero-engine rotor system according to claim 1, characterized in that: The diagnostic test point is the disc part of the rotor system.

6. A system for diagnosing rotor-stationary rubbing faults in an aero-engine rotor system according to any one of claims 1 to 5, characterized in that, It includes a displacement sensor, a computer, and a memory, wherein the computer is used to execute a computer program stored in the memory to implement the steps of the method as described in any one of claims 1 to 5.

7. A readable storage medium for storing a diagnostic method for a rotor-stationary rubbing fault in an aero-engine rotor system according to any one of claims 1 to 5, characterized in that, The readable storage medium stores a computer program that is executed by a processor to implement the steps of the method as described in any one of claims 1 to 5.

Images