Real-time sensing method and device for aircraft engine turbine tip clearance

Abstract

The present invention discloses a method for real-time sensing of aero-engine turbine tip clearance. The present invention fully considers the performance degradation of aero-engines, couples the tip clearance dynamic model with the engine onboard adaptive model, and for the first time calculates the turbine efficiency change based on the empirical formula of the influence of tip clearance on turbine efficiency, and feeds it back into the engine onboard adaptive model to correct the performance degradation estimation data of the aero-engine. The present invention also discloses a device for real-time sensing of aero-engine turbine tip clearance. Compared with the existing technology, the present invention effectively eliminates the difference in aero-engine performance estimation caused by turbine tip clearance, thereby improving the turbine tip clearance sensing accuracy and providing a real-time and accurate basis for active control of engine turbine tip clearance.

Description

Technical Field

[0001] The invention relates to a real-time sensing method for aero-engine turbine blade tip clearance, belonging to the technical field of aero-engine control. Background Art

[0002] To adapt to the rapid development of next-generation aircraft technology, aviation propulsion systems must achieve higher levels of performance, lifespan, and economy. Competitive, advanced aircraft engines must not only possess excellent power performance but also meet technical specifications such as high reliability, long life, and low fuel consumption. Under varying engine operating conditions, especially during acceleration and deceleration, differences in thermal response rates and mechanical loads on the turbine rotor and casing can cause inconsistent deformation of the turbine rotor and casing assembly. To ensure safe engine operation under these varying operating conditions, a certain clearance, known as tip clearance, must be maintained between the turbine blade tips and the casing.

[0003] The turbine is the most critical hot end component that affects the performance and life of the engine. The tip clearance between the blade tip and the casing has a significant impact on the performance and life of the engine. The leakage of gas through the tip clearance will cause the work capacity of the turbine components to decrease. The engine needs to use a higher turbine inlet temperature to generate the same thrust, which in turn leads to an increase in the engine exhaust temperature and a higher fuel consumption rate. Excessive exhaust temperature margin is also one of the main reasons why the engine needs maintenance. Studies have shown that under the premise of ensuring that the turbine rotor and stator do not rub, the benefits of reducing the gap by 0.25mm include: (1) saving fuel: the engine fuel consumption rate can be reduced by about 1%; (2) improving the performance of the aircraft engine: the turbine efficiency can be increased by about 1.6%, and the engine thrust can be increased by about 0.5%; (3) extending the service life of the engine: the engine exhaust temperature can be reduced by about 10K, and the life of the hot end components can be increased by about 1000 cycles; (4) reducing emissions: significantly reducing NO x , CO, CO2 and other pollutants and greenhouse gas emissions. Therefore, in the operation of aircraft engines, accurate perception of the turbine tip clearance in the entire flight segment and maintaining a tight tip clearance while ensuring sufficient safety margins are important ways to improve engine performance and economy. This technology is also known as active clearance control technology. Existing studies have shown that the use of active control of turbine tip clearance can achieve a maximum tip clearance gain of approximately 1.27mm. Therefore, active control of turbine tip clearance has become a hot topic and difficulty in the research of high-performance aircraft engines now and in the future. It is one of the key technologies that must be possessed, and accurate real-time perception of turbine tip clearance is the top priority.

[0004] However, current research on active turbine tip clearance control, both domestically and internationally, has primarily focused on simulation, with few reports on its practical application in engineering. To address current challenges facing active turbine tip clearance control, such as insufficient clearance sensing and the inability to address engine performance degradation, two key issues require in-depth research: 1) how to develop a high-confidence, real-time dynamic estimation model for turbine tip clearance; and 2) how to uncover the coupling mechanism between engine performance degradation and turbine tip clearance. Summary of the Invention

[0005] The purpose of the present invention is to overcome the shortcomings of existing turbine tip clearance sensing technology and provide a real-time sensing method for aircraft engine turbine tip clearance. The method fully considers the performance degradation of aircraft engines, couples the tip clearance dynamic model with the engine onboard adaptive model, and matches the engine performance changes in real time. The tip clearance with high confidence and strong real-time performance can be obtained, thereby providing a real-time and accurate basis for the active control of the engine turbine tip clearance.

[0006] The present invention specifically adopts the following technical solutions to solve the above technical problems:

[0007] A method for real-time sensing of aero-engine turbine blade tip clearance comprises the following steps:

[0008] Step 1: Based on the measurable engine parameters output by the aircraft engine sensors and the unmeasurable engine parameters output by the nonlinear airborne adaptive model with a health parameter estimation module, the total deformation of the casing, blades, and disk are calculated respectively, and then the turbine tip clearance is calculated based on the initial tip clearance and the initial lengths and total deformation of the casing, blades, and disk;

[0009] Step 2: Calculate the actual efficiency η of the turbine relative to the nominal efficiency η based on the turbine tip clearance TC obtained in step 1. nom Turbine efficiency change Δη HPT , where the actual efficiency η of the turbine is calculated according to the following formula:

[0010]

[0011] Where, turbine efficiency correction coefficient K = 1 + 0.586 (ψ Ztip 3.63 ), Zweifel load factor β1 and β2 are the flow inlet and outlet angles, respectively, and the airfoil cascade density ratio c zm and s m are the blade axial chord and the average radius of the cascade pitch, r r 、r s are the outer diameter of the wheel and the inner diameter of the casing, L b is the blade length, TCnom is the nominal value of turbine tip clearance;

[0012] Step 3: When the aircraft engine is in a dynamic state, use the turbine efficiency change Δη HPT Correct the health parameter estimation value output by the health parameter estimation module in the nonlinear airborne adaptive model, and then go to step 1.

[0013] Preferably, the total deformation of the casing, blades, and disc is calculated as follows:

[0014] Casing: Calculate the convective heat transfer coefficients of the outer and inner surfaces of the casing. Then, calculate the temperature field distribution of the casing through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and modify the temperature-dependent material properties. Then, calculate the thermal deformation of the casing through a 1-D elastic thermal deformation calculation. Finally, integrate the material strain in the radial direction to obtain the total deformation of the casing.

[0015] Blades: Calculate the average convective heat transfer coefficient on the blade surface. Then, calculate the temperature of the blade using the lumped parameter method and correct the temperature-dependent material properties to obtain the temperature field distribution. Then, calculate the thermal deformation of the blade using the 0-D elastic thermal deformation method and calculate the centrifugal deformation of the blade using the rotor dynamics method. Finally, the total deformation of the blade is obtained by superimposing the thermal deformation and the centrifugal deformation.

[0016] Impeller: The convective heat transfer coefficients of the front and rear surfaces of the impeller are calculated. The temperature field distribution of the impeller is then determined through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and correction of temperature-dependent material properties. The thermal deformation of the impeller is then calculated through a 1-D elastic thermal deformation calculation. External loads on the turbine impeller during rotation are introduced through blade rotor dynamics calculations, and the centrifugal strain of the impeller is then calculated through the impeller rotor dynamics calculation. Finally, the obtained centrifugal strain of the impeller is superimposed with the thermal strain to obtain the total deformation of the impeller.

[0017] Preferably, the health parameter estimation module is a health parameter estimation module based on unscented Kalman filtering.

[0018] Preferably, the following method is used to identify the operating condition of the aircraft engine: if the absolute value of the difference between the mean of the flight altitude, Mach number, and fan speed of the previous several sampling periods and the flight altitude, Mach number, and fan speed at the current moment is less than the corresponding difference threshold, and the standard deviation of the flight altitude, Mach number, and fan speed of the previous several sampling periods is less than the corresponding standard deviation threshold, then it is determined that the aircraft engine is in a steady-state operating condition; otherwise, it is determined that the aircraft engine is in a dynamic operating condition.

[0019] Based on the same inventive concept, the following technical solutions can also be obtained:

[0020] A real-time sensing device for turbine blade tip clearance of an aero-engine, comprising:

[0021] The tip clearance sensing model is used to calculate the total deformation of the casing, blades, and disk based on the measurable engine parameters output by the aircraft engine sensors and the unmeasurable engine parameters output by the nonlinear airborne adaptive model with a health parameter estimation module. The turbine tip clearance is then calculated using the initial tip clearance and the initial lengths and total deformations of the casing, blades, and disk.

[0022] HPT efficiency correction module is used to calculate the actual efficiency η of the turbine relative to the nominal efficiency η based on the turbine tip clearance TC obtained by the tip clearance perception model nom Turbine efficiency change Δη HPT , where the actual efficiency η of the turbine is calculated according to the following formula:

[0023]

[0024] Where, turbine efficiency correction coefficient K = 1 + 0.586 (ψ Ztip 3.63 ), Zweifel load factor β1 and β2 are the flow inlet and outlet angles, respectively, and the airfoil cascade density ratio c zm and s m are the blade axial chord and the average radius of the cascade pitch, r r 、r s are the outer diameter of the wheel and the inner diameter of the casing, L b is the blade length, TC nom is the nominal value of turbine tip clearance;

[0025] Steady-state operating condition identification module, used to identify whether the aircraft engine is in a steady-state operating condition or a dynamic operating condition;

[0026] The mode switching module is used to use the turbine efficiency change Δη when the aircraft engine is in a dynamic working state. HPT Correcting the health parameter estimation value output by the health parameter estimation module in the nonlinear airborne adaptive model.

[0027] Preferably, the total deformation of the casing, blades, and disc is calculated as follows:

[0028] Casing: Calculate the convective heat transfer coefficients of the outer and inner surfaces of the casing. Then, calculate the temperature field distribution of the casing through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and modify the temperature-dependent material properties. Then, calculate the thermal deformation of the casing through a 1-D elastic thermal deformation calculation. Finally, integrate the material strain in the radial direction to obtain the total deformation of the casing.

[0029] Blades: Calculate the average convective heat transfer coefficient on the blade surface. Then, calculate the temperature of the blade using the lumped parameter method and correct the temperature-dependent material properties to obtain the temperature field distribution. Then, calculate the thermal deformation of the blade using the 0-D elastic thermal deformation method and calculate the centrifugal deformation of the blade using the rotor dynamics method. Finally, the total deformation of the blade is obtained by superimposing the thermal deformation and the centrifugal deformation.

[0030] Impeller: The convective heat transfer coefficients of the front and rear surfaces of the impeller are calculated. The temperature field distribution of the impeller is then determined through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and correction of temperature-dependent material properties. The thermal deformation of the impeller is then calculated through a 1-D elastic thermal deformation calculation. External loads on the turbine impeller during rotation are introduced through blade rotor dynamics calculations, and the centrifugal strain of the impeller is then calculated through the impeller rotor dynamics calculation. Finally, the obtained centrifugal strain of the impeller is superimposed with the thermal strain to obtain the total deformation of the impeller.

[0031] Preferably, the health parameter estimation module is a health parameter estimation module based on unscented Kalman filtering.

[0032] Preferably, the steady-state operating condition identification module uses the following method to identify the operating condition of the aircraft engine: if the absolute value of the difference between the mean of the flight altitude, Mach number, and fan speed of the previous several sampling periods and the flight altitude, Mach number, and fan speed at the current moment is less than the corresponding difference threshold, and the standard deviation of the flight altitude, Mach number, and fan speed of the previous several sampling periods is less than the corresponding standard deviation threshold, then it is determined that the aircraft engine is in a steady-state operating condition; otherwise, it is determined that the aircraft engine is in a dynamic operating condition.

[0033] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0034] The present invention fully considers the performance degradation of aircraft engines, couples the tip clearance dynamic model with the engine onboard adaptive model, and for the first time calculates the turbine efficiency change based on the empirical formula for the influence of tip clearance on turbine efficiency, and feeds it back into the engine onboard adaptive model to correct the performance degradation estimation data of the aircraft engine, effectively eliminating the aircraft engine performance estimation difference caused by turbine tip clearance, thereby improving the turbine tip clearance perception accuracy, and providing a real-time and accurate basis for the active control of the engine turbine tip clearance. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a schematic diagram of the architecture of a blade tip clearance sensing model in a specific embodiment;

[0036] Figure 2 Schematic diagram for parameter definition of turbine rotor blade cascade;

[0037] Figure 3 Schematic diagram of the structural principle of a nonlinear airborne adaptive model in a specific embodiment;

[0038] Figure 4 Schematic diagram of the structure principle of a real-time sensing device for turbine blade tip clearance in a specific embodiment;

[0039] Figure 5 Flowchart of parameter interaction between engine / airborne model and tip clearance perception model;

[0040] Figure 6 Schematic diagram of the engine steady-state operating condition identification module based on sliding time window. DETAILED DESCRIPTION

[0041] In order to overcome the shortcomings of existing turbine tip clearance sensing technology, the solution of the present invention is to fully consider the performance degradation of the aircraft engine, couple the tip clearance dynamic model with the engine onboard adaptive model, and for the first time calculate the turbine efficiency change based on the empirical formula of the influence of tip clearance on turbine efficiency, and feed it back to the engine onboard adaptive model to correct the performance degradation estimation data of the aircraft engine, so as to eliminate the aircraft engine performance estimation difference caused by turbine tip clearance, thereby improving the turbine tip clearance sensing accuracy.

[0042] The method for real-time sensing of aero-engine turbine blade tip clearance proposed by the present invention comprises the following steps:

[0043] Step 1: Based on the measurable engine parameters output by the aircraft engine sensors and the unmeasurable engine parameters output by the nonlinear airborne adaptive model with a health parameter estimation module, the total deformation of the casing, blades, and disk are calculated respectively, and then the turbine tip clearance is calculated based on the initial tip clearance and the initial lengths and total deformation of the casing, blades, and disk;

[0044] Step 2: Calculate the actual efficiency η of the turbine relative to the nominal efficiency η based on the turbine tip clearance TC obtained in step 1. nom Turbine efficiency change Δη HPT , where the actual efficiency η of the turbine is calculated according to the following formula:

[0045]

[0046] Where, turbine efficiency correction coefficient K = 1 + 0.586 (ψ Ztip 3.63 ), Zweifel load factor β1 and β2 are the flow inlet and outlet angles, respectively, and the airfoil cascade density ratio c zm and s m are the blade axial chord and the average radius of the cascade pitch, r r 、r s are the outer diameter of the wheel and the inner diameter of the casing, L b is the blade length, TC nom is the nominal value of turbine tip clearance;

[0047] Step 3: When the aircraft engine is in a dynamic state, use the turbine efficiency change Δη HPT Correct the health parameter estimation value output by the health parameter estimation module in the nonlinear airborne adaptive model, and then go to step 1.

[0048] The real-time sensing device for aero-engine turbine blade tip clearance proposed by the present invention comprises:

[0049] The tip clearance sensing model is used to calculate the total deformation of the casing, blades, and disk based on the measurable engine parameters output by the aircraft engine sensors and the unmeasurable engine parameters output by the nonlinear airborne adaptive model with a health parameter estimation module. The turbine tip clearance is then calculated using the initial tip clearance and the initial lengths and total deformations of the casing, blades, and disk.

[0050] HPT efficiency correction module is used to calculate the actual efficiency η of the turbine relative to the nominal efficiency η based on the turbine tip clearance TC obtained by the tip clearance perception model nom Turbine efficiency change Δη HPT , where the actual efficiency η of the turbine is calculated according to the following formula:

[0051]

[0052] Where, turbine efficiency correction coefficient K = 1 + 0.586 (ψ Ztip 3.63 ), Zweifel load factor β1 and β2 are the flow inlet and outlet angles, respectively, and the airfoil cascade density ratio c zm and s m are the blade axial chord and the average radius of the cascade pitch, r r 、r s are the outer diameter of the wheel and the inner diameter of the casing, L b is the blade length, TC nom is the nominal value of turbine tip clearance;

[0053] Steady-state operating condition identification module, used to identify whether the aircraft engine is in a steady-state operating condition or a dynamic operating condition;

[0054] The mode switching module is used to use the turbine efficiency change Δη when the aircraft engine is in a dynamic working state. HPT Correcting the health parameter estimation value output by the health parameter estimation module in the nonlinear airborne adaptive model.

[0055] To facilitate public understanding, the technical solution of the present invention is described in detail below through a specific embodiment with reference to the accompanying drawings:

[0056] First, a tip clearance perception model was constructed. This model calculates the total deformation of the casing, blades, and disk based on the measurable engine parameters output by aircraft engine sensors and the unmeasurable engine parameters output by a nonlinear airborne adaptive model with a health parameter estimation module. The turbine tip clearance was then calculated using the initial tip clearance and the initial lengths and total deformations of the casing, blades, and disk.

[0057] The modeling method of the above-mentioned tip clearance perception model can adopt various existing technologies. However, considering the many factors such as load deformation between components that cannot be ignored but are often neglected in the existing tip clearance modeling research process, the present invention further extracts the main and secondary factors affecting the tip clearance change based on the first principles. While simplifying the modeling process, it supplements the influence of the centrifugal force of the blade on the deformation of the impeller, as well as the factors such as material properties that change with temperature, which were previously ignored in the modeling process. The radial deformation of the casing, impeller and blades is calculated separately to obtain a turbine tip clearance model with high confidence and strong real-time performance. Its basic structure is as follows: Figure 1 As shown, the details are as follows:

[0058] Casing: Based on the measurable engine parameters output by aircraft engine sensors and the unmeasurable engine parameters output by a nonlinear airborne adaptive model with a health parameter estimation module, the convective heat transfer coefficients of the casing's outer and inner surfaces are calculated. The casing's temperature field distribution is then determined through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and temperature-dependent material property correction. The casing's thermal deformation is then calculated using a 1-D elastic thermal deformation calculation. Finally, the material strain is integrated in the radial direction to obtain the total casing deformation.

[0059] Blades: Based on the measurable engine parameters output by aircraft engine sensors and the unmeasurable engine parameters output by a nonlinear airborne adaptive model with a health parameter estimation module, the average convective heat transfer coefficient of the blade surface is calculated. The temperature field distribution of the blade is then obtained through temperature calculation of the 0-D heat transfer process based on the lumped parameter method and correction of temperature-dependent material properties. The thermal deformation of the blade is then calculated through 0-D elastic thermal deformation calculation, and the centrifugal deformation of the blade is obtained through blade rotor dynamics calculation. Finally, the thermal deformation and centrifugal deformation are superimposed to obtain the total deformation of the blade.

[0060] Wheel: Based on the measurable engine parameters output by aircraft engine sensors and the unmeasurable engine parameters output by a nonlinear airborne adaptive model with a health parameter estimation module, the convective heat transfer coefficients of the front and rear surfaces of the wheel are calculated. The temperature field distribution of the wheel is then obtained through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and correction of temperature-dependent material properties. The thermal deformation of the wheel is then calculated through a 1-D elastic thermal deformation calculation. The external load on the turbine wheel during rotation is introduced through blade rotor dynamics calculations, and the centrifugal strain of the wheel is then calculated through wheel rotor dynamics calculations. Finally, the obtained wheel centrifugal strain is superimposed with the thermal strain to obtain the total deformation of the wheel.

[0061] Finally, based on the known initial tip clearance, the initial lengths of each component, and the calculated total deformation of the casing, blades, and disk, the dynamic change of the turbine tip clearance with engine operating conditions, TC(t), can be characterized in real time.

[0062] 2. Based on the principle of turbine tip clearance leakage flow, the present invention establishes a turbine efficiency correction model to quantitatively calculate the effect of tip clearance on turbine efficiency. The specific implementation process is as follows:

[0063] In the engine dynamic model, the turbine nominal efficiency η is usually used nomTo reflect the turbine efficiency at the design point, since the gas leakage generated by the tip clearance will cause the turbine to work less, when the actual tip clearance is different from the nominal clearance at the design point, the actual efficiency of the turbine components will also be different from the nominal efficiency η nom In order to accurately measure the change in turbine efficiency caused by the turbine tip clearance, the present invention defines the turbine efficiency correction coefficient K and constructs the following empirical relationship:

[0064]

[0065] Where TC is the tip clearance, η is the turbine efficiency, r r 、r s are the outer diameter of the wheel and the inner diameter of the casing, L b is the blade length, subscript nom Represents the nominal value. The high-pressure turbine characteristics can be corrected by the turbine efficiency correction factor K to reflect the deviation between the actual turbine efficiency and the nominal efficiency. The calculation formula is as follows:

[0066] K=1+0.586(ψ Ztip 3.63 )

[0067] where ψ Ztip is the Zweifel load factor, which is defined as the ratio of the actual tangential load in the blade cascade to the ideal load. Its value is related to the cascade density and the aerodynamic load (represented by the flow deflection angle):

[0068]

[0069] Among them, β1 and β2 are the flow inlet and outlet angles respectively (note that β2 is always negative), which can be referred to Figure 2 The rotor blade cascade in the airfoil blade cascade "thickness" ratio σ is defined as follows:

[0070]

[0071] Among them, the variable c zm and s m are the blade axial chord and the average radius of the blade pitch. The term "density" here can be understood as the density of the blades, because the lower the pitch, the more blades appear and the greater the density of the blades. The average radius value of the blade pitch s m The calculation is as follows:

[0072]

[0073] Among them, r m and z are the root mean square of the outer diameter of the wheel and the inner diameter of the casing and the number of blades respectively.

[0074] 3. Establish a high-precision nonlinear airborne adaptive model with a health parameter estimation module that can track the actual engine performance changes; the nonlinear airborne adaptive model constructed in this embodiment is as follows: Figure 3 The specific establishment process is as follows:

[0075] S1. Establish a component-level mathematical model of an aircraft engine. The discrete nonlinear system dynamic model expression is as follows:

[0076] x k+1 =f(x k ,u k ,h k )+w k

[0077] y k =g(x k ,u k ,h k )+v k

[0078] In the formula is the state quantity, including the fan speed N f and core engine speed N3; is the controlled quantity, i.e. the fuel quantity Wf; is the measurable output, including the fan speed N f and core engine speed N3, duct inlet total pressure Pt 17 , high pressure compressor inlet total temperature Tt 25 , total pressure Pt 25 and outlet total temperature Tt 30 , static pressure Ps 30 , total temperature at high pressure turbine outlet Tt 45 ; is the component health parameter degradation to be estimated, including the fan flow SW FAN and efficiency SE FAN , low pressure compressor flow SW LPC Efficiency SE LPC , high pressure compressor flow SW HPC and efficiency SE HPC , high pressure turbine flow SW HPT and efficiency SE HPT ; and is zero-mean, uncorrelated Gaussian white noise, with w ~ (0, Q) (Q is the system noise covariance matrix), v ~ (0, R) (R is the measurement noise covariance matrix);

[0079] S2. In order to obtain the degradation of the engine health parameters through the nonlinear filtering estimation method, the health parameters are augmented into the state variables of the system. The obtained augmented discrete nonlinear system dynamic model is expressed as follows:

[0080]

[0081]

[0082] According to the above expression, a nonlinear airborne adaptive model of an aircraft engine based on the UKF (unscented Kalman filter) estimator is established, which is mainly composed of an airborne model and a steady-state operating condition identification module. Under the same input and the same environmental conditions, the UKF estimator is based on the measurable output parameters y of the real engine. k+1 Estimating parameters with the onboard model Deviation Δy k+1 Estimate the degradation of the current engine health parameters and modify the onboard model in real time so that the onboard model always tracks the actual engine operating status, thereby estimating the true values ​​of unmeasurable parameters such as the high-pressure turbine inlet temperature parameter T4 and the inlet flow rate W4;

[0083] S3 and UKF estimators are prior art and will not be described here in detail. Figure 3 As shown, using an unscented Kalman filter to estimate health parameters involves two phases: time update and measurement update. Based on the posterior state estimate, the onboard model health parameters can be continuously updated to track the actual engine operating state, thereby calculating the true values ​​of unmeasurable parameters such as the current thrust of the actual engine and the surge margin of the high-pressure compressor. It is important to note that the sigma points generated during the time update can be reused during the calculation process, rather than generating new sigma points based on the prior state estimate and the prior state covariance matrix. This reduces the computational burden with a small loss of accuracy.

[0084] 4. Based on the above-mentioned blade tip clearance perception model, turbine efficiency correction model, and nonlinear airborne adaptive model, a turbine blade tip clearance real-time perception device of the present invention is constructed:

[0085] The turbine blade tip clearance real-time sensing device constructed in this embodiment is as follows: Figure 4 As shown, the specific working process is as follows:

[0086] S1. The input parameters of the real engine include the actuator control quantity u, system noise w, and flight environment parameters H, Ma, T sd , the engine health parameter h, the turbine efficiency change Δη is calculated in real time from the current tip clearance estimate HPT ;

[0087] S2. The measurable parameters y of the real engine are output through the sensor module, taking into account the influence of measurement noise v;

[0088] S3, the actual engine operating state is regulated by the engine controller, which receives the throttle angle command PLA from the flight console and adjusts the engine state according to the flight environment parameters H, Ma, T sd The corresponding fan speed is calculated and then compared with the fan speed n fed back by the sensor. f The current error is obtained by subtraction, and the control amount (fuel amount) Wf is calculated. f,cmd , acts on the actuator to achieve closed-loop speed control;

[0089] S4, Variable Bleed Valve (VBV) and Variable Area Fan Nozzle (VAFN) are controlled by the current fan speed n f and flight environment parameters H,Ma, and perform open-loop control according to the designed scheduling plan;

[0090] S5. The input of the nonlinear airborne adaptive model includes the actuator control quantity u, the flight environment parameters H, Ma, T sd , and the estimated values ​​of health parameters that are adaptively adjusted during operation

[0091] S6. During the operation of the nonlinear airborne adaptive model, the output measurable parameters The difference between the measurable parameter y output by the real engine sensor and the estimated health parameter value is obtained through the normalization process and UKF estimator.

[0092] S7, After correction, the state of the nonlinear onboard adaptive model will match the real engine, and the engine state parameters that cannot be measured by the sensor module will be calculated;

[0093] S8, the change in high-pressure turbine efficiency Δη caused by the introduction of turbine tip clearance HPT , correct the estimated health parameters in the dynamic process

[0094] S9, introduce the steady-state condition identification module, whose output ssFlag will be used as the health parameter of the nonlinear airborne adaptive model The judgment basis for estimating mode switching is that when the engine is in steady-state condition, ssFlag=0, and the health parameter estimation of the onboard adaptive model is Completely calculated by the UKF estimator; when the engine is in dynamic condition, ssFlag=1, the health parameter estimation of the onboard adaptive model The turbine efficiency change Δη caused by the turbine tip clearance is added to the result of the steady-state estimation. HPT , so that the onboard adaptive model matches the real engine during dynamic processes;

[0095] S10, such as Figure 5 As shown, input the initial state parameters and control parameters to complete the model initialization;

[0096] S11, the measurable parameters provided by the real engine and the unmeasurable parameters that affect the tip clearance change provided by the onboard adaptive model are sent to the tip clearance perception model in real time;

[0097] S12. Driven by both measurable and unmeasurable parameters, the tip clearance perception model updates the heat transfer characteristics and material properties of turbine components using thermodynamic parameters according to modeling principles, and solves for the deformation of the casing, impeller, and blades.

[0098] S13. Calculate the turbine tip clearance value and turbine efficiency change Δη based on component deformation HPT , and fed back to the real engine and the nonlinear airborne adaptive model; finally, the turbine tip clearance parameters are output and the calculation of the next time step begins.

[0099] Considering that the engine's rapid response and large dynamic deviation during acceleration and deceleration hinder accurate estimation of health parameter degradation, the present invention utilizes the UKF estimator to track engine degradation only during steady-state conditions. First, the rate of change of the PLA allows for timely and accurate identification of the onset of engine acceleration and deceleration. The UKF estimator can be disabled just before the engine is about to enter dynamic operation, preventing sudden changes in the estimated results. The actual engine operating condition is then identified using the following criteria:

[0100] (1) In the sliding time window of the first N sampling periods (i, i+1, ..., i+N-1), as Figure 6 As shown, it is determined whether the absolute value of the difference between the mean value of the flight altitude Alt, the Mach number Ma and the fan speed Nf and the data at the current moment (i+N) is less than the threshold values ​​mAltThr, mMaThr and mNfThr respectively;

[0101] (2) In the time window of the first N sampling periods, determine whether the standard deviations of Alt, Ma, and Nf are less than the thresholds stdAltThr, stdMaThr, and stdNfThr, respectively.

[0102] If all of the above conditions are met, the aircraft engine is determined to be in a steady-state operating condition; otherwise, the aircraft engine is determined to be in a dynamic operating condition.

[0103] The current working state of the engine can be accurately and timely determined based on parameters such as PLA, Alt, Ma, and Nf. The determination result will serve as an instruction to trigger the UKF estimator to run:

[0104] 1) ssFlag = 0 → Enable → The engine is in steady-state operation, and the UKF estimator estimates the degradation of engine health parameters;

[0105] 2) ssFlag = 1 → Disable → The engine is in dynamic operating condition and the UKF estimator is prohibited from running.

[0106] In this embodiment, the threshold settings of the engine steady-state operating condition judgment module are as follows: dPLAThr=0.2, mAltThr=100, mMaThr=0.05, mNfThr=5, stdAltThr=50, stdMaThr=0.05, stdNfThr=0.5, N=100.

Claims

1. A real-time sensing method for turbine tip clearance of an aircraft engine, characterized in that: The following steps are involved: Step 1: Based on the measurable engine parameters output by the aircraft engine sensors and the unmeasurable engine parameters output by the nonlinear airborne adaptive model with a health parameter estimation module, the total deformation of the casing, blades, and disk are calculated respectively, and then the turbine tip clearance is calculated based on the initial tip clearance and the initial lengths and total deformation of the casing, blades, and disk; Step 2: Calculate the actual efficiency η of the turbine relative to the nominal efficiency η based on the turbine tip clearance TC obtained in step 1. nom Turbine efficiency change Δη HPT , where the actual efficiency η of the turbine is calculated according to the following formula: Where, turbine efficiency correction coefficient K = 1 + 0.586 (ψ Ztip 3.63 ), Zweifel load factor β1 and β2 are the flow inlet and outlet angles, respectively, and the airfoil cascade density ratio c zm and s m are the blade axial chord and the average radius of the cascade pitch, r r 、r s are the outer diameter of the wheel and the inner diameter of the casing, L b is the blade length, TC nom is the nominal value of turbine tip clearance; Step 3: When the aircraft engine is in a dynamic state, use the turbine efficiency change Δη HPT Correct the health parameter estimation value output by the health parameter estimation module in the nonlinear airborne adaptive model, and then go to step 1.

2. The method for real-time sensing of aero-engine turbine blade tip clearance according to claim 1, characterized in that: The calculation method of the total deformation of the casing, blades and disc is as follows: Casing: Calculate the convective heat transfer coefficients of the outer and inner surfaces of the casing. Then, calculate the temperature field distribution of the casing through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and modify the temperature-dependent material properties. Then, calculate the thermal deformation of the casing through a 1-D elastic thermal deformation calculation. Finally, integrate the material strain in the radial direction to obtain the total deformation of the casing. Blades: Calculate the average convective heat transfer coefficient on the blade surface. Then, calculate the temperature of the blade using the lumped parameter method and correct the temperature-dependent material properties to obtain the temperature field distribution. Then, calculate the thermal deformation of the blade using the 0-D elastic thermal deformation method and calculate the centrifugal deformation of the blade using the rotor dynamics method. Finally, the total deformation of the blade is obtained by superimposing the thermal deformation and the centrifugal deformation. Impeller: The convective heat transfer coefficients of the front and rear surfaces of the impeller are calculated. The temperature field distribution of the impeller is then determined through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and correction of temperature-dependent material properties. The thermal deformation of the impeller is then calculated through a 1-D elastic thermal deformation calculation. External loads on the turbine impeller during rotation are introduced through blade rotor dynamics calculations, and the centrifugal strain of the impeller is then calculated through the impeller rotor dynamics calculation. Finally, the obtained centrifugal strain of the impeller is superimposed with the thermal strain to obtain the total deformation of the impeller.

3. The method for real-time sensing of aero-engine turbine blade tip clearance according to claim 1, characterized in that: The health parameter estimation module is a health parameter estimation module based on unscented Kalman filtering.

4. The method for real-time sensing of aero-engine turbine blade tip clearance according to claim 1, characterized in that: The following method is used to identify the operating condition of an aircraft engine: if the absolute value of the difference between the mean of the flight altitude, Mach number, and fan speed of the previous sampling periods and the current flight altitude, Mach number, and fan speed is less than the corresponding difference threshold, and the standard deviation of the flight altitude, Mach number, and fan speed of the previous sampling periods is less than the corresponding standard deviation threshold, then the aircraft engine is determined to be in a steady-state operating condition; otherwise, the aircraft engine is determined to be in a dynamic operating condition.

5. A real-time sensing device for turbine tip clearance of an aircraft engine, characterized in that: include: The tip clearance sensing model is used to calculate the total deformation of the casing, blades, and disk based on the measurable engine parameters output by the aircraft engine sensors and the unmeasurable engine parameters output by the nonlinear airborne adaptive model with a health parameter estimation module. The turbine tip clearance is then calculated using the initial tip clearance and the initial lengths and total deformation of the casing, blades, and disk. HPT efficiency correction module is used to calculate the actual efficiency η of the turbine relative to the nominal efficiency η based on the turbine tip clearance TC obtained by the tip clearance perception model nom Turbine efficiency change Δη HPT , where the actual efficiency η of the turbine is calculated according to the following formula: Where, turbine efficiency correction coefficient K = 1 + 0.586 (ψ Ztip 3.63 ), Zweifel load factor β1 and β2 are the flow inlet and outlet angles, respectively, and the airfoil cascade density ratio c zm and s m are the blade axial chord and the average radius of the cascade pitch, r r 、r s are the outer diameter of the wheel and the inner diameter of the casing, L b is the blade length, TC nom is the nominal value of the turbine blade tip clearance; the steady-state operating condition identification module is used to identify whether the aircraft engine is in a steady-state operating condition or a dynamic operating condition; The mode switching module is used to use the turbine efficiency change Δη when the aircraft engine is in a dynamic working state. HPT Correcting the health parameter estimation value output by the health parameter estimation module in the nonlinear airborne adaptive model.

6. The real-time sensing device for turbine tip clearance of an aircraft engine as claimed in claim 5, characterized in that: The calculation method of the total deformation of the casing, blades and disc is as follows: Casing: Calculate the convective heat transfer coefficients of the outer and inner surfaces of the casing. Then, calculate the temperature field distribution of the casing through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and modify the temperature-dependent material properties. Then, calculate the thermal deformation of the casing through a 1-D elastic thermal deformation calculation. Finally, integrate the material strain in the radial direction to obtain the total deformation of the casing. Blades: Calculate the average convective heat transfer coefficient on the blade surface. Then, calculate the temperature of the blade using the lumped parameter method and correct the temperature-dependent material properties to obtain the temperature field distribution. Then, calculate the thermal deformation of the blade using the 0-D elastic thermal deformation method and calculate the centrifugal deformation of the blade using the rotor dynamics method. Finally, the total deformation of the blade is obtained by superimposing the thermal deformation and the centrifugal deformation. Impeller: The convective heat transfer coefficients of the front and rear surfaces of the impeller are calculated. The temperature field distribution of the impeller is then determined through a 1-D heat transfer process temperature calculation based on the Crank-Nicolson+Thomas algorithm and correction of temperature-dependent material properties. The thermal deformation of the impeller is then calculated through a 1-D elastic thermal deformation calculation. External loads on the turbine impeller during rotation are introduced through blade rotor dynamics calculations, and the centrifugal strain of the impeller is then calculated through the impeller rotor dynamics calculation. Finally, the obtained centrifugal strain of the impeller is superimposed with the thermal strain to obtain the total deformation of the impeller.

7. The real-time sensing device for turbine tip clearance of an aircraft engine as claimed in claim 5, characterized in that: The health parameter estimation module is a health parameter estimation module based on unscented Kalman filtering.

8. The real-time sensing device for turbine tip clearance of an aircraft engine as claimed in claim 5, characterized in that: The steady-state operating condition identification module uses the following method to identify the operating condition of the aircraft engine: if the absolute value of the difference between the mean of the flight altitude, Mach number, and fan speed of the previous sampling periods and the flight altitude, Mach number, and fan speed at the current moment is less than the corresponding difference threshold, and the standard deviation of the flight altitude, Mach number, and fan speed of the previous sampling periods is less than the corresponding standard deviation threshold, then the aircraft engine is determined to be in a steady-state operating condition; otherwise, the aircraft engine is determined to be in a dynamic operating condition.

Images