CN112560340B - Method for estimating surge margin of aircraft engine and control method - Google Patents
Method for estimating surge margin of aircraft engine and control method Download PDFInfo
- Publication number
- CN112560340B CN112560340B CN202011448201.XA CN202011448201A CN112560340B CN 112560340 B CN112560340 B CN 112560340B CN 202011448201 A CN202011448201 A CN 202011448201A CN 112560340 B CN112560340 B CN 112560340B
- Authority
- CN
- China
- Prior art keywords
- surge margin
- model
- surge
- engine
- aircraft engine
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000009530 blood pressure measurement Methods 0.000 claims abstract description 7
- 238000013528 artificial neural network Methods 0.000 claims description 26
- 230000003044 adaptive effect Effects 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 20
- 230000000694 effects Effects 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 230000015556 catabolic process Effects 0.000 claims description 4
- 230000006835 compression Effects 0.000 claims description 4
- 238000007906 compression Methods 0.000 claims description 4
- 238000006731 degradation reaction Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 2
- 230000001537 neural effect Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000013507 mapping Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000010349 pulsation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Geometry (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses an aircraft engine surge margin estimation method which comprises the steps of firstly constructing an airborne self-adaptive model, taking near surge range surge margin estimation data obtained by a surge margin estimation method based on compressor tip pressure measurement as feedback, correcting a surge boundary of an engine nonlinear model in the airborne self-adaptive model, and then carrying out final surge margin estimation by using the corrected airborne self-adaptive model. The invention also discloses a method for controlling the surge margin of the aircraft engine. Compared with the prior art, the method can output wide-range high-confidence-degree surge margin estimation data which can meet the actual control requirement of the aircraft engine, so that the model-based active surge margin control technology can really enter engineering application.
Description
Technical Field
The invention relates to an aircraft engine surge margin estimation method.
Background
At present, the engine control does not get rid of the design idea of the traditional control, and the powerful computing power and the logic function of a computer are not fully utilized. The traditional engine control is based on sensor control, namely, performance parameters such as thrust, surge margin and the like of an engine are indirectly controlled through measurable engine state parameters such as rotating speed, pressure ratio, temperature and the like. Although the control mode is simple and reliable, the change situation of the surge margin in the working process of the engine is difficult to accurately reflect, the loss of the stability margin under the worst working environment of the engine needs to be considered in the design process, and even the full exertion of the performance of the engine is limited. With the updating of fighters, the fifth generation of fighters put forward high performance requirements on the engine such as high efficiency, high thrust-weight ratio, high stability and the like. However, with the rapid development of full-authority digital electronic control, the performance of traditional engine control is brought into full play, and to further improve the overall performance of the engine from the control perspective, the surge margin needs to be estimated in real time and actively controlled.
Most of the current aircraft engines feed back real-time working states on line through an airborne adaptive model, and then calculate a surge margin as a feedback quantity to form a direct surge margin closed-loop control loop. However, at present, most of the research on model base at home and abroad focuses on the simulation stage, and is rarely reported in the practical application of engineering. The reason is mainly that the airborne adaptive model cannot realize zero-error matching with a real engine due to the fact that the accuracy of the airborne adaptive model cannot realize zero-error matching, certain errors exist, the airborne adaptive model cannot carry out high-accuracy estimation on the surge margin in a wide range, and the engine instability can be caused when the engine is directly subjected to closed-loop feedback control based on the inaccurate surge margin.
Disclosure of Invention
The technical problem to be solved by the invention is to overcome the defects in the prior art, provide an aircraft engine surge margin estimation method, and provide wide-range high-confidence-degree surge margin estimation data which can meet the actual control requirements of the aircraft engine.
The invention specifically adopts the following technical scheme to solve the technical problems:
an aircraft engine surge margin estimation method comprises the steps of firstly constructing an airborne adaptive model, taking near surge range surge margin estimation data obtained by a surge margin estimation method based on compressor tip pressure measurement as feedback, correcting a surge boundary of an engine nonlinear model in the airborne adaptive model, and then carrying out final surge margin estimation by using the corrected airborne adaptive model.
Preferably, the onboard adaptive model is an onboard adaptive hybrid model based on neural network nonlinear compensation, and comprises: the engine comprises an engine nonlinear model, a Kalman filter and a neural network learning algorithm.
Further preferably, the training data used by the onboard adaptive hybrid model based on neural network nonlinearity compensation is subjected to cluster compression in advance.
Still further preferably, the clustering compression uses a gaussian clustering method.
Preferably, the specific method for correcting is as follows: and taking near surge range surge margin estimation data obtained by a surge margin estimation method based on compressor tip pressure measurement as near surge range surge margin target output of the airborne adaptive model, and correcting a surge boundary of an engine nonlinear model in the airborne adaptive model by utilizing a neural network learning algorithm.
According to the same inventive concept, the following technical scheme can be obtained:
the method for controlling the surge margin of the aircraft engine is used for controlling the aircraft engine based on the surge margin estimation data obtained by the method in any technical scheme.
Compared with the prior art, the technical scheme of the invention has the following beneficial effects:
the invention adopts the near surge range surge margin estimation data with extremely high confidence coefficient acquired by the compressor tip pressure measurement-based surge margin estimation method as feedback to correct the surge boundary of the engine nonlinear model in the airborne adaptive model, so that the corrected airborne adaptive model can output wide-range high-confidence-coefficient surge margin estimation data which can meet the actual control requirement of the aircraft engine, and further, the model-based active surge margin control technology can really enter engineering application.
Drawings
FIG. 1 is a schematic structural diagram of an aircraft engine surge margin estimation model according to the present invention;
fig. 2 is a schematic diagram of a training process of a neural network.
Detailed Description
Aiming at the problem that surge margin estimation data generated by the existing airborne self-adaptive model cannot meet the precision requirement required by engine control, the invention adopts the technical scheme that the existing airborne self-adaptive model is improved, the near surge range surge margin estimation data with extremely high confidence coefficient, which is obtained by adopting a surge margin estimation method based on the pressure measurement of the blade tips of a compressor, is used as feedback, and the surge boundary of an engine nonlinear model in the airborne self-adaptive model is corrected, so that the corrected airborne self-adaptive model can output wide-range high-confidence coefficient surge margin estimation data which can meet the actual control requirement of an aircraft engine.
The basic implementation of a surge margin estimation method based on compressor tip pressure measurement (see [ Manuj Dhingra. compressor Stability Management [ D ]// American: georgia insulation of technology.2006]) is as follows: calculating a similarity coefficient for measuring the repeatability of the pressure pulsation signal according to the pressure pulsation signal near the blade obtained by the pressure sensor; secondly, selecting a proper threshold to calculate the intersection times of two function lines of the similarity coefficient and the threshold in unit time, obtaining the inherent characteristic relation between the intersection times and the surge margin in unit time according to the off-line analysis of test data, and establishing an engine surge margin estimation model based on the pressure of the blade tips of the compressor; according to the engine surge margin estimation model, the surge margin of the engine can be accurately estimated through an interpolation method.
The surge margin estimation method based on the measurement of the pressure of the blade tip of the air compressor can achieve more reliable surge margin estimation than the existing airborne adaptive model, and the reliability of the method is not influenced by the performance degradation of the engine; however, through a large number of experiments, the scheme can meet the requirement of accuracy, but only has high confidence in a near-surge range, and cannot be used for active control of the surge margin of the engine.
Therefore, the inventor proposes that a surge margin estimation method based on the measurement of the pressure of the blade tip of the compressor is organically combined with the existing onboard adaptive model surge margin estimation method so as to realize the wide-range high-confidence surge margin estimation.
For the public to understand, the technical scheme of the invention is explained in detail by a specific embodiment and the accompanying drawings:
the structure of the aircraft engine surge margin estimation model of the embodiment is shown in fig. 1, and comprises an onboard adaptive model and a correction module (a dotted line part in the figure) for correcting a surge boundary of an engine nonlinear model in the onboard adaptive model.
The airborne adaptive Model in this embodiment adopts an airborne adaptive hybrid Model Based on neural network nonlinear compensation (see [ Csank J T, common J w. enhanced Engine Performance Using a Model-Based Engine Control Architecture [ C ]// Aiaa/sae/ase Joint Performance reference. 2015]), which can eliminate the errors between the Engine nonlinear Model and the actual Engine component Performance parameters, and expand the applicable range of the Model, and mainly includes three parts: the engine comprises an engine nonlinear model, a Kalman filter and a neural network learning algorithm.
As shown in fig. 1, the neural network learning algorithm is used to learn the true engine versus model measurable output residual offline when the switch is in the "learn" position. The input parameters for neural network learning comprise working condition parameters, engine input parameters and measurable output parameter residuals. When the switch is in an application position, a sample database obtained from the offline learning result of the neural network is selected to compensate the estimation parameters output by the airborne nonlinear model, the individual difference of the engine is eliminated, and then the Kalman filtering algorithm is adopted to estimate the performance degradation of the part so as to automatically correct the nonlinear model of the engine, thereby improving the precision of the model.
In order to solve the redundancy problem of the training data in the full-envelope online, the method of clustering (preferably adopting Gaussian clustering) is adopted to carry out airborne real-time clustering compression on the flight data in the specific embodiment, GMM (Gaussian Mixture model) is constructed in real time online, and a GMM sample database is obtained; and when the characteristic data after clustering is added to an MLP (Multi layer Perceptron) training sample set in an offline non-real-time stage, learning and updating to obtain an MLP sample database.
In the embodiment, estimated data of the surge margin in the near surge range obtained by a surge margin estimation method based on the measurement of the pressure of the blade tip of the compressor is used as the target output of the surge margin in the near surge range of the airborne adaptive model, and the surge boundary of the engine nonlinear model in the airborne adaptive model is corrected by using a neural network learning algorithm, which is specifically realized by using the neural network 2 in fig. 1.
The training process of the invention is divided into measurable parameter training and non-measurable parameter training of the aero-engine. The measurable parameter training is the training process of the airborne self-adaptive model, and the mapping relation between the neural network and a real engine is established by building the neural network 1 in the figure 1; the method comprises the steps of carrying out non-measurable parameter training, namely a training process of a surge margin model, estimating a surge margin according to a tip pressure signal measured from a real engine, if the real engine is in a near-surge range, comparing the surge margin with the surge margin estimated by a nonlinear engine model to obtain a residual error, establishing a mapping relation between the residual error and input parameters and measurable parameters of the engine by utilizing a neural network learning algorithm, and correcting a surge boundary of the nonlinear engine model through the mapping relation.
Step 1: determining input parameters u ═ { H, Ma, PLA }, i.e., altitude, mach number, and throttle lever angle, of the aircraft engine, while all switches in fig. 1 point to 1, i.e., training mode;
step 2: the method comprises the steps of building a neural network 1, wherein the neural network is of a three-layer full-connection layer model structure, an input layer is three nodes corresponding to an input parameter u, the number of nodes of a hidden layer is adjusted according to a model training effect, and an output layer is a measurable parameter y ═ N1,N2,T25,P25,T3,P3,T5,P5And estimating measurable parameters
Residual estimation of
The residual, including the effects of model degradation, measurement noise, sensor bias, and any attendant modeling errors, can be used to adjust the engine estimated measurable parameters to more closely match the actual values;
step 3: building a neural network 2 which is a three-layer full-connection layer model structure, wherein an input layer is an input parameter u plus an estimated surge margin
The number of nodes of the corresponding four nodes and the hidden layer is adjusted according to the training effect of the model, and the output layer is the undetectable parameter surge margin z and the estimated surge margin
Residual error of
Adjusting the neural network model to enable the estimated value of the surge margin of the undetectable parameter to be closer to the true value, and correcting the surge boundary of the nonlinear model of the engine;
step 4: after training of 2 neural networks is completed, the switch points to 2 from 1, namely an application mode;
step 5: residual estimation output by neural network 1
Estimating measurable parameters for a non-linear model of an engine
Compensating to obtain estimated measurable parameter correction
And calculating the residual error of the two
Inputting the residual error to a Kalman filter to obtain an estimated surge margin
And health parameters
By means of Kalman filteringCorrecting the nonlinear model of the engine by the output of the wave filter;
step 6: and adding new flight data, retraining and repeating the steps 1 to 5.
By adopting the aircraft engine surge margin estimation model, wide-range high-confidence-degree surge margin estimation data which can meet actual control requirements of the aircraft engine can be obtained, and various existing or future model-based active surge margin control technologies can be realized based on the wide-range high-confidence-degree surge margin estimation data.
Claims (5)
1. The method for estimating the surge margin of the aircraft engine is characterized by comprising the steps of firstly constructing an airborne self-adaptive model, taking near surge range surge margin estimation data obtained by a surge margin estimation method based on the measurement of the pressure of the blade tip of a gas compressor as feedback, correcting the surge boundary of an engine nonlinear model in the airborne self-adaptive model, and then carrying out final surge margin estimation by using the corrected airborne self-adaptive model; the method specifically comprises the following steps:
step 1: determining input parameters u ═ { H, Ma, PLA }, which are respectively the height, Mach number and throttle lever angle, and entering a training mode;
step 2: the method comprises the steps of building a neural network 1, wherein the neural network is of a three-layer full-connection layer model structure, an input layer is three nodes corresponding to an input parameter u, the number of nodes of a hidden layer is adjusted according to a model training effect, and an output layer is a measurable parameter y ═ N1,N2,T25,P25,T3,P3,T5,P5And estimating measurable parameters
Residual estimation of
The residual, including the effects of model degradation, measurement noise, sensor bias and any attendant modeling errors, can be used to adjust the engine estimated measurable parameters to more closely match the actual values;
step 3: building neural networks2, the model structure of three full-connected layers is adopted, and the input layer is an input parameter u plus an estimated surge margin
The number of nodes of the corresponding four nodes and the hidden layer is adjusted according to the training effect of the model, and the output layer is the undetectable parameter surge margin z and the estimated surge margin
Residual error of
Adjusting the neural network model to enable the estimated value of the surge margin of the undetectable parameter to be closer to the true value, and correcting the surge boundary of the nonlinear model of the engine;
step 4: after training is finished for 2 neural networks, entering an application mode;
step 5: residual estimation output by neural network 1
Estimating measurable parameters for a non-linear model of an engine
Compensating to obtain estimated measurable parameter correction
And calculating the residual error of the two
Inputting the residual error to the Kalman filter to obtain the estimated surge margin
And health parameters
Correcting the nonlinear engine model through the output of the Kalman filter;
step 6: and adding new flight data, retraining and repeating the steps 1 to 5.
2. The aircraft engine surge margin estimation method of claim 1, wherein training data used by the neural network nonlinearity compensation-based airborne adaptive hybrid model is pre-cluster compressed.
3. The aircraft engine surge margin estimation method of claim 2, wherein the cluster compression uses a gaussian clustering method.
4. The aircraft engine surge margin estimation method of any one of claims 1 to 3, wherein the specific method for correcting is as follows: and taking near surge range surge margin estimation data obtained by a surge margin estimation method based on compressor tip pressure measurement as near surge range surge margin target output of the airborne adaptive model, and correcting a surge boundary of an engine nonlinear model in the airborne adaptive model by utilizing a neural network learning algorithm.
5. A method for controlling the surge margin of an aircraft engine is characterized in that the aircraft engine is controlled based on the surge margin estimation data obtained by the method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011448201.XA CN112560340B (en) | 2020-12-09 | 2020-12-09 | Method for estimating surge margin of aircraft engine and control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011448201.XA CN112560340B (en) | 2020-12-09 | 2020-12-09 | Method for estimating surge margin of aircraft engine and control method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112560340A CN112560340A (en) | 2021-03-26 |
| CN112560340B true CN112560340B (en) | 2022-02-01 |
Family
ID=75061030
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011448201.XA Active CN112560340B (en) | 2020-12-09 | 2020-12-09 | Method for estimating surge margin of aircraft engine and control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112560340B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022252206A1 (en) * | 2021-06-04 | 2022-12-08 | 大连理工大学 | Aero-engine surge active control system based on fuzzy switching of controllers |
| CN113642271B (en) * | 2021-08-20 | 2022-07-01 | 南京航空航天大学 | Model-based aeroengine performance recovery control method and device |
| CN114880766B (en) * | 2022-04-22 | 2024-07-12 | 南京航空航天大学 | Aeroengine steady state performance maintaining control method and system |
| CN115099362A (en) * | 2022-07-20 | 2022-09-23 | 南京航空航天大学 | Gas compressor surge detection method based on unsupervised clustering algorithm |
| CN117574771B (en) * | 2023-11-29 | 2024-10-25 | 北京航空航天大学 | Component performance margin design method of self-adaptive variable cycle engine |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111382500A (en) * | 2020-02-20 | 2020-07-07 | 中国民航管理干部学院 | Safety analysis and verification method for turbocharging system of aircraft engine |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109184913B (en) * | 2018-10-08 | 2019-12-20 | 南京航空航天大学 | Stability estimation and prediction-based active compound control method for aerodynamic stability of aircraft engine |
| CN109344510A (en) * | 2018-10-08 | 2019-02-15 | 南京航空航天大学 | A kind of active stability control method based on the estimation of aero-engine stability margin |
-
2020
- 2020-12-09 CN CN202011448201.XA patent/CN112560340B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111382500A (en) * | 2020-02-20 | 2020-07-07 | 中国民航管理干部学院 | Safety analysis and verification method for turbocharging system of aircraft engine |
Non-Patent Citations (1)
| Title |
|---|
| 航空发动机性能恢复控制方法;黄金泉等;《航空动力学报》;20120715(第07期);第160-168页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112560340A (en) | 2021-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112560340B (en) | Method for estimating surge margin of aircraft engine and control method | |
| CN108416086B (en) | Aero-engine model self-adaptive correction method based on deep learning | |
| CN110502840B (en) | Online prediction method for gas circuit parameters of aero-engine | |
| CN110579962B (en) | Turbofan engine thrust prediction method based on neural network and controller | |
| CN112286047B (en) | NARMA-L2 multivariable control method based on neural network | |
| CN109162813A (en) | One kind being based on the modified Aeroengine Smart method for controlling number of revolution of iterative learning | |
| CN107357176B (en) | Modeling method for test run data of aero-engine | |
| CN110219736A (en) | Aero-engine Direct Thrust Control Strategy based on Nonlinear Model Predictive Control | |
| CN109164708B (en) | Neural network self-adaptive fault-tolerant control method for hypersonic aircraft | |
| CN111651940A (en) | Turboshaft engine steady-state model identification method based on PSO-NARX | |
| CN108459505A (en) | A kind of unconventional layout aircraft fast modeling method of suitable control Iterative Design | |
| CN111781942A (en) | Fault-tolerant flight control method based on self-constructed fuzzy neural network | |
| CN109696090A (en) | It is a kind of for carrier rocket guided missile online single-shot thrust discrimination method | |
| CN115146538A (en) | Power system state estimation method based on message passing graph neural network | |
| Chen et al. | Aero‐Engine Real‐Time Models and Their Applications | |
| CN111708378A (en) | Guided missile longitudinal attitude control algorithm based on reinforcement learning | |
| CN114995152A (en) | Deviation correction method for civil aviation engine performance model | |
| CN110985216B (en) | Intelligent multivariable control method for aero-engine with online correction | |
| CN114818205B (en) | Online sensing method for blade tip clearance of full life cycle of aero-engine | |
| CN117648850A (en) | Aircraft icing simulation system based on neural network training method | |
| CN116933654A (en) | Turbofan engine thrust real-time estimation method integrating UKF and neural network | |
| CN116987878A (en) | Anti-interference temperature control method for heat treatment of aerospace complex cast parts | |
| CN114609969B (en) | Numerical control machine tool track error compensation method based on cloud computing | |
| CN113219840B (en) | Self-adaptive sliding mode cross-coupling contour control method for three-axis motion platform | |
| CN113341973B (en) | Course control method based on asymmetric phase difference of flapping wings |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |