CN114077775A - Intelligent dynamic pressure measurement method for aircraft engine - Google Patents
Intelligent dynamic pressure measurement method for aircraft engine Download PDFInfo
- Publication number
- CN114077775A CN114077775A CN202111280406.6A CN202111280406A CN114077775A CN 114077775 A CN114077775 A CN 114077775A CN 202111280406 A CN202111280406 A CN 202111280406A CN 114077775 A CN114077775 A CN 114077775A
- Authority
- CN
- China
- Prior art keywords
- pressure
- short measuring
- measuring tube
- dynamic pressure
- model
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims description 28
- 238000009530 blood pressure measurement Methods 0.000 title claims description 15
- 238000004088 simulation Methods 0.000 claims abstract description 34
- 238000013528 artificial neural network Methods 0.000 claims abstract description 24
- 238000000691 measurement method Methods 0.000 claims abstract description 11
- 230000008878 coupling Effects 0.000 claims abstract description 4
- 238000010168 coupling process Methods 0.000 claims abstract description 4
- 238000005859 coupling reaction Methods 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 25
- 230000003068 static effect Effects 0.000 claims description 21
- 238000012360 testing method Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 5
- 239000000463 material Substances 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 239000012530 fluid Substances 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 239000007787 solid Substances 0.000 claims description 3
- 238000004514 thermodynamic simulation Methods 0.000 claims description 3
- 238000011084 recovery Methods 0.000 abstract description 8
- 238000005259 measurement Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 abstract description 2
- 230000000306 recurrent effect Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000002485 combustion reaction Methods 0.000 description 1
- 239000000306 component Substances 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000011144 upstream manufacturing 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]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- 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
- G06F30/23—Design optimisation, verification or simulation using finite element methods [FEM] or finite difference methods [FDM]
-
- 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]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Pure & Applied Mathematics (AREA)
- Fluid Mechanics (AREA)
- Algebra (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Computational Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an aeroengine-oriented dynamic pressure intelligent measurement method, which comprises the following steps: step 1) establishing a 3-dimensional numerical simulation model of a short measuring pipe and an outlet runner of a gas compressor based on a coupling relation among multiple physical fields; step 2), calculating flow field parameters at the outlet of the compressor and in the short measuring tube; and 3) establishing a dynamic total pressure intelligent measurement method based on the RNN neural network. Compared with the prior art, the invention has the following technical effects: (1) stagnation of the pressure signal is realized through the short measuring tube, dynamic pressure attenuation is reduced to the maximum extent, and dynamic information of the pressure is reserved. (2) The dynamic pressure recovery model based on the neural network realizes compensation fitting of dynamic pressure attenuation and further improves the measurement precision.
Description
Technical Field
The invention belongs to the technical field of control of aero-engines, and particularly relates to an aero-engine-oriented dynamic pressure intelligent measurement method.
Background
The compressor is one of the core components of an aircraft engine, and applies work to air by utilizing blades rotating at a high speed to improve air pressure and improve thermodynamic cycle efficiency for a downstream combustion chamber of the compressor. The performance of the compressor directly affects the safety of the engine, and the performance measurement of the compressor is a key link in the development and control process of the aeroengine. The parameter which can reflect the performance of the compressor most is the pressure ratio of the compressor, and the pressure ratio of the compressor is widely applied to important fields of engine modeling, overrun protection, health management and the like. In practical use, the pressure ratio of the compressor is generally obtained by directly measuring the pressure of gas at the inlet and the outlet of the compressor.
When the pressure sensor is actually applied to the outlet of the compressor, the disturbance of a flow field of a measuring mechanism and the heat resistance of a material of the pressure sensor are considered due to the large air flow speed, the overhigh temperature and the like, and the pressure sensor cannot be directly measured. But need with the help of the pitot tube, stagnate airflow velocity for zero, lead the air current that the content awaits measuring to outer duct simultaneously, cool off pressure sensor with the help of outer duct air current. At the moment, the pressure sensor and the pitot tube form a pressure measuring system, and in order to accurately measure the airflow pressure at the outlet of the compressor, the mapping relation of the airflow pressure at the outlet of the compressor and the measuring end of the sensor needs to be obtained. Meanwhile, when the engine surges, the airflow at the outlet of the compressor fluctuates greatly, and the dynamic response of the whole pressure measurement system must be considered in order to measure the dynamic pressure of the airflow at the moment without distortion as much as possible. Therefore, the dynamic characteristic of the mapping relation between the total pressure of the outlet of the compressor and the static pressure of the airflow at the sensor is researched, and the method has important significance for pressure measurement of the outlet of the compressor, engine modeling, overrun protection and health management.
The invention provides an intelligent dynamic pressure measuring method for an aircraft engine, aiming at the problems of overlarge dynamic total pressure measuring difficulty and insufficient precision of an outlet of a traditional gas compressor.
Disclosure of Invention
The purpose of the invention is as follows: the technical problem to be solved by the invention is to overcome the defects of the prior art, and the measuring system cools the total pressure measuring system by utilizing the bypass airflow. The length of the short measuring tube is shortened, the precision of dynamic total pressure measurement is greatly improved in the aspect of a geometric structure, and meanwhile, a compensation algorithm based on an RNN neural network is designed to fit dynamic pressure attenuation errors, so that the measurement precision is further improved.
The technical scheme is as follows: in order to achieve the purpose, the invention adopts the technical scheme that:
an intelligent dynamic pressure measurement method for an aircraft engine comprises the following steps:
step 1), establishing a 3-dimensional numerical simulation model of a short measuring pipe and an outlet runner of a gas compressor based on a coupling relation among multiple physical fields;
step 2), calculating flow field parameters at the outlet of the compressor and in the short measuring tube;
step 3), establishing a dynamic total pressure intelligent measurement method based on an RNN (Current Neural network) Neural network;
further, the multi-physical field model establishing platform in the step 1) is an ANSYS CFX software platform, and the specific steps are as follows:
step 1.1), determining the geometric size of a short measuring tube, the geometric size of an outlet of a gas compressor and the length of a fully developed flow passage;
step 1.2), establishing a geometric model of an outlet flow channel of the gas compressor, a geometric model of an inner casing and a geometric model of a short measuring tube based on the size data in the step 1.1);
step 1.3), establishing a mesh model of an outlet flow channel of the gas compressor, a mesh model of an inner casing and a mesh model of a short measuring pipe based on the geometric model established in the step 1.2);
further, the specific steps in the step 1.3) are as follows:
step 1.3.1), adopting ANSYS workbench Mesh software to divide the network, and gradually carrying out local encryption on the grid along the direction from the fluid domain to the fluid-solid boundary in consideration of the large air velocity gradient at the boundary layer;
step 1.3.2), in order to guarantee the reliability of the calculation results while taking into account the calculation time, it is necessary to check the grid adaptability so as to select the appropriate grid number. Continuously encrypting the short measuring tube cavity and the area 200mm before and after the inlet of the short measuring tube in the grid verification process;
further, the specific steps in the step 2) are as follows:
step 2.1), selecting 6 typical working points of an engine in an airplane envelope, and carrying out engine working process simulation at the typical working points based on an existing certain turbofan engine pneumatic thermodynamic model to obtain static pressure, static temperature and gas flow rate data of an outlet of a compressor in the engine and an outer culvert of the engine;
step 2.2), setting a three-dimensional geometric model of a 3-dimensional numerical simulation model based on the geometric models of the compressor, the inner casing and the short measuring tube established in the step 1.2);
step 2.3), setting boundary conditions of a 3-dimensional numerical simulation model of the short measuring tube and the outlet runner of the gas compressor based on the static pressure, the static temperature, the gas flow rate, preset materials and other characteristics obtained in the step 2.1), carrying out 3-dimensional numerical simulation of dynamic changes of flow field pressure, and obtaining total pressure and total temperature dynamic change data of the opening end and the measuring end of the short measuring tube;
further, the specific steps in the step 2.1) are as follows:
step 2.1.1), 6 typical operating points are selected (H ═ 0km, Ma ═ 0.8), (H ═ 2km, Ma ═ 0.8), (H ═ 4km, Ma ═ 0.2), (H ═ 10km, Ma ═ 1.4), (H ═ 14km, Ma ═ 1), and (H ═ 16km, Ma ═ 1.8);
step 2.2.2), performing aerodynamic and thermodynamic simulation of sine, random and step signals at 6 working points respectively in sequence to obtain required parameters;
further, the specific steps in the step 2.3) are as follows:
step 2.3.1), respectively defining dynamic pressure signals in a sine function form by using the amplitude of 0.3 and 0.5Mpa and the frequency of 5Hz, 18Hz and 30Hz in ANSYS CFX software, and taking the dynamic pressure signals as a boundary condition of a three-dimensional numerical simulation model;
step 2.3.2), step and random dynamic pressure signals are subjected to CFX simulation by using discrete point data obtained in the step 2.2.2) and self-defining a discrete point function, and setting a calculation step length equal to a discrete point time interval as a boundary condition;
further, the specific steps in the step 3) are as follows:
and 3.1) designing an algorithm based on a Recurrent Neural Network (RNN) to fit the measurement error based on the simulation data obtained in the step 2, thereby compensating the measurement result.
Step 3.2), selecting a network input and output and network data set, a training set and a test set based on the simulation data obtained in the step 2.1) and the step 2.3);
step 3.3), setting network training parameters, training a neural network, and forming a dynamic total pressure intelligent measurement method;
further, the specific steps in the step 3.2) are as follows:
step 3.2.1), taking total pressure at the measuring end, converted rotating speed, flying height, Mach number, ground atmospheric pressure and temperature as input, taking the difference of sampling step length among samples into consideration, taking time as input, and taking pressure at the inlet end as output;
step 3.2.2), selecting a measured total pressure, a converted rotating speed, a flying height, a Mach number, ground atmospheric pressure, time, inlet end pressure and temperature to form a data set, and selecting 75% of data as a training set and 25% of data as a testing set;
further, the specific steps in the step 3.3) are as follows:
step 3.3.1), setting network training times, learning rate, training required precision and hidden layer number, and training a neural network;
and 3.3.2) obtaining the dynamic pressure of the flow field of the compressor at the inlet end of the short measuring tube by measuring total pressure at the measuring end, flight environment and working state data based on the model obtained in the step 3.3.1), thereby establishing the intelligent measuring method for the dynamic pressure of the short measuring tube.
Has the advantages that: compared with the prior art, the intelligent engine dynamic pressure measuring method based on the neural network has the following technical effects by adopting the technical scheme:
(1) stagnation of the pressure signal is realized through the short measuring tube, dynamic pressure attenuation is reduced to the maximum extent, and dynamic information of the pressure is reserved.
(2) The dynamic pressure recovery model based on the neural network realizes compensation fitting of dynamic pressure attenuation and further improves the measurement precision.
Drawings
FIG. 1 is a flow chart of a dynamic pressure intelligent measurement method for an aircraft engine.
FIG. 2 is a flow chart for establishing a 3-dimensional numerical simulation model of a short measuring tube and an outlet runner of a compressor.
FIG. 3 is a schematic diagram of a short pipe model, wherein (a) is an isometric view of the short pipe model and (b) is a sectional view of the short pipe model.
Fig. 4 is a schematic view of the flow channels of the inner duct, the outer duct and the short measuring pipe of the outlet of the compressor, wherein (a) is a complete flow channel model, and (b) is an 1/6 flow channel model.
Fig. 5 is a schematic diagram of short pipe meshing.
FIG. 6 is a diagram of a recurrent neural network.
FIG. 7 is a sample test result of the intermediate state verification of the dynamic total pressure recovery model.
Detailed Description
The invention is further illustrated by the following figures and examples. The present invention will be better understood from the following examples. However, those skilled in the art will readily appreciate that the specific parameter selection, numerical simulation, neural network training and results thereof described in the examples are merely illustrative of the invention and should not limit the invention as detailed in the claims.
The invention relates to an aeroengine-oriented dynamic pressure intelligent measurement method, the flow is shown in figure 1, the invention establishes a 3-dimensional CFX simulation model containing gas-solid-heat multi-physical fields according to the outlet structure, the flow characteristics and the like of a short measuring pipe and a gas compressor; and calculating to obtain the static temperature, speed and static pressure of the culvert airflow at the outlet of the gas compressor and the static temperature, speed and static pressure of the culvert airflow at the outlet of the gas compressor according to the flying environment of the engine, the working state (converted rotating speed, height, Mach number, ground atmospheric pressure and the like) and the component-level model, setting the boundary conditions of the CFX simulation model according to the static temperature, speed and static pressure of the culvert airflow, and carrying out CFX simulation to obtain the temperature and pressure distribution along the tube pass in the short-distance measuring tube. According to the flight environment, the working state and the CFX simulation data, an RNN neural network is established, wherein the RNN neural network takes the measured total pressure, the converted rotating speed, the flight altitude, the Mach number, the ground atmospheric pressure and the time as input, and the pressure at the inlet end of the short measuring pipe as output, and is used as a dynamic pressure recovery model of the short measuring pipe. Based on the model, the dynamic pressure of the flow field of the compressor at the inlet end of the short measuring tube is obtained according to the total pressure of the measuring end, the flight environment and the working state data, so that the intelligent measuring method for the dynamic pressure of the short measuring tube is established, and the intelligent measuring method comprises the following steps:
step 1), a 3-dimensional numerical simulation model of a short measuring pipe and an outlet flow channel of a gas compressor is established based on a coupling relation among multiple physical fields, a model establishment platform is ANSYS CFX software, the flow is shown in figure 2, and the method comprises the following steps:
step 1.1), determining the geometric size of an outlet of the gas compressor, the geometric size of a short measuring tube and the length of a fully developed flow passage; compressor outlet geometry: the radius of the hub is 290mm, the inner diameter of the inner casing is 324mm, the outer diameter of the inner casing is 325mm, and the radius of the outer casing is 475 mm. In order to simulate real fully-developed flow, the upstream flow channel of the short measuring pipe is lengthened, so that airflow uniformly distributed at the inlet of an internal culvert and an external culvert is fully developed before reaching the short measuring pipe, and a 600mm lengthened flow channel is adopted to establish a numerical simulation model of fully-developed flow at the outlet of the gas compressor;
step 1.2), establishing a geometric model of an outlet flow channel of the gas compressor, a geometric model of an inner casing and a geometric model of a short measuring tube based on the size data in the step 1.1) as shown in figures 3 and 4;
step 1.3), establishing a mesh model of an outlet flow channel of the gas compressor, a mesh model of an inner casing and a mesh model of a short measuring pipe based on the geometric model established in the step 1.2);
step 1.3.1), network division is performed by adopting ANSYS workbench Mesh software, the grids are gradually encrypted along the direction from the fluid domain to the fluid-solid boundary under the consideration of large gas velocity gradient at the boundary layer, and the grid division of the short measuring tube is shown in FIG. 5.
Step 1.3.2), in order to guarantee the reliability of the calculation results while taking into account the calculation time, it is necessary to check the grid adaptability so as to select the appropriate grid number. The areas of 200mm before and after the short measuring tube cavity and the inlet of the short measuring tube are continuously encrypted in the grid verification process. Carrying out grid independence verification through ground test data, collecting static temperature distribution and total pressure change of a central line of the short measuring pipe, and simultaneously carrying out temperature distribution analysis on the section of the short measuring pipe cavity to finally obtain 43497 model grids;
step 2), calculating the parameters of the flow field at the outlet of the compressor and in the short measuring tube, and comprising the following steps:
step 2.1), selecting 6 typical working points of an engine in an airplane envelope, and carrying out engine working process simulation at the typical working points based on an existing certain turbofan engine pneumatic thermodynamic model to obtain static pressure, static temperature and gas flow rate data of an outlet of a compressor in the engine and an outer culvert of the engine;
step 2.1.1), 6 typical operating points are selected (H ═ 0km, Ma ═ 0.8), (H ═ 2km, Ma ═ 0.8), (H ═ 4km, Ma ═ 0.2), (H ═ 10km, Ma ═ 1.4), (H ═ 14km, Ma ═ 1), and (H ═ 16km, Ma ═ 1.8);
and 2.2.2) sequentially carrying out the aerodynamic and thermodynamic simulation of the sine, random and step form pressure signals as the input signals of the short measuring tube at 6 working points respectively to obtain the required parameters.
Step 2.2), setting a 3-dimensional geometric model of a 3-dimensional numerical simulation model according to the geometric models of the gas compressor, the inner casing and the short measuring tube established in the step 1.2);
and 2.3) setting boundary conditions of a 3-dimensional numerical simulation model of the short measuring tube and the outlet runner of the gas compressor based on the static pressure, the static temperature, the gas flow velocity, preset material characteristic parameters and other data obtained in the step 2.1), carrying out 3-dimensional numerical simulation of dynamic changes of the flow field pressure, and obtaining dynamic changes of total pressure and total temperature at the inlet end and the measuring end of the short measuring tube.
Step 2.3.1), respectively defining dynamic pressure signals in a sine function form in ANSYS CFX software with the amplitude of 0.3 and 0.5Mpa and the frequency of 5Hz, 18Hz and 30Hz as shown in table 1, and taking the dynamic pressure signals as a boundary condition of a three-dimensional numerical simulation model;
TABLE 1 sinusoidal sample compressor outlet static pressure frequency and amplitude
Step 2.3.2), in ANSYS CFX software, fitting according to the dynamic pressure data in the step and random signal forms obtained in the step 2.2.2) to obtain a nonlinear function which is used as a boundary condition of the three-dimensional numerical simulation model.
Step 3), establishing an RNN neural network-based dynamic total pressure intelligent measurement method, which comprises the following specific steps:
step 3.1), establishing a Recurrent Neural Network (RNN);
and 3.2) selecting an RNN network input and output and a network data set, a training set and a test set based on the simulation data obtained in the step 2.1) and the step 2.3).
Step 3.2.1), taking total pressure at the measuring end, converted rotating speed, flying height, Mach number and ground atmospheric pressure as input, taking the difference of sampling step length among samples into consideration, taking time as input, and taking pressure at the inlet end as output;
step 3.2.2), selecting a data set consisting of measured total pressure, converted rotating speed, flying height, Mach number, ground atmospheric pressure, time, inlet end pressure and temperature, and distributing a training set and a test set by using a weight, wherein 75% of data is the training set and 25% of data is the test set;
step 3.3), setting network training parameters, training a neural network, and forming a dynamic total pressure intelligent measurement method;
step 3.3.1), an enumeration mode is adopted, the number of hidden layers is gradually increased from one layer, the trained network is tested, regression coefficients of the test sample and the training sample are observed, the number of network layers with larger regression coefficients of the test sample is selected, and the influence of different hidden layers on the regression coefficients of the test sample is shown in table 2. From the table, when the number of hidden layers is equal to 9, the trained recurrent neural network has the highest regression coefficient for the test sample, and meanwhile, when the number of hidden layers continues to increase, although the regression coefficients of the training samples are very close, the regression coefficient of the test sample is lower than 0.91, which indicates that the recurrent neural network has been fitted. Therefore, 9 layers are selected as the number of the hidden layers of the circular neural network of the dynamic total pressure recovery model of the compressor outlet. Setting specific parameters of the RNN neural network: the number of network training times is 1000, the learning rate is 0.001, the initial training required precision is 0, the number of hidden layers is 9, and the training neural network is shown in fig. 6;
TABLE 2 regression coefficients for different implicit layer number test samples
And 3.3.2) obtaining the dynamic pressure of the flow field of the compressor at the inlet end of the short measuring tube by measuring total pressure at the measuring end, flight environment and working state data based on the model obtained in the step 3.3.1), thereby establishing the intelligent measuring method for the dynamic pressure of the short measuring tube.
Taking a group of check samples in an intermediate state as an example, fig. 7 is a group of test sample results, a straight line in the graph is a measurement error value obtained by calculating a CFX value, a circular line is an error guess value obtained by calculating a dynamic total pressure recovery model of the text, a triangular line is an error compensated by the dynamic total pressure recovery model of the text, the maximum value of an actual error is 905Pa, which is 0.08% of the total pressure of the content, and the average value of the error is 550Pa, which is 0.05% of the total pressure of the content, and the maximum value of the error is reduced to 97Pa after being compensated by the dynamic total pressure recovery model, which is 0.009% of the total pressure of the content, and the average value of the error is 13Pa, which is 0.0012% of the total pressure of the content.
Claims (9)
1. An intelligent dynamic pressure measurement method for an aircraft engine is characterized by comprising the following steps:
step 1), establishing a 3-dimensional numerical simulation model of a short measuring pipe and an outlet runner of a gas compressor based on a coupling relation among multiple physical fields;
step 2), calculating flow field parameters at the outlet of the compressor and in the short measuring tube;
and 3), establishing a dynamic total pressure intelligent measurement method based on the RNN neural network.
2. The intelligent dynamic pressure measurement method for the aircraft engine according to claim 1, characterized in that: the 3-dimensional numerical simulation model establishing platform in the step 1) is an ANSYS CFX software platform, and the specific steps are as follows:
step 1.1), determining the geometric size of a short measuring tube, the geometric size of an outlet of a gas compressor and the length of a fully developed flow passage;
step 1.2), establishing a geometric model of an outlet flow channel of the gas compressor, a geometric model of an inner casing and a geometric model of a short measuring tube based on the size data determined in the step 1.1);
step 1.3), dividing grids according to the geometric model established in the step 1.2), and establishing a grid model of an outlet flow channel of the gas compressor, a grid model of an inner casing and a grid model of a short measuring pipe.
3. The intelligent dynamic pressure measurement method for the aircraft engine as claimed in claim 2, wherein: the specific steps of step 1.3) are as follows:
step 1.3.1), adopting ANSYS workbench Mesh software in an ANSYS CFX software platform to divide the network, and gradually carrying out local encryption on the network along the direction from the fluid domain to the fluid-solid boundary in consideration of the large air velocity gradient at the boundary layer;
step 1.3.2), in order to ensure the reliability of the calculation result and simultaneously consider the calculation time, the grid adaptability is necessary to be checked, so that the proper grid number is selected; the areas of 200mm before and after the short measuring tube cavity and the inlet of the short measuring tube are continuously encrypted in the grid verification process.
4. The intelligent dynamic pressure measurement method for the aircraft engine as claimed in claim 2, wherein: the specific steps of the step 2) are as follows:
step 2.1), selecting 6 typical working points of an engine in an airplane envelope, and carrying out engine working process simulation at the typical working points based on an existing certain turbofan engine pneumatic thermodynamic model to obtain total pressure of an outlet of a compressor in the engine, static pressure of an outer culvert of the engine, static temperature and gas flow rate data;
step 2.2), setting a 3-dimensional geometric model of a 3-dimensional numerical simulation model according to the geometric model of the outlet flow channel of the gas compressor, the geometric model of the inner casing and the geometric model of the short measuring pipe established in the step 1.2);
and 2.3) setting boundary conditions of a 3-dimensional numerical simulation model of the short measuring tube and the outlet runner of the gas compressor based on the static pressure, the static temperature, the gas flow velocity and preset material characteristic parameters obtained in the step 2.1), carrying out 3-dimensional numerical simulation of dynamic changes of the flow field pressure, and obtaining total pressure and total temperature dynamic change data of the inlet end and the measuring end of the short measuring tube.
5. The intelligent dynamic pressure measurement method for the aircraft engine as claimed in claim 4, wherein: the specific steps of step 2.1) are as follows:
step 2.1.1), 6 typical operating points are selected (H ═ 0km, Ma ═ 0.8), (H ═ 2km, Ma ═ 0.8), (H ═ 4km, Ma ═ 0.2), (H ═ 10km, Ma ═ 1.4), (H ═ 14km, Ma ═ 1), and (H ═ 16km, Ma ═ 1.8);
and 2.2.2) sequentially carrying out the aerodynamic and thermodynamic simulation of the sine, random and step form pressure signals as the input signals of the short measuring tube at 6 working points respectively to obtain the required parameters.
6. The intelligent dynamic pressure measurement method for the aircraft engine as claimed in claim 5, wherein: the specific steps of step 2.3) are as follows:
step 2.3.1), respectively defining dynamic pressure signals in a sine function form by using the amplitude of 0.3 and 0.5Mpa and the frequency of 5Hz, 18Hz and 30Hz in ANSYS CFX software, and taking the dynamic pressure signals as a boundary condition of a three-dimensional numerical simulation model;
step 2.3.2), in ANSYS CFX software, fitting according to the dynamic pressure data in the step 2.2.2) and random signal forms to obtain a nonlinear function as a boundary condition of the three-dimensional numerical simulation model.
7. The intelligent dynamic pressure measurement method for the aircraft engine according to claim 1, characterized in that: the specific steps of the step 3) are as follows:
step 3.1), establishing an RNN network;
step 3.2), based on the simulation data obtained in the step 2.1) and the step 2.3), selecting an RNN network input and output, a network data set, a training set and a test set;
and 3.3) setting RNN network training parameters and training a neural network to form the dynamic total pressure intelligent measurement method.
8. The intelligent dynamic pressure measurement method for the aircraft engine according to claim 7, characterized in that: the specific steps of the step 3.2) are as follows:
step 3.2.1), selecting total pressure, converted rotating speed, flight height, Mach number and ground atmospheric pressure at a measuring end of the short measuring tube as RNN (radio network) input, taking the difference of sampling step length among samples into consideration, taking time as one input of the RNN, and taking pressure at an inlet end of the short measuring tube as output;
step 3.2.2), selecting a short measuring tube to measure total pressure, converted rotating speed, flight altitude, Mach number, ground atmospheric pressure, time and inlet end pressure of the short measuring tube to form an RNN network data set, and randomly selecting 75% of data as a training set and 25% of data as a testing set;
9. the intelligent dynamic pressure measurement method for the aircraft engine according to claim 7, characterized in that: the specific steps of step 3.3) are as follows:
step 3.3.1), setting network training times, learning rate, training requirement precision and hidden layer number, and training an RNN network;
and 3.3.2) obtaining the dynamic pressure of the flow field of the compressor at the inlet end of the short measuring tube by using the total pressure of the measuring end of the short measuring tube, the flight environment and the working state data based on the model obtained in the step 3.3.1), thereby establishing the intelligent measuring method for the dynamic pressure of the short measuring tube.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111280406.6A CN114077775A (en) | 2021-10-29 | 2021-10-29 | Intelligent dynamic pressure measurement method for aircraft engine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111280406.6A CN114077775A (en) | 2021-10-29 | 2021-10-29 | Intelligent dynamic pressure measurement method for aircraft engine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114077775A true CN114077775A (en) | 2022-02-22 |
Family
ID=80283549
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111280406.6A Pending CN114077775A (en) | 2021-10-29 | 2021-10-29 | Intelligent dynamic pressure measurement method for aircraft engine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114077775A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120197550A1 (en) * | 2010-10-05 | 2012-08-02 | GM Global Technology Operations LLC | System for diagnosing error conditions of a gas flow control system for diesel engines |
| US20200063665A1 (en) * | 2018-01-25 | 2020-02-27 | Dalian University Of Technology | Aero-engine full flight envelope model adaptive modification method based on deep learning algorithm |
| CN111625960A (en) * | 2020-05-27 | 2020-09-04 | 海南热带汽车试验有限公司 | CFD-based E10 ethanol gasoline engine combustion three-dimensional simulation method |
| CN113051661A (en) * | 2021-02-19 | 2021-06-29 | 南京航空航天大学 | High-temperature airflow dynamic total pressure intelligent soft measurement method based on micro cavity multi-dynamics |
-
2021
- 2021-10-29 CN CN202111280406.6A patent/CN114077775A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120197550A1 (en) * | 2010-10-05 | 2012-08-02 | GM Global Technology Operations LLC | System for diagnosing error conditions of a gas flow control system for diesel engines |
| US20200063665A1 (en) * | 2018-01-25 | 2020-02-27 | Dalian University Of Technology | Aero-engine full flight envelope model adaptive modification method based on deep learning algorithm |
| CN111625960A (en) * | 2020-05-27 | 2020-09-04 | 海南热带汽车试验有限公司 | CFD-based E10 ethanol gasoline engine combustion three-dimensional simulation method |
| CN113051661A (en) * | 2021-02-19 | 2021-06-29 | 南京航空航天大学 | High-temperature airflow dynamic total pressure intelligent soft measurement method based on micro cavity multi-dynamics |
Non-Patent Citations (1)
| Title |
|---|
| 王伟韬: "航空发动机压气机叶尖间隙流体流动分析及优化设计", 《CNKI硕士电子期刊》, no. 01, 31 January 2021 (2021-01-31), pages 1 - 76 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112550758B (en) | Method for obtaining actual performance of each part of engine under complete machine condition | |
| Arts et al. | Aero-thermal performance of a two-dimensional highly loaded transonic turbine nozzle guide vane: A test case for inviscid and viscous flow computations | |
| CN111044252B (en) | High-precision air inlet channel flow measuring method | |
| CN204085645U (en) | With the gas flow standard device of self calibration structure | |
| CN113051661A (en) | High-temperature airflow dynamic total pressure intelligent soft measurement method based on micro cavity multi-dynamics | |
| Gong et al. | Aerodynamic and thermodynamic analysis of an aero-engine pre-swirl system based on structure design and performance improvement | |
| CN115524134A (en) | Air entraining amount testing structure and method for aircraft engine anti-icing system | |
| CN208534819U (en) | Device for the test of fan aeroperformance | |
| CN111498141B (en) | Method and device for realizing real-time monitoring of airflow angle based on micro probe | |
| CN114077775A (en) | Intelligent dynamic pressure measurement method for aircraft engine | |
| CN106126858A (en) | A kind of method of fan type selecting in electrokinetic cell air cooling system | |
| CN115575074A (en) | High-speed wind tunnel ventilation model internal resistance measurement uncertainty analysis method | |
| CN110083902B (en) | Temperature distortion map inverse design method based on discrete sequence | |
| Arts et al. | Aero-thermal performance of a two dimensional highly loaded transonic turbine nozzle guide vane: A test case for inviscid and viscous flow computations | |
| Campbell | An investigation of distortion indices for prediction of stalling behavior in aircraft gas turbine engines | |
| Kasper et al. | Experimental Investigation of an Aggressive S-Shaped Intermediate Compressor Duct | |
| Beard et al. | Mass flow rate measurement in a transonic turbine test facility with temperature distortion and swirl | |
| CN211144878U (en) | End face nozzle for axial flow compressor model level flow measurement | |
| CN115203831A (en) | Intelligent compensation method for dynamic total pressure probe of turbofan engine compressor outlet | |
| Davis et al. | A proposal for integration of wind tunnel and engine test programs for the evaluation of airframe-propulsion compatibility using numerical simulations | |
| CN114353708B (en) | Method for detecting icing thickness of low-flow water supply working condition pipeline of pressurized water supply system | |
| Sexton | An investigation of the dynamic stalling characteristics of rotating axial-flow compressor blades | |
| Mou | Gas Turbine Engine Performance Testing-Aero | |
| Fessler et al. | Preliminary survey of compressor rotor-blade wakes and other flow phenomena with a hot-wire anemometer | |
| Anni et al. | Numerical Simulation Analysis of Flow Field in Intake System of a Core Engine Test |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20221111 Address after: No. 29, Qinhuai District, Qinhuai District, Nanjing, Jiangsu Applicant after: Nanjing University of Aeronautics and Astronautics Applicant after: AECC AERO ENGINE CONTROL SYSTEM INSTITUTE Address before: No. 29, Qinhuai District, Qinhuai District, Nanjing, Jiangsu Applicant before: Nanjing University of Aeronautics and Astronautics |
|
| TA01 | Transfer of patent application right |