CN116186899A - Data-driven supercritical airfoil pressure distribution prediction method, system and medium - Google Patents

Data-driven supercritical airfoil pressure distribution prediction method, system and medium Download PDF

Info

Publication number
CN116186899A
CN116186899A CN202310153828.XA CN202310153828A CN116186899A CN 116186899 A CN116186899 A CN 116186899A CN 202310153828 A CN202310153828 A CN 202310153828A CN 116186899 A CN116186899 A CN 116186899A
Authority
CN
China
Prior art keywords
airfoil
pressure distribution
data
coordinates
prediction
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
Application number
CN202310153828.XA
Other languages
Chinese (zh)
Inventor
程然
王为拓
孙克斌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southwest University of Science and Technology
Original Assignee
Southwest University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Southwest University of Science and Technology filed Critical Southwest University of Science and Technology
Priority to CN202310153828.XA priority Critical patent/CN116186899A/en
Publication of CN116186899A publication Critical patent/CN116186899A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/15—Vehicle, aircraft or watercraft design
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B64—AIRCRAFT; AVIATION; COSMONAUTICS
    • B64F—GROUND OR AIRCRAFT-CARRIER-DECK INSTALLATIONS SPECIALLY ADAPTED FOR USE IN CONNECTION WITH AIRCRAFT; DESIGNING, MANUFACTURING, ASSEMBLING, CLEANING, MAINTAINING OR REPAIRING AIRCRAFT, NOT OTHERWISE PROVIDED FOR; HANDLING, TRANSPORTING, TESTING OR INSPECTING AIRCRAFT COMPONENTS, NOT OTHERWISE PROVIDED FOR
    • B64F5/00—Designing, manufacturing, assembling, cleaning, maintaining or repairing aircraft, not otherwise provided for; Handling, transporting, testing or inspecting aircraft components, not otherwise provided for
    • 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
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/04—Constraint-based CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/10—Numerical modelling
    • 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
    • G06F2113/00—Details relating to the application field
    • G06F2113/28—Fuselage, exterior or interior
    • 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)
  • Geometry (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computing Systems (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Medical Informatics (AREA)
  • Biophysics (AREA)
  • Algebra (AREA)
  • Computational Mathematics (AREA)
  • Fluid Mechanics (AREA)
  • Automation & Control Theory (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Manufacturing & Machinery (AREA)
  • Transportation (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a data-driven supercritical airfoil pressure distribution prediction method, a system and a medium, wherein the method comprises the steps of obtaining airfoil data; extracting a plurality of airfoil coordinates from the airfoil data, and processing the plurality of airfoil coordinates to obtain a plurality of airfoil control point coordinates; extracting a plurality of pressure distribution curves from the airfoil data, and processing the plurality of pressure distribution curves to obtain a plurality of pressure distribution principal component amplitudes; based on the neural network technology, constructing an airfoil pressure distribution predictor according to the coordinates of a plurality of airfoil control points and the amplitudes of a plurality of pressure distribution main components; and inputting the obtained coordinates of the control points of the airfoil to be predicted into an airfoil pressure distribution predictor for prediction to obtain a pressure distribution curve. The airfoil pressure distribution predictor disclosed by the embodiment of the invention is constructed based on the neural network model, so that the high-precision prediction of the pressure distribution of the supercritical airfoil can be realized in practice, and the problem of time and labor consumption in the calculation of the pressure distribution of the traditional airfoil is solved.

Description

Data-driven supercritical airfoil pressure distribution prediction method, system and medium
Technical Field
The invention relates to the technical field of aerodynamics of airfoils, in particular to a data-driven supercritical airfoil pressure distribution prediction method, a data-driven supercritical airfoil pressure distribution prediction system and a data-driven supercritical airfoil pressure distribution prediction medium.
Background
Currently, the aerodynamics optimization design of airfoils is an important research content in aircraft design. Pneumatic optimization is a typical black box expensive optimization problem in that the mapping of its objective function to the design variables is difficult to give in analytical form, and expensive in that the process of solving the objective function from the design variables is extremely time consuming.
The pressure distribution of a supercritical airfoil is a strong design element in pneumatic design as a pneumatic performance strongly related to its geometry and has its inherent characteristics. In conventional pneumatic designs, there is a reverse design flow with pressure distribution as the final goal of airfoil design, i.e., manually designing a pressure distribution with excellent performance with artificial experience and then reversely pushing out its corresponding airfoil. In the aerodynamic optimization design of supercritical airfoils, pressure distribution is an indispensable evaluation mode.
In airfoil design and optimization, the computation from the geometry of the airfoil to its aerodynamic performance tends to be very time consuming and resource intensive. In particular, the aerodynamic performance of the supercritical airfoil is highly sensitive to the geometric shape of the supercritical airfoil, so that the flow field simulation of the supercritical airfoil often needs more huge calculation amount, and then the pressure distribution of the airfoil is extracted by means of post-treatment and the like. Calculation of aerodynamic performance for a supercritical airfoil often takes tens of hours or even a day. The process of using such Computational Fluid Dynamics (CFD) is therefore overly labor and time intensive.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides a data-driven supercritical airfoil pressure distribution prediction method, which solves the problem that the calculation process required to be carried out for acquiring the airfoil pressure distribution at present consumes too much manpower and time resources.
The invention also provides a data-driven supercritical airfoil pressure distribution prediction system and a computer-readable storage medium.
A data-driven supercritical airfoil pressure distribution prediction method according to an embodiment of the first aspect of the present invention includes the steps of:
acquiring airfoil data, wherein the airfoil data comprises a plurality of supercritical airfoils and a plurality of airfoils obtained through disturbance;
extracting a plurality of airfoil coordinates from the airfoil data, and processing the airfoil coordinates to obtain a plurality of airfoil control point coordinates;
extracting a plurality of pressure distribution curves from the airfoil data, and processing the plurality of pressure distribution curves to obtain a plurality of pressure distribution principal component amplitudes;
based on a neural network technology, constructing an airfoil pressure distribution predictor according to the airfoil control point coordinates and the pressure distribution principal component amplitudes;
and inputting the obtained airfoil control point coordinates to be predicted to the airfoil pressure distribution predictor for prediction to obtain a pressure distribution curve, wherein the pressure distribution curve represents an airfoil pressure distribution prediction result.
The data-driven supercritical airfoil pressure distribution prediction method provided by the embodiment of the invention has at least the following beneficial effects:
the airfoil coordinates and the pressure distribution curves are extracted from the airfoil data of the supercritical airfoils, so that the airfoil coordinates and the pressure distribution curves can be processed to construct an airfoil pressure distribution predictor for performing airfoil pressure distribution prediction and obtaining a prediction result, namely specifically obtaining the pressure distribution curve. The airfoil pressure distribution predictor obtained by the method of the embodiment of the invention is constructed based on a neural network model under the artificial intelligence technology, so that the high-precision prediction of the pressure distribution of the supercritical airfoil can be realized in practice, and the problem that a large amount of calculation resources and time are required to be consumed in the calculation of the pressure distribution of the traditional airfoil is solved.
According to some embodiments of the invention, the data-driven supercritical airfoil pressure distribution prediction method further comprises the steps of:
and performing performance evaluation on the pressure distribution curve to obtain an airfoil performance evaluation result.
According to some embodiments of the invention, the performance evaluation of the pressure distribution curve to obtain airfoil performance evaluation results includes the following steps:
deriving the position of the suction platform of the pressure distribution curve to obtain a derivation result;
and when the derivative result is smaller than a set threshold value, determining that the wing profile corresponding to the pressure distribution curve is a performance qualified wing profile.
According to some embodiments of the invention, the processing of the plurality of airfoil coordinates to obtain a plurality of airfoil control point coordinates includes the steps of:
and carrying out parameterization fitting on a plurality of airfoil coordinates based on a non-uniform rational B-spline model to obtain a plurality of airfoil control point coordinates.
According to some embodiments of the invention, the processing of the plurality of pressure profiles to obtain a plurality of pressure profile principal component amplitudes comprises the steps of:
and encoding a plurality of the pressure distribution curves based on the self-encoder to obtain a plurality of the pressure distribution principal component amplitudes.
According to some embodiments of the invention, the constructing an airfoil pressure distribution predictor based on the airfoil control point coordinates and the pressure distribution principal component amplitudes based on the neural network technique comprises the following steps:
based on a multi-layer perceptron model, utilizing a plurality of airfoil control point coordinates and a plurality of pressure distribution principal component amplitudes to establish a neural network;
training the neural network to obtain the airfoil pressure distribution predictor.
According to some embodiments of the present invention, the inputting the obtained coordinates of the control point of the airfoil to be predicted to the airfoil pressure distribution predictor for prediction to obtain a pressure distribution curve includes the following steps:
acquiring a plurality of to-be-predicted airfoil control point coordinates and inputting the coordinates to the airfoil pressure distribution predictor;
outputting a plurality of pressure distribution main component amplitude prediction data after being processed by the airfoil pressure distribution predictor;
and decoding a plurality of the pressure distribution principal component amplitude prediction data based on a decoder of the self-encoder to obtain the pressure distribution curve.
A data-driven supercritical airfoil pressure distribution prediction system according to an embodiment of the second aspect of the invention comprises:
the data acquisition unit is used for acquiring airfoil data, wherein the airfoil data comprise a plurality of supercritical airfoils and a plurality of airfoils obtained through disturbance;
the first data processing unit is used for extracting a plurality of airfoil coordinates from the airfoil data and processing the airfoil coordinates to obtain a plurality of airfoil control point coordinates;
a second data processing unit for extracting a plurality of pressure distribution curves from the airfoil data and processing the plurality of pressure distribution curves to obtain a plurality of pressure distribution principal component amplitudes;
the modeling unit is used for constructing an airfoil pressure distribution predictor according to the coordinates of a plurality of airfoil control points and the amplitudes of a plurality of pressure distribution principal components based on a neural network technology;
the prediction unit is used for inputting the obtained to-be-predicted airfoil control point coordinates to the airfoil pressure distribution predictor for prediction so as to obtain a pressure distribution curve, wherein the pressure distribution curve represents an airfoil pressure distribution prediction result.
The data-driven supercritical airfoil pressure distribution prediction system provided by the embodiment of the invention has at least the following beneficial effects:
by executing the method in the system of the embodiment of the invention, a plurality of airfoil coordinates and a plurality of pressure distribution curves can be extracted from the airfoil data of a plurality of supercritical airfoils, so that the airfoil coordinates and the pressure distribution curves can be processed to construct an airfoil pressure distribution predictor for carrying out airfoil pressure distribution prediction and obtaining a prediction result, namely specifically obtaining the pressure distribution curve. The airfoil pressure distribution predictor obtained by utilizing the system of the embodiment of the invention is constructed based on a neural network model under the artificial intelligence technology, so that the high-precision prediction of the pressure distribution of the supercritical airfoil can be realized in practice, and the problem that a large amount of calculation resources and time are required to be consumed in the calculation of the pressure distribution of the traditional airfoil is solved.
According to some embodiments of the invention, the data-driven supercritical airfoil pressure distribution prediction system further comprises an evaluation unit for performing a performance evaluation on the pressure distribution curve to obtain an airfoil performance evaluation result.
A computer readable storage medium according to an embodiment of the third aspect of the present invention stores computer executable instructions for causing a computer to perform a data-driven supercritical airfoil pressure distribution prediction method according to an embodiment of the first aspect of the present invention.
It will be appreciated that the advantages of the third aspect compared to the related art are the same as those of the first aspect compared to the related art, and reference may be made to the related description in the first aspect, which is not repeated here.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the invention will become apparent and may be better understood from the following description of embodiments taken in conjunction with the accompanying drawings in which:
FIG. 1 is a flow chart of a data-driven supercritical airfoil pressure distribution prediction method of an embodiment of the invention;
FIG. 2 is a schematic representation of the location of a pressure profile suction platform in accordance with an embodiment of the present invention;
FIG. 3 is a schematic view of airfoil coordinate processing according to an embodiment of the invention;
FIG. 4 is a pressure profile processing schematic of an embodiment of the present invention;
FIG. 5 is a schematic diagram of a pressure profile predictor in accordance with an embodiment of the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention.
In the description of the present invention, the description of first, second, etc. is for the purpose of distinguishing between technical features only and should not be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
In the description of the present invention, it should be understood that the direction or positional relationship indicated with respect to the description of the orientation, such as up, down, etc., is based on the direction or positional relationship shown in the drawings, is merely for convenience of describing the present invention and simplifying the description, and does not indicate or imply that the apparatus or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
In the description of the present invention, unless explicitly defined otherwise, terms such as arrangement, installation, connection, etc. should be construed broadly and the specific meaning of the terms in the present invention can be determined reasonably by a person skilled in the art in combination with the specific content of the technical solution.
The following description of the embodiments of the present invention will be made with reference to the accompanying drawings, in which it is apparent that the embodiments described below are some, but not all embodiments of the invention.
Referring to fig. 1, a flowchart of a data-driven supercritical airfoil pressure distribution prediction method according to an embodiment of the present invention is shown, where the method includes the following steps:
acquiring airfoil data, wherein the airfoil data comprises a plurality of supercritical airfoils and a plurality of airfoils obtained through disturbance;
extracting a plurality of airfoil coordinates from the airfoil data, and processing the plurality of airfoil coordinates to obtain a plurality of airfoil control point coordinates;
extracting a plurality of pressure distribution curves from the airfoil data, and processing the plurality of pressure distribution curves to obtain a plurality of pressure distribution principal component amplitudes;
based on the neural network technology, constructing an airfoil pressure distribution predictor according to the coordinates of a plurality of airfoil control points and the amplitudes of a plurality of pressure distribution main components;
and inputting the obtained coordinates of the control points of the airfoil to be predicted into an airfoil pressure distribution predictor for prediction to obtain a pressure distribution curve, wherein the pressure distribution curve represents the airfoil pressure distribution prediction result.
Specifically, as shown in fig. 1, it can be understood that, for the method of the embodiment of the present invention, the main design concept is to construct an airfoil pressure distribution predictor based on the collected plurality of airfoils and using the pressure distribution thereof, and take the pressure distribution as an airfoil evaluation method in an optimization framework. In some embodiments, five hundred supercritical airfoils are collected first, airfoil data are obtained through disturbance, coordinate points of the five hundred airfoils are parameterized, and pressure distribution of the five hundred airfoils is also preprocessed, so that control coordinate points of the five hundred airfoils and ten pressure distribution principal component amplitudes are obtained respectively. And processing the relevant airfoil parameters, so as to construct a pressure distribution predictor model based on a neural network technology, and finally, predicting the airfoil pressure distribution of the obtained airfoil control point coordinates to be predicted through a trained predictor to obtain a pressure distribution curve serving as a prediction result.
Further, it can be understood that the method of the embodiment of the invention can effectively and highly accurately predict the pneumatic performance pressure distribution of the airfoil, so that the method can be used in the pneumatic optimization design of the supercritical airfoil; the method provided by the embodiment of the invention can effectively replace the CFD effect, and remarkably saves the calculation time in the supercritical airfoil design process; the method provided by the embodiment of the invention is suitable for wide wing shapes and has stronger generalization, so that the method can be used for carrying out optimal design based on various wing shapes.
In this embodiment, the airfoil coordinates and the pressure distribution curves are extracted from the airfoil data of the supercritical airfoils, so that the airfoil coordinates and the pressure distribution curves can be processed to construct an airfoil pressure distribution predictor for performing airfoil pressure distribution prediction and obtaining a prediction result, that is, specifically, obtaining the pressure distribution curve. The airfoil pressure distribution predictor obtained by the method of the embodiment of the invention is constructed based on a neural network model under the artificial intelligence technology, so that the high-precision prediction of the pressure distribution of the supercritical airfoil can be realized in practice, and the problem that a large amount of calculation resources and time are required to be consumed in the calculation of the pressure distribution of the traditional airfoil is solved.
In some embodiments, as shown in FIG. 1, the data-driven supercritical airfoil pressure distribution prediction method further comprises the steps of:
and performing performance evaluation on the pressure distribution curve to obtain an airfoil performance evaluation result.
Specifically, referring to fig. 1, it may be understood that after the airfoil pressure distribution prediction result is obtained by using the method of the embodiment of the present invention, performance evaluation may be continuously performed on the airfoil pressure distribution prediction result, that is, on the pressure distribution curve, so as to determine whether the predicted airfoil corresponding to the predicted pressure distribution curve has better performance. Therefore, by performing the performance analysis, the airfoil pressure distribution predictor of the embodiment of the present invention can be obtained while further confirming the prediction effect thereof. Specifically, in the optimization flow framework, numerical indexes such as the rising resistance of the airfoil profile are easy to be used as optimization indexes, and are easy to be set as objective functions. For example, in an optimization framework of evolution calculation, the rising resistance index of each airfoil can be calculated, so that numerical operations such as difference and the like are performed between the rising resistance index and a set excellent index, and the index, namely the pressure distribution map, is usually used as an image for manual judgment, wherein the population is continuously evolved towards the direction with good numerical indexes, and then the pressure distribution is used. By utilizing the method steps of the embodiment of the invention, the method can be used as a numerical index to be embedded into an optimization framework for evaluation, specifically, the derivative of the suction platform can be calculated, and the better the numerical value of the difference value from 0 is, the better the obtained wing profile is.
In some embodiments, as shown in FIG. 2, performance evaluation is performed on the pressure profile to obtain airfoil performance evaluation results, including the steps of:
deriving the position of the suction platform of the pressure distribution curve to obtain a derivation result;
and when the derivative result is smaller than the set threshold value, determining that the wing profile corresponding to the pressure distribution curve is a performance qualified wing profile.
Specifically, referring to fig. 2, it can be understood that, in the performance evaluation process of the pressure distribution curve, specifically, by conducting derivative processing on the position pointed by the arrow in fig. 2, that is, the position of the suction platform of the pressure distribution curve, since the derivative obtained by the suction platform of the supercritical airfoil with better performance is almost zero, the airfoil with the derivative of the pressure distribution suction platform in a very small range in the evaluation result, that is, the derivative result is smaller than the set threshold, is the airfoil with qualified performance. In the pneumatic optimization design of the supercritical airfoil, the pressure distribution of the generated new airfoil is an important index when evaluating the performance, but the pressure distribution is usually judged to be good or bad by manual experience as a graph under the normal condition, so the embodiment of the invention can automatically embed the pressure distribution constraint into the optimization flow of the airfoil by a numerical method.
In some embodiments, as shown in FIG. 3, the processing of the plurality of airfoil coordinates to obtain a plurality of airfoil control point coordinates includes the steps of:
and carrying out parameterization fitting on the plurality of airfoil coordinates based on the non-uniform rational B-spline model to obtain a plurality of airfoil control point coordinates.
In particular, with reference to fig. 3, it can be appreciated that in aircraft development, high performance is increasingly required, and current research centers have been placed primarily on high performance airfoil designs, such as supercritical airfoils. The high-performance airfoil design refers to high-precision fitting of airfoil optimization design results, so that the purpose of fine design of the appearance of the aircraft is achieved. For the design of high-performance airfoils, airfoil parameterization is the basis of numerical computation, and is one of the important contents of the current airfoil design research, and an effective, robust and high-precision parameterization method needs to be searched. One such parameterization is Non-uniform rational B-Splines (Non-Uniform Rational B-Spines, NURBS). The specific explanation is as follows:
non-uniformity (Non-uniformity): it is meant that the range of influence of a control vertex can vary. This is useful when creating an irregular surface. Also, unified curves and surfaces are not invariant under perspective projection, which is a serious drawback for interactive 3D modeling.
Rational (ratio): meaning that each NURBS object can be defined by a rational polynomial formal expression.
B-Spline (B-Spline): refers to constructing a curve with routes that are interpolated between one or more points.
NURBS can be used to create curved objects, and because NURBS is always defined by curves and curves, it is difficult to create a faceted edge in the NURBS surface, and based on this feature, various complex curved shapes and special effects can be created with NURBS, such as human skin, face or streamline sports cars, airfoils, etc. Therefore, NURBS is a very excellent modeling mode, and is widely used in some software such as CAD, and NURBS can better control the curvilinearity of the object surface than the traditional grid modeling mode, so that a more vivid modeling can be created, and the method has very strong flexibility, so that the airfoil can be locally controlled and smooth.
In some embodiments, as shown in FIG. 4, the processing of the plurality of pressure profiles to obtain a plurality of pressure profile principal component amplitudes includes the steps of:
based on the self-encoder, a plurality of pressure profiles are encoded to obtain a plurality of pressure profile principal component amplitudes.
Specifically, referring to fig. 4, it can be understood that a self encoder (AE) is a type of artificial neural network used in semi-supervised learning and non-supervised learning, and functions to perform characterization learning on input information by taking the input information as a learning target. The self-Encoder comprises two parts, namely an Encoder (Encoder) and a Decoder (Decoder), and in the embodiment, a plurality of pressure distribution curves are input into the Encoder, so that a high-dimensional hidden variable vector can be obtained, and the hidden variable vector can be re-decoded into the original input pressure distribution curve through the Decoder. Thus, by utilizing a self-encoder, pressure profile encoding can be achieved to obtain a plurality of pressure profile principal component amplitudes.
In some embodiments, as shown in fig. 5, an airfoil pressure distribution predictor is constructed from a plurality of airfoil control point coordinates, a plurality of pressure distribution principal component amplitudes based on a neural network technique, comprising the steps of:
based on the multi-layer perceptron model, utilizing a plurality of airfoil control point coordinates and a plurality of pressure distribution principal component amplitudes to establish a neural network;
the neural network is trained to obtain the airfoil pressure distribution predictor.
Specifically, referring to FIG. 5, FIG. 5 is a diagram of the present inventionThe pressure profile predictor of the embodiment is a schematic structural diagram. It will be appreciated that embodiments of the present invention utilize a multi-layer perceptron (MLP) to build an airfoil pressure profile predictor, and thus the airfoil pressure profile predictor of embodiments of the present invention includes an input layer, a hidden layer, an output layer, with a full connection between the layers. The input to the input layer is a plurality of airfoil control point coordinates and the output of the final output layer is a plurality of pressure distribution principal component amplitudes corresponding to the airfoil, in some embodiments ten pressure distribution principal component amplitudes, with the corresponding loss function being a mean square loss function. Nodes in the figure represent fully connected nodes, W 1 、W 2 、W 3 The weights of the neural networks of each layer are respectively. Therefore, the airfoil pressure distribution predictor is finally obtained by utilizing the airfoil control point coordinates and the pressure distribution principal component amplitude as input and output parameters respectively to perform nonlinear model training of the neural network.
In some embodiments, the obtained airfoil control point coordinates to be predicted are input to an airfoil pressure distribution predictor for prediction to obtain a pressure distribution curve, and the method comprises the following steps:
acquiring a plurality of airfoil control point coordinates to be predicted and inputting the coordinates to an airfoil pressure distribution predictor;
outputting a plurality of pressure distribution main component amplitude prediction data after being processed by an airfoil pressure distribution predictor;
a decoder based on the self-encoder decodes the plurality of pressure distribution principal component amplitude prediction data to obtain a pressure distribution curve.
Specifically, it can be understood that after the airfoil pressure distribution predictor according to the embodiment of the present invention is obtained, the airfoil pressure distribution predictor can be used for pressure distribution prediction, and the specific process is as follows: firstly, acquiring a plurality of airfoil control point coordinates of an airfoil to be predicted, namely acquiring a plurality of airfoil control point coordinates to be predicted; then inputting a plurality of to-be-predicted airfoil control point coordinates into an airfoil pressure distribution predictor of the embodiment of the invention to obtain a plurality of predicted pressure distribution principal component amplitudes, namely obtaining a plurality of pressure distribution principal component amplitude prediction data; and finally, using a decoder in the self-encoder to correspondingly decode the predicted amplitudes of the main components of the plurality of pressure distribution, thereby restoring the pressure distribution curve.
In addition, the embodiment of the invention also provides a data-driven supercritical airfoil pressure distribution prediction system, which comprises: the system comprises a data acquisition unit, a first data processing unit, a second data processing unit, a modeling unit and a prediction unit. The data acquisition unit is used for acquiring airfoil data, wherein the airfoil data comprises a plurality of supercritical airfoils and a plurality of airfoils obtained through disturbance; the first data processing unit is used for extracting a plurality of airfoil coordinates from the airfoil data and processing the airfoil coordinates to obtain a plurality of airfoil control point coordinates; the second data processing unit is used for extracting a plurality of pressure distribution curves from the wing profile data and processing the plurality of pressure distribution curves to obtain a plurality of pressure distribution principal component amplitudes; the modeling unit is used for constructing an airfoil pressure distribution predictor according to the coordinates of a plurality of airfoil control points and the amplitudes of a plurality of pressure distribution main components based on the neural network technology; the prediction unit is used for inputting the obtained to-be-predicted airfoil control point coordinates to an airfoil pressure distribution predictor for prediction so as to obtain a pressure distribution curve, wherein the pressure distribution curve represents an airfoil pressure distribution prediction result.
Specifically, referring to fig. 1 in combination, it can be understood that, for the system of the embodiment of the present invention, the main design concept is to construct an airfoil pressure distribution predictor based on the collected multiple airfoils and using the pressure distribution thereof, and take the pressure distribution as an airfoil evaluation method in an optimization framework. In some embodiments, five hundred supercritical airfoils are collected by using a data acquisition unit and the airfoils are obtained as airfoil data by disturbance, then the coordinate points of the five hundred airfoils are parameterized by using a first data processing unit, and the pressure distribution of the five hundred airfoils is also preprocessed by using a second data processing unit, so as to obtain control coordinate points of the five hundred airfoils and ten pressure distribution principal component amplitudes respectively. The relevant airfoil parameters are processed, a modeling unit is utilized, a pressure distribution predictor model is constructed based on a neural network technology, and finally, the airfoil pressure distribution prediction is carried out on the obtained airfoil control point coordinates to be predicted in the trained predictor through the prediction unit, so that a pressure distribution curve serving as a prediction result is obtained.
Further, it can be appreciated that the system of the embodiment of the invention can effectively and highly accurately predict the airfoil aerodynamic performance pressure distribution, so that the system can be used in supercritical airfoil aerodynamic optimization design; the system of the embodiment of the invention can effectively replace the CFD effect, and remarkably saves the calculation time in the design process of the supercritical airfoil; the system provided by the embodiment of the invention is suitable for wide wing shapes and has stronger generalization, so that the system can be optimally designed based on various wing shapes.
In this embodiment, by executing the method according to the embodiment of the present invention in the system according to the embodiment of the present invention, a plurality of airfoil coordinates and a plurality of pressure distribution curves may be extracted from airfoil data of a plurality of supercritical airfoils, so that the plurality of airfoil coordinates and the plurality of pressure distribution curves may be processed to construct an airfoil pressure distribution predictor, which is used for performing airfoil pressure distribution prediction and obtaining a prediction result, that is, specifically, obtaining a pressure distribution curve. The airfoil pressure distribution predictor obtained by utilizing the system of the embodiment of the invention is constructed based on a neural network model under the artificial intelligence technology, so that the high-precision prediction of the pressure distribution of the supercritical airfoil can be realized in practice, and the problem that a large amount of calculation resources and time are required to be consumed in the calculation of the pressure distribution of the traditional airfoil is solved.
In some embodiments, the data-driven supercritical airfoil pressure distribution prediction system further includes an evaluation unit for performing a performance evaluation on the pressure distribution curve to obtain an airfoil performance evaluation result.
Specifically, referring to fig. 1 in combination, it may be understood that after the airfoil pressure distribution prediction result is obtained by using the system according to the embodiment of the present invention, the airfoil pressure distribution prediction result may be continuously evaluated, that is, the evaluation unit is used to evaluate the performance of the pressure distribution curve, so as to determine whether the predicted airfoil corresponding to the predicted pressure distribution curve has better performance. Therefore, by performing performance analysis using the evaluation unit, the airfoil pressure distribution predictor of the embodiment of the present invention can be obtained while further confirming the prediction effect thereof.
Further, referring to fig. 2, it can be understood that, in the specific processing procedure of the evaluation unit, the derivative of the suction platform of the supercritical airfoil with better performance is almost zero by conducting the derivative processing on the position pointed by the arrow in fig. 2, that is, the position of the suction platform of the pressure distribution curve, so that the airfoil with the derivative of the suction platform of the pressure distribution in a very small range in the evaluation result, that is, the derivative result is smaller than the set threshold, is the airfoil with qualified performance. In the pneumatic optimization design of the supercritical airfoil, the pressure distribution of the generated new airfoil is an important index when evaluating the performance, but the pressure distribution is usually judged as a graph by manual experience, so that the evaluation unit of the embodiment of the invention can automatically embed the pressure distribution constraint into the optimization flow of the airfoil by a numerical method.
Furthermore, embodiments of the present invention provide a computer-readable storage medium storing computer-executable instructions that are executed by one or more control processors to cause the one or more control processors to perform a data-driven supercritical airfoil pressure distribution prediction method of the method embodiments described above, for example, to perform the method of fig. 1 described above.
Those of ordinary skill in the art will appreciate that all or some of the steps, systems, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as known to those skilled in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Furthermore, as is well known to those of ordinary skill in the art, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
The embodiments of the present invention have been described in detail with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of one of ordinary skill in the art without departing from the spirit of the present invention.

Claims (10)

1. The data-driven supercritical airfoil pressure distribution prediction method is characterized by comprising the following steps of:
acquiring airfoil data, wherein the airfoil data comprises a plurality of supercritical airfoils and a plurality of airfoils obtained through disturbance;
extracting a plurality of airfoil coordinates from the airfoil data, and processing the airfoil coordinates to obtain a plurality of airfoil control point coordinates;
extracting a plurality of pressure distribution curves from the airfoil data, and processing the plurality of pressure distribution curves to obtain a plurality of pressure distribution principal component amplitudes;
based on a neural network technology, constructing an airfoil pressure distribution predictor according to the airfoil control point coordinates and the pressure distribution principal component amplitudes;
and inputting the obtained airfoil control point coordinates to be predicted to the airfoil pressure distribution predictor for prediction to obtain a pressure distribution curve, wherein the pressure distribution curve represents an airfoil pressure distribution prediction result.
2. The data-driven supercritical airfoil pressure distribution prediction method according to claim 1, further comprising the steps of:
and performing performance evaluation on the pressure distribution curve to obtain an airfoil performance evaluation result.
3. The method of claim 2, wherein said performing a performance evaluation on said pressure profile to obtain an airfoil performance evaluation result comprises the steps of:
deriving the position of the suction platform of the pressure distribution curve to obtain a derivation result;
and when the derivative result is smaller than a set threshold value, determining that the wing profile corresponding to the pressure distribution curve is a performance qualified wing profile.
4. The data-driven supercritical airfoil pressure distribution prediction method according to claim 1, wherein said processing of a plurality of said airfoil coordinates to obtain a plurality of airfoil control point coordinates comprises the steps of:
and carrying out parameterization fitting on a plurality of airfoil coordinates based on a non-uniform rational B-spline model to obtain a plurality of airfoil control point coordinates.
5. The data-driven supercritical airfoil pressure distribution prediction method according to claim 4, wherein said processing of a plurality of said pressure distribution curves to obtain a plurality of pressure distribution principal component amplitudes comprises the steps of:
and encoding a plurality of the pressure distribution curves based on the self-encoder to obtain a plurality of the pressure distribution principal component amplitudes.
6. The data-driven supercritical airfoil pressure distribution prediction method according to claim 5, wherein the constructing an airfoil pressure distribution predictor from a plurality of airfoil control point coordinates and a plurality of pressure distribution principal component amplitudes based on a neural network technique comprises the steps of:
based on a multi-layer perceptron model, utilizing a plurality of airfoil control point coordinates and a plurality of pressure distribution principal component amplitudes to establish a neural network;
training the neural network to obtain the airfoil pressure distribution predictor.
7. The method for predicting the pressure distribution of a data-driven supercritical airfoil according to claim 6, wherein the step of inputting the obtained coordinates of the control point of the airfoil to be predicted to the airfoil pressure distribution predictor for prediction to obtain a pressure distribution curve comprises the steps of:
acquiring a plurality of to-be-predicted airfoil control point coordinates and inputting the coordinates to the airfoil pressure distribution predictor;
outputting a plurality of pressure distribution main component amplitude prediction data after being processed by the airfoil pressure distribution predictor;
and decoding a plurality of the pressure distribution principal component amplitude prediction data based on a decoder of the self-encoder to obtain the pressure distribution curve.
8. A data-driven supercritical airfoil pressure distribution prediction system, comprising:
the data acquisition unit is used for acquiring airfoil data, wherein the airfoil data comprise a plurality of supercritical airfoils and a plurality of airfoils obtained through disturbance;
the first data processing unit is used for extracting a plurality of airfoil coordinates from the airfoil data and processing the airfoil coordinates to obtain a plurality of airfoil control point coordinates;
a second data processing unit for extracting a plurality of pressure distribution curves from the airfoil data and processing the plurality of pressure distribution curves to obtain a plurality of pressure distribution principal component amplitudes;
the modeling unit is used for constructing an airfoil pressure distribution predictor according to the coordinates of a plurality of airfoil control points and the amplitudes of a plurality of pressure distribution principal components based on a neural network technology;
the prediction unit is used for inputting the obtained to-be-predicted airfoil control point coordinates to the airfoil pressure distribution predictor for prediction so as to obtain a pressure distribution curve, wherein the pressure distribution curve represents an airfoil pressure distribution prediction result.
9. The data-driven supercritical airfoil pressure distribution prediction system according to claim 8, further comprising an evaluation unit for performing a performance evaluation on the pressure distribution curve to obtain an airfoil performance evaluation result.
10. A computer-readable storage medium storing computer-executable instructions for causing a computer to perform the data-driven supercritical airfoil pressure distribution prediction method according to any one of claims 1 to 7.
CN202310153828.XA 2023-02-09 2023-02-09 Data-driven supercritical airfoil pressure distribution prediction method, system and medium Pending CN116186899A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310153828.XA CN116186899A (en) 2023-02-09 2023-02-09 Data-driven supercritical airfoil pressure distribution prediction method, system and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310153828.XA CN116186899A (en) 2023-02-09 2023-02-09 Data-driven supercritical airfoil pressure distribution prediction method, system and medium

Publications (1)

Publication Number Publication Date
CN116186899A true CN116186899A (en) 2023-05-30

Family

ID=86445975

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310153828.XA Pending CN116186899A (en) 2023-02-09 2023-02-09 Data-driven supercritical airfoil pressure distribution prediction method, system and medium

Country Status (1)

Country Link
CN (1) CN116186899A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117350178A (en) * 2023-12-05 2024-01-05 深圳十沣科技有限公司 Airfoil lift resistance prediction method, apparatus, device and storage medium

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117350178A (en) * 2023-12-05 2024-01-05 深圳十沣科技有限公司 Airfoil lift resistance prediction method, apparatus, device and storage medium
CN117350178B (en) * 2023-12-05 2024-04-02 深圳十沣科技有限公司 Airfoil lift resistance prediction method, apparatus, device and storage medium

Similar Documents

Publication Publication Date Title
CN116167668B (en) BIM-based green energy-saving building construction quality evaluation method and system
Lu et al. Generalized radial basis function neural network based on an improved dynamic particle swarm optimization and AdaBoost algorithm
CN110223324A (en) A kind of method for tracking target of the twin matching network indicated based on robust features
CN108197736A (en) A kind of Air Quality Forecast method based on variation self-encoding encoder and extreme learning machine
CN112036463A (en) Power equipment defect detection and identification method based on deep learning
EP2817783A1 (en) Method and apparatus for mesh simplification
CN116186899A (en) Data-driven supercritical airfoil pressure distribution prediction method, system and medium
CN114332578A (en) Image anomaly detection model training method, image anomaly detection method and device
CN115983148B (en) CFD simulation cloud image prediction method, system, electronic equipment and medium
CN116227543B (en) Detection method and detection device for abnormal electricity consumption, electronic device and electronic equipment
CN113822284A (en) RGBD image semantic segmentation method based on boundary attention
CN113591566A (en) Training method and device of image recognition model, electronic equipment and storage medium
CN115587964A (en) Entropy screening-based pseudo label cross consistency change detection method
CN112949944A (en) Underground water level intelligent prediction method and system based on space-time characteristics
Cao et al. Multi-feature fusion tracking based on a new particle filter
CN116628836A (en) Data-driven supercritical airfoil performance prediction method, system and medium
CN108197368B (en) Method for simply and conveniently calculating geometric constraint and weight function of complex aerodynamic shape of aircraft
Jia et al. Sample generation of semi‐automatic pavement crack labelling and robustness in detection of pavement diseases
CN115879536A (en) Learning cognition analysis model robustness optimization method based on causal effect
CN114595751A (en) Node classification method, system, readable storage medium and computer device
CN103530886A (en) Low-calculation background removing method for video analysis
CN113570129A (en) Method for predicting strip steel pickling concentration and computer readable storage medium
CN113780644A (en) Photovoltaic output prediction method based on online learning
CN108629134A (en) A kind of similitude intensifying method in field small in manifold
CN112990304B (en) Semantic analysis method and system suitable for power scene

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