Method for predicting aerodynamic parameters of airfoil flow field based on point cloud network

Abstract

The invention belongs to the field of computational fluid mechanics and the field of deep learning, and particularly relates to an airfoil profile flow field aerodynamic parameter prediction method based on a point cloud network. And directly using an airfoil flow field grid coordinate in CFD calculation as an input, and using a flow field parameter corresponding to a flow field grid point as an output. Compared with a convolutional neural network, the point cloud neural network does not need to carry out image processing on the data, artificial errors are prevented from being introduced, and the steps of data preprocessing are simplified. Meanwhile, grid-based CFD calculation data naturally has point cloud characteristics and can be directly used for a point cloud network. Besides, in order to avoid overlarge input and output dimensions of the point cloud network, downsampling is performed on an airfoil flow field grid, and then aerodynamic parameters of non-sampling points of the flow field are filled based on flow field aerodynamic parameter predicted values of sampling grid points through a gappy POD method.

Description

Technical Field

[0001] The present invention belongs to the fields of computational fluid dynamics and deep learning, and specifically relates to a method for predicting aerodynamic parameters of an airfoil flow field based on a point cloud network. Background Art

[0002] The prediction of airfoil flow field is of great significance in aerospace, wind power generation and other fields. Airfoil flow field refers to the flow field formed by air or other fluids flowing around the airfoil. Accurate calculation of airfoil flow field is crucial for aircraft design, aerodynamic performance optimization and flight safety. Traditional flow field solution methods mainly rely on computational fluid dynamics (CFD) technology. CFD solves fluid mechanics equations (such as Navier-Stokes equations) through numerical simulation methods to simulate the flow of fluid around the airfoil. However, although the CFD method is accurate, it has high computational cost and long time consumption, especially when dealing with complex geometric shapes, the computational efficiency is further reduced. Traditional CFD methods usually use numerical methods such as finite volume method (FVM), finite element method (FEM) or finite difference method (FDM), which require the process to be discretized into a grid structure. In order to capture the flow field details on complex surfaces and walls, mesh refinement and other operations are required at the leading and trailing edges of the airfoil and the wall. These mesh generation and refinement steps themselves also require a lot of computing resources and time.

[0003] In recent years, with the development of machine learning technology, the application of machine learning in flow field prediction has gradually attracted attention. Machine learning methods learn flow field characteristics from a large amount of flow field data by training neural networks, thereby achieving efficient flow field prediction. Compared with traditional CFD methods, machine learning methods have the advantages of fast calculation speed and strong adaptability. In the current public literature, the flow field prediction method based on machine learning is mainly a convolutional neural network (CNN) based on image processing. In the prediction of airfoil flow field based on convolutional neural network, the airfoil data is usually pixelated and converted into an image as the input of the neural network. Artificial errors are introduced in the process of converting data such as airfoil coordinates into image pixels. In order for the convolutional neural network to capture the slight deformation of the airfoil, the resolution of the input image needs to be improved. The larger the resolution of the input image, the more parameters to be trained for the neural network and the longer the training time. Secondly, in order to visualize the airfoil geometry as the input of the convolutional neural network, the airfoil boundary and the background are usually set to different pixel values, or the pixels in the internal area of ​​the airfoil are masked, such as setting all the pixel values ​​in the internal area of ​​the airfoil to 0 to distinguish the boundary of the airfoil. This processing method results in that not all pixels in the input image are used to describe the airfoil geometry, wasting the computational power of the convolutional neural network. Finally, in the CFD calculation domain, the areas of interest to the researchers, such as near the wall and the leading edge of the airfoil, will be meshed to obtain more accurate calculation results. However, if pixelation is performed, any area in the calculation domain will be treated "equally" and divided into pixels at equal intervals. This results in the mesh encryption operation in CFD calculations to improve calculation accuracy being completely wasted. Summary of the invention

[0004] Purpose of the present invention: In order to solve the above problems of convolutional neural networks in the prediction of airfoil flow fields, in view of the deficiencies in the prior art, this application proposes a method for predicting aerodynamic parameters of airfoil flow fields based on point cloud networks, directly using the airfoil flow field grid coordinates in CFD calculations as input, and using the flow field parameters corresponding to the flow field grid points as output. Compared with convolutional neural networks, point cloud neural networks do not need to image the data, avoid the introduction of artificial errors, and simplify the data preprocessing steps. At the same time, the grid-based CFD calculation data naturally has point cloud characteristics and can be directly used in point cloud networks. In addition, in order to avoid the input and output dimensions of the point cloud network being too large, the airfoil flow field grid is downsampled, and then the aerodynamic parameters of the unsampled points of the flow field are filled in based on the predicted values ​​of the flow field aerodynamic parameters of the sampling grid points through the gappy POD method.

[0005] Technical solution of the present invention: In order to achieve the above-mentioned invention object, the present invention adopts the following technical solution: a method for predicting aerodynamic parameters of airfoil flow field based on point cloud network, the specific steps of which include:

[0006] Step 1: Select an initial airfoil and establish a two-dimensional coordinate system Oxy, where the coordinate origin O is the leading edge of the airfoil, the Ox axis passes through the trailing edge of the airfoil and points to the trailing edge of the airfoil, and the Oy axis is perpendicular to the Ox axis and upward; combine the airfoil parameterization method and the random sampling method to generate N random airfoils based on the initial airfoil to form an airfoil library;

[0007] Step 2: Set the flow field conditions and obtain the flow field aerodynamic parameters of each airfoil in the airfoil library based on CFD simulation calculation;

[0008] Step 3: Sample Ns grid points from the Nt grid points in the airfoil flow field; the sampling rule is: if the i-th (1≤i≤Nt) grid point is a grid point on the airfoil surface, it is sampled; otherwise, the probability of its being sampled is proportional to the distance D from the point to the airfoil surface. i Satisfies the normal distribution:

[0009]

[0010] in D i is the distance from the i-th (1≤i≤Nt) grid point to the airfoil surface;

[0011] Step 4: For all airfoils in the airfoil library, the coordinates of Ns sampling grid points of the airfoil flow field are used as input data, and the aerodynamic parameters of the flow field of the sampling grid points are used as output data to establish a neural network data set;

[0012] Step 5: Establish a point cloud network, and then randomly select 80% of the data set in step 4 as the training data set, and the remaining 20% ​​of the data set as the test data set. Use the training data set to train the point cloud network, and the test data set is used to test and verify the prediction performance of the point cloud network;

[0013] Step 6: Select an airfoil from the test data set, use the coordinates of the airfoil flow field sampling grid points as input, input the trained point cloud network, and use the network's output value as the predicted value of the aerodynamic parameters of the airfoil flow field sampling grid points.

[0014] Preferably, the prediction method further comprises the following steps:

[0015] Step 7: Create a matrix P with 0.8×N rows and Ns×2+Nt×n columns. Each row of the matrix P represents the flow field data of an airfoil sample in the training set.

[0016] The 1st to Ns×2 columns of the matrix P are the horizontal and vertical coordinates of the Ns sampling grid points, and the Ns×2+1 to Ns×(2+n) columns are the aerodynamic parameters of the airfoil flow field corresponding to the Ns sampling grid points;

[0017] The columns Ns×(2+n)+1 to the last column are the n aerodynamic parameters corresponding to the non-sampling grid points of the airfoil flow field.

[0018] Create a row vector Pr, where the data of Pr is the airfoil flow field data to be predicted, wherein the 1st to Ns×2 columns are the horizontal and vertical coordinates of the Ns sampling grid points in the airfoil flow field to be predicted, the Ns×2+1 column to the Ns×(2+n) column are the flow field parameters of the Ns sampling grid points in the airfoil flow field to be predicted, which are obtained by prediction in step 6, and the Ns×(2+n)+1 column to the last column are the n aerodynamic parameters corresponding to the non-sampling grid points in the airfoil flow field to be predicted, which are obtained by calculating the average value of the corresponding columns of the matrix P.

[0019] Step 8: Use the intrinsic orthogonal decomposition method to calculate the base mode matrix Pc, base mode coefficient Ps and mean vector Pm of the combination matrix [P; Pr]; the base mode matrix Pc, base mode coefficient Ps and mean vector Pm satisfy [P; Pr] = Ps × Pc T +Pm;

[0020] Step 9: Update Pr using formula (2):

[0021]

[0022] Where b is the principal component coefficient vector;

[0023] Step 10: Repeat step 9 to iteratively update Pr until the maximum number of iterations E is reached. The Ns×2+1 columns to the last column of Pr are the aerodynamic parameters corresponding to the flow field grid points of the airfoil to be predicted.

[0024] Preferably, in step 1, the airfoil parameterization method is a common parameter method, such as a free deformation method, a quasi-function shape function method, etc.

[0025] Preferably, in step 1, the method for generating a random airfoil library based on the initial airfoil by combining the airfoil parameterization method and the random sampling method is to parameterize the initial airfoil by applying the airfoil parameterization method, and then select N groups of geometric parameters by random sampling in the parameter space, deform the initial airfoil according to the geometric parameters, and use the set of N new airfoils after deformation as the airfoil library.

[0026] Preferably, in step 2, the aerodynamic parameters of the airfoil flow field may be aerodynamic pressure coefficient, Mach number, speed or other aerodynamic parameters of the airfoil flow field that can be obtained through CFD simulation calculation.

[0027] Preferably, in step 4, the format of the input data is a matrix of N×Ns×2, that is, there are N input samples, each sample is a matrix of Ns rows and 2 columns, and each row of data represents the horizontal and vertical coordinates of a sampling grid point. The format of the output data is a matrix of N×Ns×n, that is, there are N output samples, each sample is a matrix of Ns rows and n columns, and each row of data represents n flow field aerodynamic parameters corresponding to a sampling grid point.

[0028] Preferably, in step 5, the structure of the point cloud network is: Ns×2 input layer, 64×64 multilayer perceptron 1, 64×64 multilayer perceptron 2, 64×128 multilayer perceptron 3, maximum pooling layer 1, merging layer of multilayer perceptron 2 and maximum pooling layer 1, 128×128 multilayer perceptron 4, 128×n multilayer perceptron 5, N×n output layer. Ns is the number of sampling grid points of the airfoil flow field, and n is the number of aerodynamic parameters of the airfoil flow field.

[0029] The advantages of the present invention are: in the method of the present application, the trained point cloud network can quickly complete the prediction of the aerodynamic parameters of the airfoil flow field. Compared with the traditional CFD solution method, the calculation time is reduced from several minutes to 1 second, and the solution speed is increased by hundreds of times. Compared with the traditional convolutional neural network, this method does not involve data pixelation or data dimensionality reduction, retains the accuracy of the CFD calculation results, and avoids the introduction of artificial errors. The method of the present application provides a new idea for the rapid prediction of the aerodynamic parameters of the airfoil flow field.

[0030] Compared with a single point cloud network, the method of this application does not predict the aerodynamic parameters of all grid points in the flow field, but only models and predicts the aerodynamic parameters of the sampling grid points, and then combines the gappy POD method to calculate the aerodynamic parameters of non-sampling grid points based on the aerodynamic parameters of the sampling grid points. This reduces the data dimension of the point cloud network and improves the training speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Schematic diagram of the initial airfoil and the random airfoil library. The solid line is the initial airfoil, denoted as RAE2822, and the dark area around the initial airfoil is the coordinate range of the random airfoil library, denoted as airfoil library.

[0032] Figure 2 Schematic diagram of the process structure of the method of the present application, wherein MLP refers to a multi-layer perceptron, and the (64,128) digital label after MLP indicates that the size of the first fully connected layer in the multi-layer perceptron is 64, and the size of the second fully connected layer is 128, and the meanings of other digital labels are similar;

[0033] Figure 3 Comparison between the predicted value of the airfoil surface pressure coefficient distribution by the method described in this application and the CFD calculated value; Figure 3 (a) represents the CFD calculated value, Figure 3 (b) indicates the predicted value, Figure 3 (c) represents the prediction error. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0035] The features and illustrative embodiments of various aspects of the present invention will be described in detail below. In the detailed description below, many specific details are proposed to provide a comprehensive understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present invention by illustrating examples of the present invention. The present invention is by no means limited to any specific settings and methods proposed below, but covers any improvements, replacements and modifications of structures, methods, devices without departing from the spirit of the present invention. In the accompanying drawings and the following description, known structures and technologies are not shown to avoid unnecessary ambiguity in the present invention.

[0036] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other, and the embodiments can refer to and quote each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0037] Example 1

[0038] A method for predicting aerodynamic parameters of airfoil flow field based on point cloud network, comprising the following steps:

[0039] Step 1: Select the RAE2822 airfoil as the initial airfoil and establish a two-dimensional coordinate system Oxy, where the coordinate origin O is the leading edge of the airfoil, the Ox axis passes through the trailing edge of the airfoil and points to the trailing edge of the airfoil, and the Oy axis is perpendicular to the Ox axis and upward. Parameterize the initial airfoil based on the free deformation method. The parameterization method is: use the Latin hypercube sampling method to generate 250 sets of control point displacement values ​​of the free deformation method, deform the initial airfoil, and the set of 250 sets of new airfoils after deformation is used as the random airfoil library. The schematic diagram of the initial RAE2822 and the random airfoil library is shown in the figure. Figure 1 shown.

[0040] Step 2: Set the incoming flow Mach number to 0.45 and the angle of attack to 2°, and obtain the flow field pressure coefficient of each airfoil in the random airfoil library based on CFD simulation calculation.

[0041] Step 3: Sample 2048 grid points from the 28560 grid points in the airfoil flow field. The sampling rule is: if the i-th (1≤i≤28560) grid point is a grid point on the airfoil surface, it is sampled; otherwise, the probability of its being sampled is proportional to the distance D from the point to the airfoil surface. i Satisfies the normal distribution:

[0042]

[0043] in D i is the distance from the i-th (1≤i≤Nt) grid point to the airfoil surface.

[0044] Step 4: For all 250 airfoils in the airfoil library, the coordinates of the 28,560 sampling grid points of the airfoil flow field are used as input data, and the flow field aerodynamic parameters of the sampling grid points are used as output data to establish a neural network data set. The format of the input data is a 250×2048×2 matrix, that is, there are 250 input samples, each sample is a matrix of 2048 rows and 2 columns, and each row of data represents the horizontal and vertical coordinates of a sampling grid point. The format of the output data is a 250×2048×1 matrix, that is, there are 250 output samples, each sample is a matrix of 2048 rows and 1 column, and each row of data represents a flow field aerodynamic parameter corresponding to a sampling grid point: the flow field pressure coefficient.

[0045] Step 5: Establish a point cloud network. The structure of the point cloud network is as follows: Figure 2 As shown. The structure of the point cloud network is: 2048×2 input layer, 64×64 multilayer perceptron 1, 64×64 multilayer perceptron 2, 64×128 multilayer perceptron 3, maximum pooling layer 1, merging layer of multilayer perceptron 2 and maximum pooling layer 1, 128×128 multilayer perceptron 4, 128×1 multilayer perceptron 5, 2048×1 output layer. Then randomly select 80% of the data set in step 4 as the training data set, and the remaining 20% ​​of the data set as the test data set. The training data set is used to train the point cloud network, and the test data set is used to test and verify the prediction performance of the point cloud network. The point cloud network uses the mean square error MSE as the loss function, and MSE is defined as:

[0046]

[0047] Where n is the number of data samples; y i is the true value of the i-th data sample; is the predicted value of the ith data sample. The optimization training method is the Adam algorithm.

[0048] Step 6: Select an airfoil from the test data set, use the coordinates of 2048 sampling grid points of the airfoil flow field as input, input the trained point cloud network, and use the output value of the network as the predicted value of the pressure coefficient of the sampling grid points of the airfoil flow field.

[0049] Step 7: Create a matrix P with 0.8×250 rows and 2048×2+28560×1 columns. Each row of the matrix P represents the flow field data of an airfoil sample in the training set. The first to 2048×2 columns of the matrix P are the horizontal and vertical coordinates of the 2048 sampling grid points, the 2048×2+1 columns to the 2048×(2+1) columns are the airfoil flow field pressure coefficients corresponding to the 2048 sampling grid points, and the 2048×(2+1)+1 columns to the last column are the pressure coefficients corresponding to the non-sampling grid points of the airfoil flow field. Create a row vector Pr, where the data of Pr is the pressure coefficient of the airfoil flow field to be predicted, where the 1st to 2048×2 columns are the horizontal and vertical coordinates of the 2048 sampling grid points in the airfoil flow field to be predicted, the 2048×2+1 column to the 2048×(2+1) column are the flow field pressure coefficients of the 2048 sampling grid points in the airfoil flow field to be predicted, which are obtained by prediction in step 6, and the 2048×(2+1)+1 column to the last column are the pressure coefficients corresponding to the non-sampling grid points in the airfoil flow field to be predicted, which are obtained by calculating the average value of the corresponding columns of the matrix P.

[0050] Step 8: Use the eigenorthogonal decomposition method to calculate the base mode matrix Pc, base mode coefficient Ps and mean vector Pm of the combined matrix [P; Pr]. The base mode matrix Pc, base mode coefficient Ps and mean vector Pm satisfy [P; Pr] = Ps × Pc T +Pm, specifically:

[0051]

[0052] Among them, P i，j is the element in the i-th row and j-th column of the matrix P; Ps i，j is the element in the i-th row and j-th column of the matrix Ps; Pc i，j is the element in the i-th row and j-th column of the matrix Pc; Pm i is the i-th element of the mean vector Pm;

[0053] Step 9: Update Pr using formula (2):

[0054]

[0055] Where b is the principal component coefficient vector.

[0056] Step 10: Repeat step 9 to iteratively update Pr until the maximum iteration step is 50. The 2048×2+1 column to the last column of Pr is the pressure coefficient corresponding to the grid point of the flow field of the airfoil to be predicted. The prediction of the airfoil flow field pressure coefficient is completed.

[0057] Example 2

[0058] A method for predicting aerodynamic parameters of airfoil flow field based on point cloud network. The steps include:

[0059] Step 1: Select the RAE2822 airfoil as the initial airfoil and establish a two-dimensional coordinate system Oxy, where the coordinate origin O is the leading edge of the airfoil, the Ox axis passes through the trailing edge of the airfoil and points to the trailing edge of the airfoil, and the Oy axis is perpendicular to the Ox axis and upward. Parameterize the initial airfoil based on the free deformation method. The parameterization method is: use the Latin hypercube sampling method to generate 250 sets of control point displacement values ​​of the free deformation method, deform the initial airfoil, and the set of 250 sets of new airfoils after deformation is used as the random airfoil library. The schematic diagram of the initial RAE2822 and the random airfoil library is shown in the figure. Figure 1 shown.

[0060] Step 2: Set the incoming flow Mach number to 0.5 and the angle of attack to 1.74°, and obtain the flow field pressure coefficient of each airfoil in the random airfoil library based on CFD simulation calculation.

[0061] Step 3: Sample 2048 grid points from the 28560 grid points in the airfoil flow field. The sampling rule is: if the i-th (1≤i≤28560) grid point is a grid point on the airfoil surface, it is sampled; otherwise, the probability of its being sampled is proportional to the distance D from the point to the airfoil surface. i Satisfies the normal distribution:

[0062]

[0063] in D i is the distance from the i-th (1≤i≤Nt) grid point to the airfoil surface.

[0064] Step 4: For all 250 airfoils in the airfoil library, the coordinates of the 28,560 sampling grid points of the airfoil flow field are used as input data, and the flow field aerodynamic parameters of the sampling grid points are used as output data to establish a neural network data set. The format of the input data is a 250×2048×2 matrix, that is, there are 250 input samples, each sample is a matrix of 2048 rows and 2 columns, and each row of data represents the horizontal and vertical coordinates of a sampling grid point. The format of the output data is a 250×2048×1 matrix, that is, there are 250 output samples, each sample is a matrix of 2048 rows and 1 column, and each row of data represents a flow field aerodynamic parameter corresponding to a sampling grid point: the flow field pressure coefficient.

[0065] Step 5: Establish a point cloud network. The structure of the point cloud network is as follows: Figure 2 As shown. The structure of the point cloud network is: 2048×2 input layer, 64×64 multilayer perceptron 1, 64×64 multilayer perceptron 2, 64×128 multilayer perceptron 3, maximum pooling layer 1, merging layer of multilayer perceptron 2 and maximum pooling layer 1, 128×128 multilayer perceptron 4, 128×1 multilayer perceptron 5, 2048×1 output layer. Then randomly select 80% of the data set in step 4 as the training data set, and the remaining 20% ​​of the data set as the test data set. The training data set is used to train the point cloud network, and the test data set is used to test and verify the prediction performance of the point cloud network. The point cloud network uses the mean square error MSE as the loss function, and MSE is defined as:

[0066]

[0067] Where n is the number of data samples; y i is the true value of the i-th data sample; is the predicted value of the ith data sample. The optimization training method is the Adam algorithm.

[0068] Step 6: Select an airfoil from the test data set, use the coordinates of 2048 sampling grid points of the airfoil flow field as input, input the trained point cloud network, and use the output value of the network as the predicted value of the pressure coefficient of the sampling grid points of the airfoil flow field.

[0069] Step 7: Create a matrix P with 0.8×250 rows and 2048×2+28560×1 columns. Each row of the matrix P represents the flow field data of an airfoil sample in the training set. The first to 2048×2 columns of the matrix P are the horizontal and vertical coordinates of the 2048 sampling grid points, the 2048×2+1 columns to the 2048×(2+1) columns are the airfoil flow field pressure coefficients corresponding to the 2048 sampling grid points, and the 2048×(2+1)+1 columns to the last column are the pressure coefficients corresponding to the non-sampling grid points of the airfoil flow field. Create a row vector Pr, where the data of Pr is the pressure coefficient of the airfoil flow field to be predicted, where the 1st to 2048×2 columns are the horizontal and vertical coordinates of the 2048 sampling grid points in the airfoil flow field to be predicted, the 2048×2+1 column to the 2048×(2+1) column are the flow field pressure coefficients of the 2048 sampling grid points in the airfoil flow field to be predicted, which are obtained by prediction in step 6, and the 2048×(2+1)+1 column to the last column are the pressure coefficients corresponding to the non-sampling grid points in the airfoil flow field to be predicted, which are obtained by calculating the average value of the corresponding columns of the matrix P.

[0070] Step 8: Use the eigenorthogonal decomposition method to calculate the base mode matrix Pc, base mode coefficient Ps and mean vector Pm of the combined matrix [P; Pr]. The base mode matrix Pc, base mode coefficient Ps and mean vector Pm satisfy [P; Pr] = Ps × Pc T +Pm, specifically:

[0071]

[0072] Among them, P i，j is the element in the i-th row and j-th column of the matrix P; Ps i，j is the element in the i-th row and j-th column of the matrix Ps; Pc i，j is the element in the i-th row and j-th column of the matrix Pc; Pm i is the i-th element of the mean vector Pm;

[0073] Step 9: Update Pr using formula (2):

[0074]

[0075] Where b is the principal component coefficient vector.

[0076] Step 10: Repeat step 9 to iteratively update Pr until the maximum iteration step is 50. The 2048×2+1 column to the last column of Pr is the pressure coefficient corresponding to the grid point of the flow field of the airfoil to be predicted. The prediction of the airfoil flow field pressure coefficient is completed.

[0077] The above describes the specific embodiments of the present invention. It should be noted that, for those skilled in the art, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also fall within the scope of protection of the present invention.

Claims

1. A method for predicting aerodynamic parameters of airfoil flow field based on point cloud network, characterized in that: The specific steps include: Step 1: Select an initial airfoil, combine the airfoil parameterization method and the random sampling method, and generate N random airfoils based on the initial airfoil to form an airfoil library; Step 2: Set the flow field conditions, and obtain the flow field aerodynamic parameters of each airfoil in the airfoil library based on CFD simulation calculations; Step 3: Sample Ns grid points from the Nt grid points in the flow field of each airfoil; Step 4: For all airfoils in the airfoil library, use the coordinates of the Ns sampling grid points of the airfoil flow field as input data, and the flow field aerodynamic parameters of the sampling grid points as output data to establish a neural network data set; Step 5: Establish a point cloud network, and use the neural network data set to train the point cloud network; Step 6: Select an airfoil from the test data set, use the coordinates of the sampling grid points of the airfoil flow field as input, input the trained point cloud network, and use the output value of the network as the predicted value of the aerodynamic parameters of the sampling grid points of the airfoil flow field.

2. The method for predicting aerodynamic parameters of airfoil flow field based on point cloud network according to claim 1, characterized in that: In step 1, airfoil parameterization methods include free deformation method and quasi-function shape function method.

3. The method for predicting aerodynamic parameters of airfoil flow field based on point cloud network according to claim 1, characterized in that: In step 1, the method of generating a random airfoil library based on the initial airfoil by combining the airfoil parameterization method and the random sampling method is to parameterize the initial airfoil by applying the airfoil parameterization method, and then select N groups of geometric parameters by random sampling in the parameter space, deform the initial airfoil according to the geometric parameters, and the set of N new airfoils after deformation is used as the airfoil library.

4. The method for predicting aerodynamic parameters of airfoil flow field based on point cloud network according to claim 1, characterized in that: In step 2, the sampling rule is: if the i-th grid point is a grid point on the airfoil surface, 1≤i≤Nt, then it is sampled; otherwise, the probability of its being sampled is proportional to the distance D from the point to the airfoil surface. i Satisfies normal distribution: in D i is the distance from the i-th grid point to the airfoil surface, 1≤i≤Nt.

5. The method for predicting aerodynamic parameters of airfoil flow field based on point cloud network according to claim 1, characterized in that: In step 5, 80% of the data set in step 4 is randomly selected as a training data set, and the remaining 20% ​​of the data set is used as a test data set. The training data set is used to train the point cloud network, and the test data set is used to test and verify the prediction performance of the point cloud network.

6. The method for predicting aerodynamic parameters of airfoil flow field based on point cloud network according to claim 1, characterized in that: Step 7: Create a matrix P with 0.8×N rows and Ns×2+Nt×n columns. Each row of the matrix P represents the flow field data of an airfoil sample in the training set. Step 8: Use the intrinsic orthogonal decomposition method to calculate the base modal matrix Pc, base modal coefficient Ps and mean vector Pm of the combined matrix [P; Pr]. The base modal matrix Pc, base modal coefficient Ps and mean vector Pm satisfy [P; Pr] = Ps×Pc T +Pm; Step 9: Update Pr using formula (2): Where b is the principal component coefficient vector; Step 10: Repeat step 9 to iteratively update Pr until the maximum number of iterations E is reached. The Ns×2+1 columns to the last column of Pr are the aerodynamic parameters corresponding to the flow field grid points of the airfoil to be predicted.

7. The method for predicting aerodynamic parameters of airfoil flow field based on point cloud network according to claim 1, characterized in that: The aerodynamic parameters of the airfoil flow field include aerodynamic pressure coefficient, Mach number, speed or other aerodynamic parameters of the airfoil flow field that can be obtained through CFD simulation calculations.

8. The method for predicting aerodynamic parameters of airfoil flow field based on point cloud network according to claim 1, characterized in that: In step 4, the format of the input data is a matrix of N×Ns×2, that is, there are N input samples in total, each sample is a matrix of Ns rows and 2 columns, and each row of data represents the horizontal and vertical coordinates of a sampling grid point; the format of the output data is a matrix of N×Ns×n, that is, there are N output samples in total, each sample is a matrix of Ns rows and n columns, and each row of data represents n flow field aerodynamic parameters corresponding to a sampling grid point.

9. The method for predicting aerodynamic parameters of airfoil flow field based on point cloud network according to claim 1, characterized in that: In step 5, the structure of the point cloud network is: Ns×2 input layer, 64×64 multilayer perceptron 1, 64×64 multilayer perceptron 2, 64×128 multilayer perceptron 3, maximum pooling layer 1, merging layer of multilayer perceptron 2 and maximum pooling layer 1, 128×128 multilayer perceptron 4, 128×n multilayer perceptron 5, N×n output layer; where Ns is the number of sampling grid points of the airfoil flow field, and n is the number of aerodynamic parameters of the airfoil flow field.

Images