A method for predicting two-dimensional distribution of film cooling effectiveness based on prior knowledge embedding

Abstract

The application provides a generative gas film cooling efficiency two-dimensional distribution prediction method based on prior knowledge embedding, which learns the mapping relationship between input parameters and coefficients through multiple layers of neurons, then combines function equations to increase dimensions, and reduces the learning explosion caused by the increase of information quantity due to the increase of dimensions; and the information extracted by the feature extraction module is fused with the deep information extracted by the image encoding module, then a multi-head attention mechanism is combined to improve the prediction accuracy of the model for different input parameters. The application realizes the dimension increase in the generative learning process through the prior knowledge embedding mode, thereby reducing the large learning quantity and model complexity required by the model in the dimension increase learning process, and these learning quantities need to be supplemented by a large number of training samples in the traditional model. Therefore, the method of the application makes the model training more difficult, makes the model capture more information under the condition of limited sample quantity, and improves the training accuracy and training speed.

Description

Technical Field

[0001] The present invention relates to the field of film cooling and artificial intelligence technology, and in particular to a generative film cooling efficiency two-dimensional distribution prediction method based on prior knowledge embedding. Background Art

[0002] Film cooling is a key technology widely used in high-temperature components of aircraft engines, playing an important role in thermal protection of combustion chamber walls, turbine blades, and nozzles. In the design of film cooling structures, film cooling efficiency and distribution uniformity are extremely important performance indicators. If the film cooling efficiency is low and the cooling is uneven, it will bring many adverse effects:

[0003] 1: Increase the amount of cooling gas used;

[0004] 2. Uneven film cooling can cause uneven temperature distribution in the combustion chamber, leading to increased combustion instability and the generation of more unburned hydrocarbons (UHC) and carbon monoxide (CO), as well as other pollutants such as NOx.

[0005] 3: Long-term local uneven cooling will accelerate the wear of engine blades and hot end components, resulting in more frequent blade replacement and maintenance, and in severe cases even leading to component ablation and engine failure.

[0006] Therefore, accurate prediction of film cooling efficiency can help designers quickly iterate and optimize cooling solutions, guide the layout and improvement of cooling structures in the early stages of design, thereby improving the cooling methods of high-temperature components and increasing the iterative efficiency and speed of cooling design.

[0007] Currently, the methods for predicting two-dimensional film cooling based on one-dimensional vector parameters mainly include empirical fitting methods and artificial intelligence methods.

[0008] The empirical fitting method uses empirical correlations to fit the effectiveness and standard deviation on the centerline, thereby obtaining a two-dimensional distribution of effectiveness. However, the analytical two-dimensional distribution equation for effectiveness introduces numerous model parameters and custom intermediate variables, resulting in a high model complexity. Furthermore, both parameters and intermediate variables require calibration using experimental data, making implementation difficult and significantly limiting the method's applicability.

[0009] The artificial intelligence method is to make predictions through artificial neural network models. For example, the Chinese patent application with publication number CN112084597A, "AI prediction method for two-dimensional distribution of single-hole air film cooling efficiency based on bell curve", proposed a two-dimensional cooling efficiency prediction method based on bell curve. By building a fully connected neural network to learn the mapping relationship between the input parameters and the bell curve coefficients, this method ignores the spatial correlation of adjacent pixels in the air film distribution, and the method strictly requires that the distribution of cooling efficiency is in the form of a bell curve, which obviously limits the scope of application of this method. For example, the document "Research on prediction of single-hole adiabatic air film cooling efficiency distribution based on machine learning [J]. Advancement Technology, 2022, 43 (4): 218-228" uses the characteristics of neural network convolution to introduce the spatial correlation between vector pixels, calculates a large amount of sample data as the network input data set through CFD, and directly builds multiple fully connected layers to expand the length of the vector, and then reconstructs it into a two-dimensional matrix for multiple upsampling, and restores the scale through convolution. This achieves mapping training and learning between low dimensions and high dimensions. However, the research conclusions of the above two methods show that the network has a strong dependence on the quality and quantity of sample data. The sample data volumes of the above two methods are 1,300 and 3,900 respectively, and both are CFD calculation samples.

[0010] This raises two key issues: First, many scholars have shown that the accuracy of CFD calculations of film cooling efficiency is questionable, which challenges the quality of the dataset. Second, current two-dimensional generative learning requires a large amount of training and a sufficient number of samples to achieve accurate predictions. To ensure dataset quality, acquiring thousands of samples through experimental testing is extremely costly.

[0011] Therefore, how to accurately predict the two-dimensional distribution of film cooling efficiency under the condition of a small number of test samples is a key problem that the present invention needs to solve. Summary of the Invention

[0012] Technical issues solved:

[0013] The film cooling efficiency prediction model established by the existing method mainly directly constructs the mapping relationship between one-dimensional input and two-dimensional output through multiple convolutions. Since a large amount of learning is required to generate the two-dimensional film cooling efficiency distribution, this type of model has a great demand for the amount of training data. However, the increase in the amount of data will undoubtedly increase the cost of data acquisition. Moreover, the quality of the film cooling efficiency data obtained by the CFD method is questionable, resulting in it being obtained only through experimental testing, which undoubtedly greatly increases the cost and difficulty of data acquisition. In addition, the existing network model has many convolution kernels and neurons, and many weight parameters need to be trained, which also requires more training time and parameter adjustment cycle.

[0014] Technical solution:

[0015] In order to solve the problems existing in the existing technology, the present invention proposes a generative two-dimensional distribution prediction method for film cooling efficiency based on prior knowledge embedding, establishes a neural network model combined with prior knowledge embedding, and realizes the dimensionality upgrade process from one dimension to two dimensions in the model through prior knowledge, thereby reducing the amount of training required for the model; and improves the prediction accuracy and generalization ability of the model through image encoding and decoding modules, thereby achieving the purpose of accurately predicting the two-dimensional film cooling efficiency distribution from one-dimensional parameter input even with a small sample size.

[0016] The technical solution of the present invention is:

[0017] The method for predicting the two-dimensional distribution of film cooling efficiency based on prior knowledge embedding includes the following steps:

[0018] Step 1: Obtain training samples through experimental testing:

[0019] Each set of sample data includes a film hole parameter vector composed of structural parameters and aerodynamic parameters and a two-dimensional distribution image of film cooling efficiency obtained through experimental testing of the corresponding film hole;

[0020] Step 2: Build the network model:

[0021] The network model input is the film hole parameter vector, and the output is the two-dimensional distribution image of film cooling efficiency;

[0022] The network model includes a priori knowledge embedding module, a feature extraction module, an image encoding and decoding module, and an output module:

[0023] The input of the prior knowledge embedding module is the air film hole parameter vector, which is mapped to the coefficients A1, A2, A3, B1, B2, and B3; according to the coefficients A1, A2, A3, B1, B2, and B3, the function

[0024] F(x)=A1·log(A2·x)+A3

[0025] Q(x)=B1·(1-x) 2 +B2·2·x·(1-x)+B3·x 2

[0026]

[0027] First, the coefficients F(x) and Q(x) are calculated, and then the dimensionality is increased to restore the two-dimensional distribution cloud map of the film cooling efficiency. x represents the coordinate of each point in the two-dimensional distribution image of the film cooling efficiency along the flow direction, y represents the coordinate of each point in the two-dimensional distribution image of the film cooling efficiency along the span direction, and G(y, F(x), Q(x)) is the cooling efficiency value at the (x, y) coordinate point. After passing through the convolution block, a matrix with the same size as the model output image is obtained as the input of the image encoding module.

[0028] The feature extraction module performs high-dimensional extraction on the input air film hole parameter vector, reconstructs the extracted high-dimensional features into a two-dimensional matrix, and then expands the channels through the convolution block to obtain a matrix with the same size as the image encoding output;

[0029] The image encoding and decoding module downsamples the matrix output by the prior knowledge embedding module N times to obtain information at different levels of the image. The features extracted at each layer are then spliced ​​with the information at the same level after decoding. The matrix obtained after the final downsampling is fused with the high-dimensional features extracted by the feature extraction module and flattened into a one-dimensional vector. The matrix is ​​then reconstructed through a multi-head attention mechanism to capture key information in the image and vector features. After obtaining the key information, it is gradually upsampled to restore it to the size of the image and provided to the output module to obtain the final two-dimensional distribution output of the film cooling efficiency.

[0030] Step 3: Use the sample data obtained in step 1 to train the network model constructed in step 2 to obtain a trained prediction model;

[0031] Step 4: For the film cooling scheme to be predicted, obtain the film hole parameter vector, input the trained prediction model obtained in step 3, and obtain the two-dimensional distribution prediction result of the film cooling efficiency.

[0032] In a further preferred embodiment, the structural parameters include: the hole inclination angle α between the axis of the air film hole and the outlet plane; the expansion angle β of the air film hole expansion section; the forward inclination angle γ of the air film hole expansion section; the inner fold angle δ of the V-shaped hole outlet; the aspect ratio L / d of the air film hole; the ratio of the width to the diameter of the air film hole outlet L w / d; Air film hole inlet and outlet area ratio A out / A in ; The dimensionless pore spacing P / d of the air film pores.

[0033] In a further preferred embodiment, the aerodynamic parameters include: blowing ratio M; density ratio DR; turbulence Tu and acceleration factor k.

[0034] In a further preferred embodiment, the test is carried out by the following test system:

[0035] The main flow is generated by the blower, passes through the butterfly valve and the rectifier cavity to produce a uniform inlet flow at the inlet of the test section; then, different turbulence intensities are generated at the outlet of the film hole through turbulence grids of different densities to simulate the effects of different turbulence intensities on cooling; a suction slit is set at the front end of the measuring plate to allow the boundary layer to redevelop there, so that the boundary layer thickness at the film hole outlet is close to the local boundary layer thickness of the actual engine film hole; at the same time, a trip wire is set after the slit to promote the transition of the boundary layer from laminar flow to turbulent flow; velocity, temperature and pressure measurement points are set above the film hole outlet to monitor the test in real time The coolant system uses a high-pressure gas tank and compressor as the gas source. After passing through valves and pressure regulating devices, it is heated by a heater so that the temperature difference between the primary and secondary flows does not exceed 0.2K. After the flow is controlled by an Alicat mass flowmeter, it enters the cold air cavity and is supplied to the air film holes. Temperature and pressure measurement points are set in the cold air cavity to monitor the coolant status in real time. PSP pressure-sensitive paint is sprayed on the surface of the measuring plate, and images during the test are collected with the corresponding excitation light source and a scientific-grade CCD camera with a filter. After post-processing of the images, a two-dimensional distribution image of the air film cooling efficiency is obtained.

[0036] Beneficial effects

[0037] The generative two-dimensional distribution prediction method for film cooling efficiency based on prior knowledge embedding proposed in this invention has the following advantages:

[0038] 1. Use multiple layers of neurons to learn the mapping relationship between input parameters and coefficients, and then combine them with a system of functional equations to increase the dimension, thereby reducing the learning volume explosion caused by the increase in the amount of information due to the increase in dimension;

[0039] 2. The information extracted by the feature extraction module is integrated with the deep information extracted by the image encoding module, and then combined with the multi-head attention mechanism to improve the model's prediction accuracy for different input parameters.

[0040] In summary, the present invention achieves dimensionality increase during generative learning by embedding prior knowledge, thereby reducing the large amount of learning and model complexity required during dimensionality increase. These learning requirements require a large number of training samples in traditional models. Therefore, the method of the present invention makes model training easier, allowing the model to capture more information with a limited number of samples, thereby improving training accuracy and speed.

[0041] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0043] Figure 1 : training sample measurement system;

[0044] Figure 2 : Schematic diagram of air film pore structure parameters;

[0045] Figure 3 : Generative neural network model based on prior knowledge embedding;

[0046] Figure 4 : Schematic diagram of the prior knowledge embedding module;

[0047] Figure 5 : Schematic diagram of feature extraction module;

[0048] Figure 6 : Schematic diagram of image encoding and decoding module;

[0049] Figure 7 : Loss convergence curve during training;

[0050] Figure 8 : The physical structure of 4 test samples;

[0051] Figure 9 : Comparison of the actual results, the prediction results of the traditional model and the prediction results of the method of the present invention; (a) the actual result of sample 1, (a1) the prediction result of the traditional model for sample 1, (a2) the prediction result of the method of the present invention for sample 1; (b) the actual result of sample 2, (b1) the prediction result of the traditional model for sample 2, (b2) the prediction result of the method of the present invention for sample 2; (c) the actual result of sample 3, (c1) the prediction result of the traditional model for sample 3, (c2) the prediction result of the method of the present invention for sample 3; (d) the actual result of sample 4, (d1) the prediction result of the traditional model for sample 4, (d2) the prediction result of the method of the present invention for sample 4;

[0052] Figure 10: The difference between the prediction results of the traditional model and the prediction results proposed by the present invention and the real cloud map; (a1) the prediction result of the traditional model for sample 1, (a2) the prediction result of the method of the present invention for sample 1; (b1) the prediction result of the traditional model for sample 2, (b2) the prediction result of the method of the present invention for sample 2; (c1) the prediction result of the traditional model for sample 3, (c2) the prediction result of the method of the present invention for sample 3; (d1) the prediction result of the traditional model for sample 4, (d2) the prediction result of the method of the present invention for sample 4;

[0053] Figure 11 : Prediction errors of the traditional model and the prediction model proposed in this invention on four typical samples. DETAILED DESCRIPTION

[0054] The following describes in detail embodiments of the present invention. The embodiments are exemplary and intended to explain the present invention, but are not to be construed as limiting the present invention.

[0055] In this embodiment, a film cooling efficiency prediction model established by existing methods requires a large amount of training data, which greatly increases the data acquisition cost. In addition, the model has many weight parameters, and the model training time and parameter adjustment cycle are long. A generative film cooling efficiency two-dimensional distribution prediction method based on prior knowledge embedding is proposed. Prior knowledge is used to realize the dimensionality upgrade process from one dimension to two dimensions in the model, reducing the required training amount of the model. The image encoding and decoding modules are used to improve the prediction accuracy and generalization ability of the model, so as to achieve the purpose of accurately predicting the two-dimensional film cooling efficiency distribution from one-dimensional parameter input even when the sample size is small.

[0056] Specifically include the following:

[0057] (1) Obtain training samples:

[0058] In order to obtain high-quality test data, the training samples used in this embodiment are all experimental test results. Figure 1 The experimental system used for sample acquisition in this embodiment is demonstrated.

[0059] Depend on Figure 1As can be seen, the main flow originates from the blower and passes through a butterfly valve and a rectifier chamber to create a uniform inlet flow at the test section inlet. Turbulence grids of varying densities then generate varying turbulence at the film hole outlet to simulate the effects of varying turbulence on cooling. A suction slit is placed at the front of the measurement plate (the test surface in the figure) to allow the boundary layer to redevelop there, ensuring that the boundary layer thickness at the film hole outlet approximates the local boundary layer thickness of an actual engine film hole. A trip wire is also placed after the slit to promote the transition from laminar to turbulent flow, as film hole outlets in actual engines often reside in a turbulent boundary layer. Velocity, temperature, and pressure measurement points are located above the film hole outlet to monitor the main flow conditions in real time during testing. The coolant system is sourced from a high-pressure gas tank and compressor. After passing through a valve and pressure regulator, it is heated by a heater to ensure the temperature difference between the main and secondary flows does not exceed 0.2K. The flow rate is then controlled by an Alicat mass flowmeter before entering the cooling chamber and being supplied to the film hole. Temperature and pressure measurement points are located in the cooling chamber to monitor the coolant conditions in real time. The surface of the measuring plate is sprayed with PSP pressure-sensitive paint, and equipped with a corresponding excitation light source and a scientific-grade CCD camera with a filter to collect images during the test. After post-processing of the image, a two-dimensional distribution image of the film cooling efficiency is obtained.

[0060] Figure 2 The air film hole structure studied in this embodiment is shown. The air film hole structure parameters mainly include the hole inclination angle α between the air film hole axis and the outlet plane, the expansion angle β of the air film hole expansion section, the forward inclination angle γ of the air film hole expansion section, the inflection angle δ of the V-shaped hole outlet; the length L of the air film hole, the length L1 of the expansion section, the width L of the air film hole outlet w ; Air film hole entrance area A in , outlet area A out ; and the hole spacing P between two adjacent air film holes.

[0061] The length parameter is dimensionless using the pore diameter d, and the inlet and outlet are dimensionless using the area ratio. Moreover, the inlet and outlet area ratio and the expansion length L1 are not independent of each other. Only one of the two needs to be determined to determine the entire hole structure. In this embodiment, the area ratio is used as one of the input items. Therefore, a total of 8 structural parameters are used as input:

[0062] The inclination angle α between the axis of the air film hole and the outlet plane:

[0063] The expansion angle β of the film hole expansion section. The hole type studied in this embodiment is a bilaterally symmetrical expansion.

[0064] The forward tilt angle γ of the film hole expansion section;

[0065] The inner angle δ of the V-shaped hole outlet, where δ = 0 for the fan-shaped hole;

[0066] The aspect ratio of the air film hole L / d;

[0067] Ratio of film hole outlet width to diameter L w / d;

[0068] Air film hole inlet and outlet area ratio A out / A in ;

[0069] The dimensionless pore spacing P / d of the film pores.

[0070] Table 1 shows the input structure parameter ranges of the sample data in this embodiment:

[0071] Table 1 Structural parameter value range

[0072] Tilt angle α(°) 25，30，40 Expansion angle β(°) 0～15 Forward tilt angle γ(°) 0～15 Inward folding angle δ(°) 0～30 Aspect ratio L / d 4～7 <![CDATA[出口宽度与直径比L w / d]]> 1～3.5 <![CDATA[面积比A out / A in ]]> 1.0～8.6 Hole spacing P / d 3～8

[0073] Table 2 shows the aerodynamic parameter range of the sample group. This example selects the aerodynamic parameters commonly used in studying the film cooling efficiency: blowing ratio M = (ρ c u c ) / (ρ m u m ), density ratio DR = ρ c / ρ m , turbulence Tu=u ′ / u m And the acceleration factor k = ν × (du / dx) / u, which measures the pressure gradient 2 As input. Where ρ represents the gas density, u represents the airflow velocity, and the subscripts c and m represent the secondary flow and the main flow, respectively. ′ represents the pulsating velocity of the mainstream, v represents the dynamic viscosity coefficient of the gas, and x represents the coordinates of each point along the flow direction in the two-dimensional distribution image of film cooling efficiency.

[0074] Table 2 Value ranges of aerodynamic parameters

[0075] <![CDATA[主流速度u m (m / s)]]> 30 <![CDATA[主次流温度T m T c (K)]]> 298±0. Blowing ratio M 0.5～3 Density ratio DR 1，1.5 Turbulence Tu 2，10 Acceleration factor k <![CDATA[-2.0×10 -6 ~2.0×10 -6 ]]>

[0076] In the final training samples, each set of sample data includes a film hole parameter vector consisting of 8 structural parameters and 4 aerodynamic parameters and a two-dimensional distribution image of the film cooling efficiency obtained through experimental testing of the corresponding film hole.

[0077] (2) Building a network model

[0078] Figure 3 The generative network model architecture constructed in this embodiment is demonstrated, which takes a one-dimensional vector consisting of 8 structural parameters constituting the air film hole and 4 main aerodynamic parameters as input and a two-dimensional distribution of the air film cooling efficiency as output.

[0079] The network model primarily consists of a prior knowledge embedding module, a feature extraction module, an image encoding and decoding module, and an output module. The model framework consists of three main branches. The first branch, consisting of the prior knowledge embedding module and the feature extraction module, is used to capture and upscale the input information. The second branch is the image encoding module, which leverages the spatial recognition properties of convolution to improve the ability to predict the overall distribution and local correlation of film cooling efficiency. The third branch is the decoding module, which gradually decodes the fused deep information and restores it to its original two-dimensional size.

[0080] The following is a detailed introduction to the functions and construction framework of each module.

[0081] Prior knowledge embedding module:

[0082] like Figure 4 As shown in the figure, the prior knowledge embedding module consists of a coefficient mapping module and a learning function module. The coefficient mapping module includes an extraction layer, a hidden layer, and an output layer. The coefficient mapping module first extracts the features of the one-dimensional vector consisting of 12 elements of the input through an extraction layer composed of 16 neurons. Then, after passing through two hidden layers composed of 32 neurons each, it is mapped to the coefficients A1, A2, A3, B1, B2, and B3 through an output layer composed of 16 neurons. In the learning function module, the learned six coefficients A1, A2, A3, B1, B2, and B3 are combined with the attenuation function (Formula 1F(x)) and the diffusion function equation (Formula 2Q(x)) to obtain the coefficients F(x) and Q(x) of the Gaussian function (Formula 3G(y, F(x), Q(x))), where x and y represent the coordinate values ​​of each point in the two-dimensional distribution image of film cooling efficiency along the streamwise and spanwise directions, respectively:

[0083] F(x)=A1·log(A2·x)+A3 (1)

[0084] Q(x)=B1·(1-x) 2 +B2·2·x·(1-x)+B3·x 2 (2)

[0085]

[0086] The cooling efficiency value G(y, F(x), Q(x)) at each (x, y) coordinate point is calculated by formula 3, where x represents the coordinate of each point in the two-dimensional distribution image of film cooling efficiency along the flow direction, and y represents the coordinate of each point in the two-dimensional distribution image of film cooling efficiency along the span direction. The learned six coefficients A1, A2, A3, B1, B2, and B3 (one-dimensional) are upgraded to restore the two-dimensional distribution cloud map of film cooling efficiency. The two-dimensional distribution cloud map of film cooling efficiency is a 512×512 matrix, which serves as the input of the image encoding module.

[0087] The purpose of the prior knowledge embedding module is to learn the mapping relationship between the input vector and the coefficient vector, and to construct a two-dimensional distribution cloud map of film cooling efficiency in combination with the prior knowledge.

[0088] Feature extraction module:

[0089] like Figure 5 As shown in the figure, the feature extraction module first extracts the high-dimensional features of the one-dimensional vector composed of 12 elements of the input by a high-dimensional extraction layer with multiple neurons, and then reconstructs the extracted high-dimensional features into the form of a two-dimensional matrix and expands the channels through a 3×3 convolution block to obtain a matrix with the same size as the image encoding output.

[0090] Specifically, the input one-dimensional vector passes through the high-dimensional extraction layer consisting of 512 neurons, 1024 neurons and 4096 neurons in sequence, and the high-dimensional features are reconstructed into a 64×64 matrix. After the channel is expanded through a 3×3 convolution block, the expanded 64×64 matrix is ​​obtained, which is consistent with the image encoding output.

[0091] The purpose of the feature extraction module is to extract the high-dimensional features of the one-dimensional vector composed of 12 elements of the input, and to fuse them into the high-dimensional hierarchical information after image encoding.

[0092] Image encoding and decoding module:

[0093] Figure 6The construction logic of the image encoding and decoding module is demonstrated. The 512×512 matrix corresponding to the two-dimensional distribution image of the film cooling efficiency output by the prior knowledge embedding module is first processed by two 3×3 convolution blocks to obtain a first matrix of 512×512. After one downsampling and two 3×3 convolution blocks, the first matrix is ​​obtained to obtain a second matrix of 256×256. After a second downsampling and two 3×3 convolution blocks, a third matrix of 128×128 is obtained. After a third downsampling and two 3×3 convolution blocks, a fourth matrix of 64×64 is obtained. The 64×64 fourth matrix is ​​fused with the high-dimensional features extracted by the feature extraction module and flattened into a one-dimensional vector. It is then reconstructed into a 64×64 matrix through the multi-head attention mechanism. 4; the reconstructed 64×64 fifth matrix passes through a 3×3 convolution block and is concatenated with the 64×64 fourth matrix. After passing through two 3×3 convolution blocks and the first deconvolution, a 128×128 sixth matrix is ​​obtained. The sixth matrix is ​​concatenated with the 128×128 third matrix. After passing through two 3×3 convolution blocks and the second deconvolution, a 256×256 seventh matrix is ​​obtained. The seventh matrix is ​​concatenated with the 256×256 second matrix. After passing through two 3×3 convolution blocks and the third deconvolution, an 512×512 eighth matrix is ​​obtained. The eighth matrix is ​​concatenated with the 512×512 first matrix and output to the output module.

[0094] After obtaining the matrix corresponding to the two-dimensional distribution image of film cooling efficiency output by the prior knowledge embedding module, the encoding module performs three downsampling operations to obtain information at different levels of the image. The features extracted at each level are then concatenated with the decoded information at the same level. This embodiment integrates the high-dimensional features extracted by the feature extraction module with the deep-level information obtained after the third downsampling. A Transformer model with a multi-head attention mechanism is then used to capture key information from the image and vector features. Once this key information is obtained, it is gradually upsampled to restore it to the image size and provided to the output module to produce the final two-dimensional distribution output of film cooling efficiency.

[0095] The purpose of the image encoding and decoding module is to make up for the defect of inaccurate prediction of images in the overall space and local range obtained only through prior knowledge.

[0096] Output Module:

[0097] The output module passes the output matrix of the image encoding and decoding module through two 3×3 convolution blocks to obtain the final two-dimensional distribution output of the film cooling efficiency.

[0098] (3) Model training

[0099] Because obtaining test data through experiments is very expensive, in this example, a total of 200 sets of sample data were obtained, and this sample group was divided into a training set, a validation set, and a test set in a ratio of 7:2:1. Training was performed on an Nvidia RTX4080Super GPU with 16GB of video memory and an Intel Core i7 14700K CPU with 64GB of memory. The programming language was Python, and the optimizer was Adam. The learning rate decayed according to the natural exponential law with the cumulative number of iterations (Formula 4). The loss function was a weighted combination of MSE loss and L1 loss (Formula 5). Figure 7 The loss curve for this model's training process is shown. The loss on the validation set is comparable to the loss on the training set, and stabilizes after 400 rounds of training. The total training time is approximately 928 minutes, and the prediction time for the test sample is approximately 0.94 milliseconds.

[0100]

[0101]

[0102] where a k is the learning rate of the kth step, ψ is the decay coefficient, which is set to 0.05. c is the time scale coefficient for controlling decay, which is set to 100. i is the true value, is the predicted value, and n is the number of matrix elements.

[0103] (4) Training and testing results

[0104] For the convenience of presentation, the prediction results of four representative test samples are shown here. Table 3 shows the input structural parameters and aerodynamic parameters of these four samples. Figure 8 The physical structures of the four test samples are shown.

[0105] Table 3 Test sample parameters

[0106]

[0107]

[0108] Figure 9 and Figure 10 The comparison results of the actual results, the traditional model prediction results and the prediction results of this embodiment on these four samples are shown. It can be seen that the model proposed in this embodiment significantly improves the prediction accuracy of the overall film cooling efficiency distribution and the prediction accuracy at the local position of the hole outlet.

[0109] Figure 11The prediction errors of the traditional model and the prediction model proposed in the present invention for the surface average film cooling efficiency of the four samples are shown. It can be seen that the prediction error of the prediction model of the present invention is reduced by 4.3% to 41.2% compared with the original traditional model.

[0110] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention without departing from the principles and purpose of the present invention.

Claims

1. A generative two-dimensional distribution prediction method for film cooling efficiency based on prior knowledge embedding, characterized by: The following steps are involved: Step 1: Obtain training samples through experimental testing: Each set of sample data includes a film hole parameter vector composed of structural parameters and aerodynamic parameters and a two-dimensional distribution image of film cooling efficiency obtained through experimental testing of the corresponding film hole; Step 2: Build the network model: The network model input is the film hole parameter vector, and the output is the two-dimensional distribution image of film cooling efficiency; The network model includes a priori knowledge embedding module, a feature extraction module, an image encoding and decoding module, and an output module; The input of the prior knowledge embedding module is the air film hole parameter vector, which is mapped to the coefficients A1, A2, A3, B1, B2, and B3; according to the coefficients A1, A2, A3, B1, B2, and B3, the function F(x)=A1·log(A2·x)+A3 Q(x)=B1·(1-x) 2 +B2·2·x·(1-x)+B3·x 2 First, the coefficients F(x) and Q(x) are calculated, and then the dimensionality is increased to restore the two-dimensional distribution cloud map of the film cooling efficiency. x represents the coordinate of each point in the two-dimensional distribution image of the film cooling efficiency along the flow direction, y represents the coordinate of each point in the two-dimensional distribution image of the film cooling efficiency along the span direction, and G(y, F(x), Q(x)) is the cooling efficiency value at the (x, y) coordinate point. After passing through the convolution block, a matrix with the same size as the model output image is obtained as the input of the image encoding module. The feature extraction module performs high-dimensional extraction on the input air film hole parameter vector, reconstructs the extracted high-dimensional features into a two-dimensional matrix, and then expands the channels through the convolution block to obtain a matrix with the same size as the image encoding output; The image encoding and decoding module downsamples the matrix output by the prior knowledge embedding module N times to obtain information at different levels of the image. The features extracted at each layer are then spliced ​​with the information at the same level after decoding. The matrix obtained after the final downsampling is fused with the high-dimensional features extracted by the feature extraction module and flattened into a one-dimensional vector. The matrix is ​​then reconstructed through a multi-head attention mechanism to capture key information in the image and vector features. After obtaining the key information, it is gradually upsampled to restore it to the size of the image and provided to the output module to obtain the final two-dimensional distribution output of the film cooling efficiency. Step 3: Use the sample data obtained in step 1 to train the network model constructed in step 2 to obtain a trained prediction model; Step 4: For the film cooling scheme to be predicted, obtain the film hole parameter vector, input the trained prediction model obtained in step 3, and obtain the two-dimensional distribution prediction result of the film cooling efficiency.

2. The method for predicting the two-dimensional distribution of film cooling efficiency based on prior knowledge embedding according to claim 1, characterized in that: The structural parameters include: the hole inclination angle α between the axis of the air film hole and the outlet plane; the expansion angle β of the air film hole expansion section; the forward inclination angle γ of the air film hole expansion section; the inner fold angle δ of the V-shaped hole outlet; the aspect ratio L / d of the air film hole; the ratio of the width to the diameter of the air film hole outlet L w / d; Air film hole inlet and outlet area ratio A out / A in ; The dimensionless pore spacing P / d of the film pores.

3. The method for predicting the two-dimensional distribution of film cooling efficiency based on prior knowledge embedding according to claim 1 or 2, characterized in that: The aerodynamic parameters include: blowing ratio M; density ratio DR; turbulence Tu and acceleration factor k.

4. The method for predicting the two-dimensional distribution of film cooling efficiency based on prior knowledge embedding according to claim 1, characterized in that: The test is carried out by the following test system: The main flow is generated by the blower and passes through a butterfly valve and a rectifier chamber to produce a uniform inlet flow at the test section inlet. Turbulence grids of varying densities then generate varying turbulence intensities at the film hole outlet to simulate the effects of varying turbulence on cooling. A suction slit is placed at the front end of the measurement plate to allow the boundary layer to redevelop there, ensuring that the boundary layer thickness at the film hole outlet approaches that of an actual engine film hole. A trip wire is also placed behind the slit to promote the transition of the boundary layer from laminar to turbulent flow. Velocity, temperature, and pressure measurement points are set above the film hole outlet to monitor the mainstream status during the test in real time. The coolant system uses a high-pressure gas tank and compressor as the gas source. After passing through a valve and a pressure regulating device, it is heated by a heater so that the temperature difference between the primary and secondary flows does not exceed 0.2K. After the flow is controlled by an Alicat mass flowmeter, it enters the cold air cavity and is supplied to the film hole. Temperature and pressure measurement points are set in the cold air cavity to monitor the coolant status in real time. PSP pressure-sensitive paint is sprayed on the surface of the measuring plate, and the corresponding excitation light source and a scientific-grade CCD camera with a filter are used to collect images during the test. After post-processing of the image, a two-dimensional distribution image of the film cooling efficiency is obtained.

5. The method for predicting the two-dimensional distribution of film cooling efficiency based on prior knowledge embedding according to claim 3, characterized in that: The prior knowledge embedding module consists of a coefficient mapping module and a learning function module; the coefficient mapping module includes an extraction layer, a hidden layer and an output layer. The coefficient mapping module first extracts the characteristics of the input air film hole parameter vector through an extraction layer composed of 16 neurons, and then maps it to coefficients A1, A2, A3, B1, B2, and B3 through an output layer composed of 16 neurons after passing through two hidden layers composed of 32 neurons respectively; in the learning function module, the cooling efficiency value of each point in the two-dimensional distribution cloud map of air film cooling efficiency is calculated using the learned 6 coefficients A1, A2, A3, B1, B2, and B3.

6. The method for predicting the two-dimensional distribution of film cooling efficiency based on prior knowledge embedding according to claim 1, characterized in that: In the feature extraction module, the input air film hole parameter vector passes through a high-dimensional extraction layer consisting of 512 neurons, 1024 neurons, and 4096 neurons in sequence, and the high-dimensional features are reconstructed into a 64×64 matrix. After the channel is expanded through a 3×3 convolution block, the expanded 64×64 matrix is ​​obtained, which is consistent with the image encoding output.

7. The method for predicting the two-dimensional distribution of film cooling efficiency based on prior knowledge embedding according to claim 1, characterized in that: In the image encoding and decoding module, the 512×512 matrix corresponding to the two-dimensional distribution image of the film cooling efficiency output by the prior knowledge embedding module is first processed by two 3×3 convolution blocks to obtain a first matrix of 512×512. After the first matrix is ​​downsampled once and subjected to two 3×3 convolution blocks, a second matrix of 256×256 is obtained. After the second downsampling and two 3×3 convolution blocks, a third matrix of 128×128 is obtained. After the third downsampling and two 3×3 convolution blocks, a fourth matrix of 64×64 is obtained. The 64×64 fourth matrix is ​​fused with the high-dimensional features extracted by the feature extraction module and flattened into a one-dimensional vector. Then, the matrix is ​​subjected to the multi-head attention mechanism. The reconstructed 64×64 fifth matrix is ​​reconstructed into a 64×64 matrix. After passing through a 3×3 convolution block, the reconstructed 64×64 fifth matrix is ​​concatenated with the 64×64 fourth matrix. After passing through two 3×3 convolution blocks and the first deconvolution, a 128×128 sixth matrix is ​​obtained. The sixth matrix is ​​concatenated with the 128×128 third matrix. After passing through two 3×3 convolution blocks and the second deconvolution, a 256×256 seventh matrix is ​​obtained. The seventh matrix is ​​concatenated with the 256×256 second matrix. After passing through two 3×3 convolution blocks and the third deconvolution, a 512×512 eighth matrix is ​​obtained. The eighth matrix is ​​concatenated with the 512×512 first matrix and output to the output module.

Images