Method and system for rapidly predicting leakage characteristics of composite material

Abstract

The invention relates to the technical field of material leakage characteristic prediction, in particular to a rapid prediction method and system for leakage characteristics of a composite material. The method comprises the following steps: acquiring CT scanning data of a composite material, and constructing a plurality of composite material image models according to the CT scanning data; carrying out leakage simulation on the composite material image model by utilizing a leakage analysis model to obtain the leakage rate of the composite material; combining the composite material image model and the composite material leakage rate to establish a composite material leakage characteristic data set; constructing a composite material leakage characteristic prediction model, and training and verifying the composite material leakage characteristic prediction model through the composite material leakage characteristic data set; and on the basis of the trained composite material leakage characteristic prediction model, rapid prediction of the composite material leakage characteristics is completed. According to the method, the composite material leakage characteristic prediction model and the training data set are established based on the in-situ CT scanning result, and the leakage performance of the composite material is efficiently and accurately predicted.

Description

Technical Field

[0001] The invention relates to the technical field of material leakage characteristic prediction, and in particular to a method and system for quickly predicting leakage characteristics of composite materials. Background Art

[0002] Cryogenic propellant tanks are an important part of launch vehicles. The mainstream spacecraft in the world have tried to use composite materials to reduce the overall weight and production costs. Some developed countries have successfully applied them to cryogenic propellant tanks of aircraft. The storage temperatures of cryogenic propellants represented by liquid hydrogen and liquid oxygen are -183℃ and -252℃ respectively. Extreme ambient temperatures will have a significant impact on the mechanical properties and leakage properties of composite materials. Specifically, microcrack initiation and interface stratification caused by thermal expansion mismatch are prone to occur inside the composite structure, which can easily cause propellant leakage. This poses new challenges to the safety and reliability design of lightweight composite structures for spacecraft equipment. Therefore, it is urgent to establish the evolution law of composite material leakage characteristics in ultra-low temperature environments and develop efficient prediction methods to provide basic theory and key technical support for the design of composite tank structures.

[0003] At present, the prediction of composite material leakage characteristics is mainly based on complex numerical calculation models, such as computational fluid dynamics models, which have too high calculation costs. The use of numerical simulation and theoretical methods to predict the leakage performance of composite materials can effectively reduce the research cost, but the above methods are not suitable for the prediction of the leakage behavior of composite materials containing complex geometric cracks, and the calculation cost is still high. Therefore, it is an urgent problem to be able to be applicable to composite materials containing complex geometric cracks and to be able to easily and quickly accurately predict the leakage performance of composite materials. Summary of the invention

[0004] In view of the shortcomings of existing methods and the needs of practical applications, in order to improve the adaptability to composite materials containing complex geometric cracks, the problem of conveniently and quickly accurately predicting the leakage performance of composite materials is solved. On the one hand, the present invention provides a method for quickly predicting the leakage characteristics of composite materials, and the method for quickly predicting the leakage characteristics of composite materials includes the following steps: Obtain CT scanning data of the composite material, and construct multiple composite material image models according to the CT scanning data; use the leakage analysis model to simulate the leakage of the composite material image model to obtain the composite material leakage rate; combine the composite material image model and the composite material leakage rate to establish a composite material leakage characteristic data set; construct a composite material leakage characteristic prediction model, and train and verify the composite material leakage characteristic prediction model through the composite material leakage characteristic data set; based on the trained composite material leakage characteristic prediction model, complete the rapid prediction of composite material leakage characteristics. The present invention establishes a composite material leakage characteristic data set based on the in-situ CT scanning results, and then trains the composite material leakage characteristic prediction model. The leakage performance of the composite material is efficiently and accurately predicted through the machine learning model, which improves the adaptability to composite materials containing complex geometric cracks and solves the problem of convenient and rapid accurate prediction of composite material leakage performance.

[0005] Optionally, constructing a plurality of composite material image models according to the CT scanning data comprises the following steps: The CT scanning data is statistically analyzed to obtain crack parameters of different ply angles in the scanning area; multiple two-dimensional images are generated based on the crack parameters; the two-dimensional images are stacked in the order of ply to obtain the composite material image model. The present invention improves the adaptability of composite materials with complex geometric cracks by performing statistical analysis on the CT scanning data and then generating multiple two-dimensional images.

[0006] Optionally, combining the composite material image model and the composite material leakage rate to establish a composite material leakage characteristic data set comprises the following steps: The composite material image model is standardized to obtain a standardized processing result; the composite material leakage rate is normalized to obtain a normalized processing result; and the composite material leakage characteristic data set is established by combining the standardized processing result and the normalized processing result. The present invention is conducive to improving the stability and calculation efficiency of the present invention by performing standardized processing and normalized processing on the data.

[0007] Optionally, the composite material image model is standardized to satisfy the following formula: , in, represents the value after normalization. represents the data before standardization. represents a given mean, represents a given standard deviation; The leakage rate of the composite material is normalized to satisfy the following formula: , in, represents the normalized value. represents the value before normalization. represents the minimum leakage rate, Indicates the maximum leakage rate.

[0008] Optionally, the step of training and validating the composite material leakage characteristic prediction model using the composite material leakage characteristic data set comprises the following steps: Perform convolution operation on the composite material image model to obtain image features; vectorize the image features, add position information to the vectorized results, and then calculate the attention value; add physical information to the vectorized results to obtain physical constraint features; combine the image features, the attention value, the physical constraint features and the composite material leakage characteristic data set to train and verify the composite material leakage characteristic prediction model. The present invention fuses the position information into the data set, retains the information of the ply sequence, and thus obtains more representative image features. Fusing the physical information into the data set ensures that a sufficiently accurate model can be trained with a small amount of training data.

[0009] Optionally, the convolution operation is performed on the composite material image model to obtain image features that satisfy the following formula: ,in, Representing images The gray value after position convolution, represents the height of the convolution kernel, represents the width of the convolution kernel, Indicates height width The weight of the convolution kernel, Representing images The gray value before position convolution. The present invention uses convolution operation to implement row-by-row and column-by-column convolution operation of the convolution kernel on the image matrix.

[0010] Optionally, the adding of position information to the vectorization result satisfies the following formula: ,in, Indicates Even columns of rows, Indicates The odd columns of the row, PE represents the code containing the position information, Indicates the dimension of the input matrix. The present invention uses fixed position encoding based on sine function and cosine function to add position information, which can convert the information of any position into a linear combination of previous positions, which is conducive to retaining the information of ply order during training.

[0011] Optionally, the calculation of the attention value satisfies the following formula: , in, represents the attention value, represents a query in the Transformer model, Represents the key in the Transformer model, Represents the value in the Transformer model, represents transpose, express The transposed matrix of Indicates the length of the query, The attention value of the present invention retains the information of the ply order of the composite material laminate, which is beneficial to the training accuracy of the subsequent steps.

[0012] Optionally, the physical information is added to the vectorized result to obtain physical constraint features, satisfying the following formula: , in, represents the permeability parameter of porous media, represents the porosity, represents the Kozeny constant, The present invention adds physical information for estimating the permeability of porous media, further ensuring that a sufficiently accurate model can be trained with a small amount of training data.

[0013] In the second aspect, in order to be able to efficiently execute a composite material leakage characteristics rapid prediction method provided by the present invention, the present invention also provides a composite material leakage characteristics rapid prediction system, including a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute a composite material leakage characteristics rapid prediction method as described in the first aspect of the present invention. A composite material leakage characteristics rapid prediction system of the present invention has a compact structure and stable performance, and can stably execute a composite material leakage characteristics rapid prediction method provided by the present invention, further improving the overall applicability and practical application capabilities of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A flow chart of a method for quickly predicting leakage characteristics of a composite material provided by an embodiment of the present invention; Figure 2 A schematic diagram of a composite material image model construction process provided by an embodiment of the present invention; Figure 3 A framework diagram of a rapid prediction system for leakage characteristics of composite materials provided by an embodiment of the present invention; Figure 4 A schematic diagram of the structure of a device for quickly predicting leakage characteristics of a composite material provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not need to be adopted to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, software or methods are not specifically described.

[0016] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. In addition, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any suitable combination and / or subcombination. In addition, it should be understood by those of ordinary skill in the art that the figures provided herein are for illustrative purposes and that the figures are not necessarily drawn to scale.

[0017] See also Figure 1 In order to improve the adaptability of composite materials containing complex geometric cracks and solve the problem of conveniently and quickly predicting the leakage performance of composite materials, the present invention provides a method for quickly predicting the leakage characteristics of composite materials, such as Figure 1 As shown, the method comprises the following steps: S1. Acquire CT scanning data of a composite material, and construct a plurality of composite material image models according to the CT scanning data.

[0018] See also Figure 2 In the embodiment, step S1 of constructing a plurality of composite material image models according to the CT scan data comprises the following steps: S11, performing statistical analysis on the CT scanning data to obtain crack parameters of different ply angles in the scanning area.

[0019] Specifically, in-situ CT scanning data of the composite material is obtained from the CT equipment, and then the scanned data is statistically analyzed to obtain crack parameters at different ply angles in the scanning area. The crack parameters include size, shape, volume fraction and spatial distribution.

[0020] S12. Generate a plurality of two-dimensional images based on the crack parameters.

[0021] Matlab software was used to create a two-dimensional image containing crack features, where the length of the crack was randomly distributed within the ply, the crack width was set to the average crack width of the layer, and the crack thickness completely penetrated the corresponding ply.

[0022] Furthermore, multiple two-dimensional images are generated by random combinations of different crack densities and ply angles.

[0023] S13, stacking the two-dimensional images in a ply order to obtain the composite material image model.

[0024] Specifically, the two-dimensional images are stacked together according to the ply sequence to combine the composite material image model.

[0025] S2. Using a leakage analysis model to perform leakage simulation on the composite material image model to obtain a leakage rate of the composite material.

[0026] In the embodiment, the leakage analysis model is Avizo software. The composite material image model obtained in step S1 is imported into the Avizo software, and the image model is simulated for leakage using its leakage analysis module to obtain the absolute permeability of laminates with different layup sequences. The calculation results are converted using Darcy's law to obtain the leakage rate of the composite material.

[0027] Furthermore, the leakage rate of composite materials can be obtained by batch calculation in Avizo software by writing Python scripts.

[0028] S3. Combining the composite material image model and the composite material leakage rate, a composite material leakage characteristic data set is established.

[0029] In an embodiment, the step S3 combines the composite material image model and the composite material leakage rate to establish a composite material leakage characteristic data set, including the following steps: S31, performing standardization processing on the composite material image model to obtain a standardization processing result.

[0030] Specifically, the composite material image model is standardized to satisfy the following formula: , in, represents the value after normalization. represents the data before standardization. represents a given mean, Represents a given standard deviation.

[0031] S32, normalizing the leakage rate of the composite material to obtain a normalized result.

[0032] Specifically, the leakage rate of the composite material is normalized to satisfy the following formula:

[0033] in, represents the normalized value. represents the value before normalization. represents the minimum leakage rate, Represents the maximum leakage rate. The Min-Max normalization method is used to normalize the leakage rate data of composite plies with different plies, and the data can be mapped in the interval [0, 1], which is conducive to improving the calculation efficiency.

[0034] S33, combining the standardized processing result and the normalized processing result to establish the composite material leakage characteristic data set.

[0035] The standardized and normalized data are organized into a data set, which should include all the features used for analysis, namely the independent variables, and the measurement results of the leakage characteristics, namely the dependent variables.

[0036] S4. Construct a composite material leakage characteristic prediction model, and train and verify the composite material leakage characteristic prediction model through the composite material leakage characteristic data set.

[0037] In the embodiment, the composite material leakage characteristic prediction model constructed in step S4 is a machine learning model. Preferably, the composite material leakage characteristic prediction model is a two-dimensional convolutional neural network with an attention module.

[0038] The basic structure of a two-dimensional convolutional neural network includes an input layer, a convolutional layer, an activation function, a pooling layer, a fully connected layer, and an output layer. The attention module is a Transformer model, and its network structure is mainly composed of an encoder and a decoder, both of which contain an attention layer and a feedforward neural network layer. The attention layer and position encoding module of the Transformer model can retain the information of the ply layup order of the composite laminate.

[0039] By utilizing the attention layer and position encoding module of the Transformer model, the information of the ply layup order of the composite laminate can be retained. A two-dimensional convolutional neural network is used to learn the image and extract the crack features. Combined with the position encoding function of the Transformer model, the ply layup order is recorded to establish a composite material leakage characteristics prediction model that takes into account the composite material ply layup order.

[0040] Furthermore, the step S4 of training and verifying the composite material leakage characteristic prediction model by using the composite material leakage characteristic data set comprises the following steps: S41, performing a convolution operation on the composite material image model to obtain image features.

[0041] The composite material image model is read into the Pytorch software, and its corresponding leakage rate is recorded as the label value; the read image is scaled to 224×224 pixels and then converted into a tensor that can be processed by the computer.

[0042] For an image matrix of size M×N, a convolution kernel of size U×V is used for convolution. Further, the composite image model is convolved to obtain image features that satisfy the following formula: ,in, Representing images The gray value after position convolution, represents the height of the convolution kernel, represents the width of the convolution kernel, Indicates height width The weight of the convolution kernel, Representing images Gray value of the position before convolution.

[0043] pass and , realizing the convolution operation of the convolution kernel on the image matrix row by row and column by column; secondly, the ReLU function is selected as the activation function, and the maximum pooling method is used to retain the maximum value in the area to obtain more representative image features.

[0044] S42: vectorize the image features, add position information to the vectorization result, and then calculate the attention value.

[0045] The attention layer and position encoding module of the Transformer model are used to retain the information of the ply layup order of the composite laminate. The content of the attention calculation is the image features extracted by convolution, which is expanded into a one-dimensional vector as the input of the attention layer.

[0046] Furthermore, the position information is added to the vectorization result to satisfy the following formula: , in, Indicates Even columns of rows, Indicates The odd columns of the row, PE represents the code containing the position information, Indicates the dimensions of the input matrix.

[0047] It can be understood that, assuming the input is an n-ply, dimensional matrix, the position encoding is a matrix of the same size and is added to the input matrix. The elements of the posth row, 2ith column and 2i+1th column can be calculated using the above formula. At the same time, trigonometric functions can use the sum-difference-product formula to convert the information at any position into a linear combination of the previous positions. In this way, the information of the ply order can be retained during the training process. Furthermore, the calculation of the attention value satisfies the following formula: , in, represents the attention value, represents a query in the Transformer model, Represents the key in the Transformer model, Represents the value in the Transformer model, represents transpose, express The transposed matrix of Indicates the length of the query, represents the normalized exponential function.

[0048] S43. Add physical information to the vectorized result to obtain physical constraint features.

[0049] Specifically, in step S43, physical information is added to the vectorized result to obtain physical constraint features, which satisfy the following formula: , in, represents the permeability parameter of porous media, represents the porosity, represents the Kozeny constant, Represents specific surface area.

[0050] The physical information is added in the fully connected layer after the convolution and calculated in the loss function. After the composite image model tensor is convolved to extract features, it is expanded into a one-dimensional vector, and the porosity and specific surface area are added to the last part of the vector for physical information calculation.

[0051] S44. Combine the image features, the attention value, the physical constraint features and the composite material leakage characteristic data set to train and verify the composite material leakage characteristic prediction model.

[0052] In an embodiment, the image features, the attention values, the physical constraint features and the composite material leakage characteristic data set are combined to construct a data training set and a data verification set in proportion, and then the composite material leakage characteristic prediction model is trained and verified.

[0053] In the embodiment, the loss function of the composite material leakage characteristic prediction model is a root mean square error, which satisfies the following formula: ,in, represents the root mean square error, represents the number of observations, Indicates The true value of the observation, Indicates The predicted value of an observation.

[0054] Furthermore, the coefficient of determination is used to evaluate and verify the composite material leakage characteristics prediction model, and its value range is from 0 to 1. The closer the value is to 1, the better the model fitting effect is. The formula is as follows: , in, represents the coefficient of determination, Represents the mean of the true values.

[0055] It should be understood that in order to avoid convergence difficulties caused by order of magnitude problems, it is also necessary to perform normalization in each layer of training, that is, batch normalization. In order to avoid overfitting of the neural network and loss of generalization ability, the discard method is used to add some noise interference during training to improve the anti-interference ability of the network model.

[0056] S5. Based on the trained composite material leakage characteristic prediction model, complete the rapid prediction of composite material leakage characteristics.

[0057] The CT data of the composite material to be predicted is obtained, and the leakage characteristics of the composite material to be predicted are predicted using the trained composite material leakage characteristic prediction model, that is, the leakage characteristics of the composite material to be predicted are obtained efficiently and accurately.

[0058] The present invention uses CT image statistical information to predict the leakage performance of composite laminates, establishes a composite material leakage image data set and performs neural network training, does not require high-precision CT scanning, and does not require complex grid division processing, avoids the problems of poor grid quality and difficult calculation convergence in traditional models, greatly reduces the calculation cost, can quickly obtain the leakage performance of composite materials, and opens up new prospects for quickly estimating the leakage rate of different laminates.

[0059] Secondly, the composite material leakage characteristic prediction model established by the present invention can calculate the leakage of composite material laminates based on crack images of different orders. The embedded physical model can accelerate the convergence speed of the training process and improve the accuracy. Compared with other neural network models or machine learning models, the present invention can predict the leakage of composite materials faster while ensuring accuracy.

[0060] See also Figure 3 In an embodiment, in order to efficiently execute a composite material leakage characteristic rapid prediction method provided by the present invention, the present invention also provides a composite material leakage characteristic rapid prediction system, including: an input device, an output device, a processor, and a memory, wherein the input device, the output device, the processor, and the memory are interconnected, and the memory contains program instructions, and the program instructions are used for the steps of the composite material leakage characteristic rapid prediction method. The composite material leakage characteristic rapid prediction system of the present invention has a compact structure and stable performance, and can stably execute a composite material leakage characteristic rapid prediction method of the present invention, further improving the overall applicability and practical application ability of the present invention.

[0061] In an embodiment, the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. An input device may be used to obtain data information. An output device may be used to output the results obtained by storing program instructions contained in a computer program in a memory provided by the present invention. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory.

[0062] In yet another alternative embodiment, see Figure 4 In order to efficiently execute the composite material leakage characteristic rapid prediction method provided by the present invention, this embodiment also provides a composite material leakage characteristic rapid prediction device, such as Figure 4 As shown, including: The memory 10 is used to store computer programs; the processor 20 is used to execute the computer programs to implement the above-mentioned composite material leakage characteristic rapid prediction method. The memory 10, the processor 20, the communication interface 31 and the communication bus 32. The memory 10, the processor 20 and the communication interface 31 all communicate with each other through the communication bus 32.

[0063] In an embodiment, the memory 10 is used to store one or more program instructions. The memory 10 may store program instructions for implementing the following functions: Acquire CT scanning data of the composite material, and construct a plurality of composite material image models according to the CT scanning data; use a leakage analysis model to perform leakage simulation on the composite material image model to obtain the composite material leakage rate; combine the composite material image model and the composite material leakage rate to establish a composite material leakage characteristic data set; construct a composite material leakage characteristic prediction model, and train and verify the composite material leakage characteristic prediction model through the composite material leakage characteristic data set; and complete the rapid prediction of composite material leakage characteristics based on the trained composite material leakage characteristic prediction model.

[0064] In one possible implementation, the memory 10 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, and at least one application required for a function, etc.; the data storage area may store data created during use. In addition, the memory 10 may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include NVRAM. The memory stores an operating system and operating instructions, executable modules or data structures, or a subset thereof, or an extended set thereof, wherein the operating instructions may include various operating instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0065] The processor 20 may be a central processing unit (CPU), an application specific integrated circuit, a digital signal processor, a field programmable gate array or other programmable logic device, and the processor 20 may be a microprocessor or any conventional processor, etc. The processor 20 may call a program stored in the memory 10. The communication interface 31 may be an interface of a communication module, and is used to connect with other devices or systems.

[0066] Of course, it should be noted that Figure 4 The structure shown does not constitute a limitation on the composite material leakage characteristic rapid prediction device in this embodiment. In actual applications, the composite material leakage characteristic rapid prediction device may include Figure 4 More or fewer components than shown, or combinations of certain components.

[0067] The embodiment further provides a storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the above-mentioned method for rapidly predicting leakage characteristics of composite materials are implemented.

[0068] The storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0069] In summary, the present invention establishes a composite material leakage characteristic data set based on the in-situ CT scanning results, and then trains a composite material leakage characteristic prediction model. The composite material leakage performance is predicted efficiently and accurately through a machine learning model, which improves the adaptability to composite materials containing complex geometric cracks and solves the problem of conveniently and quickly accurately predicting the leakage performance of composite materials. Therefore, the present invention effectively overcomes various shortcomings in the prior art and has a high industrial utilization value.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope recorded in the present invention.

Claims

1. A method for quickly predicting leakage characteristics of composite materials, characterized in that: The method for rapidly predicting leakage characteristics of composite materials comprises the following steps: Acquire CT scanning data of the composite material, and construct a plurality of composite material image models according to the CT scanning data; Using a leakage analysis model to perform leakage simulation on the composite material image model to obtain a leakage rate of the composite material; Combining the composite material image model and the composite material leakage rate to establish a composite material leakage characteristic data set; Constructing a composite material leakage characteristic prediction model, and training and verifying the composite material leakage characteristic prediction model through the composite material leakage characteristic data set; Based on the trained composite material leakage characteristic prediction model, the leakage characteristics of the composite material can be quickly predicted.

2. The method for rapid prediction of composite material leakage characteristics according to claim 1, characterized in that: The method of constructing a plurality of composite material image models according to the CT scanning data comprises the following steps: Performing statistical analysis on the CT scanning data to obtain crack parameters at different ply angles within the scanning area; Based on the crack parameters, generating a plurality of two-dimensional images; The two-dimensional images are stacked in a ply order to obtain the composite material image model.

3. The method for rapid prediction of composite material leakage characteristics according to claim 1, characterized in that: The step of combining the composite material image model and the composite material leakage rate to establish a composite material leakage characteristic data set includes the following steps: Performing standardization processing on the composite material image model to obtain a standardization processing result; Performing normalization processing on the leakage rate of the composite material to obtain a normalization processing result; The composite material leakage characteristic data set is established by combining the standardized processing result and the normalized processing result.

4. The method for rapid prediction of composite material leakage characteristics according to claim 3, characterized in that: The composite material image model is standardized to satisfy the following formula: ， in, represents the value after normalization. represents the data before standardization. represents a given mean, represents a given standard deviation; The leakage rate of the composite material is normalized to satisfy the following formula: ， in, represents the normalized value. represents the value before normalization. represents the minimum leakage rate, Indicates the maximum leakage rate.

5. The method for rapid prediction of composite material leakage characteristics according to claim 1, characterized in that: The method of training and verifying the composite material leakage characteristic prediction model by using the composite material leakage characteristic data set comprises the following steps: Performing a convolution operation on the composite material image model to obtain image features; Vectorize the image features, add position information to the vectorized results, and then calculate the attention value; Add physical information to the vectorized results to obtain physical constraint features; The composite material leakage characteristic prediction model is trained and verified by combining the image features, the attention value, the physical constraint features and the composite material leakage characteristic data set.

6. The method for rapid prediction of composite material leakage characteristics according to claim 5, characterized in that: The convolution operation is performed on the composite material image model to obtain image features that satisfy the following formula: ， in, Representing images The gray value after position convolution, represents the height of the convolution kernel, represents the width of the convolution kernel, Indicates height width The weight of the convolution kernel, Representing images Gray value before position convolution.

7. The method for rapid prediction of composite material leakage characteristics according to claim 5, characterized in that: The position information is added to the vectorized result to satisfy the following formula: ， in, Indicates Even columns of rows, Indicates The odd columns of the row, PE represents the code containing the position information, Indicates the dimensions of the input matrix.

8. The method for rapid prediction of composite material leakage characteristics according to claim 5, characterized in that: The calculation of the attention value satisfies the following formula: ， in, represents the attention value, represents a query in the Transformer model, Represents the key in the Transformer model, Represents the value in the Transformer model, represents transpose, express The transposed matrix of Indicates the length of the query, represents the normalized exponential function.

9. The method for rapid prediction of composite material leakage characteristics according to claim 5, characterized in that: The physical information is added to the vectorized result to obtain the physical constraint feature, which satisfies the following formula: ， in, represents the permeability parameter of porous media, represents the porosity, represents the Kozeny constant, Represents specific surface area.

10. A rapid prediction system for leakage characteristics of composite materials, characterized in that: The system for rapid prediction of composite material leakage characteristics includes: an input device, an output device, a processor, and a memory. The input device, the output device, the processor, and the memory are interconnected. The memory includes program instructions, and the program instructions are used to execute the method for rapid prediction of composite material leakage characteristics according to any one of claims 1 to 9.

Images