A High-Temperature Heat Pipe Structure Design and Simulation Method

Abstract

This application discloses a method for designing and simulating high-temperature heat pipe structures. The method includes: constructing a two-layer neural network model; training the two-layer neural network model using high-temperature heat pipe samples; comparing the similarity between the evaluation parameters of the high-temperature heat pipe obtained from the simulation platform and the evaluation parameters obtained from the two-layer neural network model; iteratively optimizing the two-layer neural network model to form an optimized model; completing the three-dimensional structural design of the high-temperature heat pipe in a three-dimensional simulation interface; and utilizing the optimized two-layer neural network model and the simulation platform to realize the structural design and simulation calculation of the high-temperature heat pipe. This method can effectively improve the design efficiency of high-temperature heat pipes, reduce enterprise R&D costs, and improve the overall operating efficiency of the system.

Description

Technical fields:

[0001] This invention relates to the field of high-temperature heat pipe design and simulation calculation technology, and in particular to a high-temperature heat pipe structure design and simulation method. Background technology:

[0002] High-temperature heat pipe technology is one of the most efficient heat exchange technologies for utilizing high-temperature waste heat. Its unique heat transfer characteristics effectively solve many challenges in the field of high-temperature heat exchange. High-temperature heat pipes transfer heat through the continuous evaporation and condensation of liquid metal within a fully enclosed vacuum system, exhibiting high heat transfer capacity and excellent isothermal properties. High-temperature heat pipes have significant application potential in fields such as space technology and concentrated solar thermal utilization.

[0003] However, high-temperature heat pipes face numerous challenges in structural design and performance calculation. Traditional methods, using two-dimensional planar diagrams for high-temperature heat pipe structural design, demand extremely high levels of expertise from designers. Furthermore, performance parameters can only be obtained after the heat pipe design is completed, followed by production on the production line and offline testing. This results in lengthy, inefficient, and costly R&D, design, and production processes for high-temperature heat pipes. Moreover, the accuracy of design parameters needs improvement due to the influence of various external factors. To address these issues, this invention proposes a method for high-temperature heat pipe structural design and simulation. Specifically, Figure 1 This is a structural diagram of a high-temperature heat pipe. Summary of the Invention:

[0004] The technical solution adopted by this invention to solve the above-mentioned technical problems is to propose a high-temperature heat pipe structure design and simulation method, which includes the following steps:

[0005] S1. Log in to the high-temperature heat pipe structure design and simulation platform, which includes a three-dimensional simulation interface;

[0006] S2. Complete the three-dimensional structural design of the high-temperature heat pipe in the three-dimensional simulation interface;

[0007] S3. Construct a two-layer neural network model, train the two-layer neural network model using high-temperature heat pipe samples in the experience base, and output the high-temperature heat pipe evaluation parameter array C1.

[0008] S4. Start the simulation platform, run the high-temperature heat pipe sample in step S3, and output the high-temperature heat pipe evaluation parameter array C2.

[0009] S5. Iteratively update the parameters of the two-layer neural network model so that the similarity threshold between the high-temperature heat pipe evaluation parameter array C1 and the high-temperature heat pipe evaluation parameter array C2 is less than the preset threshold T, and finally form an optimized two-layer neural network model.

[0010] S6. Utilize an optimized two-layer neural network model and simulation platform to realize the structural design and simulation calculation of high-temperature heat pipes.

[0011] Further, step S2 specifically includes:

[0012] The three-dimensional simulation interface includes a three-dimensional high-temperature heat pipe component structure library;

[0013] Users can assemble high-temperature heat pipe components by dragging and dropping to form a three-dimensional high-temperature heat pipe structure;

[0014] Furthermore, after selecting a three-dimensional high-temperature heat pipe component, the user can modify its structure; the modifications specifically include:

[0015] After selecting the three-dimensional high-temperature heat pipe component, the user can enter the structural parameters of the three-dimensional high-temperature heat pipe component on the interface;

[0016] After selecting a 3D high-temperature heat pipe component, users can add or delete components on the interface.

[0017] Furthermore, after selecting a three-dimensional high-temperature heat pipe component, the user can modify the structure of the three-dimensional high-temperature heat pipe component, including the following methods:

[0018] The user inputs voice, which is a descriptive language for modifying the structure of three-dimensional high-temperature heat pipe components;

[0019] Parse the user's voice input and generate text;

[0020] The above text keywords will be extracted, including keywords related to component structural parts and modified variables;

[0021] Convert the keywords of component structural parts into professional technical keywords of component structure;

[0022] Based on the converted technical keywords for the component structure, link them to the modification slider;

[0023] The structure of the three-dimensional high-temperature heat pipe components is modified using the aforementioned modified variables and modified sliders.

[0024] Furthermore, step S3 specifically includes:

[0025] S31. Construct the first layer neural network model matrix based on the n three-dimensional structural parameters of the high-temperature heat pipe;

[0026] The first layer neural network model matrix includes n neural network models;

[0027] The step of constructing the first layer neural network model matrix based on the n three-dimensional structural parameters of the high-temperature heat pipe specifically includes: constructing a corresponding neural network model for each of the three-dimensional structural parameters, such that parameter i corresponds to neural network model i;

[0028] The n neural network models are arranged in parallel;

[0029] The three-dimensional structural parameters include, but are not limited to, working fluid type, working fluid filling rate, working tilt angle, suction core hole shape, and suction core hole shape parameters.

[0030] The first layer neural network model matrix includes, but is not limited to, working fluid type neural network model, working fluid filling rate neural network model, working tilt angle neural network model, and suction core hole shape neural network model.

[0031] The above three-dimensional structural parameters are used as input data for each neural network model, that is, the working fluid type, working fluid filling rate, working tilt angle, liquid suction core hole shape, and liquid suction core hole shape parameters are input into each neural network model.

[0032] The weight values ​​of each input parameter are different in each neural network model; specifically, among all the input parameters of neural network model i, the weight of input parameter i is greater than the weight of other input parameters.

[0033] The output data of each neural network model is an array of evaluation parameters for high-temperature heat pipes;

[0034] S32. Construct the second-layer perceptual neural network model and connect it to the matrix of the first-layer neural network model;

[0035] The second-layer perceptual neural network model is connected to the matrix of the first-layer neural network model, specifically by using the output data of each neural network model in the first-layer neural network model as the input data of the second-layer perceptual neural network model.

[0036] S33. Initialize the parameters of each neural network model in the first layer neural network model matrix and the parameters of the second layer perceptual neural network model;

[0037] S34. Obtain high-temperature heat pipe sample data from the experience database, the sample data including the three-dimensional structural parameters of the high-temperature heat pipe; set environmental parameter data;

[0038] S35. Input the high-temperature heat pipe sample data and environmental parameter data from step S34 into each neural network model in the first layer neural network matrix;

[0039] S36. The evaluation parameter array C1 of the high-temperature heat pipe is output by the second-layer perceptual neural network model.

[0040] Further, step S4 specifically includes:

[0041] S41. Obtain the high-temperature sample data from step S34, wherein the sample data includes the three-dimensional structural parameters of the high-temperature heat pipe; set environmental parameter data;

[0042] S42. Start the simulation platform, run the high-temperature heat pipe sample in step S41, and output the high-temperature heat pipe evaluation parameter array C2.

[0043] Furthermore, step S5 specifically includes:

[0044] S51. Calculate the similarity between the high-temperature heat pipe evaluation parameter array C1 and the high-temperature heat pipe evaluation parameter array C2. If the similarity is greater than the preset threshold T, proceed to step S52; otherwise, proceed to step S53.

[0045] S52. Modify the parameters of the first-layer neural network model matrix and the second-layer perceptual neural network model, and re-execute steps S35-S36 and S51 until the similarity is less than the preset threshold T.

[0046] The parameters of the modified first-layer neural network model matrix and the second-layer perceptual neural network model include weights, biases, and activation functions.

[0047] S53. Obtain the optimized two-layer neural network model.

[0048] Further, step S6 specifically includes:

[0049] S61. After the three-dimensional structure design of the high-temperature heat pipe in step 2 is completed, the two-layer neural network model outputs an array of high-temperature heat pipe evaluation parameters for the designed three-dimensional structure of the high-temperature heat pipe.

[0050] S62. If the high-temperature heat pipe evaluation parameters output in step S61 do not meet the design target value, the user modifies the three-dimensional structure design of the high-temperature heat pipe until the high-temperature heat pipe evaluation parameters output by the dual-layer neural network model meet the design target value, and obtains the three-dimensional structure of the high-temperature heat pipe that meets the design target value.

[0051] S63. Start the simulation platform and run the three-dimensional structure of the high-temperature heat pipe that meets the design target value in step S62. Obtain the high-temperature heat pipe evaluation parameter array as the simulation calculation result of the high-temperature heat pipe.

[0052] The beneficial effects of this invention are as follows:

[0053] 1. This invention reduces the complexity of design by setting up a three-dimensional high-temperature heat pipe component structure library in the three-dimensional simulation interface, allowing users to design the structure of high-temperature heat pipes by dragging and dropping.

[0054] 2. In the process of designing the structure of three-dimensional high-temperature heat pipe components, this invention allows modification of the high-temperature heat pipe component structure via voice, and can convert the user's descriptive language into professional technical keywords, enabling users without design experience to participate in the design of high-temperature heat pipe components; furthermore, when users use different descriptive languages ​​for the same part of the same component structure, the system can convert different descriptive languages ​​for the same part into the same professional technical keywords through the association library of descriptive language and professional technical keywords.

[0055] 3. By constructing a first-layer neural network model matrix and a second-layer perceptual neural network model, a two-layer neural network model is formed, which improves the accuracy of the neural network model output. Furthermore, the first-layer neural network model corresponds one-to-one with the three-dimensional structural parameters of the high-temperature heat pipe, forming a neural network model matrix. By setting different weights for the three-dimensional structural parameters corresponding to different neural network models, the accuracy of the neural network model output is further improved.

[0056] 4. This invention compares the similarity between the high-temperature heat pipe evaluation parameter array C2 output by the simulation platform and the high-temperature heat pipe evaluation parameter array C1 output by the two-layer neural network model, and dynamically corrects the parameters of the two-layer neural network model, thereby optimizing the two-layer neural network model and improving the accuracy of the output results of the two-layer neural network model.

[0057] 5. After designing the three-dimensional structure of the high-temperature heat pipe, the present invention first outputs evaluation parameters using a two-layer neural network model. If the evaluation parameters do not meet the preset requirements, the three-dimensional structure of the high-temperature heat pipe is modified, thereby improving the efficiency of the high-temperature heat pipe design. If the evaluation parameters meet the preset requirements, the three-dimensional structure of the high-temperature heat pipe is run through a simulation platform to obtain the high-temperature heat pipe evaluation parameter array as the simulation calculation result of the high-temperature heat pipe, thereby improving the overall operating efficiency of the system.

[0058] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above description and other objects, features and advantages of the present invention more obvious and understandable, preferred embodiments are provided and described in detail below. Attached Figure Description

[0059] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention.

[0060] Furthermore, the same reference numerals are used to denote the same parts throughout the accompanying drawings. In the drawings:

[0061] Figure 1 The structural diagram of the high-temperature heat pipe provided by the present invention;

[0062] Figure 2 A flowchart of a high-temperature heat pipe structure design and simulation method provided by the present invention. Detailed Implementation

[0063] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0064] In the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," "fixing," etc., should be interpreted broadly. For example, they can refer to a connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0065] A method for designing and simulating high-temperature heat pipe structures includes the following steps:

[0066] S1. Log in to the high-temperature heat pipe structure design and simulation platform, which includes a three-dimensional simulation interface;

[0067] S2. Complete the three-dimensional structural design of the high-temperature heat pipe in the three-dimensional simulation interface;

[0068] Further, step S2 specifically includes:

[0069] The three-dimensional simulation interface includes a three-dimensional high-temperature heat pipe component structure library;

[0070] Users can assemble high-temperature heat pipe components by dragging and dropping to form a three-dimensional high-temperature heat pipe structure;

[0071] Furthermore, after selecting a three-dimensional high-temperature heat pipe component, the user can modify its structure; the modifications specifically include:

[0072] After selecting the three-dimensional high-temperature heat pipe component, the user can enter the structural parameters of the three-dimensional high-temperature heat pipe component on the interface;

[0073] After selecting a 3D high-temperature heat pipe component, users can add or delete components on the interface.

[0074] Furthermore, after selecting a three-dimensional high-temperature heat pipe component, the user can modify the structure of the three-dimensional high-temperature heat pipe component, including the following methods:

[0075] The user inputs voice, which is a descriptive language for modifying the structure of three-dimensional high-temperature heat pipe components;

[0076] Parse the user's voice input and generate text;

[0077] The above text keywords will be extracted, including keywords related to component structural parts and modified variables;

[0078] Convert the keywords of component structural parts into professional technical keywords of component structure;

[0079] Based on the converted technical keywords for the component structure, link them to the modification slider;

[0080] The structure of the three-dimensional high-temperature heat pipe components is modified using the aforementioned modified variables and modified sliders.

[0081] S3. Construct a two-layer neural network model, train the two-layer neural network model using high-temperature heat pipe samples in the experience base, and output the high-temperature heat pipe evaluation parameter array C1.

[0082] Furthermore, step S3 specifically includes:

[0083] S31. Construct the first layer neural network model matrix based on the n three-dimensional structural parameters of the high-temperature heat pipe;

[0084] The first layer neural network model matrix includes n neural network models;

[0085] The step of constructing the first layer neural network model matrix based on the n three-dimensional structural parameters of the high-temperature heat pipe specifically includes: constructing a corresponding neural network model for each of the three-dimensional structural parameters, such that parameter i corresponds to neural network model i;

[0086] The n neural network models are arranged in parallel;

[0087] The three-dimensional structural parameters include, but are not limited to, working fluid type, working fluid filling rate, working tilt angle, suction core hole shape, and suction core hole shape parameters.

[0088] The first layer neural network model matrix includes, but is not limited to, working fluid type neural network model, working fluid filling rate neural network model, working tilt angle neural network model, and suction core hole shape neural network model.

[0089] The above three-dimensional structural parameters are used as input data for each neural network model, that is, the working fluid type, working fluid filling rate, working tilt angle, liquid suction core hole shape, and liquid suction core hole shape parameters are input into each neural network model.

[0090] The weight values ​​of each input parameter are different in each neural network model; specifically, among all the input parameters of neural network model i, the weight of input parameter i is greater than the weight of other input parameters.

[0091] The output data of each neural network model is an array of evaluation parameters for high-temperature heat pipes;

[0092] S32. Construct the second-layer perceptual neural network model and connect it to the matrix of the first-layer neural network model;

[0093] The second-layer perceptual neural network model is connected to the matrix of the first-layer neural network model, specifically by using the output data of each neural network model in the first-layer neural network model as the input data of the second-layer perceptual neural network model.

[0094] S33. Initialize the parameters of each neural network model in the first layer neural network model matrix and the parameters of the second layer perceptual neural network model;

[0095] S34. Obtain high-temperature heat pipe sample data from the experience database, the sample data including the three-dimensional structural parameters of the high-temperature heat pipe; set environmental parameter data;

[0096] S35. Input the high-temperature heat pipe sample data and environmental parameter data from step S34 into each neural network model in the first layer neural network matrix;

[0097] S36. The evaluation parameter array C1 of the high-temperature heat pipe is output by the second-layer perceptual neural network model.

[0098] S4. Start the simulation platform, run the high-temperature heat pipe sample in step S3, and output the high-temperature heat pipe evaluation parameter array C2.

[0099] Further, step S4 specifically includes:

[0100] S41. Obtain the high-temperature sample data from step S34, wherein the sample data includes the three-dimensional structural parameters of the high-temperature heat pipe; set environmental parameter data;

[0101] S42. Start the simulation platform, run the high-temperature heat pipe sample in step S41, and output the high-temperature heat pipe evaluation parameter array C2.

[0102] S5. Iteratively update the parameters of the two-layer neural network model so that the similarity threshold between the high-temperature heat pipe evaluation parameter array C1 and the high-temperature heat pipe evaluation parameter array C2 is less than the preset threshold T, and finally form an optimized two-layer neural network model.

[0103] Furthermore, step S5 specifically includes:

[0104] S51. Calculate the similarity between the high-temperature heat pipe evaluation parameter array C1 and the high-temperature heat pipe evaluation parameter array C2. If the similarity is greater than the preset threshold T, proceed to step S52; otherwise, proceed to step S53.

[0105] S52. Modify the parameters of the first-layer neural network model matrix and the second-layer perceptual neural network model, and re-execute steps S35-S36 and S51 until the similarity is less than the preset threshold T.

[0106] The parameters of the modified first-layer neural network model matrix and the second-layer perceptual neural network model include weights, biases, and activation functions.

[0107] S53. Obtain the optimized two-layer neural network model.

[0108] S6. Utilize an optimized two-layer neural network model and simulation platform to realize the structural design and simulation calculation of high-temperature heat pipes.

[0109] Further, step S6 specifically includes:

[0110] S61. After the three-dimensional structure design of the high-temperature heat pipe in step 2 is completed, the two-layer neural network model outputs an array of high-temperature heat pipe evaluation parameters for the designed three-dimensional structure of the high-temperature heat pipe.

[0111] S62. If the high-temperature heat pipe evaluation parameters output in step S61 do not meet the design target value, the user modifies the three-dimensional structure design of the high-temperature heat pipe until the high-temperature heat pipe evaluation parameters output by the dual-layer neural network model meet the design target value, and obtains the three-dimensional structure of the high-temperature heat pipe that meets the design target value.

[0112] S63. Start the simulation platform and run the three-dimensional structure of the high-temperature heat pipe that meets the design target value in step S62. Obtain the high-temperature heat pipe evaluation parameter array as the simulation calculation result of the high-temperature heat pipe.

[0113] The beneficial effects of this invention are as follows:

[0114] 1. This invention reduces the complexity of design by setting up a three-dimensional high-temperature heat pipe component structure library in the three-dimensional simulation interface, allowing users to design the structure of high-temperature heat pipes by dragging and dropping.

[0115] 2. In the process of designing the structure of three-dimensional high-temperature heat pipe components, this invention allows modification of the high-temperature heat pipe component structure via voice, and can convert the user's descriptive language into professional technical keywords, enabling users without design experience to participate in the design of high-temperature heat pipe components; furthermore, when users use different descriptive languages ​​for the same part of the same component structure, the system can convert different descriptive languages ​​for the same part into the same professional technical keywords through the association library of descriptive language and professional technical keywords.

[0116] 3. By constructing a first-layer neural network model matrix and a second-layer perceptual neural network model, a two-layer neural network model is formed, which improves the accuracy of the neural network model output. Furthermore, the first-layer neural network model corresponds one-to-one with the three-dimensional structural parameters of the high-temperature heat pipe, forming a neural network model matrix. By setting different weights for the three-dimensional structural parameters corresponding to different neural network models, the accuracy of the neural network model output is further improved.

[0117] 4. This invention compares the similarity between the high-temperature heat pipe evaluation parameter array C2 output by the simulation platform and the high-temperature heat pipe evaluation parameter array C1 output by the two-layer neural network model, and dynamically corrects the parameters of the two-layer neural network model, thereby optimizing the two-layer neural network model and improving the accuracy of the output results of the two-layer neural network model.

[0118] 5. After designing the three-dimensional structure of the high-temperature heat pipe, the present invention first outputs evaluation parameters using a two-layer neural network model. If the evaluation parameters do not meet the preset requirements, the three-dimensional structure of the high-temperature heat pipe is modified, thereby improving the efficiency of the high-temperature heat pipe design. If the evaluation parameters meet the preset requirements, the three-dimensional structure of the high-temperature heat pipe is run through a simulation platform to obtain the high-temperature heat pipe evaluation parameter array as the simulation calculation result of the high-temperature heat pipe, thereby improving the overall operating efficiency of the system.

[0119] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for designing and simulating high-temperature heat pipe structures, characterized in that, The method includes the following steps: S1. Log in to the high-temperature heat pipe structure design and simulation platform, which includes a three-dimensional simulation interface; S2. Complete the three-dimensional structural design of the high-temperature heat pipe in the three-dimensional simulation interface; The three-dimensional simulation interface includes a three-dimensional high-temperature heat pipe component structure library; Users can assemble high-temperature heat pipe components by dragging and dropping to form a three-dimensional high-temperature heat pipe structure; After selecting a three-dimensional high-temperature heat pipe component, the user can modify its structure; the modifications specifically include: After selecting the three-dimensional high-temperature heat pipe component, the user can enter the structural parameters of the three-dimensional high-temperature heat pipe component on the interface; After selecting a 3D high-temperature heat pipe component, users can add or delete components on the interface. The user inputs voice, which is a descriptive language for modifying the structure of three-dimensional high-temperature heat pipe components; Parse the user's voice input and generate text; The above text keywords will be extracted, including keywords related to component structural parts and modified variables; Convert the keywords of component structural parts into professional technical keywords of component structure; Based on the converted technical keywords for the component structure, link them to the modification slider; The structure of the three-dimensional high-temperature heat pipe components is modified using the aforementioned modified variables and modified sliders; S3. Construct a two-layer neural network model, train the model using high-temperature heat pipe samples from the experience base, and output the high-temperature heat pipe evaluation parameter array C1; specifically including: S31. Construct the first layer neural network model matrix based on the n three-dimensional structural parameters of the high-temperature heat pipe; The first layer neural network model matrix includes n neural network models; The step of constructing the first layer neural network model matrix based on the n three-dimensional structural parameters of the high-temperature heat pipe specifically includes: constructing a corresponding neural network model for each of the three-dimensional structural parameters, such that parameter i corresponds to neural network model i; The n neural network models are arranged in parallel; The three-dimensional structural parameters include, but are not limited to, working fluid type, working fluid filling rate, working tilt angle, suction core hole shape, and suction core hole shape parameters. The first layer neural network model matrix includes, but is not limited to, working fluid type neural network model, working fluid filling rate neural network model, working tilt angle neural network model, and suction core hole shape neural network model. The above three-dimensional structural parameters are used as input data for each neural network model, that is, the working fluid type, working fluid filling rate, working tilt angle, liquid suction core hole shape, and liquid suction core hole shape parameters are input into each neural network model. The weight values ​​of each input parameter are different in each neural network model; specifically, among all the input parameters of neural network model i, the weight of input parameter i is greater than the weight of other input parameters. The output data of each neural network model is an array of evaluation parameters for high-temperature heat pipes; S32. Construct the second-layer perceptual neural network model and connect it to the matrix of the first-layer neural network model; The second-layer perceptual neural network model is connected to the matrix of the first-layer neural network model, specifically by using the output data of each neural network model in the first-layer neural network model as the input data of the second-layer perceptual neural network model. S33. Initialize the parameters of each neural network model in the first layer neural network model matrix and the parameters of the second layer perceptual neural network model; S34. Obtain high-temperature heat pipe sample data from the experience database, the sample data including the three-dimensional structural parameters of the high-temperature heat pipe; set environmental parameter data; S35. Input the high-temperature heat pipe sample data and environmental parameter data from step S34 into each neural network model in the first layer neural network matrix; S36. The evaluation parameter array C1 of the high-temperature heat pipe is output by the second-layer perceptual neural network model; S4. Start the simulation platform, run the high-temperature heat pipe sample in step S3, and output the high-temperature heat pipe evaluation parameter array C2. S5. Iteratively update the parameters of the two-layer neural network model so that the similarity threshold between the high-temperature heat pipe evaluation parameter array C1 and the high-temperature heat pipe evaluation parameter array C2 is less than the preset threshold T, and finally form an optimized two-layer neural network model. S6. Utilize an optimized two-layer neural network model and simulation platform to realize the structural design and simulation calculation of high-temperature heat pipes.

2. The high-temperature heat pipe structure design and simulation method as described in claim 1, characterized in that, Step S4 specifically includes: S41. Obtain the high-temperature sample data from step S34, wherein the sample data includes the three-dimensional structural parameters of the high-temperature heat pipe; set environmental parameter data; S42. Start the simulation platform, run the high-temperature heat pipe sample in step S41, and output the high-temperature heat pipe evaluation parameter array C2.

3. The high-temperature heat pipe structure design and simulation method as described in claim 2, characterized in that, Step S5 specifically includes: S51. Calculate the similarity between the high-temperature heat pipe evaluation parameter array C1 and the high-temperature heat pipe evaluation parameter array C2. If the similarity is greater than the preset threshold T, proceed to step S52; otherwise, proceed to step S53. S52. Modify the parameters of the first-layer neural network model matrix and the second-layer perceptual neural network model, and re-execute steps S35-S36 and S51 until the similarity is less than the preset threshold T. The parameters of the modified first-layer neural network model matrix and the second-layer perceptual neural network model include weights, biases, and activation functions; S53. Obtain the optimized two-layer neural network model.

4. The high-temperature heat pipe structure design and simulation method as described in claim 3, characterized in that, Step S6 specifically includes: S61. After the three-dimensional structure design of the high-temperature heat pipe in step 2 is completed, the two-layer neural network model outputs an array of high-temperature heat pipe evaluation parameters for the designed three-dimensional structure of the high-temperature heat pipe. S62. If the high-temperature heat pipe evaluation parameters output in step S61 do not meet the design target value, the user modifies the three-dimensional structure design of the high-temperature heat pipe until the high-temperature heat pipe evaluation parameters output by the dual-layer neural network model meet the design target value, and obtains the three-dimensional structure of the high-temperature heat pipe that meets the design target value. S63. Start the simulation platform and run the three-dimensional structure of the high-temperature heat pipe that meets the design target value in step S62. Obtain the high-temperature heat pipe evaluation parameter array as the simulation calculation result of the high-temperature heat pipe.

Images