Aviation three-dimensional aided design integration method and system based on large language model

Abstract

The invention discloses an aviation three-dimensional aided design integration method and system based on a large language model. The method comprises the steps of obtaining multi-modal design demand data submitted by a user; analyzing the multi-modal design demand data submitted by the user; and calling a large language model to generate a corresponding modeling command according to the analyzed multi-modal design demand data, calling a computer aided design system by utilizing the modeling command to carry out modeling design, and optimizing the large language model at the same time. The system comprises a user interaction module for receiving multi-modal design demand data input of a user; the data analysis module is used for analyzing the multi-modal design demand data and extracting a design demand; the modeling instruction generation module is used for calling a large language model to generate a modeling instruction according to a design requirement; the modeling execution module is used for calling a computer aided design system API to complete modeling by utilizing the modeling instruction; and the model optimization module is used for feeding back the model data modified by the user to the large language model for optimization.

Description

Technical Field

[0001] The present invention belongs to the field of computer-aided design and artificial intelligence technology, and in particular relates to an aviation three-dimensional auxiliary design integration method and system based on a large language model. Background Art

[0002] Aviation 3D design is a complex and highly sophisticated process, typically requiring designers to possess deep expertise and extensive design experience. While traditional computer-aided design (CAD) tools enable parametric modeling and 3D design, their complex user interfaces and high learning curve often limit design efficiency and accuracy. Furthermore, the extensive design requirements involved in aviation design (such as the design of wings, fuselages, landing gear, and other components) require frequent model iteration and optimization, further increasing the design workload.

[0003] In recent years, large language models (LLMs) have achieved breakthroughs in natural language processing, task planning, and knowledge reasoning, opening up new possibilities for intelligent design tasks. However, there is currently no mature method to efficiently integrate LLMs with CAD systems to achieve automated modeling and optimization of complex design requirements. Summary of the Invention

[0004] This paper proposes an integrated method and system for aviation 3D-assisted design based on a large language model. This method establishes a data transmission interface between existing CAD systems and LLMs by designing a standard API middle layer. A standard file transfer format is specified to enable bidirectional data interaction between the CAD system and LLMs. The forward data flow captures design requirements from multimodal inputs such as text, images, and existing 3D models, and generates CAD modeling instructions through the LLM to complete model creation in the CAD system. The reverse data flow extracts user-modified model data from the CAD system and parses it into the LLM's standard training data format for fine-tuning and optimization of the model. This integrated method includes a user interaction module, a data parsing module, and a fine-tuning and optimization module, enabling bidirectional data interaction between the CAD system and LLMs.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] An integrated method for aviation 3D-aided design based on a large language model includes the following steps:

[0007] Step 1: Obtain multimodal design requirement data submitted by users, including text descriptions, pictures, and 3D models;

[0008] Step 2: Analyze the multimodal design requirement data submitted by the user;

[0009] Step three: call the large language model to generate corresponding modeling commands based on the parsed multimodal design requirement data, use the modeling commands to call the computer-aided design system for modeling and design, and optimize the large language model at the same time.

[0010] A further improvement of the present invention is that, in step 1, the specific method of obtaining the multimodal design requirement data submitted by the user includes:

[0011] receiving a text-formatted description of design requirements through a text input window in a user interface;

[0012] Receive design reference pictures provided by users through a picture upload window in the user interface;

[0013] The model upload window in the user interface receives existing 3D model files provided by the user in STL format.

[0014] A further improvement of the present invention is that, in step 2, the specific method for parsing the multimodal design requirement data submitted by the user includes:

[0015] Parse the text-formatted requirement description:

[0016] Pre-process the text requirements submitted by users, segment the complete text into independent phrases, and remove stop words that have no practical significance for the design task;

[0017] Create a professional terminology dictionary for the aviation design field, match and identify entity information in the text based on this dictionary, and extract component type, parameter name, design data, and design unit information;

[0018] Parse the requirements description of the image format:

[0019] Preprocess the input image, including adjusting the image size and resolution, using noise reduction and enhancement technology to improve image quality, and converting it into a grayscale image to facilitate subsequent edge contour information extraction;

[0020] Use the ResNet convolutional neural network to identify the geometric contour features in the current image, and use the Hough transform to detect the basic geometric elements in the geometric contour, including points, lines, and arcs;

[0021] Use OCR to identify text annotations in images to obtain dimension annotation information and convert it into corresponding design parameters;

[0022] Parse the requirement description of the 3D model format:

[0023] Read STL file data, extract vertex and triangle normal vector information contained in the file, and store it in a point cloud array;

[0024] Traverse all vertices and triangle face information, establish the connection relationship between triangle faces based on the angle threshold of adjacent normal vectors of triangle faces, reconstruct the model's geometric topology and extract key geometric feature information of points, lines, surfaces and bodies;

[0025] Traverse all line and surface elements in sequence, determine their spatial position relationship, and perform dimensional measurements to obtain design data of geometric features.

[0026] A further improvement of the present invention is that, in step 3, the specific method of calling the large language model to generate corresponding modeling commands according to the parsed multimodal design requirement data includes:

[0027] The parsed multimodal design requirement data is stored and represented in a unified JSON structured data format, including the design component name, geometric features, and feature design parameters. Natural language prompts are generated based on the structured data using the set encoding method, and the API interface is called to send it to the LLM, and the modeling instructions returned by the LLM are received.

[0028] A further improvement of the present invention is that, in step 3, the specific method of using the modeling command to call the computer-aided design system to perform modeling design includes:

[0029] Start and initialize the CAD system environment, configure the relevant working directory, unit system and file format;

[0030] Parse the modeling instructions generated by LLM and map them to the parametric modeling API of the CAD system, including: sketch creation, sketch drawing, feature stretching / removal, Boolean operations, and call the CAD system API to execute the corresponding modeling processes and commands in sequence, and update and display the created model through the CAD system graphics engine.

[0031] A further improvement of the present invention is that the natural language prompt specifically includes:

[0032] Textual description of design requirements; extracted geometric features; parametric design data; functional requirements and constraints of the target model.

[0033] A further improvement of the present invention is that, in step 3, the specific method of calling the computer-aided design system to optimize the large language model includes:

[0034] After the user modifies and adjusts the model created by the LLM modeling quality, the CAD system saves the current 3D model, saves the design requirement information submitted by the user in the current step one and the CAD model feature information modified by the user in a unified manner, and outputs them in a JSON structured data format as training data for model fine-tuning optimization. The API is called to use the Adam optimizer to send the exported training data to the large language model for fine-tuning, and the accuracy of the modeling instruction generation is evaluated using the BLEU and ROUGE indicators.

[0035] A further improvement of the present invention is that the parsing results of the multimodal design requirement data are expressed and stored in the JSON data format, and the JSON data format specifically includes:

[0036] Design object name; geometric feature description; parametric design data; material information.

[0037] A further improvement of the present invention is that the method is applicable to a three-dimensional aviation design scenario, including the following design tasks:

[0038] Parametric modeling of aircraft wings, fuselage, and landing gear; optimization design of aircraft internal structures; 3D modeling and assembly simulation of aircraft parts.

[0039] An integrated aviation 3D-aided design system based on a large language model, comprising:

[0040] User interaction module: receives multimodal design requirement data input from users, including text descriptions, pictures, and 3D models;

[0041] Data analysis module: parses multimodal design requirement data and extracts design requirements;

[0042] Modeling instruction generation module: calls the large language model to generate modeling instructions according to design requirements;

[0043] Modeling execution module: uses modeling instructions to call the computer-aided design system API to complete modeling;

[0044] Model optimization module: Feeds back user-modified model data to the large language model for optimization.

[0045] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0046] Aiming at the intelligent needs of aviation 3D design, the present invention defines a bidirectional data interaction interface between a computer-aided design (CAD) system and a large language model (LLM) through an API interface and a JSON structured data format. By parsing the multimodal design input submitted by the user, including text, images, and 3D models, the interface is parsed into structured design requirements and converted into natural language prompts for the LLM. The API enables the LLM to generate corresponding CAD modeling commands based on the design prompts, and the modified CAD modeling data is sent back to the LLM for optimization. This improves the efficiency of aviation 3D design to a certain extent and simplifies the design process. At the same time, reverse training optimization also improves the accuracy of the LLM's design instruction generation. By constructing a bidirectional data interaction interface between the CAD system and the LLM, the present invention realizes intelligent modeling and automatic model optimization in the CAD system. By using a dedicated dataset for aviation 3D design for training, the efficiency and accuracy of aviation 3D design modeling are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 This is an architectural diagram of an integrated system for aviation 3D-aided design based on a large language model.

[0049] Figure 2 This is a flowchart of an integrated method for aviation 3D-aided design based on a large language model.

[0050] Figure 3 A specific flow chart for data parsing required for multimodal design.

[0051] Figure 4 Flowchart for bidirectional data interaction between a computer-aided design system and a large language model.

[0052] Figure 5 This is a rendering of an embodiment of the present invention. DETAILED DESCRIPTION

[0053] Hereinafter, only certain exemplary embodiments are briefly described. As will be appreciated by those skilled in the art, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and description are to be considered as illustrative in nature and not restrictive.

[0054] In the description of the present invention, it is to be understood that when used in this specification and the appended claims, the terms “include” and “comprise” indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0055] It should also be understood that the terms used in the present specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0056] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0057] The accompanying drawings illustrate various schematic diagrams of structures according to embodiments disclosed herein. These figures are not drawn to scale; for clarity, some details are exaggerated and some details may be omitted. The shapes of the various regions and layers shown in the figures, as well as their relative sizes and positional relationships, are merely exemplary and may deviate in practice due to manufacturing tolerances or technical limitations. Those skilled in the art may design regions / layers with different shapes, sizes, and relative positions as needed.

[0058] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0059] Example 1

[0060] See also Figure 1 The present invention provides an aviation 3D auxiliary design integrated system based on a large language model, comprising:

[0061] User interaction module: receives multimodal design requirement data input from users, including text descriptions, pictures, and 3D models;

[0062] Data analysis module: parses multimodal design requirement data and extracts design requirements;

[0063] Modeling instruction generation module: calls the large language model to generate modeling instructions according to design requirements;

[0064] Modeling execution module: uses modeling instructions to call the computer-aided design system API to complete modeling;

[0065] Model optimization module: Feeds back user-modified model data to the large language model for optimization.

[0066] Example 2

[0067] See also Figure 2 The present invention provides an integrated method for aviation 3D auxiliary design based on a large language model, and the specific steps are as follows:

[0068] Step 1: Obtain multimodal design requirement data submitted by users, including text descriptions, pictures, and 3D models;

[0069] Before integrating the computer-aided design system with the large language model, the present invention needs to first obtain the user's multimodal design requirement data through the user graphical interface of the computer-aided system, including a descriptive text of the design object (such as "Design a passenger aircraft wing with a wingspan of 50m and a length of 40m"), a picture of the design object (JPEG format) or an existing 3D model file (STL format). The text is directly input through the software text box, and the picture and 3D model are uploaded through the file system.

[0070] Step 2: Analyze the multimodal design requirement data submitted by the user;

[0071] See also Figure 2 The data parsing module proposed in the present invention processes the multimodal design requirement data information submitted by the user to obtain specific design requirements. The specific process is as follows:

[0072] Parsing design requirements in text format includes: pre-processing the text requirements submitted by users, segmenting the complete text into independent phrases and removing stop words that have no practical significance for the design task; matching and identifying entity information in the text based on a professional terminology dictionary for the aviation design field, and extracting information such as component type, parameter name, design data and design unit.

[0073] Analysis of design requirements for image formats includes: preprocessing the input image, including adjusting the image size and resolution, using noise reduction and enhancement technology to improve image quality, and converting it into a grayscale image to facilitate subsequent edge contour information extraction; using aviation 3D design image data to train the ResNet convolutional neural network, and based on the trained model, identifying and extracting the geometric contour features in the current image, using Hough transform detection to detect basic geometric elements in the geometric contour, including points, lines, arcs, etc.; using OCR to identify text annotations in the image to obtain dimension annotation information, and convert it into corresponding design parameters.

[0074] Parsing the design requirements of the three-dimensional model format includes: reading STL file data, extracting the vertex and triangle normal vector information contained in the file, and storing it in a point cloud array; traversing all vertex and triangle information, establishing the connection relationship between triangles based on the angle threshold of adjacent normal vectors of triangles, reconstructing the model's geometric topology and extracting key geometric feature information such as geometric points, lines, surfaces, and bodies; traversing all line and surface elements in turn, determining their spatial position relationship, and performing dimensional measurements to obtain design data of geometric features.

[0075] See also Figure 3 The instruction generation module of the present invention stores and represents the parsed design requirements in a unified JSON structured data format, including the parsed core design requirements, including geometric parameters and material selection. The JSON data format includes the following content:

[0076] geometry: (object) contains parameters related to the 3D geometric design of the aircraft

[0077] wing: (object) Wing characteristic parameters:

[0078] airfoil_type: (string) Airfoil definition, e.g. "NACA0012".

[0079] span: (float) Wingspan length in meters.

[0080] chord_length: (float) chord length in meters.

[0081] twist_angle: (float) The twist angle of the airfoil, in degrees (°).

[0082] leading_edge_radius: (float) Leading edge curvature radius in meters.

[0083] surface_finish: (string) Surface finish type.

[0084] additional_features: (array) Contains additional 3D design features.

[0085] feature_type: (string) The type of feature.

[0086] quantity: (integer) The quantity of the feature.

[0087] location: (float) The spatial location of the feature.

[0088] parameters: (float) Feature creation parameters in meters.

[0089] Material: (object) contains parameters related to the aircraft material

[0090] type: (string) Material type.

[0091] density: (floating point number) material density in kg / m 3 .

[0092] yield_strength: (float) Yield strength of the material in MPa.

[0093] Step three: call the large language model to generate corresponding modeling commands based on the parsed multimodal design requirement data, use the modeling commands to call the computer-aided design system for modeling and design, and optimize the large language model at the same time.

[0094] The modeling instruction generation module uses the specified encoding method to generate natural language prompts based on the parsed multimodal design requirement data JSON structured data, which includes the design object name, geometric features, material information, etc. The format is as follows:

[0095] Please generate a 3D model of an aircraft wing according to the following design requirements:

[0096] 1. Name of design object: aircraft wing.

[0097] 2. Geometric features:

[0098] - Airfoil profile: 30 meters in length, 5 meters in width, 2 meters in leading edge camber, and 1.5 meters in trailing edge camber.

[0099] -Wing tip: Arc type with a radius of 0.5 meters.

[0100] 3. Material information:

[0101] -Material name: Aluminum alloy.

[0102] -Material density: 2.7g / cm 3 .

[0103] -Tensile strength: 450MPa.

[0104] Please ensure that the model design meets the above geometric features, material properties and functional requirements, and generate the corresponding modeling instructions.

[0105] Call the API interface to send the above prompt information to LLM and receive the modeling instructions returned by LLM; after the modeling module parses the modeling instructions and calls the CAD system to perform the modeling operation, the modified and adjusted data is sent to LLM through the model optimization module for model fine-tuning and optimization.

[0106] See also Figure 4 The specific process of bidirectional data exchange between the computer design-aided system and the large language model in the present invention is as follows:

[0107] The modeling execution module of the present invention starts and initializes the CAD system environment, configures the relevant working directory and unit system, and maps the modeling execution module to the parametric modeling API of the CAD system, including: sketch creation, sketch drawing, feature stretching / removal, Boolean operations, etc., and calls the CAD system API to execute the corresponding modeling processes and commands in sequence, and updates and displays the created model through the CAD system graphics engine; after the user modifies and adjusts the model created by the LLM modeling quality, the model optimization module of the present invention saves the current three-dimensional model of the CAD system, and uniformly saves the design requirement information submitted by the user in the current step one and the CAD model feature information modified by the user into a JSON structured data format as training data for model fine-tuning optimization, calls the API to use the Adam optimizer to send the exported training data to the large language model for fine-tuning, and combines the BLEU and ROUGE indicators to evaluate the accuracy of the modeling instruction generation.

[0108] Example 3

[0109] See also Figure 5 This embodiment takes the design of an aircraft wing structure as an example to specifically illustrate the steps and effects of the method of the present invention in practical applications. The design of an aircraft wing structure needs to comprehensively consider multiple factors such as strength, stiffness, weight, and manufacturability while meeting the aerodynamic shape. Traditional design methods are highly dependent on engineering experience and have long iteration cycles. Taking the design of the main load-bearing structure of the wing of a high-subsonic unmanned reconnaissance aircraft as an example, it is required to meet the strength and stiffness requirements under specific flight loads. The traditional design method requires a senior engineer with many years of design experience to perform manual design and iterative optimization through traditional CAD software modeling and CAE simulation analysis software.

[0110] Using the method of the present invention, first submit multimodal design requirement data to LLM, including Figure 5Enter a text description in area A (for example: Design the wing structure of an unmanned reconnaissance aircraft. Main parameters: wingspan 12m, average aerodynamic chord 1.5m, airfoil NACA 64-215, cruise Mach number 0.75, maximum overload 2.5g, minimum overload -1g. Material selected: 7075-T6 aluminum alloy. A double-beam single-compartment structure is desired, with a rib spacing of approximately 0.5m). Figure 5 Image input for area B (a conceptual sketch of the wing planform, spar, and rib distribution);

[0111] Secondly, the method of the present invention parses the multimodal design requirements input by the user and extracts the design requirements as unified JSON structured data, which includes:

[0112]

[0113] Finally, the JSON structured data containing the design requirements is sent to LLM, which generates the corresponding CAD modeling instructions and a series of parameterized CAD modeling instructions for creating the wing structure and skin and automatically generates a preliminary 3D model of the wing internal structure and Figure 5 The middle C area shows that the designer reviews the initial structural model generated by LLM in the CAD system and manually adjusts the spar web height, rib thickness, and addition of stiffeners based on the preliminary results. The adjusted structural parameters are reconverted into training data in JSON format and fed back to LLM for fine-tuning. For example, if the designer repeatedly changes the standard T-section ribs generated by LLM to I-sections in a specific load area to improve bending stiffness, LLM will give priority to I-sections in subsequent similar tasks. Through user feedback and fine-tuning optimization of a large number of aviation structure design cases and rules, LLM's BLEU / ROUGE indicator evaluation scores have been significantly improved, and its ability to understand complex structural design requirements and generate accurate and efficient modeling instructions has been continuously enhanced. This accumulation and evolution of knowledge for specific design fields is unmatched by traditional tools.

[0114] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, from all points of view, the embodiments should be regarded as illustrative and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and range of equivalents of the claims are included in the present invention. Any reference signs in the claims should not be construed as limiting the claim to which they relate.

[0115] In addition, it should be understood that although this specification describes the embodiments, not every embodiment contains only one independent technical solution. This description is for clarity only. Those skilled in the art should consider the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for the purpose of illustrating the technical concept of the present invention and cannot be used to limit the scope of protection of the present invention. Any changes made based on the technical solution in accordance with the technical concept proposed by the present invention fall within the scope of protection of the claims of the present invention.

Claims

1. An integrated method for aviation 3D-aided design based on a large language model, characterized in that: The following steps are involved: Step 1: Obtain multimodal design requirement data submitted by users, including text descriptions, pictures, and 3D models; Step 2: Analyze the multimodal design requirement data submitted by the user; Step three: call the large language model to generate corresponding modeling commands based on the parsed multimodal design requirement data, use the modeling commands to call the computer-aided design system for modeling and design, and optimize the large language model at the same time.

2. The aviation 3D-aided design integration method based on a large language model according to claim 1, characterized in that: In step 1, the specific method for obtaining the multimodal design requirement data submitted by the user includes: receiving a text-formatted description of design requirements through a text input window in a user interface; Receive design reference pictures provided by users through a picture upload window in the user interface; The model upload window in the user interface receives existing 3D model files provided by the user in STL format.

3. The aviation 3D-aided design integration method based on a large language model according to claim 2, characterized in that: In step 2, the specific method for parsing the multimodal design requirement data submitted by the user includes: Parse the text-formatted requirement description: Pre-process the text requirements submitted by users, segment the complete text into independent phrases, and remove stop words that have no practical significance for the design task; Create a professional terminology dictionary for the aviation design field, match and identify entity information in the text based on this dictionary, and extract component type, parameter name, design data, and design unit information; Parse the requirements description of the image format: Preprocess the input image, including adjusting the image size and resolution, using noise reduction and enhancement technology to improve image quality, and converting it into a grayscale image to facilitate subsequent edge contour information extraction; Use the ResNet convolutional neural network to identify the geometric contour features in the current image, and use the Hough transform to detect the basic geometric elements in the geometric contour, including points, lines, and arcs; Use OCR to identify text annotations in images to obtain dimension annotation information and convert it into corresponding design parameters; Parse the requirement description of the 3D model format: Read STL file data, extract vertex and triangle normal vector information contained in the file, and store it in a point cloud array; Traverse all vertices and triangle face information, establish the connection relationship between triangle faces based on the angle threshold of adjacent normal vectors of triangle faces, reconstruct the model's geometric topology and extract key geometric feature information of points, lines, surfaces and bodies; Traverse all line and surface elements in sequence, determine their spatial position relationship, and perform dimensional measurements to obtain design data of geometric features.

4. The aviation 3D-aided design integration method based on a large language model according to claim 3, characterized in that: In step 3, the specific method of calling the large language model to generate corresponding modeling commands based on the parsed multimodal design requirement data includes: The parsed multimodal design requirement data is stored and represented in a unified JSON structured data format, including the design component name, geometric features, and feature design parameters. Natural language prompts are generated based on the structured data using the set encoding method, and the API interface is called to send it to the LLM, and the modeling instructions returned by the LLM are received.

5. The aviation 3D-aided design integration method based on a large language model according to claim 3, characterized in that: In step 3, the specific method of using the modeling command to call the computer-aided design system to perform modeling and design includes: Start and initialize the CAD system environment, configure the relevant working directory, unit system and file format; Parse the modeling instructions generated by LLM and map them to the parametric modeling API of the CAD system, including: sketch creation, sketch drawing, feature stretching / removal, Boolean operations, and call the CAD system API to execute the corresponding modeling processes and commands in sequence, and update and display the created model through the CAD system graphics engine.

6. The aviation 3D-aided design integration method based on a large language model according to claim 5, characterized in that: The natural language prompt specifically includes: Textual description of design requirements; extracted geometric features; parametric design data; functional requirements and constraints of the target model.

7. The aviation 3D-aided design integration method based on a large language model according to claim 3, characterized in that: In step 3, the specific method of using the computer-aided design system to optimize the large language model includes: After the user modifies and adjusts the model created by the LLM modeling quality, the CAD system saves the current 3D model, saves the design requirement information submitted by the user in the current step one and the CAD model feature information modified by the user in a unified manner, and outputs them in a JSON structured data format as training data for model fine-tuning optimization. The API is called to use the Adam optimizer to send the exported training data to the large language model for fine-tuning, and the accuracy of the modeling instruction generation is evaluated using the BLEU and ROUGE indicators.

8. The aviation 3D-aided design integration method based on a large language model according to claim 3 is characterized in that: The results of the multimodal design requirement data analysis are expressed and stored in the JSON data format, which specifically includes: Design object name; geometric feature description; parametric design data; material information.

9. The aviation 3D-aided design integration method based on a large language model according to claim 1, characterized in that: The method is applicable to aviation 3D design scenarios, including the following design tasks: Parametric modeling of aircraft wings, fuselage, and landing gear; optimization design of aircraft internal structures; 3D modeling and assembly simulation of aircraft parts.

10. An aviation 3D auxiliary design integrated system based on a large language model, characterized by: include: User interaction module: receives multimodal design requirement data input from users, including text descriptions, pictures, and 3D models; Data analysis module: parses multimodal design requirement data and extracts design requirements; Modeling instruction generation module: calls the large language model to generate modeling instructions according to design requirements; Modeling execution module: uses modeling instructions to call the computer-aided design system API to complete modeling; Model optimization module: Feeds back user-modified model data to the large language model for optimization.

Images