3D modeling calculation auxiliary system based on natural language driving

Abstract

The invention discloses a 3D modeling calculation auxiliary system based on natural language driving, in the system, a basic information input unit is used for receiving a modeling request input by a user, and the modeling request comprises description of specific scene modeling requirements of the user in a natural language form; the semantic recognition unit performs semantic recognition on natural language input information to extract structural modeling calculation parameters, and performs standardized output; a corresponding modeling tool flow is called through a 3D modeling unit according to the semantic recognition result of the large model, and parametric modeling is carried out; the finite element calculation unit can perform data interaction with the 3D modeling unit, geometric information established by the 3D modeling unit is input into finite element calculation software, calculation parameter structures serve as other parameters to be input into the finite element calculation software, corresponding finite element models are established according to the calculation parameter structures, and solving calculation is performed. According to the system, a hybrid architecture of natural language semantic link parameterized workflow is adopted, so that the learning cost of parameterized modeling, finite element calculation and special calculation requirements can be effectively reduced, and the efficiency of modeling and calculation is improved.

Description

Technical Field

[0001] The present invention relates to the field of construction engineering technology, and in particular to a three-dimensional (3D) modeling and structural calculation and analysis solution driven by natural language processing (NLP) technology. Background Art

[0002] With the increasing complexity of civil engineering structures, 3D modeling technology has been widely used in architectural and structural design.

[0003] In existing 3D modeling technologies, users typically use mainstream modeling tools or platforms to model specific engineering structures, then import them into finite element calculation programs for calculations. Some commercial modeling software also includes calculation modules. For modeling purposes, users often manually model and calculate specific structures by dragging and clicking with a mouse, or manually write program code to execute modeling scripts to achieve rapid modeling.

[0004] However, in existing 3D modeling technologies, when the specific structure to be built changes, users are required to manually adjust or rewrite the corresponding modeling code to meet the new modeling requirements. This process is cumbersome, time-consuming and labor-intensive.

[0005] To this end, parametric modeling technology is currently widely used to achieve rapid modeling of typical structures. However, even if parametric modeling tools are adopted, the following pain points still need to be addressed: 1) Parametric tools still require users to master basic knowledge such as data structures and visual programming, and the learning cost is relatively high; 2) Taking the typical parametric modeling platform Grasshopper as an example, it mainly focuses on the establishment of complex 3D models, and its computing power still has a large gap with conventional finite element calculation programs in terms of specification adaptation and calculation accuracy.

[0006] Furthermore, in the field of construction, the current application of 3D modeling technology still has the following problems: 1) When it comes to the calculation of common structures such as scaffolding, in addition to modeling, commercial finite element software in the civil engineering field still requires users to master basic knowledge such as the basic principles of finite elements, which is still difficult for general construction personnel to master; 2) Currently, most civil engineering software is closed-source, and external calls to various software are difficult. In addition to finite element software, different software still needs to be selected for specific construction problems, and users need to understand the calculation requirements of construction scenarios and the adaptation of specific software.

[0007] Furthermore, the current technical path is to understand user natural language input based on a large language model and then directly generate modeling and calculation code for the corresponding software platform. However, direct code generation is affected by a series of uncertain factors such as user input, code rules, and model training corpus, and often contains errors, and the modeling and calculation are not robust enough.

[0008] It can be seen that how to simplify the modeling process, lower the professional threshold, and achieve deep integration of modeling and computational analysis while ensuring the improvement of the robustness of modeling and computation is an urgent problem that needs to be solved in this field. Summary of the Invention

[0009] In response to the problems existing in the existing 3D modeling and calculation solutions in the field of construction engineering, the purpose of the present invention is to provide a 3D modeling and calculation assistance system driven by natural language. This system adopts a hybrid architecture of natural language semantically linked parametric workflows, which can effectively reduce the learning cost of parametric modeling, finite element calculations and special computing requirements, and achieve improved modeling and calculation efficiency.

[0010] In order to achieve the above-mentioned object, the present invention provides a natural language driven 3D modeling computer-assisted system, comprising:

[0011] A basic information input unit, the basic information input unit is used to receive a modeling request input by a user, the modeling request including a description of a specific scenario modeling requirement by the user in a natural language form;

[0012] A semantic recognition unit, which is based on the semantic recognition processing module of the AI ​​large model and can perform semantic recognition, extract structural modeling, calculate parameters, and perform standardized output based on natural language input information;

[0013] The 3D modeling unit interacts with the semantic recognition unit data and can call the corresponding modeling tool flow to perform parametric modeling based on the semantic recognition results of the large model;

[0014] The finite element calculation unit can interact with the 3D modeling unit data, input the geometric information established by the 3D modeling unit into the finite element calculation software, input the calculation parameter structure as the remaining parameters into the finite element calculation software, and establish the corresponding finite element model based on this, and perform solution calculations.

[0015] Furthermore, the semantic recognition processing module based on the AI ​​large model extracts and recognizes specific tasks based on preset structured prompt words for natural language input information, and extracts structural modeling calculation parameters.

[0016] Furthermore, the structured prompt words include:

[0017] A user role setting unit, wherein the user role setting unit is used to set basic identity information for the AI ​​large model;

[0018] A task processing logic setting unit, which sets task identification, parameter identification and standard output logic;

[0019] A standard output unit is used to integrate and output specific parameters in the format of "variable name = specific value".

[0020] Furthermore, the basic information input unit is configured with a first information interaction window for receiving a natural language description input by a user.

[0021] Furthermore, the basic information input unit is provided with a second information interaction window for storing structured prompt words of the large model.

[0022] Furthermore, the basic information input unit is provided with a third information interaction window for feeding back the large model information extraction content.

[0023] Furthermore, the basic information input unit is provided with a fourth information interaction window for inputting the AI ​​large model and finite element calculation software authorization key (API Key) and uniform resource locator (URL).

[0024] Furthermore, the basic information input unit is provided with a fifth information interaction window, which is used as an external expansion window supporting the calling of external exe files.

[0025] Furthermore, the 3D modeling unit performs keyword recognition and segmentation processing on the structural modeling calculation parameters output by the basic information input unit through a keyword segmentation module.

[0026] Furthermore, the 3D modeling computing assistance system also includes an external program calling unit, which is configured to call a corresponding external program based on a program identification code extracted by the basic information input unit through semantic recognition of natural language input information and a preset program address.

[0027] The solution provided by the present invention has the following beneficial technical effects compared with the prior art:

[0028] (1) Reduce learning costs: Engineering technicians only need to use non-standardized natural language input to achieve modeling and calculation, without having to master complex professional software operations, which greatly reduces the learning costs of front-line construction personnel.

[0029] (2) Improve modeling and computing efficiency: The built-in general large model can autonomously call the parametric modeling workflow based on the understanding of natural language, realizing a complete closed loop of common structure intelligent modeling and finite element calculation, and significantly improving modeling and computing efficiency.

[0030] (3) Support for functional expansion: It can flexibly access special verification programs such as roof travel verification and support natural language calling to further enhance the system's computing capabilities. At the same time, it is compatible with Grasshopper's existing parametric modeling resources, providing reliable support for the iterative upgrade of subsequent engineering scenario applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The present invention is further described below with reference to the accompanying drawings and specific embodiments.

[0032] Figure 1 This is a schematic diagram of the composition of the natural language-driven 3D modeling computer-assisted system of the present invention;

[0033] Figure 2 Schematic diagram of the structured prompt words in the present invention;

[0034] Figure 3 This is a schematic diagram of keyword segmentation processing of a standard text in the present invention;

[0035] Figure 4 Schematic diagram of the natural language-driven parametric modeling calculation process in the present invention;

[0036] Figure 5 A flow chart of the natural language-based external program call in the present invention;

[0037] Figure 6 This is an example diagram of the user input interface in Example 1 of the present invention;

[0038] Figure 7 This is an example diagram of the natural language call parameterized modeling calculation for the spherical shell in Example 1 of the present invention;

[0039] Figure 8 This is an example diagram of the natural language call parameterized modeling calculation for the scaffold in Example 1 of the present invention;

[0040] Figure 9 This is an example diagram of calling an external program in natural language in Example 2 of the present invention. DETAILED DESCRIPTION

[0041] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below with reference to specific illustrations.

[0042] First, let’s explain the terms involved in this application:

[0043] (1) Large language model: A deep learning model trained using a large amount of text data that can understand the meaning of natural language text and generate and output corresponding text based on user settings.

[0044] (2) Structured prompts: Instructions or frameworks used to guide language models to output specific types of results. They constrain the model’s behavior through clear formats, logic, or rules, making the generated content more in line with user needs.

[0045] (3) Parametric modeling: A series of parameter-driven automated modeling processes that enable rapid iteration and optimization of design models by defining variables, constraints, and logical relationships. It associates design elements with parameters (such as dimensions, material properties, loads, etc.) and automatically generates new design versions by adjusting parameters. The most typical parametric tool flow is implemented using Grasshopper, a parametric modeling plug-in in Rhino 3D software. The "Battery Pack" is a component library built into Grasshopper or customized by the user. These components (called "batteries") encapsulate specific functions (such as geometry generation, structural analysis, data processing, etc.). Users can build complex parametric tool flows by dragging and connecting these components.

[0046] (4) Finite Element Analysis (FEA): The finite element method (FEA) is a numerical calculation method that discretizes a complex structure into multiple simple units (such as beams, plates, and shells), establishes mechanical equations for each unit, and then analyzes the stress, strain, displacement, and other responses of the structure through overall solution. It is a core tool for structural analysis and design in civil engineering.

[0047] In response to the problems existing in the existing 3D modeling and calculation solutions in the field of construction engineering in terms of modeling process simplification and robustness of modeling and calculation, the present invention innovatively adopts a hybrid architecture of natural language semantically linked parametric workflows to form a 3D modeling and calculation assistance system driven by natural language. This 3D modeling and calculation assistance system handles the relationship between user needs and specific execution in a robust manner, reduces the learning cost of parametric modeling, finite element calculation and special calculation needs, and achieves improved efficiency in modeling and calculation.

[0048] Specifically, the natural language driven 3D modeling calculation auxiliary system mainly includes four functional units: basic information input unit, semantic recognition unit, 3D modeling unit, finite element calculation unit and external program calling unit.

[0049] The basic information input unit in the system is used to receive a modeling request input by a user, wherein such a modeling request includes a description of the user's specific scenario modeling requirements in a natural language.

[0050] The semantic recognition unit in the system is based on the semantic recognition processing module of the AI ​​large model. It can perform semantic recognition on natural language input information, extract structural modeling calculation parameters and / or program identification codes, and output them in a standardized manner.

[0051] The 3D modeling unit and semantic recognition unit in the system interact with each other and can call the corresponding modeling tool flow to perform parametric modeling based on the semantic recognition results of the large model;

[0052] The finite element calculation unit in the system can interact with the finite element calculation unit data, input the geometric information established by the 3D modeling unit into the finite element calculation software, input the calculation parameter structure as the remaining parameters into the finite element calculation software, and establish the corresponding finite element model based on this, and perform solution calculations.

[0053] The external program calling unit in the system is configured to call a corresponding external program according to a program identification code extracted by the basic information input unit through semantic recognition of natural language input information and a preset program address.

[0054] The following describes the specific implementation scheme of each functional unit in this 3D modeling computer-aided system and the component devices that may be involved.

[0055] See also Figure 1 , which is a schematic diagram of the application scenario of the natural language-driven 3D modeling computer-assisted system.

[0056] like Figure 1 As shown, the 3D modeling calculation auxiliary system includes a terminal device 100 and a server 200 when in use. The server 200 is deployed with a demand processor 210, a parametric modeling platform 220, a finite element calculation platform 230 and other special software 240.

[0057] The basic information input unit running in the terminal device 100 can obtain natural language input information.

[0058] Furthermore, the basic information input unit running on the terminal device 100 is further configured with a plurality of information interaction windows, wherein the first information interaction window (such as the AI-cscec8 window) is used to receive a natural language description input by the user;

[0059] The second information interaction window (such as the sys_prompt window) is used to store structured prompt words of the general large model;

[0060] The third information interaction window is used as a large model information extraction window to feedback the large model information extraction content;

[0061] The fourth information interaction window is used to input the AI ​​large model and finite element calculation software authorization key (API Key) and uniform resource locator (URL);

[0062] The fifth information interaction window is used as an external expansion window to support the call of external exe files.

[0063] The demand processor 210 is deployed and runs on the server 200. It performs semantic recognition on the natural language input information based on the semantic recognition processing module of the AI ​​large model to extract structural modeling calculation parameters and / or program identification code, and outputs them in a standardized manner.

[0064] Furthermore, the semantic recognition processing module based on the AI ​​large model here can extract and recognize specific tasks based on preset structured prompt words for natural language input information, and extract structural modeling calculation parameters and / or program identification codes.

[0065] See also Figure 2 , which shows an example of the structured prompt word. The structured prompt word here mainly includes three parts:

[0066] (1) Use the role setting unit to set basic identity information for the AI ​​model, which helps the output of the AI ​​model to be closer to the target domain. This application is set as a professional civil engineering engineer;

[0067] (2) Task processing logic setting unit, which can set task identification, parameter identification and standard output logic.

[0068] As an example, the task identification logic here can be used to identify Task 1: modeling calculation task and Task 2: program call task.

[0069] Under Task 1: Modeling Calculation Task, further set the structure type identification logic, and further set the structure parameter identification logic under the structure type identification logic.

[0070] Furthermore, the structural parameter identification logic here includes geometric parameters, other parameters, fault tolerance rules, unit conversion, etc.

[0071] Under Task 2: Program Call Task, a program type identification logic is further set. Under the program type identification logic, multiple sets of external program identification logic are specifically set, including program identification information and program path information.

[0072] The standard output unit here integrates and outputs the specific parameters generated by the task processing logic setting unit identification processing in the format of "variable name = specific value" so that it can be recognized and segmented by the keyword segmenter of the next process.

[0073] The 3D modeling unit is run in the parametric modeling platform 220 deployed on the server 200, and can call the corresponding modeling tool flow to perform parametric modeling based on the semantic recognition results of the modeling requirements by calling the large model by the requirement processor 210.

[0074] Here, the keyword segmentation module is used to perform keyword recognition and segmentation processing on the structural modeling calculation parameters output by the semantic recognition processing module of the AI ​​large model.

[0075] See also Figure 3 ,After receiving the standardized format output by the semantic recognition ,processing module of the AI ​​large model, the keyword segmentation module here ,uses regular expressions to segment the output text, establishes specific ,variable names, assigns corresponding modeling and calculation parameters, and ,outputs them, including modeling type variables with structural type codes as ,assignment content.

[0076] On this basis, the 3D modeling unit further calls the corresponding modeling tool flow according to the structure type code. The tool flow only needs the necessary modeling parameters to realize 3D modeling. The tool flow reads the modeling parameters output by the keyword recognition and segmentation module to complete the parametric modeling.

[0077] For example, the 3D modeling unit can use the Grasshopper workflow based on the semantic recognition results of the model to perform parametric modeling in Rhino. Grasshopper is a built-in parametric modeling tool in Rhino 3D modeling software. It integrates modeling commands in the form of battery packs and supports Python and C# languages ​​for custom battery development.

[0078] A finite element calculation unit is run in the finite element calculation platform 230 deployed on the server 200. The finite element calculation unit can interact with the 3D modeling unit data running in the parametric modeling platform 220, input the geometric information established by the 3D modeling unit into the finite element calculation software, and input the calculation parameter structure as the remaining parameters into the finite element calculation software, thereby establishing a corresponding finite element model and performing solution calculations.

[0079] For example, the finite element calculation unit here can perform corresponding bar structure linkage calculations based on natural language after connecting to the API service. This can be done through the API tab in the Tools menu. After successful connection, the corresponding URL and api_key will appear in the information window and need to be entered in the corresponding window of the basic information input terminal.

[0080] An external program calling unit is run in other dedicated software 240 deployed on the server 200. The external program calling unit can call a corresponding external program based on the program identification code extracted by semantic recognition of natural language input information by the demand processor 210 and the preset program address.

[0081] It should be noted here that although Figure 1 Only one terminal device and one server are shown in FIG, but it should be understood that there may be two or more terminal devices and servers.

[0082] When the natural language-driven 3D modeling computing-assisted system is in operation, the terminal device 100 is primarily used to receive modeling requests input by the user and send the modeling requests in real time to the demand processor 210 deployed on the server 200. The modeling request includes the user's description of the specific scene modeling requirements in natural language.

[0083] When the server-based demand processor 210 receives input from a terminal device, it first extracts the corresponding task requirements and specific parameters from the natural language. Based on the specific task, it then transmits key parameters to various executors deployed on the server, such as the modeling platform 220, the finite element calculation platform 230, and other tools 240, to fulfill the user's needs. After modeling and calculations are completed, the corresponding executors deployed on the server return the 3D scene to the terminal device in real time for the user to view.

[0084] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0085] See also Figure 4 , which shows a flow chart of implementing parametric modeling calculation based on natural language drive by the natural language-based modeling processing method of the 3D modeling calculation auxiliary system provided by the present invention.

[0086] Based on the diagram, the process of the natural language-based modeling processing method here can include the following three main processing stages:

[0087] (1) Semantic recognition stage:

[0088] 1) The user enters his or her modeling and computing requirements text in natural language.

[0089] 2) The semantic recognition processor (i.e., processing module) based on the AI ​​large model accepts the modeling text input by the user. Since the user uses natural language input, there is a certain ambiguity in the language. The processor needs to extract and recognize specific tasks as much as possible based on the structured prompt words preset by the user, extract the structural modeling calculation parameters, and output them in a specified standardized format.

[0090] Among them, semantic recognition based on AI big model is the core, here we use Figure 2 The structured prompt words shown mainly include three parts: 1) Using role settings to set basic identity information for the AI ​​large model, which helps the output of the AI ​​large model to move closer to the target domain. This application is set as a professional civil engineering engineer; 2) Setting the logical relationship of task processing. This application constructs three links in this link: task identification, parameter identification and standardized output.

[0091] For example, the AI ​​big model will first identify whether the user's target requirement is modeling calculation. If so, it will enter Task 1. After entering Task 1, it will first identify the specific structural type input by the user, such as a square dome. Then, it will identify specific structural parameters, such as geometric parameters such as height loss and span, as well as other calculation parameters such as cross-section type, material properties, and load. Based on the fault tolerance rules, it will unify the units and complete the necessary missing information:

[0092] 3) According to the output specifications, the specific parameters are integrated and output in the format of "variable name = specific value" so that they can be recognized and segmented by the keyword segmenter of the next process.

[0093] (2) 3D modeling stage:

[0094] 1) First, the keyword segmentation module is used to perform keyword recognition and segmentation processing on the standardized data output from the semantic recognition stage.

[0095] Combine Figure 3 As shown, after the keyword segmentation module accepts the standardized format output by the semantic recognition processor, it uses regular expressions to segment the output text, establish specific variable names, assign corresponding modeling and calculation parameters, and output, which includes modeling type variables with structural type codes as the assignment content.

[0096] 2) According to the structure type code, the corresponding modeling tool flow is called. The tool flow only needs the necessary modeling parameters to realize 3D modeling. The tool flow reads the modeling parameters output by the keyword recognition and segmentation module to complete the parametric modeling.

[0097] (3) Model calculation order:

[0098] The geometric information established by the 3D modeling module is input into the finite element calculation software, and the calculation parameter structure divided by keyword recognition is input into the finite element calculation software as the remaining parameters. The above information can be integrated to establish the corresponding finite element model, and the corresponding calculation results can be obtained after solving.

[0099] See also Figure 5, which is a flow chart showing the implementation of the natural language-based external program call by the natural language-based modeling processing method of the 3D modeling calculation-assisted system provided by the present invention.

[0100] based on Figure 5 As shown in the figure, the implementation process of natural language-based external program calls can include the following three main stages:

[0101] (1) Semantic recognition stage: Similar to the implementation path of parametric modeling calculation driven by natural language, but the logical chain of recognition has changed. According to the natural language input by the user, the recognition task is "Task 2: Program call". According to the external program information built into the prompt word (i.e., a brief description of the external program software function, application scenario and corresponding software identification code), the special calculation program identification code corresponding to the user input scenario is output in a standardized format.

[0102] (2) External program call: According to the program identification code, the corresponding special program is called up. All program addresses to be referenced have been stored in this module in advance.

[0103] In order to further illustrate the solution of the present invention, the following uses specific examples to illustrate the use of this application and the data transmission process.

[0104] Prior to implementation, the natural language-driven 3D modeling computing assistance system solution provided by the present invention can be implemented as a corresponding software program, forming corresponding natural language-driven 3D modeling computing assistance software. When running, this software program will execute the above-mentioned natural language-driven 3D modeling computing assistance method and store it in a corresponding storage medium for processor access and execution.

[0105] Example 1: Natural language driven parametric modeling calculation

[0106] The steps and data transfer process for natural language-driven parametric modeling calculations in this example are as follows:

[0107] (1) Step 1:

[0108] Start the software (open Rhino software in sequence, start Grasshopper in Rhino, and then open this system file such as CSCEC8AI.gh in Grasshopper); open the finite element calculation software (such as Midas Gen), connect to the network key, obtain the software key and URL; fill in the general large model key, finite element calculation software key and URL in the corresponding window of this system), the user input interface is as follows Figure 6 shown.

[0109] (2) Step 2:

[0110] In the AI-cscec8 window on the interface, enter the natural language related to the model to be built, such as Figure 6 shown.

[0111] (3) Step 3:

[0112] ① The program first automatically identifies the structural parameters of natural language according to the prompt words and outputs the corresponding text. Figure 7 Taking the natural language information entered as an example ("Build a linked square lattice shell with latitudinal members, a span of 72m, a height of 23.4m, 13 concentric rings, 60 circumferential sections, using a box section B200x100x8x8, Q355, and a node live load of 2.3kN downward"), the standard text output after semantic recognition processing is: "L=72000, f=23400, m=60, n=13, typeG=1, mal=Q355, H=200, B=100, TW=8, TF=8, LLload=-2.300, DLload=-0.001, structure type code=LFG", which shows that the units have been unified to kN / mm.

[0113] ②After receiving the string, the keyword segmenter in the system matches the characters through regular expressions, creates the corresponding variable name, and assigns the corresponding value, thus completing the conversion of the string into a variable;

[0114] ③ The keyword splitter identifies the structural type code LFG and invokes the corresponding parametric modeling tool flow for the linked square dome. This parametric modeling tool flow reads the relevant parameters, namely L = 72000, f = 23400, m = 60, n = 13, and typeG = 1, and creates a 3D line model of the structure in professional 3D modeling software such as Rhino.

[0115] ④ The 3D line model is automatically imported into the corresponding finite element software as geometric information. At the same time, the finite element software reads another part of the variables, namely mal = Q355, H = 200, B = 100, TW = 8, TF = 8, LLload = -2.300, DLload = -0.001 as supplementary calculation parameters. The finite element model is established with Q355 steel, box section B200x100x8x8 and a node live load of 2.3kN. The boundary condition defaults to the dome bottom hinge, and the self-weight load is added by default.

[0116] ⑤After the finite element model is established, the calculation is automatically performed and the user can view the calculation results in the finite element software interface.

[0117] Similarly, scaffolding can be implemented through the same architecture ( Figure 8 ) and other typical engineering structures modeling and calculation.

[0118] Example 2: External program calls supporting natural language

[0119] The steps involved in calling an external program in natural language and the data transfer process in this example are as follows:

[0120] (1) Step 1:

[0121] Start the software (open Rhino software, start Grasshopper in Rhino, and then open this system file such as CSCEC8AI.gh in Grasshopper. You only need to call external programs without starting the finite element calculation software);

[0122] (2) Step 2:

[0123] Taking the floor trolley as an example, fill in the path of the program exe executable file in the external expansion window.

[0124] (3) Step 3: Input the natural language related to the floor slab carriage in the AI-cscec8 window of the system operation interface ("We are currently entering the basement top slab construction stage and need to solve the floor slab carriage problem"), such as Figure 9 shown.

[0125] (4) Step 4:

[0126] The system program automatically recognizes the meaning of natural language, understands user needs, automatically matches the required software according to the information of the access software, and outputs the standardized text: "ZC". After the keyword segmenter receives the text, it calls up the corresponding program and opens the corresponding program. Figure 9 shown.

[0127] Similarly, this framework can be used to call special computing software when permissions allow.

[0128] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products of the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0129] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0130] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0131] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0132] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0133] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0134] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0135] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0136] The aforementioned methods, or specific system units, or portions thereof, of the present invention are purely software-based and can be implemented as program code on physical media, such as a hard drive, optical disk, or any electronic device (e.g., a smartphone or computer-readable storage medium). When a machine loads and executes the program code (e.g., a smartphone), the machine becomes a device for implementing the present invention. The aforementioned methods and devices of the present invention can also be transmitted in program code form via some transmission medium, such as a cable, optical fiber, or any other transmission method. When the program code is received, loaded, and executed by a machine (e.g., a smartphone), the machine becomes a device for implementing the present invention.

[0137] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A 3D modeling computer-aided system driven by natural language, characterized in that: include: A basic information input unit, the basic information input unit is used to receive a modeling request input by a user, the modeling request including a description of a specific scenario modeling requirement by the user in a natural language form; A semantic recognition unit, which is based on the semantic recognition processing module of the AI ​​large model and can perform semantic recognition, extract structural modeling, calculate parameters, and perform standardized output based on natural language input information; The 3D modeling unit interacts with the semantic recognition unit data and can call the corresponding modeling tool flow to perform parametric modeling based on the semantic recognition results of the large model; The finite element calculation unit can interact with the 3D modeling unit data, input the geometric information established by the 3D modeling unit into the finite element calculation software, input the calculation parameter structure as the remaining parameters into the finite element calculation software, and establish the corresponding finite element model based on this, and perform solution calculations.

2. The natural language driven 3D modeling computer-aided system according to claim 1, characterized in that: The semantic recognition processing module based on the AI ​​large model extracts and recognizes specific tasks based on preset structured prompt words for natural language input information, and extracts structural modeling calculation parameters.

3. The natural language driven 3D modeling computer-aided system according to claim 2, characterized in that: The structured prompt words include: A user role setting unit, wherein the user role setting unit is used to set basic identity information for the AI ​​large model; A task processing logic setting unit, which sets task identification, parameter identification and standard output logic; The standard output unit integrates and outputs specific parameters in the format of "variable name = specific value".

4. The natural language driven 3D modeling computer-aided system according to claim 1, characterized in that: The basic information input unit is provided with a first information interaction window for receiving a natural language description input by a user.

5. The natural language driven 3D modeling computer-aided system according to claim 1, characterized in that: The basic information input unit is provided with a second information interaction window for storing structured prompt words of the large model.

6. The natural language driven 3D modeling computer-aided system according to claim 1, characterized in that: The basic information input unit is provided with a third information interaction window for feeding back the large model information extraction content.

7. The natural language driven 3D modeling computer-aided system according to claim 1, characterized in that: The basic information input unit is provided with a fourth information interaction window for inputting the AI ​​large model and finite element calculation software authorization key and uniform resource locator.

8. The natural language driven 3D modeling computer-aided system according to claim 1, characterized in that: The basic information input unit is provided with a fifth information interaction window, which is used as an external expansion window to support the calling of external exe files.

9. The natural language driven 3D modeling computer-aided system according to claim 1, characterized in that: The 3D modeling unit performs keyword recognition and segmentation processing on the structural modeling calculation parameters output by the basic information input unit through the keyword segmentation module.

10. The natural language driven 3D modeling computer-aided system according to claim 1, characterized in that: The 3D modeling calculation-assisted system also includes an external program calling unit, which is configured to call a corresponding external program based on a program identification code extracted by the basic information input unit through semantic recognition of natural language input information and a preset program address.

Images