CAD model generation system and method based on natural language modeling

Abstract

The invention relates to a CAD model generation system and method based on natural language modeling. The CAD model generation system comprises a data layer, an application layer and a middle layer. The application layer is used for acquiring request information of a user; the middle layer is used for analyzing the request information through a natural language processing method to obtain parameters corresponding to the design intention of the user; mapping the corresponding parameters of the user design intention to a predefined structure template, and associating a standard; according to the structure template and the constraint condition, generating an automatic script for constructing a two-dimensional / three-dimensional model through a depth model generation method; based on the automatic script, adopting Siemens NX software to generate a CAD (Computer Aided Design) model; the data layer is used for querying corresponding standard terms in a knowledge base according to the association standard; and based on the standard terms, extracting constraint conditions corresponding to the association standards. The method is suitable for rapid modeling and design verification of complex parts in a weapon system.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a CAD model generation system and method based on natural language modeling. Background Art

[0002] In the field of weapons manufacturing, especially during the design of complex structures, the construction of 2D / 3D CAD models relies primarily on manual labor. This process relies heavily on the designer's experience, resulting in low efficiency, long turnaround times, and susceptibility to human error. Furthermore, traditional modeling methods are poorly adaptable to irregular structures, making it difficult to quickly respond to the demands of changing combat missions and model iterations.

[0003] Existing intelligent modeling methods are mostly used in the production and manufacturing processes of industrial parts. Due to the complex structure of weapons, harsh service conditions and high precision requirements, it is difficult to directly apply them to high-precision and high-complexity modeling tasks in the military field, which affects the safety and stability of combat missions. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a CAD model generation system and method based on natural language modeling, which can effectively improve the efficiency and quality of weapon structure model construction and effectively solve problems such as long modeling cycle and poor adaptability.

[0005] In a first aspect, the present invention provides a CAD model generation system based on natural language modeling, comprising: a data layer, an application layer, and an intermediate layer; the intermediate layer is located between the data layer and the application layer;

[0006] The application layer is used to obtain the user's request information and send it to the middle layer;

[0007] The middle layer is configured to parse the request information sent by the application layer using a natural language processing method to obtain parameters corresponding to the user's design intent; map the parameters corresponding to the user's design intent to a predefined structural template, obtain an associated standard, and send the associated standard to the data layer; generate an automated script for constructing a two-dimensional / three-dimensional model using a deep model generation method based on the structural template and the constraints sent by the data layer; and generate a CAD model using Siemens NX software based on the automated script;

[0008] The data layer is used to query the corresponding standard terms in the knowledge base according to the association standard sent by the middle layer; extract the constraint conditions corresponding to the association standard based on the standard terms and send them to the middle layer.

[0009] Preferably, the application layer is further configured to automatically generate a preview model based on the CAD model sent by the intermediate layer, and use the preview model for the user to view and confirm or modify some parameters of the preview model until confirmed; use the NXOpen script to parse the confirmed preview model and generate a feature tree; save the feature tree as a model editing record and send it to the data layer;

[0010] The data layer is used to display the model editing records.

[0011] Preferably, the request information includes: requirement information for instructing the user to generate a CAD two-dimensional / three-dimensional model and corresponding design specifications and standards.

[0012] Preferably, the parameters corresponding to the user's design intention include: design object type, design structure characteristics and design structure parameters.

[0013] Preferably, the model editing record includes: features of the model modification, the operation sequence of the model modification and corresponding model parameters.

[0014] Preferably, the data layer is further configured to save the model editing record as model fine-tuning sample data and send the model editing record to the intermediate layer according to the model editing record sent by the application layer;

[0015] The intermediate layer is further used to fine-tune the sample data according to the model sent by the data layer, and fine-tune the preview model.

[0016] Preferably, the model fine-tuning sample data includes: a model structure template, model constraints, fine-tuning process information and fine-tuning editing records.

[0017] In a second aspect, the present invention also proposes a CAD model generation method based on natural language modeling, comprising:

[0018] Use the application layer to obtain user request information;

[0019] Utilizing the middle layer to parse the request information through a natural language processing method to obtain parameters corresponding to the user's design intent; mapping the parameters corresponding to the user's design intent to a predefined structure template to obtain an associated standard;

[0020] Using the data layer to query the corresponding standard terms in the knowledge base according to the association standard; extracting the constraint conditions corresponding to the association standard based on the standard terms;

[0021] The intermediate layer generates an automated script for constructing a two-dimensional / three-dimensional model according to the structural template and the constraints through a deep model generation method; and based on the automated script, a CAD model is generated using Siemens NX software.

[0022] In a third aspect, the present invention also proposes an electronic device, comprising: at least one processor and a memory; the memory and the processor are connected via a bus;

[0023] The memory is used to store one or more programs;

[0024] When the one or more programs are executed by the at least one processor, the method for generating a CAD model based on natural language modeling is implemented.

[0025] In a fourth aspect, the present application further proposes a readable storage medium having an execution program stored thereon, which, when executed, implements the CAD model generation method based on natural language modeling.

[0026] In summary, due to the adoption of the above technical features, the beneficial effects of the present invention are:

[0027] The present invention generates CAD two-dimensional / three-dimensional models based on natural language modeling. It can intelligently generate weapon component structural models that meet the requirements according to the design requirements and relevant standards input by the user, and realize the automation of the entire process from demand analysis, standard matching to model construction, greatly improving modeling efficiency and design accuracy, reducing manual intervention, and is suitable for rapid modeling and design verification of complex components in weapon systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0029] Figure 1 This is a flow chart of the CAD model generation method based on natural language modeling provided by the present invention. DETAILED DESCRIPTION

[0030] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0031] Example 1

[0032] The embodiment of the present invention relates to a CAD model generation system based on natural language modeling, comprising: a data layer, an application layer, and an intermediate layer; the intermediate layer is located between the data layer and the application layer;

[0033] The application layer is used to obtain the user's request information and send it to the middle layer; build a user modification interface, a user request interface and a user preview interface; the user modification interface is used to receive the user's modification request and modify the CAD model; the user preview interface is used to preview the CAD model.

[0034] The middle layer is configured to parse the request information sent by the application layer using a natural language processing method to obtain parameters corresponding to the user's design intent; map the parameters corresponding to the user's design intent to a predefined structural template to obtain an associated standard and send it to the data layer; generate an automated script for constructing a two-dimensional / three-dimensional model using a deep model generation method based on the structural template and the constraints sent by the data layer; and generate a CAD model based on the automated script using Siemens NX software and send it to the application layer;

[0035] The data layer is used to query the corresponding standard terms in the knowledge base according to the association standard sent by the middle layer; extract the constraint conditions corresponding to the association standard based on the standard terms and send them to the middle layer.

[0036] In the above, the middle layer for semantic parsing and modeling reasoning is introduced between the data layer and the application layer; the application layer is used to obtain the request information collected by the user-end interface when the user request interface receives a request sent by the user; at the same time, the application layer sends the request information to the middle layer; the middle layer generates a Python automation script for constructing a two-dimensional / three-dimensional model based on the structural template and the constraints sent by the data layer through a deep model generation method, and records process information; based on the automation script, a CAD model that can be edited online in Siemens NX software is generated, and the process information is embedded in the CAD model metadata as interpretability information; at the same time, the middle layer sends the CAD model to the application layer;

[0037] Furthermore, the application layer is further configured to automatically generate a preview model based on the CAD model sent by the intermediate layer, and use the preview model for the user to view and confirm or modify some parameters of the preview model until confirmed; use the NXOpen script to parse the confirmed preview model and generate a feature tree; save the feature tree as a model editing record and send it to the data layer;

[0038] The data layer is used to display the model editing records.

[0039] As mentioned above, the application layer automatically generates a preview based on the CAD model sent by the middle layer, allowing users to view the model and confirm or modify some parameters; uses the NX Open script to parse the feature tree in the .prt file generated after editing and saves it as a model editing record; at the same time, the application layer sends the model editing record to the data layer.

[0040] Furthermore, the request information includes: requirement information for instructing the user to generate a CAD two-dimensional / three-dimensional model and corresponding design specifications and standards.

[0041] Furthermore, the corresponding parameters of the user design intention include: design object type, design structure characteristics and design structure parameters.

[0042] Furthermore, the model editing record includes: features of the model modification, an operation sequence of the model modification, and corresponding parameters of the model.

[0043] Furthermore, the data layer is further configured to save the model editing record as model fine-tuning sample data and send the model editing record to the intermediate layer according to the model editing record sent by the application layer;

[0044] The intermediate layer is further used to fine-tune the sample data according to the model sent by the data layer, and fine-tune the preview model.

[0045] Furthermore, the model fine-tuning sample data includes: a model structure template, model constraints, fine-tuning process information and fine-tuning editing records.

[0046] The following is a detailed description of the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems using a specific embodiment.

[0047] For the described embodiment, the application layer processes user requests for missile warhead structural design;

[0048] First, the application layer needs to obtain the user's request for missile warhead structural design. For example, the user inputs: "I want a flanged cylindrical shell for an armor-piercing missile warhead. The inner diameter is 120 mm, the height is 250 mm, the wall thickness is 4 mm, the front end is for installing a focused cone, and the rear end is reserved for a fuze installation port. It must meet the missile-borne structural strength requirements." The application layer sends the request to the middle layer. The middle layer uses natural language processing to parse the semantics, extract the user's design intent (for armor-piercing, installation of a focused cone, and reservation of a fuze port) and corresponding parameters, and map them to a predefined structural template. The middle layer also sends the association criteria to the data layer.

[0049] The corresponding parameters include: object type (cylinder), structural features (with flange), structural parameters (inner diameter, height and wall thickness) and associated standards (missile-borne structural strength requirements).

[0050] Next, the data layer queries the knowledge base for the corresponding standard clauses based on the missile-borne structural strength requirements sent by the middle layer, extracting the corresponding constraints, including relevant military material parameters, structural constraints, and ballistic mechanics standards. Simultaneously, the data layer sends the constraints to the middle layer. Based on the structural template and the constraints sent by the data layer, the middle layer generates a Python automated script for constructing a two-dimensional / three-dimensional model using a deep model generation method. The script also records information such as the semantic parsing results, structural parameter sources, and standard adaptation process during the model generation process.

[0051] Next, the middle layer uses Python automation scripts for building 2D / 3D models, integrates structural parameters and functional requirements, automatically completes detailed modeling and feature constraints, and generates CAD 3D models that can be edited online in Siemens NX software (2D formats such as .dwg, .dxf, 3D formats such as .step, .igs, .stl, .sldprt, etc.). The process information is embedded into the CAD model metadata as interpretable information. At the same time, the middle layer returns it to the application layer. After receiving the missile warhead shell CAD model generated by the middle layer, the application layer automatically generates a 3D model preview interface for users to view structural details (such as flange position, shell height, fuze installation port size, etc.) and confirm or modify key design parameters. The NX Open script is used to parse the feature tree in the .prt file generated after editing and save it as a model editing record. At the same time, the application layer sends the editing record to the data layer.

[0052] The editing record includes: modified features (flange, cylindrical shell, fuse mounting hole, chamfer, fillet, etc.), operation sequence (first build the cylinder, then add the flange, then open the hole, and finally fillet) and corresponding parameters (inner diameter 120mm, height 250mm, wall thickness 4mm, flange width 20mm, hole diameter 8mm, chamfer radius 2mm).

[0053] Finally, the data layer organizes and archives the user edit records sent by the application layer, constructing fine-tuning sample data for intelligent model optimization. Simultaneously, the data layer sends this fine-tuning sample data to the middle layer. After receiving the fine-tuning sample data from the data layer, the middle layer fine-tunes the deep model generation method based on typical feedback samples from missile warhead casing design (such as frequently modified areas and common parameter combinations), continuously optimizing the model's structural understanding and modeling accuracy. Through this fine-tuning process, the system can more accurately identify user intent and improve the efficiency and accuracy of CAD modeling for complex structures such as missile warhead casings.

[0054] The model fine-tuning sample data includes: the model fine-tuning sample data includes: a structural template defining the basic geometric shape and design framework of the missile warhead, constraints such as material performance parameters and structural strength limits, process information from semantic parsing to model generation, and user editing records during model preview and adjustment.

[0055] Example 2:

[0056] An embodiment of the present invention further relates to a CAD model generation method based on natural language modeling, comprising:

[0057] Step 1: Use the application layer to obtain the user's request information;

[0058] Step 2: Utilize the middle layer to parse the request information through a natural language processing method to obtain parameters corresponding to the user's design intent; map the parameters corresponding to the user's design intent to a predefined structure template to obtain an associated standard;

[0059] Step 3: Using the data layer to query the corresponding standard terms in the knowledge base according to the association criteria; extracting the constraint conditions corresponding to the association criteria based on the standard terms;

[0060] Step 4: Using the intermediate layer to generate an automated script for constructing a two-dimensional / three-dimensional model according to the structural template and the constraints through a deep model generation method; based on the automated script, using Siemens NX software to generate a CAD model.

[0061] Furthermore, the method further includes: automatically generating a preview model based on the CAD model using the application layer, allowing the user to view the preview model and confirm or modify some parameters of the preview model; parsing the confirmed preview model using an NX Open script and generating a feature tree; and saving the feature tree as a model editing record;

[0062] The model editing record is displayed using the application layer.

[0063] Furthermore, the request information includes: requirement information for instructing the user to generate a CAD two-dimensional / three-dimensional model and corresponding design specifications and standards.

[0064] Furthermore, the corresponding parameters of the user design intention include: design object type, design structure characteristics and design structure parameters.

[0065] Furthermore, the model editing record includes: features of the model modification, an operation sequence of the model modification, and corresponding parameters of the model.

[0066] Furthermore, the method further includes: saving the model editing record as model fine-tuning sample data;

[0067] The intermediate layer is used to fine-tune the sample data according to the model to fine-tune the preview model.

[0068] Furthermore, the model fine-tuning sample data includes: a model structure template, model constraints, fine-tuning process information and fine-tuning editing records.

[0069] Example 3:

[0070] The present invention also provides an electronic device, which may be a computer, a single-chip microcomputer, a smart mobile device, or the like. The electronic device in this embodiment may include a processor, a memory, a transceiver component, and the like. The memory, processor, and transceiver component are connected via a bus; the memory may be used to store an execution program, which may include instructions; and the processor may be used to execute the instructions stored in the memory. The memory may also be used to store data, which may be accessed and / or modified during the execution of the instructions.

[0071] The processor may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the storage medium to implement the corresponding method flow or corresponding function, so as to realize the steps of a CAD model generation method based on natural language modeling in the above embodiment.

[0072] Example 4:

[0073] Based on the same inventive concept, the present invention also provides a readable storage medium, specifically an electronic device readable storage medium (Memory), which is a memory device in an electronic device for storing programs and data. It can be understood that the storage medium here can include both built-in storage media in the electronic device and, of course, extended storage media supported by the electronic device. The storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space. These instructions can be one or more execution programs (including program codes). It should be noted that the storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage. The processor loads and executes one or more instructions stored in the storage medium, which can implement the steps of a CAD model generation method based on natural language modeling in the above embodiment.

[0074] Those skilled in the art will appreciate that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention can adopt 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.) that contain computer-usable program code.

[0075] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, 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 flowchart and / or block diagram. 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.

[0076] 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 1The function specified in one or more boxes.

[0077] 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.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that after reading the present invention, those skilled in the art can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the application, but these changes, modifications or equivalent substitutions are all within the scope of protection of the pending claims.

Claims

1. A CAD model generation system based on natural language modeling, characterized in that: include: Data layer, application layer, middle layer; The middle layer is located between the data layer and the application layer; The application layer is used to obtain the user's request information and send it to the middle layer; The intermediate layer is configured to parse the request information sent by the application layer through a natural language processing method to obtain parameters corresponding to the user's design intent; map the parameters corresponding to the user's design intent to a predefined structure template, obtain an associated standard, and send the associated standard to the data layer; Generating an automated script for constructing a two-dimensional / three-dimensional model using a deep model generation method according to the structural template and the constraints sent by the data layer; and generating a CAD model using Siemens NX software based on the automated script; The data layer is used to query the corresponding standard terms in the knowledge base according to the association standard sent by the middle layer; extract the constraint conditions corresponding to the association standard based on the standard terms and send them to the middle layer.

2. The system according to claim 1, wherein: The application layer is further configured to automatically generate a preview model based on the CAD model sent by the intermediate layer, and use the preview model for the user to view and confirm or modify some parameters of the preview model until confirmed; and use the NX Open script to parse the confirmed preview model and generate a feature tree; Saving the feature tree as a model editing record and sending it to the data layer; The data layer is also used to display the model editing records.

3. The system according to claim 1, wherein: The request information includes: requirement information for instructing the user to generate a CAD two-dimensional / three-dimensional model and corresponding design specifications and standards.

4. The system according to claim 1, wherein: The corresponding parameters of the user's design intention include: design object type, design structure characteristics and design structure parameters.

5. The system according to claim 2, wherein: The model editing record includes: features of the model modification, the operation sequence of the model modification and corresponding parameters of the model.

6. The system according to claim 2, wherein: The data layer is further configured to save the model editing record sent by the application layer as model fine-tuning sample data and send the model editing record to the intermediate layer; The intermediate layer is further used to fine-tune the sample data according to the model sent by the data layer, and fine-tune the preview model.

7. The system according to claim 6, characterized in that The model fine-tuning sample data includes: model structure template, model constraint conditions, fine-tuning process information and fine-tuning editing records.

8. A CAD model generation method based on natural language modeling, characterized in that: include: Use the application layer to obtain user request information; Utilize the middle layer to parse the request information through natural language processing methods to obtain parameters corresponding to the user's design intent; Mapping the corresponding parameters of the user design intention to a predefined structure template to obtain an associated standard; Using the data layer to query the knowledge base for corresponding standard terms based on the association criteria; Extracting constraints corresponding to the associated standards based on the standard terms; Using the intermediate layer to generate an automated script for constructing a two-dimensional / three-dimensional model based on the structural template and the constraints through a deep model generation method; Based on the automation script, Siemens NX software was used to generate the CAD model.

9. An electronic device, characterized in that: include: at least one processor and memory; The memory and the processor are connected via a bus; The memory is used to store one or more programs; When the one or more programs are executed by the at least one processor, the CAD model generation method based on natural language modeling as claimed in claim 8 is implemented.

10. A readable storage medium, characterized in that: An execution program is stored thereon, and when the execution program is executed, a CAD model generation method based on natural language modeling as claimed in claim 8 is implemented.

Images