Design drawing generation system and design drawing generation method

Abstract

The invention provides a design drawing generation system and a design drawing generation method. The design drawing generation system comprises a storage device and a processor. The processor is coupled to the storage device and is used for executing the parameter set generation module, the macro code generation module and the drawing rendering module. And the parameter set generation module analyzes the demand file to obtain design demand information. And the parameter set generation module performs correlation analysis on the design demand information and the sample data to obtain an analysis result, and matches a corresponding drawing type according to the analysis result to obtain a parameter set of the design drawing. And the macro code generation module generates a drawing macro code according to the design logic of the corresponding drawing type and the parameter set of the design drawing. And the drawing rendering module performs drawing rendering according to the drawing macro code to generate a design drawing.

Description

TECHNICAL FIELD

[0001] The present application relates to a design drawing generation technology, and in particular to a design drawing generation system and a design drawing generation method. BACKGROUND

[0002] In the existing design drawing generation process, there are problems such as low efficiency, complicated modification process and time-consuming. Generally speaking, when the design needs to be modified several times, the engineers and designers need to communicate and adjust repeatedly several times, and the engineers need to adjust the parameters and verify the design several times. These all lead to slow feedback mechanism, and it is difficult to realize the rapid adjustment of drawing parameters and the optimization of design drawings. At the same time, the repetitive work causes the designers to have no time to invest in innovation, which limits the ability of enterprises to respond to market changes and development. SUMMARY

[0003] The present application relates to a design drawing generation technology, and in particular to a design drawing generation system and a design drawing generation method.

[0004] According to an embodiment of the present application, the design drawing generation system of the present application comprises a storage device and a processor. The storage device is used to store a parameter set generation module, a macro code generation module and a drawing rendering module. The processor is coupled to the storage device and is used to execute the parameter set generation module, the macro code generation module and the drawing rendering module. The processor receives a requirement file and executes the parameter set generation module to analyze and process the requirement file, thereby obtaining design requirement information. The parameter set generation module analyzes the correlation between the design requirement information and the sample data of the design experience library to obtain an analysis result, and matches the corresponding drawing type according to the analysis result to obtain the parameter set of the design drawing. The macro code generation module generates the drawing macro code according to the design logic of the corresponding drawing type and the parameter set of the design drawing. The drawing rendering module performs drawing rendering according to the drawing macro code to generate the design drawing.

[0005] According to an embodiment of the present application, the design drawing generation method of the present application comprises the following steps: a parameter set generation module is executed by a processor to analyze and process a requirement file, so that the parameter set generation module obtains design requirement information; the parameter set generation module is executed by the processor to analyze the correlation between the design requirement information and the sample data of the design experience library, so that the parameter set generation module obtains an analysis result; the parameter set generation module matches the corresponding drawing type according to the analysis result to obtain the parameter set of the design drawing; a macro code generation module is executed by the processor, so that the macro code generation module generates drawing macro code according to the design logic of the drawing type and the parameter set of the design drawing; and a drawing rendering module is executed by the processor, so that the drawing rendering module performs drawing rendering according to the drawing macro code, thereby generating the design drawing.

[0006] Based on the above, the design drawing generation system and the design drawing generation method of the present application effectively provide a system and method for generating design drawings through the setting of the code generation model and the enterprise experience library, thereby improving the efficiency and accuracy of design drawings.

[0007] In order to make the above features and advantages of the present application more obvious and easy to understand, the following specific examples are described in detail below, together with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0008] Figure 1 is a schematic diagram of a design drawing generation system of an embodiment of the present application;

[0009] Figure 2 is a flowchart of a design drawing generation method of an embodiment of the present application;

[0010] Figure 3 is a schematic diagram of a design drawing generation system and multiple modules of an embodiment of the present application;

[0011] Figure 4 is a flowchart of strong correlation coefficient identification and completion of an embodiment of the present application;

[0012] Figure 5 is a flowchart of generating design logic and parameters of an embodiment of the present application;

[0013] Figure 6 is a flowchart of generating macro code of an embodiment of the present application.

[0014] BRIEF DESCRIPTION OF DRAWINGS

[0015] 100: design drawing generation system;

[0016] 110: processor;

[0017] 120: storage device;

[0018] 121: parameter set generation module;

[0019] 122: macro code generation module;

[0020] 123: drawing rendering module;

[0021] 124: model training module;

[0022] 400: design experience library;

[0023] 600: generated code large model;

[0024] 601: parameter set;

[0025] 602: design logic;

[0026] 603: macro code;

[0027] S210-S250, S410-S480, S510-S550, S610-S650: steps. DETAILED DESCRIPTION

[0028] Reference will now be made in detail to the exemplary embodiments of the present application, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers are used in the drawings and the description to refer to the same or like parts.

[0029] Figure 1 is a schematic diagram of a design drawing generation system according to an embodiment of the present application. Referring to Figure 1 , the design drawing generation system 100 includes a processor 110 and a storage device 120. The processor 110 is coupled to the storage device 120. The design drawing generation system 100 can further include a communication interface or a data transmission interface with actual circuit components, so that the processor 110 can communicate or transmit data with external business systems, user interfaces, application programming interfaces (APIs) or databases. In the present embodiment, the design drawing generation system 100 can be implemented in a cloud server, a private server within an enterprise or a data center within an enterprise, for example.

[0030] In the present embodiment, the processor 110 of the design drawing generation system 100 can include a central processing unit (CPU), or other programmable general purpose or special purpose microprocessors (MPUs), digital signal processors (DSPs), application specific integrated circuits (ASICs), programmable logic devices (PLDs), other similar processing circuitry or combinations of these devices, for example.

[0031] The storage device 120 can also be a remote cloud storage service or a local data storage service. The storage device 120 can include a memory and / or a database, where the memory can be, for example, a non-volatile memory (NVM). The storage device 120 can store programs, modules, systems, or algorithms related to implementing embodiments of the present application for the processor 110 to access and execute to implement the related functions and operations described in embodiments of the present application.

[0032] Figure 2 is a flowchart of a design drawing generation method of an embodiment of the present application. Figure 3 is a schematic diagram of a design drawing generation system and modules of an embodiment of the present application. First referring to Figure 1 and Figure 2 The storage device 120 can store a parameter set generation module 121, a macro code generation module 122, a drawing rendering module 123, and a model training module 124. The parameter set generation module 121, the macro code generation module 122, the drawing rendering module 123, and the model training module 124 can be implemented, for example, in a program language such as JSON (JavaScript Object Notation), Extensible Markup Language (XML), or YAML, without being limited thereto.

[0033] Referring again to Figures 1 to 2 The design drawing generation system 100 performs the following steps S210-S250. In step S210, the processor 110 can receive a requirement file and execute the parameter set generation module 121 to parse the requirement file and obtain design requirement information. In an embodiment, the requirement file is a specification file recording the design specifications and parameters of the design drawing, such as a Portable Document Format (PDF) file or a file order. In this way, the parameter set generation module 121 parses the PDF file to obtain document information, and parses the document information and the design experience library, and integrates the design requirement information. In another embodiment, the requirement file is text information. For example, the requirement file is text information of a conversation between a customer and a designer. In this way, the parameter set generation module 121 processes and analyzes the conversation text information to obtain the design requirement information related to the design drawing in the conversation text information. Specifically, the design drawing generation system 100 can parse the conversation text information by a pre-trained large model, and summarize and analyze the design requirement information to achieve extraction of design parameters.

[0034] At step S220, the processor 11 executes the parameter set generation module 121 to perform a correlation analysis on the design requirement information and the sample data of the design experience library, and to obtain an analysis result. The analysis result includes a correlation probability and a matched drawing type. In an embodiment, the parameter set generation module 121 determines the correlation between the design requirement information and the sample data of the design experience library by a correlation algorithm, and when the correlation probability is greater than a set threshold, the drawing type with the correlation probability greater than the threshold is matched. The drawing type can be, for example, a design drawing, an engineering drawing or an explanatory drawing, such as a structural part drawing of a shaft sleeve type, a disc cover type, a fork type and a box type. Specifically, the drawing type can be a Computer-Aided Design (CAD) design drawing of a screw, a guide pin, an ejection pin, a machine tool spindle box, a machine tool feed box, a gearbox, a speed reducer, an engine cylinder, a machine base, a bearing seat, a support, a cylinder or a bottom plate.

[0035] At step S230, the parameter set generation module 121 matches a corresponding drawing type according to the analysis result, to obtain a parameter set of a design drawing. Specifically, when the correlation probability generated after the correlation analysis of step S220 is greater than a threshold, the drawing type matched in the design experience library is obtained. Then, the parameter set generation module 121 supplements related parameters according to the matched drawing type by a nearest neighbor algorithm, to obtain a parameter set and a design logic. In an embodiment, the threshold is 0.9.

[0036] At step S240, the processor 110 executes the macro code generation module 122 to generate a drawing macro code according to the design logic of the corresponding drawing type and the parameter set of the design drawing. At step S250, the drawing rendering module 123 performs drawing rendering according to the drawing macro code, to generate a design drawing. Moreover, the processor 110 stores the design drawing and the drawing macro code in the macro code database and the target drawing database in the data layer, respectively.

[0037] Figure 4 is a flowchart of strong correlation coefficient identification and completion of an embodiment of the present application. In this embodiment, the parameter set generation module 121 includes a correlation analysis module and a parameter completion module, and the storage device 120 further stores the correlation analysis module and the parameter completion module. Then, referring to Figure 2 、 Figure 3 and Figure 4, the design drawing generation system 100 performs steps S410-S480 as follows. In an embodiment, the parameter set generation module 121 can perform step S410 before performing the steps of the correlation analysis of step S220. In step S410, the correlation analysis module performs file normalization on the design requirement information, and adjusts the design requirement information to a format consistent with the design experience library, so that the design requirement information is consistent with the data of the design experience library. For example, the correlation analysis module adjusts the design requirement information "I want a car axle" through file normalization to a format consistent with the record mode in the design experience library, thereby improving the accuracy of the correlation analysis. In another example, the correlation analysis module converts the design requirement information into JavaScript Object Notation (JSON) format data, thereby improving the matching and processing efficiency of values in the correlation analysis.

[0038] In step S420, the correlation analysis module performs correlation analysis on the design requirement information and sample data of the design experience library to obtain an analysis result. The sample data is, for example, engineering drawing data of various parts, drawing types of various parts, and data and parameters required to generate various parts.

[0039] In step S430, the correlation analysis module compares the analysis result with a threshold value, and determines whether the analysis result is greater than the threshold value. In other words, the correlation analysis module compares the correlation probability in the analysis result with the threshold value. In step S450, in response to the correlation probability of the analysis result being greater than the threshold value, the correlation analysis module performs drawing type matching based on the drawing parameter set in the design experience library to obtain design logic (step S460). Specifically, when the correlation probability in the analysis result is greater than the threshold value, the correlation analysis module matches the design logic of the drawing type with the highest correlation probability in the design experience library. In this embodiment, the design logic is logical information of natural language-based design and generation of drawings (such as design drawings of parts).

[0040] In step S440, in response to the analysis result being less than the threshold value, the correlation analysis module performs slicing processing on the design requirement information to obtain a minimum subset of the design requirement information. Then, the correlation analysis module performs step S420 again on the minimum subset of the design requirement information, that is, the correlation analysis module performs correlation analysis again on the minimum subset to obtain a new analysis result. Moreover, the correlation analysis module determines whether the correlation of the new analysis result is greater than the threshold value, and performs corresponding step S440 or step S450 until the correlation probability in the analysis result is greater than the threshold value.

[0041] At step S470, the parameter completion module completes the relevant parameters based on the matched drawing type through a nearest neighbor algorithm to obtain a parameter set (step S480). Specifically, the relevant parameters are completed through a nearest neighbor algorithm based on the matched drawing type, the design experience library, and the parameters in the design requirement information.

[0042] Figure 5 is a flowchart of generating design logic and parameters of an embodiment of the present application. Referring to Figure 3 and Figure 5 , the design drawing generation system 100 performs the following steps S510-S550. In an embodiment, the parameter set generation module 121 includes a parsing module. At step S510, the parsing module performs a parsing process on the requirement file to determine the file type of the requirement file. The file type includes a text requirement and a file requirement. The text requirement is, for example, a part generation instruction or a word input by the user. The file requirement can be, for example, a PDF file including part drawing generation parameters.

[0043] At step S520, in response to the file type being a text type, the parsing module analyzes the requirement file through a large language model to obtain requirement information. In an embodiment, the requirement file of the text type is a multi-round dialogue. For example, the user inputs an instruction to the design drawing generation system 100 through a human-computer interface coupled with the processor, and when the design drawing generation system 100 determines that the instruction is not specific enough, the design drawing generation system 100 displays a prompt message on the display. The prompt message is, for example, “input the size of the target part” or “input the design drawing type”, etc. After multiple prompts of the system and inputs of the user, the design drawing generation system 100 summarizes the multiple information and summarizes the user's intention according to the multi-round dialogue.

[0044] At step S530, in response to the file type being a file type, the parsing module analyzes the requirement file to obtain document information, and analyzes the document information through a large language model. At step S540, the large model summarizes the document information to obtain requirement information. At step S550, the parameter set generation module 121 identifies and completes the strongly related parameters based on the design experience library 400.

[0045] In one embodiment, the macro code generation module 122 comprises a rule comparison module and a code generation module. Before generating the macro code of the drawing, the macro code generation module 122 can first determine whether the parameters are in compliance with the rules. Specifically, the rule comparison module finds the corresponding design rules from the design experience library according to the drawing type. Then, the rule comparison module compares the parameter set with the design rules to determine whether the parameter set complies with the design rules. In response to the parameter set not complying with the design rules, the rule comparison module generates an error prompt and displays the error prompt message on the display. In this embodiment, the display is communicatively connected to the processor 110. In response to the parameter set complying with the design rules, the macro code generation module 122 generates the macro code of the drawing according to the design logic and the parameter set through the code generation model.

[0046] Figure 6 is a flowchart of generating the macro code of one embodiment of the present application. As shown in Figure 6 the design drawing generation system 600 can perform the following steps S610-S650 before generating the macro code of the drawing. In step S610, the macro code generation module 122 maps the parameter set 601 into machine language parameters. Specifically, the macro code generation module 122 maps the natural language-based parameter set 601 into machine language-based parameters, thereby improving the stability and accuracy of generating the macro code.

[0047] Then, step S620 is performed, in which the macro code generation module 122 sequentially performs the steps of parameter deduplication, external parameter conversion into internal parameter, and parameter reorganization on the mapped parameters. First, the macro code generation module 122 performs parameter deduplication to remove duplicate or semantically identical parameters in the mapped parameters, so as to avoid redundancy or mutual conflict in the subsequent generated macro code. Then, in the process of converting external parameters into internal parameters, the macro code generation module 122 converts the deduplicated parameters into internal parameters that can be processed by the code generation model. In other words, the parameters processed by converting external parameters into internal parameters can correspond to the parameters required by the CAD system, thereby complying with the semantics and execution logic of the code generation model. In the parameter reorganization step, the macro code generation module 122 rearranges and formats the converted parameters according to the structure of the code generation model, so that the parameters comply with the data format recognizable and executable by the code generation model.

[0048] In step S630, the macro code generation module 122 performs a logic mapping process on the design logic 602, and converts the design logic 602 into execution instructions and parameters understandable and processable by the code generation model. In step S640, the macro code generation module 122 further maps the design logic 602 into prompt words. That is, the macro code generation module 122 converts the design logic 602 into prompt words for inputting into the code generation model. For example, the originally natural language-based design logic 602 is "design a shaft", which is converted into prompt words such as "CreateComponent", "Component:Shaft", and "Material:Steel" by the logic mapping process of the macro code generation module 122.

[0049] In step S650, the macro code generation module 122 generates the macro code based on the prompt words generated in step S640 and the parameters generated in step S620 by the code generation model 600 (i.e., the code generation model).

[0050] In an embodiment, the storage device 120 further stores a verification module. The processor 110 executes the verification module to perform a verification process on the drawing macro code (i.e., the macro code generated in step S650). In response to the drawing macro code passing the verification process, the verification module inputs the drawing macro code into the drawing rendering module 123. In an embodiment, the verification module determines whether the code conforms to the drawing generation rules in the design experience library 400 based on the drawing macro code and the design experience library 400. In another embodiment, the verification module generates a drawing according to the drawing macro code, and determines that the verification fails and generates an execution error message on the display when the design drawing cannot be generated. In response to the drawing macro code failing the verification process, the verification module outputs a verification prompt to the display according to the verification rule that fails. For example, the verification prompt is a message such as "the length-width ratio is not in proportion", "the shaft length is missing", or "the diameter of the rotating shaft should be 1 to 2 centimeters".

[0051] Then, the drawing rendering module 123 performs drawing rendering on the drawing macro code that passes the verification to generate a design drawing (step S250). In an embodiment, the design drawing includes a three-dimensional design drawing and a two-dimensional sectional drawing. The design drawing generation system 100 can receive a drawing adjustment instruction input by a user through a human-computer interface, and correspondingly adjust the design requirement information based on the drawing adjustment instruction by the parameter set generation module 121. The drawing adjustment instruction can be a user intention summarized by the design drawing generation system 100 according to multiple rounds of dialogue with the user. In this way, the design drawing generation system 100 can correspondingly generate a new design drawing according to the adjusted design requirement information.

[0052] In another aspect, the drawing rendering module 123 can integrate the design drawing into a third-party software. For example, the drawing rendering module 123 executes the design drawing through a three-dimensional computer-aided design software to present the design drawing on a display. The drawing rendering module 123 can also receive editing instructions and adjust the design drawing according to the editing instructions. In other words, when the design drawing is opened in the third-party software, the user can perform secondary editing on the design drawing by operating the third-party software. The three-dimensional computer-aided design software can be a SolidWorks software.

[0053] In an embodiment, the storage device further stores a model training module 124. To generate the model, the design drawing generation system 100 first executes the model training module 124 through the processor 110 to generate a data set according to design rules and design documents. The design documents can be enterprise design experience documents stored in the storage device 120. The design rules can be rules related to drawing design summarized in the enterprise experience library. The enterprise experience library is a database coupled to the processor 110.

[0054] In this embodiment, the model training module 124 then slices the data set and inputs the sliced data set into the large model to iteratively train the large model. It is worth noting that the model training module 124 inputs fine data into the large model for reinforcement learning to generate the code generation model (i.e., the code generation large model 600). Specifically, the fine data is high-quality data annotated by structure, such as image structure or text structure. That is, the data set used to iteratively train the large model belongs to a large range of settings and parameters, while the fine data used for reinforcement learning is more specific and detailed settings and parameters. The model training module 124 establishes an enterprise experience library based on the sliced data set and the fine data.

[0055] In an embodiment, the model training module 124 further includes the following steps in the step of iteratively training the large model: the model training module 124 inputs the sliced data into the large model at a ratio of 9:1 to perform N times of iterative training, where N is a hyperparameter. Then, the model training module 124 performs spot check processing on the code generated by the large model after X times of iterations to generate a spot check result. X is a positive integer. When the spot check result of the code is not passed, the model training module 124 changes the ratio of the data to 8:2 and inputs it into the large model again, and this cycle is repeated N times. When the spot check result is passed, the model training module 124 then performs reinforcement learning on the large model.

[0056] In the present embodiment, the model training module 124 performs the sampling inspection process, wherein the sampling inspection process is performed by the model training module 124 to obtain the code quality based on the test set in the data. The model training module 124 determines the code quality based on the correctness, stability and logical integrity of the code. In response to the code quality being greater than a threshold value, the model training module generates a passing sampling inspection result. The threshold value can be 95% or 80%, and the present case should not be limited thereto. For example, after the model training module 124 samples the code based on the test set, the code quality is obtained as a result of a correctness rate of 95%, and the threshold value is set to 95%, so that the code generated by the code generation model passes the sampling inspection. Conversely, in response to the code quality not being greater than the threshold value, the model training module generates a failing sampling inspection result.

[0057] In summary, the design drawing generation system and the design drawing generation method of the present application can provide a service of generating design drawings through a design experience library and a code generation model. Moreover, the user only needs to input a requirement file into the code generation model to generate planar and stereoscopic design drawings, thereby effectively improving the design efficiency and shortening the delivery cycle. In addition, the design drawing generation system and the method thereof also effectively improve the accuracy of the design drawings by setting the judgment parameter set regularity, macro code verification and correlation analysis, and effectively utilize the parameters and rules in the enterprise experience library.

[0058] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A design drawing generation system characterized by comprising: The system comprises: a storage device for storing a parameter set generation module, a macro code generation module, and a drawing rendering module; and a processor coupled to the storage device, executing the parameter set generation module, the macro code generation module, and the drawing rendering module, and receiving a requirement file, wherein the parameter set generation module parses the requirement file to obtain design requirement information, wherein the parameter set generation module performs correlation analysis on the design requirement information and sample data in a design experience library to obtain an analysis result, and matches a corresponding drawing type according to the analysis result to obtain a parameter set of a design drawing, wherein the macro code generation module generates a drawing macro code according to a design logic corresponding to the drawing type and the parameter set of the design drawing, wherein the drawing rendering module performs drawing rendering according to the drawing macro code to generate the design drawing.

2. The design drawing generation system according to claim 1, characterized by, The parameter set generation module comprises a parsing module, wherein the parsing module parses the requirement file to determine the file type of the requirement file, wherein in response to the file type being a file type, the parsing module parses the requirement file to obtain document information, and summarizes the document information through a large language model to obtain the requirement information, wherein in response to the file type being a text type, the parsing module arranges and analyzes the requirement file through the large language model to obtain the requirement information.

3. The design sheet generating system according to Claim 1, wherein The parameter set generation module comprises a correlation analysis module and a parameter completion module, wherein the correlation analysis module compares the analysis result with a threshold value, in response to the analysis result being greater than the threshold value, the correlation analysis module performs drawing type matching according to drawing parameter sets in the design experience library to obtain the design logic, in response to the analysis result being less than the threshold value, the correlation analysis module performs slicing processing on the design requirement information to obtain a minimum subset of the design requirement information, and performs the correlation analysis again on the minimum subset to obtain a new analysis result until the analysis result is greater than the threshold value, wherein the parameter completion module completes relevant parameters of the design requirement information based on the matched drawing type through a nearest neighbor algorithm to obtain the parameter set.

4. The design sheet generation system according to Claim 1, characterized by The macro code generation module comprises a rule comparison module and a code generation module, wherein the rule comparison module finds corresponding design rules from the design experience library according to the design logic, wherein the rule comparison module compares the parameter set with the design rules to determine whether the parameter set conforms to the design rules, in response to the parameter set not conforming to the design rules, the rule comparison module generates an error prompt, in response to the parameter set conforming to the design rules, the code generation module generates the drawing macro code through a code generation model according to the design logic and the parameter set.

5. The design sheet generation system according to Claim 1, characterized by The storage device further stores a verification module, wherein the processor executes the verification module to perform verification processing on the drawing macro code, In response to the drawing macro code passing the checking process, the checking module inputs the drawing macro code to the drawing rendering module, In response to the drawing macro code failing the checking process, the checking module outputs a checking prompt to a display according to the failed checking rule.

6. The design sheet generation system according to Claim 1, characterized by The design drawing includes a three-dimensional design drawing and a two-dimensional sectional drawing, The parameter set generation module receives a drawing adjustment instruction and adjusts the design requirement information based on the drawing adjustment instruction.

7. The design drawing generation system according to Claim 3, wherein The correlation analysis module performs file normalization on the design requirement information, and then adjusts the design requirement information to conform to the format of the design experience library.

8. The design sheet generation system according to Claim 1, characterized by The storage device also stores a model training module, The processor executes the model training module to collect and organize design rules and design document generation data sets, The model training module slices the data set to input the sliced data set to a large model, and then iteratively trains the large model, The model training module inputs fine data to the large model for reinforcement learning to generate a code generation model, The model training module establishes a design experience library according to the data set and the fine data.

9. The design sheet generation system according to claim 8, wherein, The model training module inputs the data to the large model at a coarse-fine ratio of 9:1 for N times of iterative training, where N is a hyperparameter; The model training module performs sampling inspection on the code generated by the large model after X times of iteration to generate a sampling inspection result; When the sampling inspection result is not passed, the model training module changes the coarse-fine ratio of the data to 8:2 to input to the large model, and so on for N times.

10. The design sheet generation system according to Claim 9, wherein According to the sampling inspection result, the model training module then performs reinforcement learning on the large model; The model training module performs the sampling inspection steps, including: The model training module samples the code based on a test set in the data to obtain code quality; In response to the code quality being greater than a threshold, the model training module generates a passed sampling inspection result; In response to the code quality not being greater than the threshold, the model training module generates a failed sampling inspection result.

11. The design sheet generation system according to Claim 1, characterized by The drawing rendering module integrates the design drawing into a third-party software three-dimensional computer-aided design software to present the design drawing, The drawing rendering module receives an editing instruction and adjusts the design drawing according to the editing instruction.

12. A design drawing generation method characterized by comprising: It includes: The parameter set generation module parses the requirement file through the processor to obtain design requirement information; The parameter set generation module performs correlation analysis on the design requirement information and sample data of the design experience library through the processor to obtain analysis results; The parameter set generation module matches the corresponding drawing type according to the analysis results to obtain the parameter set of the design drawing; The macro code generation module is executed by the processor, so that the macro code generation module generates a drawing macro code according to the design logic of the drawing type and the parameter set of the design drawing; The drawing rendering module is executed by the processor, so that the drawing rendering module performs drawing rendering according to the drawing macro code, thereby generating the design drawing.

13. The method of claim 12, wherein, The parameter set generation module includes an analysis module, The parameter set generation module includes an analysis module, The analysis module analyzes the requirement file to obtain the requirement information. The parameter set generation module includes a correlation analysis module and a parameter completion module, The correlation analysis module compares the analysis result with a threshold value.

14. The method of claim 12, wherein, When the analysis result is greater than the threshold value, the correlation analysis module performs drawing type matching according to the drawing parameter set in the design experience library to obtain the design logic. When the analysis result is less than the threshold value, the correlation analysis module performs slicing processing on the design requirement information to obtain a minimum subset of the design requirement information, and performs the correlation analysis on the minimum subset again to obtain a new analysis result, until the analysis result is greater than the threshold value. The parameter completion module completes related parameters of the design requirement information based on the matched drawing type by a nearest neighbor algorithm to obtain the parameter set. The macro code generation module includes a rule comparison module and a code generation module, The rule comparison module finds out corresponding design rules from the design experience library according to the design logic. The rule comparison module compares the parameter set with the design rules to determine whether the parameter set conforms to the design rules.

15. The method of claim 12, wherein, When the parameter set does not conform to the design rules, the rule comparison module generates an error prompt. When the parameter set conforms to the design rules, the code generation module generates the drawing macro code by a code generation model according to the design logic and the parameter set. Before the step of performing drawing rendering according to the drawing macro code, the following steps are included: The verification module is executed by the processor to perform verification processing on the drawing macro code. When the drawing macro code passes the verification processing, the verification module inputs the drawing macro code to the drawing rendering module; and When the drawing macro code passes the verification processing, the verification module inputs the drawing macro code to the drawing rendering module; and 16. The method of claim 12, wherein, ​ ​ ​ In response to the drawing macro code failing the checking process, the checking module outputs a checking prompt to a display according to the failed checking rule.

17. The method of claim 12, wherein, The design drawing includes a three-dimensional design drawing and a two-dimensional sectional drawing, The method further includes the following steps: the parameter set generation module receives a drawing adjustment instruction, and the parameter set generation module adjusts the design requirement information based on the drawing adjustment instruction.

18. The method of claim 14, wherein, Before performing the correlation analysis, further comprising: The design requirement information is file-normalized by the correlation analysis module, and the design requirement information is adjusted to a format consistent with the design experience library.

19. The method of claim 12, wherein, Further comprising: The model training module is executed by the processor to generate a data set according to design rules and design documents; The data set is sliced by the model training module, and the sliced data is input into a large model to iteratively train the large model; Fine data is input into the large model by the model training module for reinforcement learning to generate a code generation module; And The design experience library is established by the model training module according to the data set and the fine data.

20. The method of claim 19, wherein, The step of iteratively training the large model includes: The data is input into the large model by the model training module at a ratio of 9:1, and the large model is iteratively trained N times, where N is a hyperparameter; After X iterations, the model training module performs spot check processing on the code generated by the large model to generate a spot check result; and In response to the spot check result being failed, the model training module changes the ratio of the data to 8:2 and inputs it into the large model, and this cycle is repeated N times.

21. The method of claim 20, wherein, In response to the spot check result being passed, the model training module then performs reinforcement learning on the large model; The step of performing the spot check processing includes: The model training module performs spot check on the code based on a test set in the data to obtain code quality; In response to the code quality being greater than a threshold, the model training module generates a passed spot check result; and In response to the code quality not being greater than the threshold, the model training module generates a failed spot check result.

22. The method of claim 12, wherein, Further comprising: The design drawing is integrated into third-party software by the drawing rendering module to present the design drawing, The drawing rendering module receives an editing instruction, and the drawing rendering module adjusts the design drawing according to the editing instruction.

Images