Conversational CAD drawing design method and system based on artificial intelligence

Abstract

The invention discloses a dialogue type CAD drawing design method and system based on artificial intelligence, and the method employs a separated framework of a front-end client, a WEB server and a calculation background, and comprises the steps: firstly inputting any drawing demands by a user; the AI large model of the WEB server side identifies a drawing task according to drawing requirements, and when the drawing task is a drawing action, the AI large model searches a corresponding calculation function in a calculation background according to the drawing task and reads related calculation parameters of the calculation function from a drawing; when it is checked that the calculation parameters are not lacked, the calculation background calls a corresponding calculation function, and a calculation result is obtained according to the calculation parameters; a calculation result is sent to the front-end client for display; and finally, the user confirms whether to modify the drawing according to the calculation result, and if so, automatically modifies the graph. The problem that a scheme design drawing is difficult to generate quickly is solved.

Description

Technical Field

[0001] The present invention relates to the field of BIM technology, and in particular to a conversational CAD drawing design method and system based on artificial intelligence. Background Art

[0002] In current architectural design, experienced designers are mainly used to operate CAD software to carry out architectural design. First, the operation is complicated and the design efficiency is low; second, it is unable to quickly respond to the client's requirements and quickly provide a preliminary version of the plan. There is no such system in the existing technology, and it all depends on the designer himself to design manually, which has low technical efficiency. Summary of the invention

[0003] 1. Technical issues to be resolved

[0004] Based on the above problems, the present invention provides a conversational CAD drawing design method and system based on artificial intelligence to solve the problem that design drawings are difficult to generate quickly.

[0005] (II) Technical solution

[0006] Based on the above technical problems, the present invention provides a conversational CAD drawing design method based on artificial intelligence, and the operation method includes:

[0007] S0. Training the AI ​​big model input corpus, establishing the corresponding relationship between the AI ​​big model input drawing requirements and output drawing tasks, and the associated calculation functions;

[0008] S1. The user inputs any drawing requirements on the front-end client;

[0009] S2. The AI ​​big model on the WEB server identifies the drawing task according to the drawing requirements and inquires for confirmation;

[0010] S3, the AI ​​big model determines whether the drawing task is a drawing action, if so, proceeds to step S4, otherwise, gives an answer according to the answer in the knowledge expert database in the AI ​​big model;

[0011] S4. The AI ​​big model searches for the corresponding calculation function in the calculation background according to the drawing task, and reads the relevant calculation parameters of the calculation function from the drawing;

[0012] S5. Check whether calculation parameters are missing. If yes, perform parameter polling through the AI ​​big model. The user enters the missing parameters on the front-end client, or modifies the drawing and re-enters S5. If no, enter step S6.

[0013] S6. Send the calculation parameters to the calculation backend through the WEB server, and the calculation backend calls the corresponding calculation function to obtain the calculation results according to the calculation parameters;

[0014] S7, the calculation backend sends the calculation results to the WEB server, and the WEB server sends them to the front-end client for display;

[0015] S8. The user is asked through the front-end client whether to modify the drawing according to the calculation result. If so, the calculation background automatically modifies the drawing; otherwise, no modification is required.

[0016] Furthermore, in S1, the user inputs any drawing requirement in the form of text or voice.

[0017] Further, in S5, the parameter polling includes:

[0018] S51, the AI ​​large model queries all calculation parameters in turn;

[0019] S52, when missing calculation parameters are found, the user inputs a statement containing the parameters;

[0020] S53, matching and recording calculation parameters according to the statement containing parameters;

[0021] S54, determine whether the query of all calculation parameters is completed, if not, return to step S51, if yes, end.

[0022] Furthermore, the WEB server and computing backend perform searches and return results through message middleware.

[0023] Furthermore, in S6, the message middleware first queries whether there is a calculation function in the calculation background, then calls the calculation function, and finally returns the calculation result obtained according to the calculation parameters and the calculation function to the WEB server.

[0024] Furthermore, the drawing tasks include: designing prefabricated areas as required, designing the splitting of composite panels as required, designing interior partition walls and enclosure walls as required, performing cost estimate statistics as required, calculating the assembly rate as required, and outputting prefabricated calculation books as required.

[0025] The present invention also discloses a conversational CAD drawing design system based on artificial intelligence, which runs the conversational CAD drawing design method based on artificial intelligence, including a separate architecture of a front-end client, a WEB server and a computing background, wherein the front-end client is connected to the WEB server, and the WEB server is connected to the computing background, and the front-end client includes a user interface; the WEB server includes an AI big model, which is used to interact with the user, can identify the user input information, and determine the task type and necessary task parameters; the computing background is used for function calculation, and the computing background includes a .NET function library, which completes the task execution after receiving the task type and task parameters, and returns the function calculation results and modified drawings.

[0026] (III) Beneficial effects

[0027] The above technical solution of the present invention has the following advantages:

[0028] (1) Based on artificial intelligence technology, the present invention proposes a conversational CAD drawing design system architecture, which uses artificial intelligence technology to understand requirements and quickly conduct preliminary scheme design. It can accurately understand Party A's requirements and complete the initial version of the scheme design drawings, which greatly improves the response speed on the one hand and the design efficiency on the other hand.

[0029] (2) The present invention adopts a separate architecture, which can be implemented across platforms. The established callback function and the corresponding table of the output results of the AI ​​large model enable the large model to have function calling capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The features and advantages of the present invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the present invention in any way. In the accompanying drawings:

[0031] Figure 1 This is a schematic diagram of the architecture of a conversational CAD drawing design system based on artificial intelligence according to an embodiment of the present invention;

[0032] Figure 2 The figure is a flowchart of an artificial intelligence-based conversational CAD drawing design method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0034] The embodiment of the present invention is a conversational CAD drawing design method and system based on artificial intelligence. The system adopts a separate architecture, such as Figure 1As shown, it includes a front-end client, a WEB server and a computing background. The front-end client is connected to the WEB server via the http network protocol, and the WEB server is connected to the computing background through communication. Among them, the front-end client includes a user interface; the WEB server includes an AI big model, and the AI ​​big model is used to interact with users. The interactive form includes text questions or voice interaction, which can identify user input information, determine the task type and necessary task parameters; the computing background is used for function calculation, including a .NET function library. After receiving the task type and task parameters, the .NET function library completes the task execution and returns the task calculation results and modifies the drawings. The running method is as follows Figure 2 As shown, including:

[0035] S0. Training the AI ​​big model input corpus, establishing the input drawing requirements and output drawing tasks of the AI ​​big model, and the corresponding table of the associated calculation functions;

[0036] By establishing the corresponding table, in the subsequent steps, the drawing task can be identified according to the drawing requirements of the AI ​​large model input by the user, and the calculation function associated with the drawing task can be called.

[0037] Among them, the system of this embodiment is mainly used for the split design of prefabricated buildings. The drawing tasks include: designing prefabricated areas as required, splitting and designing composite panels as required, designing internal partition walls and enclosure walls as required, performing cost estimation statistics as required, calculating assembly rates as required, and outputting prefabricated calculation books as required;

[0038] S1. The user inputs any drawing requirements on the front-end client;

[0039] For example, the user may say “I want to prefabricate an area” or “I want to prefabricate 80%”. The user may input any drawing requirements, including drawing actions and drawing questions.

[0040] S2. The AI ​​big model on the WEB server identifies the drawing task according to the drawing requirements and inquires for confirmation;

[0041] For example, after the AI ​​large model recognizes "I want to make a prefabricated area", it obtains the associated drawing task, that is, to design the prefabricated area as required, and asks "Do you want to confirm the prefabricated area" to determine the accuracy of the recognition drawing task.

[0042] S3, the AI ​​big model determines whether the drawing task is a drawing action, if so, proceeds to step S4, otherwise, gives an answer according to the answer in the knowledge expert database in the AI ​​big model;

[0043] Any drawing requirement raised by the user may be a drawing action or a drawing question. After the AI ​​big model recognizes the language, it first determines the type of drawing task. If it is a drawing action, it continues with the following steps; if it is a drawing question, it ends after answering it.

[0044] S4. The AI ​​big model searches for the corresponding calculation function in the calculation background according to the drawing task, and reads the relevant calculation parameters of the calculation function from the drawing;

[0045] The AI ​​big model searches for the corresponding calculation function in the calculation background according to the drawing task, determines the necessary calculation parameters according to the calculation function, and reads the required calculation parameters in the CAD drawing.

[0046] S5. Check whether calculation parameters are missing. If yes, perform parameter polling through the AI ​​big model. The user enters the missing calculation parameters on the front-end client, or modifies the drawing and re-enters S5. If no, enter step S6.

[0047] Since the calculation parameters may be read from the CAD drawing in step S3, or may be extracted from the input statement by the AI ​​large model; therefore, checking whether the calculation parameters are missing includes checking the CAD drawing parameters and checking the large model source parameters. If the missing parameters are obtained from the drawing, modify the drawing; if the missing parameters are extracted from the input statement, enter the missing calculation parameters in the front-end client.

[0048] For example, if the user asks "I want to prefabricate 80%", some parameters are extracted from the user's input statement, that is, the prefabrication ratio is 80%, and there is no missing parameter; if the user asks "I want to prefabricate the area", the prefabrication ratio cannot be obtained from the drawing, and the AI ​​big model is used to poll the parameters to obtain the prefabrication ratio from the user;

[0049] Among them, parameter polling includes:

[0050] S51, the AI ​​large model queries all calculation parameters in turn;

[0051] S52, when missing calculation parameters are found, the user inputs a statement containing the parameters;

[0052] S53, matching and recording calculation parameters according to the statement containing parameters;

[0053] S54, determine whether the query of all calculation parameters is completed, if not, return to step S51, if yes, end.

[0054] During this process, the user has the right to terminate the parameter polling. If the user terminates the task, the loop will be exited.

[0055] S6. Send the calculation parameters to the calculation backend through the WEB server, and the calculation backend calls the corresponding calculation function to obtain the calculation results according to the calculation parameters;

[0056] The WEB server and computing backend perform queries and return results through the message middleware. First, the message middleware queries whether there is a calculation function in the computing backend, then calls the calculation function, and finally returns the calculation result obtained based on the calculation parameters and the calculation function to the WEB server.

[0057] S7, the calculation backend sends the calculation results to the WEB server, and the WEB server sends them to the front-end client for display;

[0058] S8. The user is asked through the front-end client whether to modify the drawing according to the calculation result. If so, the calculation background automatically modifies the drawing; otherwise, no modification is required.

[0059] In summary, the above-mentioned conversational CAD drawing design method and system based on artificial intelligence has the following advantages:

[0060] (1) Based on artificial intelligence technology, the present invention proposes a conversational CAD drawing design system architecture, which uses artificial intelligence technology to understand requirements and quickly conduct preliminary scheme design. It can accurately understand Party A's requirements and complete the initial version of the scheme design, which greatly improves the response speed on the one hand and the design efficiency on the other hand.

[0061] (2) The present invention adopts a separate architecture, which can be implemented across platforms. The established callback function and the corresponding table of the output results of the AI ​​large model enable the large model to have function calling capabilities.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the embodiments of the present invention are described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations shall fall within the scope defined by the appended claims.

Claims

1. A conversational CAD drawing design method based on artificial intelligence, characterized in that: The operation methods include: S0. Training the AI ​​big model input corpus, establishing the corresponding relationship between the AI ​​big model input drawing requirements and output drawing tasks, and the associated calculation functions; S1. The user inputs any drawing requirements on the front-end client; S2. The AI ​​big model on the WEB server identifies the drawing task according to the drawing requirements and inquires for confirmation; S3, the AI ​​big model determines whether the drawing task is a drawing action, if so, proceeds to step S4, otherwise, gives an answer according to the answer in the knowledge expert database in the AI ​​big model; S4. The AI ​​big model searches for the corresponding calculation function in the calculation background according to the drawing task, and reads the relevant calculation parameters of the calculation function from the drawing; S5. Check whether calculation parameters are missing. If yes, perform parameter polling through the AI ​​big model. The user enters the missing parameters on the front-end client, or modifies the drawing and re-enters S5. If no, enter step S6. S6. Send the calculation parameters to the calculation backend through the WEB server, and the calculation backend calls the corresponding calculation function to obtain the calculation results according to the calculation parameters; S7, the calculation backend sends the calculation results to the WEB server, and the WEB server sends them to the front-end client for display; S8. The user is asked through the front-end client whether to modify the drawing according to the calculation result. If so, the calculation background automatically modifies the drawing; otherwise, no modification is required.

2. The artificial intelligence-based conversational CAD drawing design method according to claim 1, characterized in that: In S1, the user inputs any drawing requirement in the form of text or voice.

3. The artificial intelligence-based conversational CAD drawing design method according to claim 1, characterized in that: In S5, the parameter polling includes: S51, the AI ​​large model queries all calculation parameters in turn; S52, when missing calculation parameters are found, the user inputs a statement containing the parameters; S53, matching and recording calculation parameters according to the statement containing parameters; S54, determine whether the query of all calculation parameters is completed, if not, return to step S51, if yes, end.

4. The artificial intelligence-based conversational CAD drawing design method according to claim 1, characterized in that: The WEB server and computing backend perform searches and return results through the message middleware.

5. The artificial intelligence-based conversational CAD drawing design method according to claim 4, characterized in that: In S6, the message middleware first queries whether there is a calculation function in the calculation background, then calls the calculation function, and finally returns the calculation result obtained according to the calculation parameters and the calculation function to the WEB server.

6. The conversational CAD drawing design method based on artificial intelligence according to claim 1 is characterized in that: The drawing tasks include: designing prefabricated areas as required, designing the splitting of composite panels as required, designing internal partition walls and retaining walls as required, making cost estimates as required, calculating the assembly rate as required, and outputting prefabricated calculation books as required.

7. A conversational CAD drawing design system based on artificial intelligence, characterized in that: The conversational CAD drawing design method based on artificial intelligence as described in any one of claims 1 to 6 includes a separate architecture of a front-end client, a WEB server and a computing background, wherein the front-end client is connected to the WEB server, and the WEB server is connected to the computing background, and the front-end client includes a user interface; the WEB server includes an AI big model, and the AI ​​big model is used to interact with the user, and can identify the user input information, determine the task type and necessary task parameters; the computing background is used for function calculation, and the computing background includes a .NET function library, and after receiving the task type and task parameters, the .NET function library completes the task execution and returns the function calculation result and modifies the drawing.

Images