Method for acquiring target building information based on LLM model, medium and equipment

By introducing an information processing method based on the LLM model in BIM technology, the operation inconvenience caused by excessive component information in BIM technology is solved, and more efficient and accurate information retrieval and BIM model operation are achieved.

CN119557960AInactive Publication Date: 2025-03-04CHINA TRANSPORT INFORMATION TECH GRP CO LTD
View PDF 16 Cites 0 Cited by

Patent Information

Application Number
CN202510104816.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-01-23
Publication Date
2025-03-04
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In the existing BIM technology, too much component information makes it difficult for professionals to add, delete, and modify during the modeling process, rather than professionals who have knowledge barriers when learning BIM software operations, resulting in inconvenience in information retrieval and component operation.

Method used

Using the LLM model-based method, the target text and its corresponding operation type are obtained, and the answer text is generated after processing, thereby obtaining the target building information. This method combines the LLM model with the BIM model, simplifying the information retrieval and BIM model construction process.

Benefits of technology

It improves the accessibility and interactivity of BIM data, simplifies the process of obtaining target building information, and enhances the accuracy and efficiency of information retrieval without prior data integration.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119557960A_ABST
    Figure CN119557960A_ABST
Patent Text Reader

Abstract

The invention provides a method, medium and device for obtaining target building information based on an LLM model, and relates to the technical field of data processing, and the method comprises the steps: obtaining a target text, obtaining a target operation type corresponding to the target text, processing the target text in a first mode when the target operation type is a first BIM operation type, and obtaining an answer text, and when the target operation type is a second BIM operation type, processing the target text in a second mode to obtain an answer text corresponding to the target text, and obtaining the target building information corresponding to the target text according to the target text and the answer text, so that the process of information retrieval and BIM model building is simplified, and the efficiency of information retrieval is improved. According to the method, the accessibility and the interactivity of the BIM data are improved, the BIM model is combined with the LLM model, and the integration of prior data is not needed, so that the accuracy of the obtained target building information is relatively high.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method, medium and device for acquiring target building information based on an LLM model. Background Art

[0002] With the gradual maturity of artificial intelligence technology, the application scenarios of artificial intelligence are becoming more and more abundant. At present, with the development of BIM technology, more and more engineers have joined the ranks of BIM technology applications, and more and more components have entered the BIM three-dimensional model component library, making the BIM model contain rich building component information, such as physical properties, functional characteristics, and the entire life cycle of construction, operation and maintenance. Therefore, some problems in the development of BIM model data have also followed. Too many components have brought troubles to professionals in the addition, deletion, query and modification of BIM modeling process, and the knowledge barriers faced by non-professionals in learning the specific professional operations of BIM software have also brought great inconvenience to entry and rapid participation in engineering projects. Therefore, how to propose a method for accurate information retrieval and component operation to obtain target building information has become an urgent problem to be solved. Summary of the invention

[0003] In view of the above technical problems, the technical solution adopted by the present invention is a method for obtaining target building information based on the LLM model, and the method comprises the following steps:

[0004] A target text is obtained, wherein the target text is a question text or an operation text associated with the construction field input by a target user.

[0005] A target operation type corresponding to the target text is obtained, wherein the target operation type is a type of operation on the BIM data represented by the target text.

[0006] When the target operation type is the first BIM operation type, the target text is processed in a first manner to obtain an answer text corresponding to the target text.

[0007] When the target operation type is the second BIM operation type, the target text is processed in the second manner to obtain an answer text corresponding to the target text.

[0008] According to the target text and the answer text, the target building information corresponding to the target text is obtained.

[0009] The present invention also provides a non-transitory computer-readable storage medium, in which at least one instruction or at least one program is stored, and the at least one instruction or at least one program is loaded and executed by a processor to implement the above-mentioned method for obtaining target building information based on the LLM model.

[0010] The present invention also provides an electronic device, comprising a processor and the above-mentioned non-transitory computer-readable storage medium.

[0011] The present invention has at least the following beneficial effects: obtaining a target text, wherein the target text is a question text or an operation text associated with the construction field input by a target user, obtaining a target operation type corresponding to the target text, wherein the target operation type is a type of operation on BIM data represented by the target text, when the target operation type is a first BIM operation type, processing the target text to obtain an answer text corresponding to the target text, and when the target operation type is a second BIM operation type, processing the target text to obtain an answer text corresponding to the target text, and obtaining target building information corresponding to the target text based on the target text and the answer text. It can be seen that the process of information retrieval and BIM model construction is simplified, the accessibility and interactivity of BIM data are improved, the process of obtaining target building information is improved, and the BIM model is combined with the LLM model without the need for integration of prior data, so that user needs can be obtained more accurately, so that the accuracy of the obtained target building information is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A flow chart of a method for obtaining target building information based on an LLM model provided in Embodiment 1 of the present invention;

[0014] Figure 2 This is a flow chart of S300 provided in the first embodiment of the present invention;

[0015] Figure 3 This is a flow chart of S400 provided in the first embodiment of the present invention;

[0016] Figure 4 This is a flow chart of S500 provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0018] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0019] Embodiment 1

[0020] This embodiment provides a method for obtaining target building information based on the LLM model, the method comprising the following steps: Figure 1 As shown:

[0021] S100, obtaining a target text, wherein the target text is a question text or an operation text associated with the construction field input by a target user.

[0022] Specifically, the question text is a text that raises questions about specific information, for example, what is BIM, what is the material of Pier 1 of Bridge A, and other question texts.

[0023] Specifically, the operation text is a text for performing a certain operation on specific information, for example, an operation text for modifying the height of the No. 1 pier of Bridge A to 5 m.

[0024] Specifically, the target user is a user who interacts with the target BIM software, wherein the target BIM software is BIM software embedded with a target LLM model.

[0025] Furthermore, the target LLM model is a trained LLM model, wherein those skilled in the art know that the methods of selecting any LLM model for model training in the prior art all fall within the protection scope of the present invention and will not be repeated here, for example: GPT and other LLM models.

[0026] Furthermore, a target box is set in the target BIM model, wherein the target box is used for the target user to input the target text.

[0027] Specifically, the target LLM model obtains the target text through the curl interface.

[0028] S200, obtaining a target operation type corresponding to a target text, wherein the target operation type is a type of operation on BIM data represented by the target text.

[0029] S300: When the target operation type is a first BIM operation type, a first method is used to process the target text to obtain an answer text corresponding to the target text.

[0030] Specifically, the first BIM operation type is a BIM knowledge question and answer type.

[0031] Specifically, S300 also includes the following steps: Figure 2 As shown:

[0032] S301, obtaining a first designated text according to a target text, wherein the first designated text is a text obtained by standardizing the target text through a standardized template.

[0033] Specifically, the standardized template is "You are a senior expert in G, and you can answer H professional knowledge and questions and answers very accurately. Please use your professional knowledge in combination with specific questions to answer J." In this standard template, "G aspect" is the professional field summarized by the target LLM model for user content, "H professional knowledge" is the specific professional direction of the target LLM model for user content, and "J" is the target text input by the target user. For example, when the target text is "Please introduce the bridge structure", the corresponding designated text obtained is "You are a senior expert in the field of transportation construction, and you can answer bridge professional knowledge and questions and answers very accurately. Please use your professional knowledge in combination with specific questions to answer Please introduce the bridge structure."

[0034] S302, obtaining an answer text according to the first specified text, wherein the answer text is a text generated by answering the first specified text based on the target LLM model.

[0035] S400: When the target operation type is a second BIM operation type, a second method is used to process the target text to obtain an answer text corresponding to the target text.

[0036] Specifically, the second BIM operation type includes: querying BIM information type, modifying BIM information type and calculating BIM information type.

[0037] Specifically, S400 also includes the following steps: Figure 3 As shown:

[0038] S401, obtaining a second designated text according to a target text, wherein the second designated text includes a plurality of initial fields decomposed from the target text and initial field values ​​corresponding to the plurality of initial fields.

[0039] Specifically, the initial field is a field corresponding to a field value generated after decomposing the target text, which is consistent with a field included in a preset BIM database.

[0040] Furthermore, the preset BIM database is a database that stores a number of projects and building information corresponding to the projects, wherein the building information includes project fields, facility fields, attribute fields, sub-facility fields, value fields, operation fields, calculation fields, and field values ​​corresponding to each field.

[0041] Furthermore, the field value corresponding to the project field is a specific project name, the field value corresponding to the facility field is the name of a specific facility in the project, the field value corresponding to the attribute field is a specific attribute in a preset BIM database, the field value corresponding to the sub-facility field is the name of a specific sub-facility in the project, the field value corresponding to the value field is an attribute value corresponding to a specific attribute in the preset BIM database, the field value corresponding to the operation field is the target operation type, and the field value corresponding to the calculation field is the name of the operation involved in the natural language.

[0042] Specifically, the initial field value is a field value corresponding to the initial field extracted from the target text.

[0043] For example: when the target text is "Query the material of Pier 1 of Bridge A", the corresponding second specified text is "Operation field: Query; Facility field: Bridge A; Sub-facility field: Pier 1; Attribute field: Material"; when the target text is "Modify the height of Pier 1 of Bridge A to 5m", the corresponding second specified text is "Operation field: Modify; Facility field: Bridge A; Sub-facility field: Pier 1; Attribute field: Height; Value field: 5m".

[0044] S402: Acquire a designated field corresponding to the second designated text according to the target operation type.

[0045] Specifically, S402 also includes the following steps:

[0046] S4021: When the second BIM operation type is a BIM information query type, obtaining designated fields, wherein the designated fields include a project field, a facility field, and an attribute field.

[0047] S4022: When the second BIM operation type is to modify the BIM information type, obtain designated fields, wherein the designated fields include a project field, a facility field, an attribute field, and a value field.

[0048] S4023: When the second BIM operation type is a calculation BIM information type, obtaining designated fields, wherein the designated fields include a project field, a facility field, an attribute field, a value field, and a calculation field.

[0049] S403, input the specified field into the target API to obtain an answer text, wherein the answer text is the text output by the target API calling a preset BIM database according to the specified field.

[0050] Specifically, the target LLM model calls a preset BIM database through a curl interface.

[0051] S500, obtaining target building information corresponding to the target text according to the target text and the answer text.

[0052] Specifically, S500 also includes the following steps: Figure 4 As shown:

[0053] S501, when the target operation type corresponding to the target text is the first type of target operation type, obtaining target building information, wherein the target building information includes a target building text, and the target building text is a text generated by recompiling the target text and the answer text through a target LLM model.

[0054] Specifically, the first type of target operation types includes BIM knowledge question and answer type, BIM information modification type and BIM information calculation type.

[0055] S502, when the target operation type corresponding to the target text is the second type of target operation type, the target building information is obtained, wherein the target building information includes a target building text and a target building visualization interface, the target building text is a text generated by recompiling the target text and the answer text through the target LLM model, and the target building visualization interface is a facility interface generated based on the field value corresponding to the engineering field and the field value corresponding to the facility field in the target text.

[0056] Specifically, the second target operation type includes querying BIM information type.

[0057] Specifically, the target BIM software will display the component content in a visualization window, and the mouse can be dragged in the window to move, rotate, scale, and perform other operations on the component.

[0058] Specifically, after step S500, the following steps are also included:

[0059] S51, obtaining candidate text, wherein the candidate text is the text inputted into the target box by the target user again.

[0060] S52, when the candidate text is an empty set, the context of this conversation is cached as a historical record, and the reference to the context is terminated, thereby terminating this conversation.

[0061] S53, when the candidate text is the first candidate text, the existing information of the context is integrated to complete the model information query modification and calculation, wherein the first candidate text is the question text for the target user to further query the previous component information or the operation text implemented on the current BIM model.

[0062] S54, when the candidate text is the second candidate text, the relevant information of the previous model is discarded and the method of obtaining the target building information is re-executed, wherein the second candidate text is the question text of the target user's new query component information or the operation text implemented on the new BIM model.

[0063] From the above content, we can see that the process of information retrieval and BIM model building has been simplified, the accessibility and interactivity of BIM data have been improved, and the process of obtaining target building information has been improved. The BIM model is combined with the LLM model, and there is no need to integrate prior data. It can more accurately obtain user needs and make the accuracy of the target building information obtained higher.

[0064] This embodiment can obtain a target text, wherein the target text is a question text or an operation text associated with the construction field input by a target user, and obtain a target operation type corresponding to the target text, wherein the target operation type is a type of operation on BIM data represented by the target text. When the target operation type is a first BIM operation type, the target text is processed to obtain an answer text corresponding to the target text. When the target operation type is a second BIM operation type, the target text is processed to obtain an answer text corresponding to the target text. According to the target text and the answer text, the target building information corresponding to the target text is obtained. It can be seen that the process of information retrieval and BIM model construction is simplified, the accessibility and interactivity of BIM data are improved, the process of obtaining target building information is improved, and the BIM model is combined with the LLM model. There is no need to integrate prior data, and the user's needs can be obtained more accurately, so that the accuracy of the obtained target building information is higher.

[0065] Embodiment 2

[0066] An embodiment of the present invention provides a non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the non-transitory computer-readable storage medium, and the at least one instruction or at least one program is loaded and executed by a processor to implement the steps:

[0067] A target text is obtained, wherein the target text is a question text or an operation text associated with the construction field input by a target user.

[0068] A target operation type corresponding to the target text is obtained, wherein the target operation type is a type of operation on the BIM data represented by the target text.

[0069] When the target operation type is the first BIM operation type, the target text is processed in a first manner to obtain an answer text corresponding to the target text.

[0070] When the target operation type is the second BIM operation type, the target text is processed in the second manner to obtain an answer text corresponding to the target text.

[0071] According to the target text and the answer text, the target building information corresponding to the target text is obtained.

[0072] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0073] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0074] Embodiment 3

[0075] An embodiment of the present invention provides an electronic device, which includes a processor and the non-transitory computer-readable storage medium in the second embodiment of the present invention.

[0076] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modifications to equivalent embodiments of equivalent changes using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for obtaining target building information based on the LLM model, characterized in that: The method comprises the following steps: Acquire a target text, wherein the target text is a question text or an operation text associated with the construction field input by a target user, and the target user is a user interacting with a target BIM software, wherein the target BIM software is a BIM software embedded with a target LLM model; Obtaining a target operation type corresponding to the target text, wherein the target operation type is a type of operation on the BIM data represented by the target text; When the target operation type is the first BIM operation type, the target text is processed in a first manner to obtain an answer text corresponding to the target text; When the target operation type is the second BIM operation type, the target text is processed in the second manner to obtain an answer text corresponding to the target text; According to the target text and the answer text, the target building information corresponding to the target text is obtained.

2. The method for acquiring target building information based on the LLM model according to claim 1, characterized in that: The first BIM operation type is a BIM knowledge question and answer type.

3. The method for acquiring target building information based on the LLM model according to claim 1, characterized in that: When the target operation type is the first BIM operation type, processing the target text to obtain the answer text corresponding to the target text includes the following steps: According to the target text, a first designated text is obtained, wherein the first designated text is a text obtained by standardizing the target text through a standardized template; According to the first designated text, an answer text is obtained, wherein the answer text is a text generated by answering the first designated text based on the target LLM model.

4. The method for acquiring target building information based on the LLM model according to claim 1, characterized in that: The second BIM operation type includes: querying BIM information type, modifying BIM information type and calculating BIM information type.

5. The method for acquiring target building information based on the LLM model according to claim 1, characterized in that: When the target operation type is the second BIM operation type, the second method is used to process the target text to obtain the answer text corresponding to the target text, including the following steps: According to the target text, a second designated text is obtained, wherein the second designated text includes a plurality of initial fields decomposed from the target text and initial field values ​​corresponding to the plurality of initial fields; According to the target operation type, obtain the specified field corresponding to the second specified text; The specified field is input into the target API to obtain the answer text, wherein the answer text is the text output by the target API calling the preset BIM database according to the specified field.

6. The method for acquiring target building information based on the LLM model according to claim 5, characterized in that: The initial field is a field corresponding to a field value generated after decomposing the target text and is consistent with a field included in a preset BIM database.

7. The method for acquiring target building information based on the LLM model according to claim 1, characterized in that: According to the target text and the answer text, obtaining the target building information corresponding to the target text includes the following steps: When the target operation type corresponding to the target text is the first type of target operation type, obtaining target building information, wherein the target building information includes a target building text, and the target building text is a text generated by recompiling the target text and the answer text through the target LLM model; When the target operation type corresponding to the target text is the second type of target operation type, the target building information is obtained, wherein the target building information includes a target building text and a target building visualization interface, the target building text is a text generated by recompiling the target text and the answer text through the target LLM model, and the target building visualization interface is a facility interface generated based on the field values ​​corresponding to the engineering field and the field values ​​corresponding to the facility field in the target text.

8. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the non-transitory computer-readable storage medium, characterized in that: The at least one instruction or the at least one program is loaded and executed by the processor to implement the method for acquiring target building information based on the LLM model as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: The invention comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 8.

Citation Information

Patent Citations

  • Building information modeling method and device and electronic equipment

    CN116861507A

  • Natural language processing-based BIM (Building Information Modeling) construction method, equipment and storage medium

    CN118246115A

  • Interaction method, device and equipment based on large language model and readable storage medium

    CN118363501A

  • Question and answer type BIM (Building Information Modeling) query interaction method and system

    CN118536601A

  • Engineering consultation management method, system and device based on BIM (Building Information Modeling) and medium

    CN118708732A