Architectural drawing version update content automatic generation method based on large model

Abstract

The invention discloses an architectural drawing version update content automatic generation method based on a large model. Comprising the following steps: providing a certain number of primitive list information of a plurality of groups of architectural drawings before and after version updating, respective text description information of the primitive list information and different content text information of each group of architectural drawings after version updating relative to the architectural drawings before version updating as a fine tuning data set of a large language model; performing fine tuning on the large language model through the fine tuning data set; primitive list information of the target architectural drawing before and after version updating and respective text description information thereof are input into the fine-tuned large language model; and automatically outputting text information of the updated content of the target architectural drawing after the version is updated relative to the target architectural drawing before the version is updated through the large language model, and automatically attaching the text information to the target architectural drawing after the version is updated. The method has the advantages of high efficiency, high speed, high accuracy and intelligence.

Description

Technical Field

[0001] The present invention relates to the technical field of building engineering, and in particular to a method for automatically generating updated content of building drawing versions based on a large model. Background Art

[0002] In the process management of building structure construction, when a building drawing is revised and drawn for many times and multiple versions appear, it is necessary to compare the differences between the versions before and after the update, that is, in the later version, the updated content compared with the previous version is usually annotated next to the building drawing in the form of text description. At present, the text description of the updated content before and after the update is manually compared, described and annotated by draftsmen based on the building drawings before and after the modification, which is very likely to cause omissions, and it is time-consuming and laborious to compare and identify the modified content with the naked eye. Summary of the invention

[0003] The purpose of the present invention is to provide a method for automatically generating updated content of architectural drawings based on a large model, so as to solve the problem of low efficiency and low accuracy of manually comparing the previous and next versions, manually describing the updated content and manually annotating after the architectural drawings are updated.

[0004] In order to solve the above technical problems, the present invention provides a method for automatically generating updated content of architectural drawing versions based on a large model, comprising:

[0005] Step 1: Provide a certain number of sets of element list information of architectural drawings before and after version update and their respective text description information, as well as text information of the difference between each set of architectural drawings after version update and architectural drawings before version update as a fine-tuning data set for the large language model;

[0006] Step 2: Fine-tune the large language model by fine-tuning the dataset;

[0007] Step three, input the element list information of the target architectural drawings before and after the version update and their respective text description information into the fine-tuned large language model, and automatically output the text information of the updated content of the target architectural drawings after the version update relative to the target architectural drawings before the version update through the large language model, and automatically annotate the updated content on the target architectural drawings after the version update.

[0008] Furthermore, in the method for automatically generating updated content of architectural drawing versions based on a large model provided by the present invention, in step 2, the method for fine-tuning the large language model by fine-tuning the data set includes:

[0009] Input the element lists and text description information of the architectural drawings before and after the version update of each group in the fine-tuning dataset into the large language model. The large language model automatically compares the element lists and text description information of the architectural drawings before and after the version update of each group, and automatically outputs the difference text information between the two.

[0010] The large language model is used to compare the difference text information of the architectural drawings before and after each group of version updates with the difference content text information in the corresponding fine-tuning dataset. When the two are inconsistent, the parameter information of the large language model is fine-tuned based on the difference content text information in the fine-tuning dataset until the difference text information of the architectural drawings before and after each group of version updates output by the large language model is consistent with the difference content text information in the corresponding fine-tuning dataset, thereby obtaining a fine-tuned large language model.

[0011] Furthermore, the method for automatically generating updated content of architectural drawings based on a large model provided by the present invention gives a striking color to the text information of the updated content attached to the target architectural drawings after the version update.

[0012] Furthermore, the method for automatically generating updated content of architectural drawings based on a large model provided by the present invention assigns text information of updated content attached to the target architectural drawings after the version update to a separate layer.

[0013] Furthermore, the method for automatically generating version update content of architectural drawings based on a large model provided by the present invention annotates the text information of the update content automatically output by the large language model next to the graphic element or on one side of the frame of the target architectural drawing after the version update.

[0014] Compared with the prior art, the present invention has the following beneficial effects:

[0015] The method for automatically generating updated content of architectural drawings based on a large model provided by the present invention utilizes the large language model technology of artificial intelligence, and uses the element list information of several groups of architectural drawings before and after the version update and their respective text description information, as well as the text information of the difference content of each group of architectural drawings after the version update relative to the architectural drawings before the version update as the fine-tuning data set of the large language model to fine-tune the large language model, and automatically compares the element list information of the target architectural drawings before and after the version update and their respective text description information through the fine-tuned large language model, automatically outputs the text information of the updated content of the target architectural drawings after the version update relative to the target architectural drawings before the version update, and automatically annotates it on the target architectural drawings after the version update, thereby realizing the automatic generation and automatic annotation of the text information of the updated content on the target architectural drawings after the version update. Compared with manual comparison, manual description and manual annotation of updated content, it has the advantages of high efficiency, high speed, high accuracy and intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of a method for automatically generating updated content of architectural drawing versions based on a large model. DETAILED DESCRIPTION

[0017] The present invention is described in detail below in conjunction with the accompanying drawings: The advantages and features of the present invention will become more apparent from the following description. It should be noted that the accompanying drawings are in very simplified form and in non-precise proportions, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.

[0018] Please refer to Figure 1 The embodiment of the present invention provides a method for automatically generating updated content of architectural drawing versions based on a large model, which may include the following steps:

[0019] Step 1: Provide a certain number of sets of element list information of architectural drawings before and after version update and their respective text description information, as well as the difference content text information of each set of architectural drawings after version update relative to the architectural drawings before version update as a fine-tuning data set for the large language model. The large language model can select GPT or other Transformer-based large language models. The text description information here includes but is not limited to the relative position relationship, connection relationship, filling, block reference, text annotation, axis grid, color, layer and other information of each building component in the element information of the architectural drawing. The building components include but are not limited to walls, beams, plates, columns, etc. Some example text description information can be: "There is a structural plate at axis grid 36 to axis grid 37, and the elevation of the structural plate is 12.80m", "There is a circular filling at the position (100, 123) in the drawing", etc. The element list information, text description information and difference content text information here are used as input content of the large language model.

[0020] Step 2: fine-tune the large language model through the fine-tuning dataset. Specifically, the list of primitives and text description information of the architectural drawings before and after each group of version updates in the fine-tuning dataset are input into the large language model, and the large language model automatically compares the list of primitives and text description information of the architectural drawings before and after each group of version updates and automatically outputs the difference text information between the two; the large language model compares the difference text information of the architectural drawings before and after each group of version updates with the difference content text information in the corresponding fine-tuning dataset. When the two are inconsistent, the parameter information of the large language model is fine-tuned based on the difference content text information in the fine-tuning dataset until the difference text information of the architectural drawings before and after each group of version updates output by the large language model is consistent with the difference content text information in the corresponding fine-tuning dataset, and the fine-tuned large language model is obtained. An example of the automatic comparison output text can be "there is a structural beam at the axis grid 15 to the axis grid 16 in text 1, but it does not exist in text 2", and an example of the comparison result data between the two can be "the elevation of the structural plate at the axis grid 36 to the axis grid 37 in text 2 is 300mm lower than that in text 1", where the comparison result is the updated content. In order to avoid confusion in the version order of each group of updated architectural drawings, the architectural drawings before and after the version update can be automatically numbered or prompted with words such as "Drawing 1 is the previous version" and "Drawing 2 is the next version".

[0021] Step 3: Input the element list information of the target building drawings before and after the version update and their respective text description information into the fine-tuned large language model, and automatically output the text information of the updated content of the target building drawings after the version update relative to the target building drawings before the version update through the large language model and automatically annotate it on the target building drawings after the version update. That is, the update log of the target building drawings after the version update relative to the target building drawings before the version update is automatically obtained through the fine-tuned large language model. An example of the version update log output text may be "the structural beam at the axis grid 15 to the axis grid 16 is cancelled".

[0022] The method for automatically generating updated content of architectural drawings based on a large model provided by the embodiment of the present invention utilizes the large language model technology of artificial intelligence, and uses the element list information of several groups of architectural drawings before and after the version update and their respective text description information, as well as the text information of the difference content of each group of architectural drawings after the version update relative to the architectural drawings before the version update as the fine-tuning data set of the large language model to fine-tune the large language model, and automatically compares the element list information of the target architectural drawings before and after the version update and their respective text description information through the fine-tuned large language model, automatically outputs the text information of the updated content of the target architectural drawings after the version update relative to the target architectural drawings before the version update, and automatically annotates it on the target architectural drawings after the version update, thereby realizing the automatic generation and automatic annotation of the text information of the updated content on the target architectural drawings after the version update. Compared with manual comparison, manual description and manual annotation of updated content, it has the advantages of high efficiency, high speed, high accuracy and intelligence.

[0023] In order to facilitate identification and viewing of updated content, the large-model-based architectural drawing version updated content automatic generation method provided in an embodiment of the present invention gives a striking color to the text information of the updated content attached to the target architectural drawing after the version update.

[0024] To facilitate later editing and management, the method for automatically generating architectural drawing version update content based on a large model provided in an embodiment of the present invention assigns text information of the update content attached to the target architectural drawing after the version update to a separate layer.

[0025] In order not to affect the readability of the graphic element information of the target architectural drawing after the version update, the method for automatically generating the updated content of the architectural drawing based on the big model provided in the embodiment of the present invention will automatically output the text information of the updated content of the target architectural drawing after the version update through the big language model and annotate it next to the graphic element of the target architectural drawing after the version update or on one side of the drawing frame.

[0026] The method for automatically generating updated content of architectural drawings based on a large model provided in an embodiment of the present invention can avoid omissions caused by negligence in manual comparison and visual fatigue caused by manual comparison, thus saving time and effort.

[0027] The present invention is not limited to the above-mentioned specific implementation modes. Obviously, the above-mentioned embodiments are only some embodiments of the embodiments of the present invention, but not all embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention. Those skilled in the art can make other levels of modifications and changes to the present invention. In this way, if these modifications and changes of the present invention fall within the scope of the claims of the present invention, the present invention is also intended to include these changes and changes.

Claims

1. A method for automatically generating updated content of architectural drawings based on a large model, characterized in that: include: Step 1: Provide a certain number of sets of element list information of architectural drawings before and after version update and their respective text description information, as well as text information of the difference between each set of architectural drawings after version update and architectural drawings before version update as a fine-tuning data set for the large language model; Step 2: Fine-tune the large language model by fine-tuning the dataset; Step three, input the element list information of the target architectural drawings before and after the version update and their respective text description information into the fine-tuned large language model, and automatically output the text information of the updated content of the target architectural drawings after the version update relative to the target architectural drawings before the version update through the large language model, and automatically annotate the updated content on the target architectural drawings after the version update.

2. The method for automatically generating updated content of architectural drawings based on a large model according to claim 1 is characterized in that: In step 2, the method of fine-tuning the large language model by fine-tuning the dataset includes: Input the element lists and text description information of the architectural drawings before and after the version update of each group in the fine-tuning dataset into the large language model. The large language model automatically compares the element lists and text description information of the architectural drawings before and after the version update of each group, and automatically outputs the difference text information between the two. The large language model is used to compare the difference text information of the architectural drawings before and after each group of version updates with the difference content text information in the corresponding fine-tuning dataset. When the two are inconsistent, the parameter information of the large language model is fine-tuned based on the difference content text information in the fine-tuning dataset until the difference text information of the architectural drawings before and after each group of version updates output by the large language model is consistent with the difference content text information in the corresponding fine-tuning dataset, thereby obtaining a fine-tuned large language model.

3. The method for automatically generating updated content of architectural drawings based on a large model according to claim 1 is characterized in that: The text information of the updated content attached to the target building drawing after the version update is given a striking color.

4. The method for automatically generating updated content of architectural drawings based on a large model according to claim 1 is characterized in that: The text information of the updated contents attached to the target architectural drawing after the version update is assigned to a separate layer.

5. The method for automatically generating updated content of architectural drawings based on a large model according to claim 1 is characterized in that: The text information of the updated content automatically output by the large language model is annotated beside the graphic element or on one side of the frame of the target architectural drawing after the version update.

Images