Construction scheme output method and device, electronic equipment and storage medium

Abstract

The invention relates to a construction scheme output method and device, electronic equipment and a storage medium, and relates to the technical field of engineering. The method comprises the following steps: acquiring a name, a type, a directory and a design book of a construction scheme; performing data slicing on the design book to determine a first slice set; determining the content of the construction scheme based on the name, the type, the directory, the first slice set and a pre-trained multi-modal large model; and filling the construction scheme according to the content, and outputting the filled construction scheme. According to the invention, technical personnel do not need to manually compile all construction schemes, and the filling efficiency and quality of the construction schemes are effectively improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a construction plan output method, device, electronic device and storage medium. Background Art

[0002] Currently, construction plans in the construction industry are mostly completed manually, which is not only time-consuming but also requires high professional skills from technicians. With the rapid development of artificial intelligence technology, it has become possible to assist in completing construction plans based on large-scale model technology. Large-scale models have achieved significant results in areas such as natural language processing, image recognition, and long text generation. This application is filed to improve the efficiency and quality of completing construction plans. Summary of the Invention

[0003] Embodiments of the present disclosure provide a construction plan output method, device, electronic device, and storage medium.

[0004] In the first aspect, an embodiment of the present disclosure provides a method for outputting a construction plan, including: obtaining the name, type, directory and design book of the construction plan; slicing the design book to determine a first slice set; determining the content of the construction plan based on the name, type, directory, first slice set and a pre-trained multimodal large model; filling in the construction plan according to the content, and outputting the filled-in construction plan.

[0005] In the second aspect, an embodiment of the present disclosure provides a construction plan output device, including: a data acquisition unit, configured to obtain the name, type, directory and design book of the construction plan; a data slicing unit, configured to slice the design book and determine a first slice set; a content determination unit, configured to determine the content of the construction plan based on the name, type, directory, first slice set and a pre-trained multimodal large model; a plan output unit, configured to fill in the construction plan according to the content and output the filled-in construction plan.

[0006] In a third aspect, an embodiment of the present disclosure provides an electronic device comprising a memory, a processor, a bus, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the construction plan output method described in the first aspect is implemented.

[0007] In a fourth aspect, an embodiment of the present disclosure provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the construction plan output method described in the first aspect.

[0008] Applying the technical solution of the present disclosure, one can first obtain the name, type, directory, and design document of the construction plan. Then, the design document is sliced ​​to obtain a first slice set. Based on the name, type, directory, first slice set, and a pre-trained multimodal large model, the content of the construction plan is determined. Finally, the construction plan is filled in based on the above content, and the completed construction plan is output. Applying the present disclosure eliminates the need for technicians to manually compile the entire construction plan, effectively improving the efficiency and quality of completing the construction plan.

[0009] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings are provided to facilitate a better understanding of the present invention and do not constitute a limitation of the present disclosure.

[0011] Figure 1 An exemplary system architecture diagram is shown in which an embodiment of the construction plan output method disclosed herein can be applied;

[0012] Figure 2 A flowchart of an embodiment of the construction plan output method disclosed herein;

[0013] Figure 3 This is a flow chart of another embodiment of the construction plan output method disclosed herein;

[0014] Figure 4 This is a structural diagram of an embodiment of a construction plan output device disclosed herein;

[0015] Figure 5 This is a schematic structural diagram of an embodiment of an electronic device disclosed herein. DETAILED DESCRIPTION

[0016] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present disclosure belongs.

[0017] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0018] In the absence of conflict, the embodiments of the present disclosure and the features thereof may be combined with each other.

[0019] In order to make the technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] Figure 1 An exemplary system architecture 100 is shown to which an embodiment of the construction plan output method or construction plan output device disclosed herein can be applied.

[0021] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. Network 104 is a medium for providing communication links between terminal devices 101, 102, 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0022] Users can use terminal devices 101, 102, 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications, such as browser applications and document editing applications, can be installed on terminal devices 101, 102, 103.

[0023] Terminal devices 101, 102, and 103 can be hardware or software. When terminal devices 101, 102, and 103 are hardware, they can be various electronic devices, including but not limited to smartphones, tablet computers, car computers, laptop computers, and desktop computers. When terminal devices 101, 102, and 103 are software, they can be installed in the electronic devices listed above. They can be implemented as multiple software or software modules (for example, to provide distributed services), or as a single software or software module. No specific limitations are given here.

[0024] The server 105 may be a server that provides various services, such as a backend server that processes the construction plan information sent by the terminal devices 101, 102, and 103. The backend server may determine the content of the construction plan based on the information sent by the user through each terminal device 101, 102, and 103, fill in the content, and feed the completed construction plan back to the terminal devices 101, 102, and 103.

[0025] It should be noted that the server 105 can be hardware or software. When the server 105 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server 105 is software, it can be implemented as multiple software or software modules (for example, to provide distributed services), or it can be implemented as a single software or software module. No specific limitations are given here.

[0026] It should be noted that the construction plan output method provided in the embodiment of the present disclosure can be executed by the terminal devices 101, 102, 103, or by the server 105. Accordingly, the construction plan output device can be set in the terminal devices 101, 102, 103, or in the server 105.

[0027] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0028] Figure 2 FIG2 shows a process 200 of an embodiment of the construction plan output method disclosed in the present invention. Figure 2 As shown, the construction plan output method of this embodiment may include the following steps:

[0029] Step 201: Obtain the name, type, catalog and design document of the construction plan.

[0030] In this embodiment, the execution subject of the construction plan output method (for example Figure 1 The terminal devices 101, 102, 103 or the server 105 shown can obtain the name, type, directory and design document of the construction plan in various ways. For example, the user can enter the above information through a specified page, or the execution entity can obtain the above information from a specified path. The name of the construction plan may include information such as the location and type of the construction plan. The type of the construction plan may be one of multiple pre-set categories. The directory can be understood as the outline of the construction plan, which can be a three-level directory. The design document can be a document issued by a design institute, which may include guidance information for the construction plan.

[0031] Step 202: Slice the design document to determine a first slice set.

[0032] After obtaining the design document, the information in the design document can be extracted more finely, that is, the design document data can be sliced ​​to obtain a first slice set. Specifically, the execution entity can divide the text in the design document. For example, the division can be based on the number of paragraphs or the number of words. The charts in the design document can also be divided. Each slice in the first slice set can be text information or a vector of a chart.

[0033] Step 203: Determine the content of the construction plan based on the name, type, directory, first slice set, and pre-trained multimodal large model.

[0034] After obtaining the first slice set, the above-mentioned name, type, directory, and first slice set can be input into a pre-trained multimodal large model. Multimodal Large Language Models (MLLMs) are a type of model that combines the natural language processing capabilities of large language models (LLMs) with the ability to understand and generate data from other modalities (such as vision, audio, etc.). These models provide a richer and more natural interactive experience by integrating multiple types of inputs and outputs such as text, images, and sounds. Specifically, the execution subject can be divided according to the name, type, directory, and first slice set to obtain multiple subtasks. Then, each subtask is input into the above-mentioned multimodal large language model to obtain the output result of each subtask. Finally, the output results are spliced ​​together to obtain the content of the construction plan.

[0035] Step 204: Fill in the construction plan according to the above content and output the filled-in construction plan.

[0036] After obtaining the contents of each part, the construction plan can be filled in and the filled-in construction plan can be output so that the user can view the filled-in construction plan.

[0037] The construction plan output method provided by the above-mentioned embodiment of the present disclosure can first obtain the name, type, directory, and design document of the construction plan. Then, the design document is sliced ​​to obtain a first slice set. Based on the name, type, directory, first slice set, and a pre-trained multimodal large model, the content of the construction plan is determined. Finally, the construction plan is filled in based on the above content, and the completed construction plan is output. Application of the present disclosure eliminates the need for technicians to manually compile the entire construction plan, effectively improving the efficiency and quality of filling in the construction plan.

[0038] Continue to see Figure 3 , which shows a process 300 of another embodiment of the construction plan output method according to the present disclosure. Figure 3As shown, the method in this embodiment may include the following steps:

[0039] Step 301: Obtain the name, type, catalog and design document of the construction plan.

[0040] In some optional implementations of this embodiment, the type of the construction plan can be determined by a pre-trained model. Specifically, the name of the construction plan can be input into the model, and the model outputs the type of the construction plan. The model can be trained using the names and corresponding types of historical construction plans as a training set.

[0041] In some optional implementations of the present embodiment, the above-mentioned directory can also be determined by a pre-trained model, or can be determined by a pre-stored directory template. Specifically, the name and type of the above-mentioned construction plan can be input into the above-mentioned model, and the output of the model is the directory of the construction plan. The above-mentioned model can be trained using the name and type of the historical plan and the corresponding directory as a training set. The preset knowledge base can store historical construction plans of different categories. After determining the type of the construction plan, the directory of the construction plan can be determined based on the directory of historical construction plans of the same category. For example, the directory of the historical construction plan can be directly used as the directory of the construction plan. Alternatively, the directories of each historical construction plan of the same category can be fused to obtain the directory of the construction plan.

[0042] In some optional implementations of this embodiment, a directory template may also be stored in the preset knowledge base. That is, different categories correspond to different directory templates, which can be determined by professional and technical personnel based on experience and actual construction plans. After determining the category of the current construction plan, the directory template corresponding to the category can be used as the directory of the current construction plan. The above-mentioned knowledge base can be a memory index and a distributed database, which stores each historical document vector in the database to prepare for subsequent accurate retrieval. Through the inverted index technology of the distributed database, cross-document association of elements such as construction technology, material parameters, and safety standards can be achieved.

[0043] Step 302: Slice the design document to determine a first slice set.

[0044] Step 303: Determine target-related documents of the construction plan based on the name, type, directory, and a preset knowledge base.

[0045] The executing entity can also search a preset knowledge base based on name, type, and directory to identify target-related documents for the construction plan. Specifically, keywords can be extracted from the name, type, and directory, and the preset knowledge base can be searched using these keywords. Alternatively, vectors corresponding to the name, type, and directory can be determined, and the preset knowledge base can be searched using these vectors. The preset knowledge base can include multiple types of knowledge, for example, historical construction plan documents, vectors corresponding to text in historical plan documents, historical design books, and so on.

[0046] Through the above search, multiple documents can be determined. The execution entity can sort the documents according to their relevance and use the top N documents as target related documents.

[0047] Step 304: Slice the target-related documents to determine a second slice set.

[0048] In this embodiment, long text in a design document or target-related document can be segmented into short texts whose numerical value is less than a preset threshold. This threshold can be 200. For charts in a design document or target-related document, the chart's address, title, and description can be used as data slices. This description can be obtained from a multimodal macromodel. Specifically, the chart is directly input into the multimodal macromodel, which generates the chart's description information.

[0049] Step 305 : Determine the content of the construction plan based on the name, type, directory, first slice set, second slice set, and multimodal large model.

[0050] After obtaining the second slice set, it can be input into the multimodal large model together with the name, type, directory and the first slice set to obtain the content of the construction plan.

[0051] In some optional implementations of this embodiment, after receiving the above input, the multimodal macromodel can generate a subtask set. Each subtask in the above subtask set has a task identifier. The multimodal macromodel can execute multiple subtasks in parallel. After completing a single subtask, the content corresponding to the subtask can be output and filled in. The user can query the execution progress of each subtask based on the task identifier.

[0052] When executing a single subtask, you can first set the status of the subtask to the initial state. Then, obtain the relevant documents of the subtask. Traverse the directory structure of each relevant document to determine whether each part in the target structure corresponds to the template content. If so, use the template content as a reference to fill in the construction plan. If there is no template content, you can traverse all relevant documents to determine whether the corresponding chapter name is completely matched. If it is completely matched, use the content of the chapter as a reference to fill in the construction plan. If it is not completely matched, you can return to the step of traversing the relevant documents. If all relevant documents have been traversed and no completely matching chapter name is found, the content in the partially matching chapter can be used as a reference. After determining the reference content, the language model can be used to intelligently fill in the construction plan.

[0053] Step 306 , obtaining the user's polishing requirements and an overview of the implementation plan for the above content; polishing the above content based on the above content, the polishing requirements, the above overview, the first slice set, and the multimodal large model.

[0054] In this embodiment, if the user is dissatisfied with the content output by the multimodal macromodel, they can further input polishing requirements and an outline of the implementation plan into the multimodal macromodel. In this way, the multimodal macromodel can further polish, expand, and diverge the content based on the polishing requirements and the outline of the implementation plan, combined with the first slice set, thereby achieving polishing of the content.

[0055] Step 307: Fill in the construction plan based on the polished content and output the filled-in construction plan.

[0056] The construction plan output method provided by the above-mentioned embodiment of the present disclosure can make full use of artificial intelligence algorithms to fill in various parts of the construction plan, effectively improving the filling efficiency and filling quality of the construction plan; and the present disclosure supports multiple rounds of iterative optimization, that is, the content output by the multimodal large model can be polished, expanded and diverged according to the polishing requirements, further improving the quality of the output text.

[0057] Further references Figure 4 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a construction plan output device. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0058] like Figure 4 As shown, the construction plan output device 400 of this embodiment includes: a data acquisition unit 401, a data slicing unit 402, a content determination unit 403 and a plan output unit 404.

[0059] The data acquisition unit 401 is configured to acquire the name, type, directory and design document of the construction plan.

[0060] The data slicing unit 402 is configured to perform data slicing on the design document and determine a first slice set.

[0061] The content determination unit 403 is configured to determine the content of the construction plan based on the name, type, directory, first slice set and pre-trained multimodal large model.

[0062] The plan output unit 404 is configured to fill in the construction plan according to the content and output the filled-in construction plan.

[0063] In addition, in the technical solution of this application, an electronic device is also proposed.

[0064] Figure 5 A schematic structural diagram of an electronic device provided by an embodiment of the present disclosure is shown.

[0065] like Figure 5 As shown, the electronic device may include a processor 501, a memory 502, a bus 503, and a computer program stored in the memory 502 and executable on the processor 501, wherein the processor 501 and the memory 502 communicate with each other via the bus 503. When the processor 501 executes the computer program, the steps of the above method are implemented, for example, including: obtaining the name, type, directory, and design document of the construction plan; slicing the data of the design document to determine a first slice set; determining the content of the construction plan based on the name, type, directory, first slice set, and a pre-trained multimodal large model; filling in the construction plan according to the content, and outputting the filled-in construction plan.

[0066] In addition, in one embodiment of the present disclosure, a non-transitory computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented, for example, including: obtaining the name, type, directory and design book of the construction plan; slicing the design book to determine the first slice set; determining the content of the construction plan based on the name, type, directory, first slice set and a pre-trained multimodal large model; filling in the construction plan according to the content, and outputting the filled-in construction plan.

[0067] In summary, in the technical solution disclosed herein, the name, type, directory, and design document of the construction plan can be first obtained. Then, the design document is sliced ​​to obtain a first slice set. Based on the name, type, directory, first slice set, and a pre-trained multimodal large model, the content of the construction plan is determined. Finally, the construction plan is filled in based on the above content, and the completed construction plan is output. Applying this disclosure eliminates the need for technicians to manually compile the entire construction plan, effectively improving the efficiency and quality of completing the construction plan.

[0068] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A construction plan output method, comprising: Obtain the name, type, catalog and design document of the construction plan; Slicing the design document to determine a first slice set; Determining content of the construction plan based on the name, the type, the directory, the first slice set, and a pre-trained multimodal large model; The construction plan is filled in according to the content, and the filled construction plan is output.

2. The method according to claim 1, wherein The determining of the content of the construction plan based on the name, the type, the directory, the first slice set, and the pre-trained multimodal large model includes: Determining target-related documents of the construction plan according to the name, the type, the directory, and a preset knowledge base; Slicing the target-related documents to determine a second slice set; The content of the construction plan is determined based on the name, the type, the directory, the first slice set, the second slice set, and the pre-trained multimodal large model.

3. The method according to claim 1, wherein The method further comprises: Obtaining the user's requirements for polishing the content and an overview of the construction plan; The content is polished according to the content, the polishing requirements, the profile, the first slice set, and the multimodal macro model.

4. The method according to claim 1, wherein The data slicing of the design book includes: Cutting the long text of the design document into short texts whose numerical value is less than a preset threshold; For the chart in the design book, the address, title and description information of the chart are determined, and the data slice is determined according to the address, title and description information.

5. The method according to claim 2, wherein: The method further comprises: According to the category of the construction plan, determining historical construction plans of the same category from the knowledge base; According to the catalog of the historical construction plans, the catalog of the construction plans is determined.

6. The method according to claim 5, wherein: The method further comprises: In response to determining that the category corresponds to a catalog template, a catalog of the construction plan is determined based on the catalog template.

7. The method according to claim 2, wherein: Determining target-related documents of the construction plan based on the name, the type, the directory, and a preset knowledge base includes: Determine a keyword and vector set according to the name, the type, and the directory; Searching the knowledge base based on the keywords and the vector set to determine the similarity between the documents in the knowledge base and the construction plan; The target-related document is determined according to the similarity.

8. A construction plan output device, comprising: A data acquisition unit is configured to acquire the name, type, directory and design document of the construction plan; a data slicing unit configured to slice the design document to determine a first slice set; a content determination unit configured to determine the content of the construction plan based on the name, the type, the directory, the first slice set, and a pre-trained multimodal large model; The plan output unit is configured to fill in the construction plan according to the content and output the filled-in construction plan.

9. An electronic device comprising a memory, a processor, a bus, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the construction plan output method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the construction plan output method according to any one of claims 1 to 7 is implemented.

Images