Intelligent compilation system and method of building construction scheme

Abstract

The invention discloses an intelligent compilation system and method of a building construction scheme, and the system comprises a data preprocessing module which is used for obtaining historical construction scheme text data of each building engineering project category; the knowledge base construction module is used for constructing a construction scheme knowledge base of each construction engineering project category by establishing a vector index of a construction scheme vector database of each construction engineering project category; the retrieval information extraction module is used for performing keyword extraction on the construction scheme retrieval statement of the construction engineering project input by the user, and mapping and recalling the construction scheme text data of a plurality of corresponding construction engineering projects; and the construction scheme compiling module is used for obtaining a construction scheme compiling basis of the construction engineering project. Associated storage and value utilization of previous engineering construction schemes are achieved, the scheme compiling efficiency and reliability are improved through intelligent generation of the scheme compiling basis, and effective guarantee is provided for the scheme compiling quality.

Description

Technical Field

[0001] The present invention relates to the technical field of building construction, and in particular to an intelligent compilation system and method for a building construction plan. Background Art

[0002] In the field of construction engineering, as all sectors of society have higher and higher requirements for the safety of construction sites, the standardization of construction processes, and the reliability of engineering quality, the importance of engineering construction plan preparation has become increasingly prominent, and major infrastructure companies have established construction plan preparation departments for different engineering fields. At the same time, due to the rapid growth in the scale and number of engineering projects, the contradiction between the demand for engineering construction plans and the insufficient number of plan preparation personnel has become increasingly prominent. Such problems directly lead to the difficulty in ensuring the compliance of newly prepared engineering construction plans, especially the comprehensiveness and timeliness of the preparation basis, so that the quality of engineering construction plan preparation and the efficient management of existing engineering construction plans are greatly challenged. Therefore, the development of intelligent generation of engineering construction plan preparation based on historical engineering construction plan data is of great significance.

[0003] At present, in the preparation of construction plans for building projects, in order to ensure the validity of relevant standard and specification documents in the basis for the preparation of construction plans for building projects, it is necessary to repeatedly query and compare the relevant document numbers to see if they have been updated or abolished. At the same time, due to the uneven manual experience, it is easy to have missing, outdated or cross-contradictory references to standards and specifications; secondly, the management of historical construction plans mainly uses disks or paper files for storage, and the storage type is single, and the association between documents cannot be established, and the management of plan versions is chaotic; the formation of new construction plans mainly relies on experienced plan compilers to compile the content, and the compilation efficiency is low. It is difficult to use the reference value of previous construction plans of the same type of building projects, and the basis for plan compilation is prone to omissions or replacement and abolition. At the same time, with the continuous increase in the number of construction plans for building projects, it will undoubtedly cause idle knowledge resources and inconvenience in finding and using them. The preparation of construction plans for building projects mainly depends on the generation of compilation basis. Therefore, in view of the huge contradiction between the traditional basis for plan compilation and management technology and the actual needs of the basis for project plan compilation, it is urgent to propose a more efficient intelligent compilation system and method for building construction plans. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent preparation system and method for building construction plans, which realizes the associated storage and value utilization of previous engineering construction plans. The intelligent generation of the basis for plan preparation also improves the efficiency and reliability of plan preparation, assists plan preparation personnel to improve compliance efficiency, and provides effective guarantee for the quality of plan preparation.

[0005] The intelligent compilation system of the building construction plan designed by the present invention to achieve one of the above-mentioned purposes is special in that it includes:

[0006] A data preprocessing module is used to preprocess the historical construction plan data of the construction project to obtain the preprocessed historical construction plan text data of the construction project, and classify the preprocessed historical construction plan text data of the construction project based on the construction project category to obtain the historical construction plan text data of each construction project category;

[0007] A knowledge base construction module is used to convert the historical construction plan text data of each construction project category into a historical construction plan vector representation of each construction project category, store the historical construction plan vector representation of each construction project category in a construction plan vector database of each construction project category, and construct a construction plan knowledge base of each construction project category by establishing a vector index of the construction plan vector database of each construction project category;

[0008] A retrieval information extraction module is used to extract keywords from a construction plan retrieval statement of a construction project input by a user, and convert the extracted construction plan retrieval keywords of the construction project into vector representations of the construction plan retrieval keywords of the construction project; calculate the similarity between the vector representations of the construction plan retrieval keywords of the construction project and the vector representations of historical construction plans of each construction project category in the construction plan knowledge base of each construction project category, sort the vector representations of historical construction plans of each construction project category using the similarity calculation results, and screen the sorted vector representations of historical construction plans of each construction project category based on the similarity calculation results to obtain multiple vector representations of historical construction plans of each construction project category that meet the similarity requirements; and recall the construction plan text data of multiple construction projects corresponding to the multiple vector representations of construction plans of each construction project category that meet the similarity requirements through the reverse index mapping of the construction plan vector database of each construction project category;

[0009] The construction plan preparation module is used to re-sort the construction plan text data of the multiple construction projects and the construction plan search sentences of the construction projects input by the user according to their relevance, and to perform vocabulary summarization processing on the re-sorted construction plan text data of the multiple construction projects and the construction plan search sentences of the construction projects input by the user based on a large-scale language model and preset prompt words to obtain a basis for preparing the construction plan of the construction project.

[0010] Furthermore, the data preprocessing module preprocesses the historical construction plan data of the construction project, including: when the historical construction plan data of the construction project is audio, converting the audio into a document by transcribing speech into text; when the historical construction plan data of the construction project is video, extracting frames from the video into pictures, transcribing it into text by using image recognition, and converting the video into a document; when the historical construction plan data of the construction project is a document, converting the text language in the document into simplified Chinese, and converting the document format into Word or markdown format.

[0011] Furthermore, the data preprocessing module preprocesses the historical construction plan data of the construction project, and also includes: segmenting the text data of the document in the historical construction plan data of the construction project based on a text block segmentation strategy, and filtering and removing sensitive words, label characters, numbers and rich text data in the text data to obtain the preprocessed historical construction plan text data of the construction project.

[0012] Furthermore, the knowledge base construction module converts the historical construction plan text data of each construction project category into a vector representation of the historical construction plan of each construction project category, including: dividing the historical construction plan text data of each construction project category into words, and constructing an index vocabulary table through all the words after division, and each word has a unique index in the index vocabulary table; creating a word frequency vector for the historical construction plan text data of each construction project category, the length of the word frequency vector is equal to the size of the index vocabulary table, and the value at each position of the word frequency vector corresponds to the frequency of occurrence of the word in the historical construction plan text data of each construction project category; assigning word weights by calculating the word frequency and inverse document frequency of each word, mapping each word to a high-dimensional vector space, and performing weighted averaging on all word vectors contained in the historical construction plan text data of each construction project category to obtain a vector representation of the historical construction plan of each construction project category.

[0013] Furthermore, the retrieval information extraction module performs keyword extraction on the construction plan retrieval sentence of the construction project input by the user, including: performing semantic structure analysis on the construction plan retrieval sentence of the construction project input by the user to obtain construction plan retrieval keyword data of the construction project containing hierarchical relationships or construction plan retrieval keyword data of the construction project containing non-hierarchical relationships, and unequally weighting the main part of the construction plan retrieval keyword of the construction project and the subordinate part of the construction plan keyword of the construction project in the construction plan retrieval keyword data of the construction project containing hierarchical relationships, and equally weighting the construction plan retrieval keywords of the construction project in the construction plan retrieval keyword data of the construction project containing non-hierarchical relationships to obtain unequally weighted construction plan retrieval keyword data of construction projects or equally weighted construction plan retrieval keyword data of construction projects.

[0014] Furthermore, the search information extraction module calculates the similarity between the vector representation of the construction plan search keyword of the construction project and the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category, as described in the following formula:

[0015]

[0016] Among them, similarity(A,B) represents the similarity between the vector representation of the construction plan retrieval keywords of the construction project and the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category, A represents the vector representation of the construction plan retrieval keywords of the construction project; B represents the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category, ‖A‖ represents the modulus length of the vector representation of the construction plan retrieval keywords of the construction project, and ‖B‖ represents the modulus length of the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category.

[0017] Furthermore, the construction plan preparation module reorders the construction plan text data of the multiple construction projects and the construction plan search statement of the construction project input by the user, including: calculating the correlation between the construction plan text data of the multiple construction projects and the construction plan search statement of the construction project input by the user based on a cross encoder, as described in the following formula:

[0018] score(A,B)=E(A)·E(B)

[0019] Among them, score(A,B) represents the correlation score between the construction plan text data of multiple construction engineering projects and the construction plan retrieval statement of the construction engineering project input by the user, A represents the construction plan text data of multiple construction engineering projects, B represents the construction plan retrieval statement of the construction engineering project input by the user, and E represents the encoder function; the correlation score of the construction plan text data of the multiple construction engineering projects and the construction plan retrieval statement of the construction engineering project input by the user is obtained, and the construction plan text data of the multiple construction engineering projects are reordered based on the correlation score to obtain the reordered construction plan text data of the construction engineering projects.

[0020] As a preferred solution, the large-scale language model in the construction plan compilation module adopts one or more of the following: QwenLM, Wenxinyiyan, ChatGLM, and ChatGPT.

[0021] The intelligent preparation method of the building construction plan designed by the present invention to achieve the second purpose above is special in that it includes the following steps:

[0022] Preprocessing the historical construction plan data of the construction project to obtain the preprocessed historical construction plan text data of the construction project, and classifying the preprocessed historical construction plan text data of the construction project based on the construction project category to obtain the historical construction plan text data of each construction project category;

[0023] Converting the historical construction plan text data of each construction project category into a historical construction plan vector representation of each construction project category, storing the historical construction plan vector representation of each construction project category in a construction plan vector database of each construction project category, and constructing a construction plan knowledge base of each construction project category by establishing a vector index of the construction plan vector database of each construction project category;

[0024] Keyword extraction is performed on the construction plan search statement of the construction project input by the user, and the extracted construction plan search keyword of the construction project is converted into a vector representation of the construction plan search keyword of the construction project; the similarity between the vector representation of the construction plan search keyword of the construction project and the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category is calculated, and the historical construction plan vector representation of each construction project category is sorted by using the similarity calculation result, and the sorted historical construction plan vector representation of each construction project category is screened based on the similarity calculation result to obtain a plurality of historical construction plan vector representations of each construction project category that meet the similarity requirements; the construction plan text data of the plurality of construction projects corresponding to the plurality of construction plan vector representations of each construction project category that meet the similarity requirements are recalled through the reverse index mapping of the construction plan vector database of each construction project category;

[0025] The construction plan text data of the multiple construction engineering projects and the construction plan search sentences of the construction engineering projects input by the user are re-ranked according to their relevance, and based on a large-scale language model and preset prompt words, the re-ranked construction plan text data of the multiple construction engineering projects and the construction plan search sentences of the construction engineering projects input by the user are subjected to vocabulary induction processing to obtain a basis for compiling the construction plan of the construction engineering project.

[0026] A computing device designed by the present invention to achieve the third objective above includes a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0027] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned intelligent compilation method of the building construction plan.

[0028] The present invention has the following beneficial effects:

[0029] (1) The intelligent compilation system and method of building construction plans preprocesses historical construction plan data through a data preprocessing module, including audio to text, video frame extraction to text, and document format unification, ensuring data consistency and availability; the knowledge base construction module converts the preprocessed text data into vector representation and constructs a knowledge base through the index mechanism of the vector database; the retrieval information extraction module hierarchically processes and differentially weights the search keywords entered by the user, effectively capturing the semantic complexity of the search intent. By calculating vector similarity and filtering through thresholds, the most relevant candidate construction plans can be screened; the relevance of the search results is improved, and the interference of irrelevant information is reduced, allowing users to obtain the required construction plan information more quickly and accurately.

[0030] (2) The intelligent compilation system and method of the building construction plan uses a cross encoder model to finely reorder the candidate construction plans through the construction plan compilation module, and calculates the deep semantic correlation between the user input and the candidate plans to ensure that the plan finally output to the user is highly matched with his or her needs. With the help of large-scale language models (such as QwenLM, Wenxinyiyan, ChatGLM or ChatGPT) and preset professional prompts, the system can deeply understand the semantics and reorganize the content of the reordered construction plan text; it can extract key information from multiple related historical plans according to the specific needs of the user, and generate a customized construction plan in combination with the special requirements of the current project; it improves the efficiency and quality of construction plan compilation and reduces the omissions and subjective biases that may occur in the manual compilation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A module schematic diagram showing a specific embodiment of an intelligent compilation system for a building construction plan of the present invention.

[0032] Figure 2 A schematic flow chart showing a specific embodiment of a method for intelligently compiling a building construction plan according to the present invention. DETAILED DESCRIPTION

[0033] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0034] like Figure 1 to Figure 2As shown, the embodiment of the present invention discloses an intelligent compilation system and method for a building construction plan, which can meet the needs of high-efficiency and high-accuracy automated green ball crack detection and ensure the quality of green balls and the stability of the production process.

[0035] Example 1

[0036] This embodiment discloses an intelligent compilation system for a building construction plan, the system comprising:

[0037] A data preprocessing module is used to preprocess the historical construction plan data of the construction project to obtain the preprocessed historical construction plan text data of the construction project, and classify the preprocessed historical construction plan text data of the construction project based on the construction project category to obtain the historical construction plan text data of each construction project category;

[0038] A knowledge base construction module is used to convert the historical construction plan text data of each construction project category into a vector representation of the historical construction plan of each construction project category, store the vector representation of the historical construction plan of each construction project category in a construction plan vector database of each construction project category, and construct a construction plan knowledge base of each construction project category by establishing a vector index of the construction plan vector database of each construction project category. Specifically, for example, the knowledge bases that can be selected include Milvus vector library and ES library;

[0039] The retrieval information extraction module is used to extract keywords from the construction plan retrieval statement of the construction project input by the user, and convert the extracted construction plan retrieval keywords of the construction project into vector representations of the construction plan retrieval keywords of the construction project; calculate the similarity between the vector representations of the construction plan retrieval keywords of the construction project and the vector representations of the historical construction plans of each category of the construction project in the construction plan knowledge base of each category of the construction project, and use the similarity calculation results to sort the vector representations of the historical construction plans of each category of the construction project. Specifically, for example, the vector representations of the historical construction plans of each category of the construction project can be sorted in descending order according to the overall similarity, and based on the similarity The calculation results are used to screen the sorted historical construction plan vector representations of each construction project category. The similarity requirements and the TopK strategy can be used to control the number of historical construction plan vector representations of each construction project category that are screened. For example, if the screening condition is that the similarity is in the top 10, then the number of historical construction plan vector representations of each construction project category that are screened is 10, and multiple historical construction plan vector representations of each construction project category that meet the similarity requirements are obtained; the construction plan text data of multiple construction projects corresponding to the multiple construction plan vector representations of each construction project category that meet the similarity requirements are recalled through the reverse index mapping of the construction plan vector database of each construction project category;

[0040] The construction plan preparation module is used to re-rank the construction plan text data of multiple construction projects and the construction plan retrieval sentences of the construction projects input by the user in order to further improve the semantic relevance and knowledge accuracy, and to perform vocabulary induction processing on the re-ranked construction plan text data of multiple construction projects and the construction plan retrieval sentences of the construction projects input by the user based on the large-scale language model and preset prompt words to obtain the basis for the preparation of the construction plan of the construction project.

[0041] Specifically, the data preprocessing module preprocesses the historical construction plan data of the construction project, including: when the historical construction plan data of the construction project is audio, converting the audio into a document by transcribing speech into text; when the historical construction plan data of the construction project is video, extracting frames from the video into pictures, transcribing it into text by image recognition, and converting the video into a document; when the historical construction plan data of the construction project is a document, converting the text language in the document into simplified Chinese, and converting the document format into Word or markdown format.

[0042] In this embodiment, the data preprocessing module preprocesses the historical construction plan data of the construction project, and also includes: segmenting the text data of the document in the historical construction plan data of the construction project based on the text block segmentation strategy, and filtering and removing sensitive words, label characters, numbers and rich text data in the text data to obtain the preprocessed historical construction plan text data of the construction project.

[0043] In this embodiment, the knowledge base construction module converts the historical construction plan text data of each construction project category into a vector representation of the historical construction plan of each construction project category, including: dividing the historical construction plan text data of each construction project category into words, and constructing an index vocabulary table through all the words after the division, and each word has a unique index in the index vocabulary table; creating a word frequency vector for the historical construction plan text data of each construction project category, the length of the word frequency vector is equal to the size of the index vocabulary table, and the value at each position of the word frequency vector corresponds to the frequency of occurrence of the word in the historical construction plan text data of each construction project category; assigning word weights by calculating the word frequency and inverse document frequency of each word, mapping each word to a high-dimensional vector space, and performing weighted averaging on all word vectors contained in the historical construction plan text data of each construction project category to obtain a vector representation of the historical construction plan of each construction project category.

[0044] In this embodiment, the retrieval information extraction module performs keyword extraction on the construction plan retrieval statement of the construction project input by the user, including: performing semantic structure analysis on the construction plan retrieval statement of the construction project input by the user to obtain construction plan retrieval keyword data of the construction project containing hierarchical relationships or construction plan retrieval keyword data of the construction project containing non-hierarchical relationships, and unequally weighting the main part of the construction plan retrieval keyword of the construction project and the subordinate part of the construction plan keyword of the construction project in the construction plan retrieval keyword data of the construction project containing hierarchical relationships, and equally weighting the construction plan retrieval keywords of the construction project in the construction plan retrieval keyword data of the construction project containing non-hierarchical relationships to obtain unequally weighted construction plan retrieval keyword data of construction projects or equally weighted construction plan retrieval keyword data of construction projects.

[0045] Take the sample data of "Specific preparation basis of **highway project **bid, including the following aspects: (1) survey, design and owner documents; (2) laws, regulations, standards, specifications (including local standards and specifications); (3) overall construction organization design; (4) relevant company management documents; (5) actual site investigation (projects in environmentally sensitive areas must investigate the impact of the surrounding environment, if any); (6) requirements of the environmental impact assessment report and environmental impact assessment approval for this special construction (if any)." as an example. In this embodiment, "**highway project **bid" corresponds to the main part of the construction plan search keyword of the construction project, and the rest are subordinate parts of the construction plan keyword of the construction project;

[0046] Specifically, the retrieval information extraction module calculates the similarity between the vector representation of the construction plan retrieval keyword of the construction project and the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category, as described in the following formula:

[0047]

[0048] Among them, similarity(A,B) represents the similarity between the vector representation of the construction plan retrieval keywords of the construction project and the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category, A represents the vector representation of the construction plan retrieval keywords of the construction project; B represents the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category, ‖A‖ represents the modulus length of the vector representation of the construction plan retrieval keywords of the construction project, and ‖B‖ represents the modulus length of the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category.

[0049] Specifically, the construction plan preparation module reorders the relevance of the construction plan text data of multiple construction projects and the construction plan search statements of the construction projects input by the user, including: calculating the relevance of the construction plan text data of multiple construction projects and the construction plan search statements of the construction projects input by the user based on a cross encoder, as described in the following formula:

[0050] score(A,B)=E(A)·E(B)

[0051] Among them, score(A,B) represents the correlation score between the construction plan text data of multiple construction projects and the construction plan retrieval statement of the construction project input by the user, A represents the construction plan text data of multiple construction projects, B represents the construction plan retrieval statement of the construction project input by the user, and E represents the encoder function; the correlation scores of the construction plan text data of the multiple construction projects and the construction plan retrieval statement of the construction project input by the user are obtained, and the construction plan text data of the multiple construction projects are reordered based on the correlation scores to obtain the reordered construction plan text data of the construction projects.

[0052] In this embodiment, the large-scale language model in the construction plan preparation module adopts one or more of the following: QwenLM, Wenxinyiyan, ChatGLM, and ChatGPT.

[0053] In this embodiment, a large-scale language model is used for inductive processing to ensure the rigor and semantic integrity of the answer content.

[0054] Example 2

[0055] This embodiment discloses a method for intelligently compiling a building construction plan, which includes the following steps:

[0056] Step 1: preprocess the historical construction plan data of the construction project to obtain the preprocessed historical construction plan text data of the construction project, and classify the preprocessed historical construction plan text data of the construction project based on the construction project category to obtain the historical construction plan text data of each construction project category;

[0057] Step 2: convert the historical construction plan text data of each construction project category into a historical construction plan vector representation of each construction project category, store the historical construction plan vector representation of each construction project category in a construction plan vector database of each construction project category, and construct a construction plan knowledge base of each construction project category by establishing a vector index of the construction plan vector database of each construction project category;

[0058] Step 3, extract keywords from the construction plan search sentence of the construction project input by the user, and convert the extracted construction plan search keywords of the construction project into vector representations of the construction plan search keywords of the construction project; calculate the similarity between the vector representations of the construction plan search keywords of the construction project and the historical construction plan vector representations of each construction project category in the construction plan knowledge base of each construction project category, sort the historical construction plan vector representations of each construction project category using the similarity calculation result, and screen the sorted historical construction plan vector representations of each construction project category based on the similarity calculation result to obtain multiple historical construction plan vector representations of each construction project category that meet the similarity requirements; recall the construction plan text data of multiple construction projects corresponding to the multiple construction plan vector representations of each construction project category that meet the similarity requirements through the reverse index mapping of the construction plan vector database of each construction project category;

[0059] Step 4: re-rank the construction plan text data of multiple construction projects and the construction plan search sentences of the construction projects input by the user according to their relevance, and perform vocabulary induction processing on the re-ranked construction plan text data of multiple construction projects and the construction plan search sentences of the construction projects input by the user based on the large-scale language model and preset prompt words to obtain the basis for compiling the construction plan of the construction project.

[0060] Example 3

[0061] This embodiment discloses a schematic diagram of the structure of a computing device, and the specific embodiment of the present invention does not limit the specific implementation of the computing device.

[0062] The computing device may include: a processor, a communication interface, a memory, and a communication bus.

[0063] Wherein: the processor, the communication interface, and the memory communicate with each other via a communication bus.

[0064] Communication interface, used to communicate with other devices such as client or other server network elements.

[0065] The processor is used to execute the program, and specifically can execute the relevant steps in the above-mentioned embodiment of the intelligent compilation method of the building construction plan.

[0066] Specifically, the program may include program codes including computer operation instructions.

[0067] The processor may be a central processing unit (CPU), or a specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. The one or more processors included in the computing device may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0068] The memory is used to store programs. The memory may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.

[0069] The program can be specifically used to enable the processor to execute the intelligent compilation method of the building construction plan in any of the above-mentioned method embodiments. The specific implementation of each step in the program can refer to the corresponding descriptions in the corresponding steps and units in the above-mentioned intelligent compilation system and method embodiments of the building construction plan, which will not be repeated here. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described equipment and modules can refer to the corresponding process description in the above-mentioned method embodiments, which will not be repeated here.

[0070] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to realize the content of the present invention described herein, and the description of the above specific languages ​​is for disclosing the best mode of the present invention.

[0071] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0072] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting the intention that the claimed invention requires more features than those expressly recited in each claim. Rather, as reflected in the claims, inventive aspects lie in less than all of the features of the individual embodiments previously disclosed. Therefore, the claims that follow the detailed description are hereby expressly incorporated into the detailed description, with each claim itself serving as a separate embodiment of the present invention.

Claims

1. An intelligent system for compiling construction plans, characterized in that: include: A data preprocessing module is used to preprocess the historical construction plan data of the construction project to obtain the preprocessed historical construction plan text data of the construction project, and classify the preprocessed historical construction plan text data of the construction project based on the construction project category to obtain the historical construction plan text data of each construction project category; A knowledge base construction module is used to convert the historical construction plan text data of each construction project category into a historical construction plan vector representation of each construction project category, store the historical construction plan vector representation of each construction project category in a construction plan vector database of each construction project category, and construct a construction plan knowledge base of each construction project category by establishing a vector index of the construction plan vector database of each construction project category; A retrieval information extraction module is used to extract keywords from a construction plan retrieval statement of a construction project input by a user, and convert the extracted construction plan retrieval keywords of the construction project into vector representations of the construction plan retrieval keywords of the construction project; calculate the similarity between the vector representations of the construction plan retrieval keywords of the construction project and the vector representations of historical construction plans of each construction project category in the construction plan knowledge base of each construction project category, sort the vector representations of historical construction plans of each construction project category using the similarity calculation results, and screen the sorted vector representations of historical construction plans of each construction project category based on the similarity calculation results to obtain multiple vector representations of historical construction plans of each construction project category that meet the similarity requirements; and recall the construction plan text data of multiple construction projects corresponding to the multiple vector representations of construction plans of each construction project category that meet the similarity requirements through the reverse index mapping of the construction plan vector database of each construction project category; The construction plan preparation module is used to re-sort the construction plan text data of the multiple construction projects and the construction plan search sentences of the construction projects input by the user according to their relevance, and to perform vocabulary summarization processing on the re-sorted construction plan text data of the multiple construction projects and the construction plan search sentences of the construction projects input by the user based on a large-scale language model and preset prompt words to obtain a basis for preparing the construction plan of the construction project.

2. The intelligent compilation system for building construction plans according to claim 1 is characterized in that: The data preprocessing module preprocesses the historical construction plan data of the construction project, including: when the historical construction plan data of the construction project is audio, converting the audio into a document by transcribing voice into text; when the historical construction plan data of the construction project is video, extracting frames from the video into pictures, transcribing it into text by using image recognition, and converting the video into a document; when the historical construction plan data of the construction project is a document, converting the text language in the document into simplified Chinese, and converting the document format into Word or markdown format.

3. The intelligent compilation system for building construction plans according to claim 2 is characterized in that: The data preprocessing module preprocesses the historical construction plan data of the construction project, and also includes: dividing the text data of the document in the historical construction plan data of the construction project into text blocks based on a text block segmentation strategy, and filtering and removing sensitive words, label characters, numbers and rich text data in the text data to obtain the preprocessed historical construction plan text data of the construction project.

4. The intelligent compilation system for building construction plans according to claim 1 is characterized in that: The knowledge base construction module converts the historical construction plan text data of each construction project category into a vector representation of the historical construction plan of each construction project category, including: dividing the historical construction plan text data of each construction project category into words, and constructing an index vocabulary table through all the words after division, and each word has a unique index in the index vocabulary table; creating a word frequency vector for the historical construction plan text data of each construction project category, the length of the word frequency vector is equal to the size of the index vocabulary table, and the value at each position of the word frequency vector corresponds to the frequency of occurrence of the word in the historical construction plan text data of each construction project category; assigning word weights by calculating the word frequency and inverse document frequency of each word, mapping each word to a high-dimensional vector space, and performing weighted averaging on all word vectors contained in the historical construction plan text data of each construction project category to obtain a vector representation of the historical construction plan of each construction project category.

5. The intelligent compilation system of building construction plan according to claim 1 is characterized in that: The retrieval information extraction module extracts keywords from the construction plan retrieval statement of the construction project input by the user, including: performing semantic structure analysis on the construction plan retrieval statement of the construction project input by the user to obtain construction plan retrieval keyword data of the construction project containing hierarchical relationships or construction plan retrieval keyword data of the construction project containing non-hierarchical relationships, and unequally weighting the main part of the construction plan retrieval keywords of the construction project and the subordinate part of the construction plan keywords of the construction project in the construction plan retrieval keyword data of the construction project containing hierarchical relationships, and equally weighting the construction plan retrieval keywords of the construction project in the construction plan retrieval keyword data of the construction project containing non-hierarchical relationships to obtain construction plan retrieval keyword data of unequally weighted construction projects or construction plan retrieval keyword data of equally weighted construction projects.

6. The intelligent compilation system for building construction plans according to claim 1 is characterized in that: The search information extraction module calculates the similarity between the vector representation of the construction plan search keyword of the construction project and the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category, as described in the following formula: Among them, similarity(A,B) represents the similarity between the vector representation of the construction plan retrieval keywords of the construction project and the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category, A represents the vector representation of the construction plan retrieval keywords of the construction project; B represents the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category, ‖A‖ represents the modulus length of the vector representation of the construction plan retrieval keywords of the construction project, and ‖B‖ represents the modulus length of the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category.

7. The intelligent compilation system for building construction plans according to claim 1 is characterized in that: The construction plan preparation module reorders the relevance of the construction plan text data of the multiple construction projects and the construction plan search statement of the construction project input by the user, including: calculating the relevance of the construction plan text data of the multiple construction projects and the construction plan search statement of the construction project input by the user based on a cross encoder, as described in the following formula: score(A,B)=E(A)·E(B) Among them, score(A,B) represents the correlation score between the construction plan text data of multiple construction engineering projects and the construction plan retrieval statement of the construction engineering project input by the user, A represents the construction plan text data of multiple construction engineering projects, B represents the construction plan retrieval statement of the construction engineering project input by the user, and E represents the encoder function; the correlation score of the construction plan text data of the multiple construction engineering projects and the construction plan retrieval statement of the construction engineering project input by the user is obtained, and the construction plan text data of the multiple construction engineering projects are reordered based on the correlation score to obtain the reordered construction plan text data of the construction engineering projects.

8. The intelligent compilation system for building construction plans according to claim 1 is characterized in that: The large-scale language model in the construction plan preparation module adopts one or more of the following: QwenLM, Wenxinyiyan, ChatGLM, and ChatGPT.

9. An intelligent method for compiling a building construction plan, characterized in that: The steps include: Preprocessing the historical construction plan data of the construction project to obtain the preprocessed historical construction plan text data of the construction project, and classifying the preprocessed historical construction plan text data of the construction project based on the construction project category to obtain the historical construction plan text data of each construction project category; Converting the historical construction plan text data of each construction project category into a historical construction plan vector representation of each construction project category, storing the historical construction plan vector representation of each construction project category in a construction plan vector database of each construction project category, and constructing a construction plan knowledge base of each construction project category by establishing a vector index of the construction plan vector database of each construction project category; Keyword extraction is performed on the construction plan search statement of the construction project input by the user, and the extracted construction plan search keyword of the construction project is converted into a vector representation of the construction plan search keyword of the construction project; the similarity between the vector representation of the construction plan search keyword of the construction project and the vector representation of the historical construction plan of each construction project category in the construction plan knowledge base of each construction project category is calculated, and the historical construction plan vector representation of each construction project category is sorted by using the similarity calculation result, and the sorted historical construction plan vector representation of each construction project category is screened based on the similarity calculation result to obtain a plurality of historical construction plan vector representations of each construction project category that meet the similarity requirements; the construction plan text data of the plurality of construction projects corresponding to the plurality of construction plan vector representations of each construction project category that meet the similarity requirements are recalled through the reverse index mapping of the construction plan vector database of each construction project category; The construction plan text data of the multiple construction engineering projects and the construction plan search sentences of the construction engineering projects input by the user are re-ranked according to their relevance, and based on a large-scale language model and preset prompt words, the re-ranked construction plan text data of the multiple construction engineering projects and the construction plan search sentences of the construction engineering projects input by the user are subjected to vocabulary induction processing to obtain a basis for compiling the construction plan of the construction engineering project.

10. A computing device comprising a processor, a memory, a communication interface and a communication bus, characterized in that: The processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the intelligent compilation method of building construction plans as described in claim 9.

Images