Intelligent analysis and budget calculation method and system for constructional engineering drawings

Abstract

The invention relates to an intelligent analysis and budgetary calculation method and system.The method comprises the steps that an obtained building engineering drawing is analyzed through a geometric figure recognition algorithm, and project structured data of initial design is obtained; analyzing the obtained user demand instruction through a large language model to obtain a demand structured instruction; based on the demand structured instruction, modifying the project structured data through a large language model to obtain demand structured data; and on the basis of the demand structured data, cost calculation of the constructional engineering after demand modification is carried out through an engineering budget algorithm, and a demand budget report is obtained. According to the method and the device, the LLM can be instructed to modify and obtain the budget report by directly inputting the instruction by a user under the condition that the requirements of the construction project are changed, and budget does not need to be performed after manual modification in professional software again, so that the efficiency of automatic engineering budget is greatly improved, and the problem of how to improve the efficiency of existing construction engineering budget is solved.

Description

Technical Field

[0001] This application relates to the field of computer graphics technology, and in particular, to an intelligent parsing and estimation method and system for construction engineering drawings. Background Art

[0002] Engineering estimate (Design Estimate) refers to the preliminary calculation of the total investment and cost of a project at the initial stage of the project, based on technical materials such as design drawings, estimate norms or estimate indicators, and cost norms. It directly affects the project's financing, cost control, and scheme comparison.

[0003] Traditional engineering estimates rely on manual interpretation of drawings and reference to norms in combination with BIM quantity calculation software for calculation. That is, it requires manual input of parameters and establishment of a 3D model for engineering quantity calculation, which is time-consuming and laborious and prone to calculation errors; and it is difficult to dynamically modify the design drawings during the engineering estimate process, with low flexibility, reducing the efficiency of engineering estimates.

[0004] Currently, there is no effective solution to the problem of how to improve the efficiency of existing construction engineering estimates in related technologies. Summary of the Invention

[0005] Embodiments of this application provide an intelligent parsing and estimation method and system for construction engineering drawings, to at least solve the problem of how to improve the efficiency of existing construction engineering estimates in related technologies.

[0006] In a first aspect, embodiments of this application provide an intelligent parsing and estimation method for construction engineering drawings, the method including: Parsing the obtained construction engineering drawings through a geometric figure recognition algorithm to obtain the project structured data of the initial design; Parsing the obtained user requirement instructions through a large language model to obtain requirement structured instructions; Based on the requirement structured instructions, modifying the project structured data through the large language model to obtain requirement structured data; Based on the requirement structured data, calculating the cost of the construction project after the requirement modification through an engineering estimate algorithm to obtain a requirement estimate report after the requirement modification.

[0007] In some of these embodiments, the method further includes: Calculating the cost of the initially designed construction project through an engineering estimate algorithm based on the project structured data to obtain a project estimate report of the initial design; The requirement estimate report after the requirement modification and the project estimate report of the initial design are used for the user to compare and select construction engineering schemes.

[0008] In some of these embodiments, the large language model is used to parse the obtained user requirement instructions, and the obtained requirement structured instructions include: Obtain the user requirement instructions input by the user, where the user requirement instructions include design change instructions, cost optimization instructions, and functional area modification instructions; Parse the user requirement instructions through the trained large language model to generate requirement structured instructions corresponding to the building engineering industry standards and specifications.

[0009] In some of these embodiments, after parsing the obtained user requirement instructions through the large language model to obtain requirement structured instructions, the method includes: Collect the requirement structured instructions generated by the large language model to obtain an instruction data set, and verify the instruction data set to obtain a verification data set, where the verification data set includes requirement structured instructions and annotations on whether each requirement structured instruction is generated correctly; Under the condition of meeting the preset iteration update condition, use the verification data set to iteratively update the large language model to obtain an iteratively updated large language model.

[0010] In some of these embodiments, collecting the requirement structured instructions generated by the large language model to obtain an instruction data set, and verifying the instruction data set to obtain a verification data set includes: Collect the requirement structured instructions generated by the large language model to obtain an instruction data set, where the instruction data set includes a basic instruction set and an extended instruction set, the basic instructions correspond to a single general user requirement instruction, and the extended instructions correspond to continuous and complex user requirement instructions; Verify each basic instruction in the basic instruction set one by one to verify whether the large language model correctly understands the user requirement instructions; Verify each extended instruction in the extended instruction set one by one to verify whether the logical relationship between the basic instructions that make up the extended instruction is correct; Based on the results of the two verifications, obtain a verification data set.

[0011] In some of these embodiments, under the condition of meeting the preset iteration update condition, using the verification data set to iteratively update the large language model includes: Set the preset iteration update condition for triggering the iterative update of the large language model, where the preset iteration update condition includes frequent errors in the parsing of user requirement instructions and new articles in the building engineering industry standards and specifications; Under the condition of meeting the preset iterative update condition, based on the verification data set, the large language model is iteratively updated by means of incremental learning.

[0012] In some embodiments, based on the requirement structured instruction, the large language model modifies the project structured data to obtain the requirement structured data, including: Based on the multi-dimensional features of the building components in the construction engineering drawings, the large language model generates labels for the project structured data of each building component to obtain the multi-dimensional labels of the building components; Based on the multi-dimensional labels and the requirement structured instruction, the large language model modifies the project structured data of the corresponding building component to obtain the requirement structured data.

[0013] In some embodiments, based on the multi-dimensional features of the building components in the construction engineering drawings, the large language model generates labels for the project structured data of each building component, including: Based on the position features and shape features of the building components in the construction engineering drawings, the large language model generates geometric feature labels for the project structured data of each building component; Based on the usage features and design specification features of the building components in the construction engineering drawings, the large language model generates functional attribute labels for the project structured data of each building component; Based on the material type features and construction technology features of the building components in the construction engineering drawings, the large language model generates material construction labels for the project structured data of each building component; Based on the inter-component topological features of the building components in the construction engineering drawings, the large language model generates relational labels for the project structured data of each building component.

[0014] In some embodiments, based on the multi-dimensional labels and the requirement structured instruction, the large language model modifies the project structured data of the corresponding building component, including: Based on the multi-dimensional labels of different building components, an inverted index of the project structured data is established; The large language model locates the target building component corresponding to the requirement structured instruction, retrieves the project structured data of the target building component by using the inverted index, and then modifies the project structured data.

[0015] Second aspect, an embodiment of the present application provides an intelligent parsing and budget estimation system for construction engineering drawings. The system is used to execute the method described in any one of the first aspects above. The system includes a drawing parsing module, a requirement modification module, and a project budget estimation module; The drawing parsing module is used to parse the obtained construction engineering drawings through a geometric figure recognition algorithm to obtain the project structured data of the initial design; The requirement modification module is used to parse the obtained user requirement instructions through a large language model to obtain requirement structured instructions; The requirement modification module is used to modify the project structured data through the large language model according to the requirement structured instructions to obtain requirement structured data; The project budget estimation module is used to calculate the cost of the construction project after requirement modification through a project budget estimation algorithm according to the requirement structured data to obtain a requirement budget estimation report after requirement modification.

[0016] Compared with the related technology, an intelligent parsing and budget estimation method for construction engineering drawings provided by an embodiment of the present application, wherein the method parses the obtained construction engineering drawings through a geometric figure recognition algorithm to obtain the project structured data of the initial design; parses the obtained user requirement instructions through a large language model to obtain requirement structured instructions; based on the requirement structured instructions, modifies the project structured data through the large language model to obtain requirement structured data; based on the requirement structured data, calculates the cost of the construction project after requirement modification through a project budget estimation algorithm to obtain a requirement budget estimation report, realizing the use of a large language model to modify the requirements of the initially designed construction engineering drawings and automatically calculating the requirement budget estimation report of the modified project. That is, in the case of changes in the requirements of a construction project, the user can directly input instructions to instruct the LLM to make modifications and obtain a budget estimation report, without having to manually modify it in professional software and then perform budget estimation again, greatly improving the efficiency of automatic project budget estimation and solving the problem of how to improve the efficiency of existing construction project budget estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the drawings: Figure 1 is a step flowchart of an intelligent parsing and budget estimation method for construction engineering drawings according to an embodiment of the present application; Figure 2 is an internal structure schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be described and explained below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments provided in the present application without making creative efforts belong to the scope of protection of the present application.

[0019] Obviously, the accompanying drawings in the following description are only some examples or embodiments of the present application. For those of ordinary skill in the art, without making creative efforts, the present application can also be applied to other similar scenarios based on these drawings. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for those of ordinary skill in the art related to the content disclosed in the present application, some designs, manufacturing or production changes made on the basis of the technical content disclosed in the present application are only conventional technical means and should not be understood as the content disclosed in the present application being insufficient.

[0020] Referring to "embodiments" in the present application means that the specific features, structures or characteristics described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those of ordinary skill in the art explicitly and implicitly understand that the embodiments described in the present application can be combined with other embodiments without conflict.

[0021] Unless otherwise defined, the technical terms or scientific terms involved in this application shall have the ordinary meanings understood by those with ordinary skills in the technical field to which this application belongs. The words such as "a", "an", "one kind", "the" and the like involved in this application do not indicate a quantity limitation and may represent a singular or plural number. The terms "include", "comprise", "have" and any variations thereof involved in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may further include steps or units not listed, or may further include other steps or units inherent to these processes, methods, products or devices. The words such as "connect", "be connected", "couple" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "plurality" involved in this application means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships may exist. For example, "A and / or B" may represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the front and rear associated objects. The terms "first", "second", "third" and the like involved in this application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0022] An embodiment of this application provides an intelligent analysis and estimation method for construction engineering drawings. Figure 1 It is a flowchart of the steps of the intelligent analysis and estimation method for construction engineering drawings according to the embodiment of this application, as Figure 1 shown, the method includes: Step S102, parse the obtained construction engineering drawings through a geometric figure recognition algorithm to obtain the project structured data of the initial design; Step S102 specifically includes the following steps: Step S1021, first obtain the full set of construction engineering drawings of the construction project, and these construction engineering drawings are professional engineering software drawings (such as CAD drawings); Step S1022, parse the full set of construction engineering drawings through a geometric figure recognition algorithm, and its specific parsing process is as follows: First, find the floor plan in the drawings according to the geometric features and table content features of the floor plan, and read and analyze it to generate floor information; then split the whole drawing according to the geometric features of the drawing frame table and the axis network to generate drawing sub-diagrams; then extract information such as drawing names and drawing numbers from each drawing sub-diagram to obtain the types of the drawing sub-diagrams, and set the floor attribution information of each sub-diagram in combination with the above floor information and the types of the drawing sub-diagrams; finally, on the basis of the floor attribution of each sub-diagram, generate the project structured data of the whole project, including the structured attribute data of building components (such as axis network, columns, walls, beams, slabs, etc.).

[0023] Step S104: Parse the obtained user requirement instruction through a large language model to obtain a structured requirement instruction. Step S104 specifically includes the following steps: Step S1041: Obtain the user requirement instruction input by the user. The user requirement instruction includes a design change instruction (such as material replacement, floor height adjustment), a cost optimization instruction (such as project quantity compression, unit price adjustment), and a functional area modification instruction (such as change of functional area use). It should be noted by way of example that the user requirement instruction covers all specialties of the construction project (architecture, structure, mechanical and electrical, etc.). For example: "Replace the exterior wall dry-hanging stone with paint", "Adjust the standard floor height from 3.6 meters to 3.3 meters", "Change the office area to a meeting room", etc. Extract the design change records, engineering order forms, and corresponding budget adjustment data from the historical projects of the cooperation unit to form an original instruction library. Through user interviews and questionnaire surveys, collect the high-frequency requirement instructions of engineers in actual operations.

[0024] Step S1042: Parse the user requirement instruction through the trained large language model to generate a structured requirement instruction corresponding to the construction industry standard specifications.

[0025] It should be noted by way of example that the large language model in this embodiment is a large language model dedicated to engineering budget for construction projects, which integrates construction industry standard specifications (such as industry standard manuals like "Code for Valuation with Bill of Quantities of Construction Projects" and "Budget Quota for Construction Projects"). The large language model parses the user requirement instruction input by the user and generates a structured requirement instruction corresponding to the manual provisions of the specification (such as engineering quantity calculation rules, material unit price list). For example: Manual provision: "The engineering quantity of the concrete beam is calculated by volume" → Structured requirement instruction: "Calculate the volume of all concrete beams".

[0026] It should be noted that when making a budget for an actual construction project, it is often necessary to consider how to save costs or be more economical. At this time, it is often handled by modifying the decorative materials or compressing the floor height. The traditional method requires manual modification of the material floor height in the budget software model and recalculation to obtain the budget statement, which often consumes a lot of labor. However, through the embodiment of the present application, the modified plan can be input into the large language model in a natural language manner. After understanding, the large language model retrieves and judges the structured data and modifies the relevant data.

[0027] After step S104, the method further includes step S105 of collecting the requirement structured instructions generated by the large language model to obtain an instruction dataset, and validating the instruction dataset to obtain a validation dataset, where the validation dataset includes the requirement structured instructions and the annotation of whether each requirement structured instruction is generated correctly; under the condition of meeting the preset iterative update condition, the large language model is iteratively updated through the validation dataset to obtain an iteratively updated large language model.

[0028] Step S105 specifically includes the following steps: Step S1051, collect the requirement structured instructions generated by the large language model to obtain an instruction dataset, where the instruction dataset includes a basic instruction set and an extended instruction set, the basic instructions correspond to a single general user requirement instruction, and the extended instructions correspond to continuous and complex user requirement instructions; It should be noted by way of example that the requirement structured instructions generated by the large language model are divided into a basic instruction set and an extended instruction set. The basic instructions correspond to a single general user requirement instruction, such as "Modify the material of the wall", and the extended instructions correspond to continuous and complex user requirement instructions, such as "Change the thickness of the exterior wall from 200mm to 150mm and change the exterior material to stone".

[0029] Step S1052, verify each basic instruction in the basic instruction set one by one to verify whether the large language model correctly understands the user requirement instruction; verify each extended instruction in the extended instruction set one by one to verify whether the logical relationship between the basic instructions that make up the extended instruction is correct; based on the results of the two verifications, obtain a validation dataset.

[0030] It should be noted by way of example that for the verification of the basic instruction set, the method of manual review (such as forming a review group consisting of engineering cost experts, designers, and representatives of the construction party) can be adopted to review each basic instruction one by one to verify whether the user requirement instruction can be parsed by the large language model and converted into a structured instruction. For example, for the single general user requirement instruction "Adjust the standard floor height from 3.6 meters to 3.3 meters", check whether the large model can generate the correct structured instruction to modify the floor plan and the coordinate data of related components; for the verification of the extended instruction set, it is necessary to detect whether there is a logical conflict between the basic instructions in the extended instruction. For example, for the continuous and complex user requirement instruction "Change the thickness of the exterior wall from 200mm to 150mm and change the exterior material to stone", it is necessary to verify whether the construction specification of the stone material is supported by the 150mm exterior wall thickness.

[0031] Step S1053, set the preset iterative update condition for triggering the iterative update of the large language model, where the preset iterative update condition includes that there are frequent errors in the parsing of user requirement instructions and new articles appear in the building engineering industry standard specifications; For example, there are frequent errors in the parsing of user requirement instructions, such as the number of times the large language model fails to recognize the user requirement instruction "adjust the boundary of the fire compartment" reaches the preset threshold; new articles are added to the industry standards and specifications of the construction engineering industry, such as the green building standard updates new articles.

[0032] Step S1054, under the condition of meeting the preset iterative update condition, based on the validation data set, iteratively update the large language model by means of incremental learning.

[0033] For example, use the validation data set as the incremental training set, fine-tune the large model using incremental learning technology, adapt to new instructions while retaining the original knowledge, and at the same time use contrastive learning to strengthen the large language model's ability to distinguish similar instructions (such as "replace paint" and "replace paint brand").

[0034] Step S106, based on the requirement structured instructions, modify the project structured data through the large language model to obtain the requirement structured data; Step S106 specifically includes the following steps: Step S1061, based on the multi-dimensional features of building components in the construction engineering drawings, generate labels for the project structured data of each building component through the large language model to obtain the multi-dimensional labels of building components, where the multi-dimensional labels include geometric feature labels, functional attribute labels, material construction labels, and relational labels; Specifically for Step S1061: ① Based on the position and shape features of building components in the construction engineering drawings, generate geometric feature labels for the project structured data of each building component through the large language model; For example, for a building component with coordinates located on the outermost layer of the building contour and of the wall type, its geometric feature label is "exterior wall"; for a building component with a cantilever length greater than 1 meter and less than 2 support points, its geometric feature label is "cantilever structure".

[0035] ② Based on the usage and design specification features of building components in the construction engineering drawings, generate functional attribute labels for the project structured data of each building component through the large language model; For example, for a building component with a wall material of fireproof gypsum board and located at the partition boundary, its functional attribute label is "fire compartment"; for a building component with a load attribute of a structural column or beam greater than the threshold, its functional attribute label is "load-bearing component"; ③ Based on the material type and construction process features of building components in the construction engineering drawings, generate material construction labels for the project structured data of each building component through the large language model; For example, for building components with a wall material property of stone and an installation method of dry hanging, the material installation label is "dry-hung stone"; for building components with a production property of factory prefabrication, the material installation label is "prefabricated component". ④ Based on the topological features between building components in the construction engineering drawings, use a large language model to generate relational labels for the project structured data of each building component.

[0036] For example, for building components located at the junction of upper and lower floors, the relational label is "adjacent floors".

[0037] Step S1062: Based on the multi-dimensional labels and requirement structured instructions, use a large language model to modify the project structured data of the corresponding building component to obtain requirement structured data.

[0038] Specifically, in step S1062, based on the multi-dimensional labels of different building components, establish an inverted index of the project structured data; use the large language model to locate the target building component corresponding to the requirement structured instruction, and use the inverted index to retrieve the project structured data of the target building component, and then modify the project structured data to obtain requirement structured data.

[0039] For example, retrieve building components through multi-dimensional labels, that is, use multi-dimensional labels to establish an inverted index for the project structured data of each building component, support logical operators (such as AND, OR) to combine multiple labels (such as walls that meet both #facade AND #dry-hung stone at the same time); after retrieving the results, further filter by attributes (such as the floor is 3F), and use a batch update statement to modify the attribute fields of the components according to the user's instructions. If the modification affects associated components, use a graph database to maintain the component dependency relationship, achieve topological update, and automatically trigger cascading update (such as when the floor height is adjusted, the heights of columns and walls should be modified accordingly).

[0040] Step S108: Based on the requirement structured data, perform cost calculation for the construction project after the requirement modification through an engineering budget algorithm to obtain a requirement budget statement after the requirement modification.

[0041] After step S108, the method includes: based on the project structured data, perform cost calculation for the initially designed construction project through an engineering budget algorithm to obtain a project budget statement for the initial design; the requirement budget statement after the requirement modification and the project budget statement for the initial design are used for the user to compare and select construction project plans.

[0042] It should be noted that by calling the engineering budget algorithm through FunctionCall, the component attributes in the structured data are matched and mapped with the entries in the bill of quantities (for example, "wall volume" is matched with the quota item of "masonry project - brick wall", and "concrete beam volume" is matched with the quota item of "concrete project - cast-in-place beam"); then the quantities of work are calculated item by item, and according to the quota rules and material unit prices, the costs of sub-projects are calculated to obtain the demand budget statement after the demand is modified.

[0043] Through the above steps in the embodiments of the present application, the structured analysis of the complete drawings of the entire project is realized by using the geometric figure recognition algorithm, all the drawing information of the entire project can be recognized and structured, avoiding the problem of missing key information during manual interpretation; using the large language model to support users to modify project parameters through natural language instructions, such as replacing materials, adjusting storey height, changing functional areas, etc., can dynamically respond to design changes, solving the defect that the fixed-template software cannot update the interpretation results in time, improving the interpretation efficiency and accuracy; at the same time, it can output the multi-version budget comparison in real time, providing sufficient cost analysis basis for users, facilitating users to compare options, and assisting in scientific engineering decision-making.

[0044] It should be noted that the steps shown in the above process or the flowchart of the drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0045] The embodiments of the present application provide an intelligent parsing and budget system for construction project drawings, which includes a drawing parsing module, a demand modification module, and an engineering budget module; The drawing parsing module is used to parse the obtained construction project drawings through the geometric figure recognition algorithm to obtain the structured data of the initially designed project; The demand modification module is used to parse the obtained user demand instructions through the large language model to obtain the demand structured instructions; The demand modification module is used to modify the project structured data through the large language model according to the demand structured instructions to obtain the demand structured data; The engineering budget module is used to calculate the cost of the construction project after the demand is modified through the engineering budget algorithm according to the demand structured data to obtain the demand budget statement after the demand is modified.

[0046] Through the drawing analysis module, requirement modification module, and engineering budget estimation module in the embodiments of the present application, it is possible to use a large language model to modify the requirements of the initial designed construction engineering drawings and automatically calculate the requirement budget estimation report after modification. That is, in the case of changes in the requirements of a construction project, the user can directly input an instruction to instruct the LLM to make modifications and obtain the budget estimation report, without the need to manually modify and then estimate in professional software again, greatly improving the efficiency of automatic engineering budget estimation and solving the problem of how to improve the efficiency of existing construction engineering budget estimation.

[0047] It should be noted that the above-mentioned each module can be a functional module or a program module, and can be implemented either by software or by hardware. For the modules implemented by hardware, the above-mentioned each module can be located in the same processor; or the above-mentioned each module can also be located in different processors in any combined form.

[0048] This embodiment provides an electronic device, including a memory and a processor. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0049] Optionally, the above-mentioned electronic device may further include a transmission device and an input / output device. Among them, the transmission device is connected to the above-mentioned processor, and the input / output device is connected to the above-mentioned processor.

[0050] Optionally, the above-mentioned electronic device may further include a processor, a memory, a network interface, a display screen, and an input device connected through a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the electronic device is used to communicate with an external terminal through a network connection. The computer program, when executed by the processor, implements an intelligent parsing and budget estimation method for construction engineering drawings. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic device can be a touch layer covered on the display screen, or a button, trackball, or touchpad provided on the housing of the electronic device, or an external keyboard, touchpad, or mouse, etc.

[0051] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation manners, and will not be elaborated here.

[0052] In addition, in combination with the intelligent parsing and estimation method for construction engineering drawings in the above embodiments, an embodiment of the present application can provide a storage medium to implement. A computer program is stored on the storage medium; when the computer program is executed by a processor, the intelligent parsing and estimation method for any one of the construction engineering drawings in the above embodiments is implemented.

[0053] In one embodiment, Figure 2 is a schematic internal structure diagram of an electronic device according to an embodiment of the present application. As Figure 2 shown, an electronic device is provided. The electronic device may be a server, and its internal structure diagram may be as Figure 2 shown. The electronic device includes a processor, a network interface, an internal memory, and a non-volatile memory connected through an internal bus. Among them, the non-volatile memory stores an operating system, a computer program, and a database. The processor is used to provide computing and control capabilities. The network interface is used to communicate with an external terminal through a network connection. The internal memory is used to provide an environment for the operation of the operating system and the computer program. When the computer program is executed by the processor, an intelligent parsing and estimation method for construction engineering drawings is implemented. The database is used to store data.

[0054] Those skilled in the art can understand that Figure 2 the structure shown in

[0055] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. This computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0056] Those skilled in the art should understand that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0057] The above embodiments only represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. An intelligent parsing and budget estimation method for construction engineering drawings, characterized in that, The method includes: Parsing the obtained construction engineering drawings through a geometric figure recognition algorithm to obtain the project structured data of the initial design; Parsing the obtained user requirement instructions through a large language model to obtain the requirement structured instructions; Based on the requirement structured instructions, modifying the project structured data through the large language model to obtain the requirement structured data; Based on the requirement structured data, calculating the cost of the construction project after the requirement modification through an engineering cost estimation algorithm to obtain the requirement cost estimation report after the requirement modification.

2. The method according to claim 1, wherein The method further includes: Calculating the cost of the initially designed construction project through an engineering cost estimation algorithm based on the project structured data to obtain the project cost estimation report of the initial design; The requirement cost estimation report after the requirement modification and the project cost estimation report of the initial design are used for the user to compare and select the construction project plan.

3. The method according to claim 1, characterized in that Parsing the obtained user requirement instructions through a large language model to obtain the requirement structured instructions includes: Obtaining the user requirement instructions input by the user, where the user requirement instructions include design change instructions, cost optimization instructions, and functional area modification instructions; Parsing the user requirement instructions through a trained large language model to generate requirement structured instructions corresponding to the construction engineering industry standards and specifications.

4. The method according to claim 1, characterized in that, After parsing the obtained user requirement instructions through a large language model to obtain the requirement structured instructions, the method includes: Collecting the requirement structured instructions generated by the large language model to obtain an instruction dataset, and verifying the instruction dataset to obtain a verification dataset, where the verification dataset includes the requirement structured instructions and the annotation of whether each requirement structured instruction is generated correctly; Under the condition of meeting the preset iteration update condition, iteratively updating the large language model through the verification dataset to obtain an iteratively updated large language model.

5. The method according to claim 4, wherein Collecting the requirement structured instructions generated by the large language model to obtain an instruction dataset, and verifying the instruction dataset to obtain a verification dataset includes: Collecting the requirement structured instructions generated by the large language model to obtain an instruction dataset, where the instruction dataset includes a basic instruction set and an extended instruction set, the basic instructions correspond to a single general user requirement instruction, and the extended instructions correspond to continuous and complex user requirement instructions; Verifying each basic instruction in the basic instruction set one by one to verify whether the large language model correctly understands the user requirement instructions; Verifying each extended instruction in the extended instruction set one by one to verify whether the logical relationship between the basic instructions constituting the extended instruction is correct; Based on the results of the two verifications, obtaining the verification dataset.

6. The method according to claim 4, wherein Under the condition of meeting the preset iteration update condition, iteratively updating the large language model through the verification dataset includes: Setting the preset iteration update condition for triggering the iterative update of the large language model, where the preset iteration update condition includes high-frequency errors in the parsing of user requirement instructions and new articles in the construction engineering industry standards and specifications. Under the condition of meeting the preset iterative update condition, based on the validation data set, the large language model is iteratively updated by means of incremental learning.

7. The method according to claim 1, characterized in that, Based on the requirement structured instruction, the large language model is used to modify the project structured data, and the obtained requirement structured data includes: Based on the multi-dimensional features of the building components in the construction engineering drawings, the large language model is used to generate labels for the project structured data of each building component, and the multi-dimensional labels of the building components are obtained; Based on the multi-dimensional labels and the requirement structured instruction, the large language model is used to modify the project structured data of the corresponding building component, and the requirement structured data is obtained.

8. The method according to claim 7, characterized in that, Based on the multi-dimensional features of the building components in the construction engineering drawings, the large language model is used to generate labels for the project structured data of each building component, including: Based on the position features and shape features of the building components in the construction engineering drawings, the large language model is used to generate geometric feature labels for the project structured data of each building component; Based on the usage features and design specification features of the building components in the construction engineering drawings, the large language model is used to generate functional attribute labels for the project structured data of each building component; Based on the material type features and construction process features of the building components in the construction engineering drawings, the large language model is used to generate material construction labels for the project structured data of each building component; Based on the inter-component topological features of the building components in the construction engineering drawings, the large language model is used to generate relational labels for the project structured data of each building component.

9. The method according to claim 8, wherein Based on the multi-dimensional labels and the requirement structured instruction, the large language model is used to modify the project structured data of the corresponding building component, including: Based on the multi-dimensional labels of different building components, an inverted index of the project structured data is established; The large language model is used to locate the target building component corresponding to the requirement structured instruction, and the project structured data of the target building component is retrieved by using the inverted index, and then the project structured data is modified.

10. An intelligent parsing and estimation system for construction engineering drawings, characterized in that, The system is used to execute the method according to any one of claims 1 to 9, and the system includes a drawing parsing module, a requirement modification module and an engineering cost estimation module; The drawing parsing module is used to parse the obtained construction engineering drawings through a geometric figure recognition algorithm to obtain the project structured data of the initial design; The requirement modification module is used to parse the obtained user requirement instruction through a large language model to obtain a requirement structured instruction; The requirement modification module is used to modify the project structured data through the large language model according to the requirement structured instruction to obtain the requirement structured data; The engineering cost estimation module is used to calculate the cost of the construction project after the requirement modification through an engineering cost estimation algorithm according to the requirement structured data to obtain a requirement cost estimation report after the requirement modification.

Images