Construction engineering quality control rule set generation method fusing large model and process data chain

Abstract

The invention discloses a construction engineering quality control rule set generation method fusing a large model and a process data chain. The method comprises the following steps: constructing a process data chain data structure and a quality control rule data structure of each construction process of construction engineering; performing fine adjustment on the basic large model through professional corpora, a process data chain and a quality control rule to establish a building large model; constructing / modifying a structured cue word of the construction process; gradually inputting the data into the large building model and outputting a structured construction handicraft control rule set; checking an output result of the large building model to judge whether each construction handicraft control rule is complete or not, if not, modifying a structured prompt word of a corresponding construction process, and outputting a new structured construction handicraft control rule set through the large building model; and if so, outputting a structured construction handicraft control rule set by the large building model. The method has the advantages of being high in speed, high in efficiency, intelligent, good in integrity, high in accuracy and high in specialty.

Description

Technical Field

[0001] The present invention relates to the field of intelligent construction of building projects, and in particular to a method for generating a construction project quality control rule set integrating a large model and a process data chain. Background Art

[0002] There are many factors that affect the quality of construction projects, including raw material properties, raw material mix ratios, project characteristics and requirements, construction environment, etc., and quality control is difficult. With the development of digital construction technology, many information systems have accumulated rich monitoring data, but due to the lack of structured construction process quality control rules, the information system cannot provide real-time quality problem warnings and support dynamic quality control, resulting in frequent quality problems in construction projects. The quality control rules of construction projects involve thousands of standards, covering professional terms and data in terms of structure, materials, mix design, construction technology, etc. It is extremely difficult to build quality control rules by relying on manual reading of specifications. Therefore, there is still a lack of a complete set of construction quality control rules on the market, which brings great difficulties to the research and development of intelligent construction technology and equipment. With the development of artificial intelligence, the use of large models can quickly learn relevant specifications, but currently large models can only be output in an unstructured manner, and cannot accurately output the structured rules of relevant specifications. In addition, the rules for quality control in the construction process involve professional terms in terms of component information, material information, construction process information, etc. The existing large models do not clearly understand the professional named entities and their logical relationships. Therefore, it is difficult for large models to accurately output construction project quality control rules and warning rules that match the intelligent construction system. In addition, for the same property parameter, the requirements of different process steps may be different; for example, the requirements for slump when the concrete is discharged are different from those when it is transported to the site for pouring, and the loss of slump during transportation must be considered. The existing large model also cannot meet the establishment of quality control rules. Summary of the invention

[0003] The purpose of the present invention is to provide a method for generating a construction engineering quality control rule set that integrates a large model and a process data chain, so as to solve the problems of extremely high difficulty in manually reading specifications to construct quality control rules, large data volume, and the inability of traditional large models to directly construct structured construction engineering quality control rules according to specifications.

[0004] In order to solve the above technical problems, the technical solution provided by the present invention is: a method for generating a construction engineering quality control rule set integrating a large model and a process data chain, comprising: Construct the process data chain data structure of each construction process of the building project; Construct the data structure of quality control rules for each construction process of the building project; Establish a large-scale building model, fine-tune the basic large-scale model through the professional corpus of the construction process, and input the process data chain data structure and quality control rule data structure of the construction process into the basic large-scale model, so that the basic large-scale model contains professional knowledge of the construction process and can understand the steps of the construction process and its quality control rules, thus forming a large-scale building model; Construct / modify the structured prompt words for each construction process according to the process data chain of each construction process and its corresponding quality control rule data structure; Gradually input the structured prompt words of each construction process into the large building model, and output the structured construction process quality control rule set through the large building model; The output results of the large building model are reviewed to determine whether the construction process quality control rules are complete. If not, the structured prompt words of the corresponding construction process are modified to output a new structured construction process quality control rule set through the large building model; if complete, the large building model outputs a structured construction process quality control rule set.

[0005] Furthermore, the invention provides a method for generating a construction engineering quality control rule set integrating a large model and a process data chain, wherein the process data chain data structure of the construction process is: P = {step p i =(Step Name pn i , input data set Ip i ={ip ij}, output data set Op i ={op ij})}, where P is a process data chain data structure, referred to as process data chain; the attributes of the elements of input data and output data in the process data chain data structure are integrated to form a parameter table TPP; the columns of the parameter table TPP include parameter ID, data element and value type.

[0006] Furthermore, the present invention provides a method for generating a construction project quality control rule set that integrates a large model and a process data chain, wherein the quality control rule data structure of the construction process includes a quality control rule table and a rule condition table; wherein the columns of the quality control rule table include rule ID, rule name, rule description, rule level, and step name; wherein the columns of the rule condition table include condition ID, rule ID of the rule table, parameter ID of the parameter table, calculation symbol, threshold, logical relationship, and parent condition ID.

[0007] Furthermore, the invention provides a method for generating a construction engineering quality control rule set integrating a large model and a process data chain, wherein the method for establishing a large building model comprises: Select a basic model; Use the collected professional corpus of construction technology to fine-tune the basic large model; The definitions of professional terms extracted from quality control rules are input into the basic big model, so that the basic big model can understand the concepts of rules, rule levels, rule conditions, parameters, calculation symbols, and thresholds; Input the process data chain into the basic large model, so that the basic large model can understand the steps and relevant data elements of the construction process; Create a large model of the building with professional knowledge of construction technology.

[0008] Furthermore, the method for generating a construction engineering quality control rule set integrating a large model and a process data chain provided by the present invention, wherein the method for constructing / modifying structured prompt words for each construction process according to the process data chain of each construction process and its corresponding quality control rule data structure comprises: Write role prompt words: pro={"You are an expert in [construction technology]"}, where [construction technology] is entered by the user; Traverse each step p in P i ; Traverse each step p in P i The input dataset Ip i The ip of each element ij , construct the prompt word pp ij = "Please extract the quality requirements of the [data element] data input by [step name] and output it as structured data, including [data element], [calculation symbol], [threshold], [rule level], [constraint condition]", and store it in the prompt word set PPJ={ ppj k = (process step p i , prompt word pp ij )}; Traverse each step p in P i The output dataset Op i Each element op ij , construct the prompt word pp ij = "Please extract the [data element] data requirements of the [step name] output and output them as structured data, including [data element], [calculation symbol], [threshold], [rule level], [constraint condition]", and store them in the prompt word set PPJ={ppj k = (process step p i , prompt word pp ij )}.

[0009] Furthermore, the method for generating a construction engineering quality control rule set integrating a large model and a process data chain provided by the present invention, wherein the method of gradually inputting structured prompt words of each construction process into the building large model and outputting a structured construction process quality control rule set through the building large model comprises: Traverse each element ppj in the prompt word set PPJ k = (process step pk , prompt word pp k ); The role definition word pro and the prompt word pp ij Input the large model of the building; The output results are analyzed by the large building model to obtain the output [data elements], [calculation symbols], [thresholds], [rule levels] and [constraints] information; Add a new record in the quality control rule table and store the [rule level] information and p k Step name pn i Go to the Rule Level column and determine [Rule ID]; According to the [data element], determine the corresponding [parameter ID] from the parameter table TPP; Add a new record to the rule condition table, store [Rule ID], [Parameter ID], [Calculation Symbol], [Threshold], [Constraint] in the corresponding columns, and determine [Condition ID]; Transform thresholds and constraints into structured data through a large building model; Furthermore, the method for generating a construction engineering quality control rule set integrating a large model and a process data chain provided by the present invention includes: Traverse each record rtr in the rule condition table i ; Matching record rtr in parameter table TPP i The parameter ID corresponding to [threshold] of the ; if there is no match, the threshold is considered to be a simple value; If rtr i If [Constraints] is not empty, a prompt word ppys for constraint conversion is generated = "Please convert [Constraints] into structured constraint rules. The output content includes [Data elements], [Calculation symbols], [Thresholds], and [Rule levels]"; Parse the output results of the building model, including [data elements], [calculation symbols], [thresholds], [rule levels] and other information; match the corresponding [parameter ID] from the parameter table TPP according to [data elements] and [attribute parameters]; add a record to the rule condition table and store rtr i [Rule ID], [Parameter ID], [Calculation symbol], [Threshold], "AND", rtr i The [Condition ID] information is divided into rule ID, parameter ID, calculation symbol, threshold, logical relationship, and parent condition ID; The large building model outputs a structured set of construction process quality control rules.

[0010] Compared with the prior art, the present invention has the following beneficial effects: The method for generating a construction engineering quality control rule set integrating a large model and a process data chain provided by the present invention constructs or modifies the structured prompt words of the construction process by constructing a process data chain data structure and a quality control rule data structure of the construction process, thereby outputting a structured construction process quality control rule set through a large building model, thereby realizing the intelligent and rapid establishment of the construction process quality control rule set. Compared with the method of manually reading a large number of specifications to manually construct construction process quality control rules, it has the advantages of fast speed and high efficiency, and realizes the intelligent generation of the construction engineering quality control rule set. The process data chain is used to digitally characterize a certain sub-project, including process steps, step sequences and input and output data of each step, and structured descriptions of named entities such as process steps and key data, as well as the logical relationship between them, which can well make up for the lack of professional knowledge of the basic large model. The building large model established by combining the process chain can intelligently construct and review the construction process quality control rules, which has the advantages of good integrity, high accuracy and high professionalism, and avoids the problems of poor integrity, poor accuracy and poor professionalism caused by the traditional basic large model to construct structured construction engineering quality control rules.

[0011] The method for generating a construction project quality control rule set that integrates a big model and a process data chain provided by the present invention aims to solve the problem of heavy workload in combing out the quality control rules of construction processes by applying a big model to learn relevant specifications to realize rapid combing of the quality control rules. In view of the problems of many professional terms in various construction processes of construction projects and difficulty in identifying synonyms, the process data chain is manually combed and converted into structured prompt words for the process and its quality control rules and input into the big model, thereby ensuring the standardization of the big model output and reducing ambiguity. This improves the accuracy of the quality control rules of the construction process. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a flow chart of a method for generating a quality control rule set for construction engineering that integrates a large model and a process data chain; DETAILED DESCRIPTION

[0013] The present invention is described in detail below in conjunction with the accompanying drawings: The advantages and features of the present invention will become more apparent from the following description. It should be noted that the accompanying drawings are in very simplified form and in non-precise proportions, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.

[0014] Please refer to Figure 1 The embodiment of the present invention provides a method for generating a construction engineering quality control rule set integrating a large model and a process data chain, which may include: Step 1: Construct the process data chain data structure of each construction process of the building project. The process data chain data structure P of the construction process = {step p i =(Step Name pn i , input data set Ipi ={ip ij}, output data set Op i ={op ij})}; The process data chain data structure is constructed according to relevant specifications, construction plans, design instructions and other professional corpora, and specifically includes the following steps: where i and j are positive integers.

[0015] Step 1.1: Step 2.1: Determine the step list P = {p i =(Step Name pn i )}. The sum of the step names of each process of each construction process is the step list.

[0016] Step 1.2: Determine the input data and output data of each step in the process data chain data structure P of the construction process, P={p i =(Step Name pn i Input dataset IP i ={data element ip ij}, output data set Op i ={data element op ij For example, for concrete preparation process P={p1=(mix design, Ip1, Op1), p2=(raw material preparation and testing, Ip2, Op2), p3=(weighing ingredients, Ip3, Op3), p4=(mixing control, Ip4, Op4), p5=(discharging detection, Ip5, Op5), p6=(concrete transportation, Ip6, Op6)}; Op1 of mix design includes data elements such as {water-binder ratio, sand ratio, admixture dosage, cement content, sand content, gravel content, admixture content}.

[0017] Step 1.3: Integrate the attributes of the data elements of the input data and output data in the process data chain data structure P to form a parameter table TPP; the columns of the parameter table TPP include parameter ID, data element and value type. For example, the parameter table is {001, water-cement ratio, float}. "ID" is "address".

[0018] Step 2: Construct the quality control rule data structure of each construction process of the building project. The quality control rule data structure of each construction process includes a quality control rule table and a rule condition table.

[0019] The columns of the quality control rule table include rule ID, rule name, rule description, rule level, and step name. The rule level can include three categories: should, appropriate, and possible.

[0020] The columns of the rule condition table include condition ID, rule ID of the rule table, parameter ID of the parameter table, calculation symbol, threshold, logical relationship and parent condition ID. The logical relationship and parent condition ID are used for logical operations of multiple conditions to form a composite condition. The calculation symbols may include ">, <, =, ≥, ≤, between, in", etc. "between" means between..., and "in" means within the range of...

[0021] Step 3: Establish a large building model. Fine-tune the basic large model through professional corpus of construction technology, and input the process data chain data structure and quality control rule data structure of the construction technology into the basic large model, so that the basic large model contains professional knowledge of construction technology and can understand the steps of construction technology and its quality control rules, forming a large building model. Specifically, it can be: Step 3.1: Select a basic large model M0; such as deepseek or Tongyi Qianwen, etc.

[0022] Step 3.2: Fine-tune the basic large model using the collected professional corpus of construction technology.

[0023] Step 3.3: Input the definition of the professional terminology extracted from quality control rules into the fine-tuned basic large model, so that the fine-tuned basic large model can understand the concepts of terms such as rules, rule levels, rule conditions, parameters, calculation symbols, thresholds, etc.; the professional terminology extracted from quality control rules refers to the overall content of the quality control rule data structure of the construction process.

[0024] Step 3.4: Input the process data chain P into the fine-tuned basic large model, so that the fine-tuned post-basic large model can understand the steps and related data elements of the construction process, and focus more on the above data elements in the output.

[0025] Step 3.5: Form a large building model Mp with relevant construction technology expertise i .

[0026] Step 4: Construct / modify the structured prompt words of each construction process according to the process data chain of each construction process and its corresponding quality control rule data structure. The initial step is construction, and when the prompt words are inaccurate or incomplete, they are modified. That is, the quality control rule data of the construction process is integrated into the construction of the structured prompt words of the construction process. Specifically, it may include: Step 4.1: Write the role prompt: pro={"You are an expert in [construction technology]"}, where [construction technology] is entered by the user.

[0027] Step 4.2: Traverse each step p in P i ; Step 4.3: Traverse each step p in P iThe input dataset Ip i The ip of each element ij , construct the prompt word pp ij = "Please extract the quality requirements of the [data element] data input by [step name] and output it as structured data, including [data element], [calculation symbol], [threshold], [rule level], [constraint condition]", and store it in the prompt word set PPJ={ ppj k = (process step p i , prompt word pp ij )}. Where k is a positive integer.

[0028] Step 4.4: Traverse each step p in P i The output dataset Op i Each element op ij , construct the prompt word pp ij = "Please extract the [data element] data requirements of the [step name] output and output them as structured data, including [data element], [calculation symbol], [threshold], [rule level], [constraint condition]", and store them in the prompt word set PPJ={ppj k = (process step p i , prompt word pp ij )}. For example, pp 11 =“Please extract the quality requirements for the water-cement ratio of the mix confirmation data output by the mix design, and output it as structured data, including [data element], [calculation symbol], [threshold], [rule level], and [constraint condition]”.

[0029] Step 5: gradually input the structured prompt words of each construction process into the large building model, and output the structured construction process quality control rule set through the large building model. Specifically include: Step 5.1: Traverse each element ppj in the prompt word set PPJ k = (process step p k , prompt word pp k ).

[0030] Step 5.2: If the Kth modified prompt word mppn in MPPN k If it is not empty, the prompt word mppn will be modified k Append to the prompt word, i.e. pp k = pp k +mppn k ; to avoid output errors.

[0031] Step 5.3: Replace the role definition word pro and the prompt word pp ij Input the building model MP i ; Step 5.4: Analyze the output results through the large building model to obtain the output [data elements], [calculation symbols], [thresholds], [rule levels] and [constraints] information.

[0032] Step 5.5: Add a new record to the quality control rule table and store the [rule level] information and p k Step name pn i Go to the Rule Level column and identify [Rule ID].

[0033] Step 5.6: Based on the [data element], determine the corresponding [parameter ID] from the parameter table TPP.

[0034] Step 5.7: Add a new record to the rule condition table, and store [Rule ID], [Parameter ID], [Calculation Symbol], [Threshold], [Constraint] in the corresponding columns, and determine [Condition ID].

[0035] Step 5.8: Convert thresholds and constraints into structured data using the building model. Specifically: Step 5.8.1: Traverse each record rtr in the rule condition table i .

[0036] Step 5.8.2: Match record rtr in parameter table TPP i The parameter ID corresponding to [threshold] of the specified field. If no match is found, the threshold is considered to be a simple value (numeric, text, or enumeration value).

[0037] Step 5.8.3: If rtr i If [Constraints] is not empty, a prompt word ppys for constraint conversion is generated = "Please convert [Constraints] into structured constraint rules. The output content includes [Data elements], [Calculation symbols], [Thresholds], and [Rule levels]".

[0038] Step 5.8.4: Analyze the output results of the building model, including [data element], [calculation symbol], [threshold], [rule level] and other information; match the corresponding [parameter ID] from the parameter table TPP according to [data element] and [attribute parameter]; add a record to the rule condition table and store rtr i [Rule ID], [Parameter ID], [Calculation symbol], [Threshold], "AND", rtr i The [Condition ID] information is respectively assigned to the rule ID, parameter ID, calculation symbol, threshold, logical relationship, and parent condition ID. "AND" is an AND logic operation.

[0039] Step 6: Review the output results of the large building model to determine whether the construction process quality control rules are complete. If not, modify the structured prompt words of the corresponding construction process and output a new structured construction process quality control rule set through the large building model; if complete, the large building model outputs a structured construction process quality control rule set. During the review, you can add information such as the rule name and description. When modifying the prompt words, experts can enter supplementary prompt words mppn for the prompt words in the PPJ feedback results that are incomplete. k , put into the set MPPN={ mppn k}, it is used to prompt the building model to improve certain aspects of the output.

[0040] The method for generating a construction engineering quality control rule set integrating a large model and a process data chain provided in an embodiment of the present invention constructs or modifies the structured prompt words of the construction process by constructing a process data chain data structure and a quality control rule data structure of the construction process, thereby outputting a structured construction process quality control rule set through a large building model, thereby realizing the intelligent and rapid establishment of the construction process quality control rule set. Compared with the method of manually reading a large number of specifications to manually construct construction process quality control rules, it has the advantages of fast speed and high efficiency, and realizes the intelligent generation of the construction engineering quality control rule set. The process data chain is used to digitally characterize a certain sub-project, including process steps, step sequences and input and output data of each step, and structured descriptions of named entities such as process steps and key data, as well as the logical relationship between them, which can well make up for the lack of professional knowledge of the basic large model. The building large model established by combining the process chain can intelligently construct and review the construction process quality control rules, which has the advantages of good integrity, high accuracy and high professionalism, and avoids the problems of poor integrity, poor accuracy and poor professionalism caused by the traditional basic large model to construct structured construction engineering quality control rules, wherein the quality control rule set includes more than one construction process rule.

[0041] The method for generating a construction project quality control rule set that integrates a big model and a process data chain provided in an embodiment of the present invention, aims to solve the problem of heavy workload in combing out the quality control rules of construction processes by applying a big model to learn relevant specifications to realize rapid combing of the quality control rules; in view of the problems of many professional terms in various construction processes of construction projects and difficulty in identifying synonyms, the process data chain is manually combed and converted into structured prompt words for the process and its quality control rules and input into the big model, so as to ensure the standardization of the big model output and reduce ambiguity; thereby improving the accuracy of the quality control rules of the construction process.

[0042] The method for generating a construction engineering quality control rule set that integrates a large model and a process data chain provided by an embodiment of the present invention aims to solve the problem of complex relationships between various data in the construction engineering quality control rules, establish an attribute parameter table in advance, and support the conversion of the text output of the large model into professional parameters reviewed by experts, thereby ensuring the professionalism and integrity of the results. In view of the problem that different process steps have different requirements for the same attribute parameter, constraint conditions are constructed for different steps respectively, thereby improving the accuracy and matching of the quality control rules of the construction process output by the building large model.

[0043] The method for generating a construction project quality control rule set that integrates a big model and a process data chain provided in an embodiment of the present invention integrates the professional knowledge of the process data chain and the text comprehension ability of the big model, and realizes automatic construction of a quality control rule set for the entire process according to relevant specifications of the construction project, supports the construction information system to realize intelligent early warning and advance control of project quality, and achieves the goal of improving quality and increasing efficiency.

[0044] The method for generating a construction project quality control rule set that integrates a large model and a process data chain provided in an embodiment of the present invention can, by being introduced into an intelligent building system, issue an early warning for process nodes in the construction that do not comply with the quality control rules, so as to facilitate monitoring of the quality of the construction process.

[0045] The present invention is not limited to the above-mentioned specific implementation modes. Obviously, the above-mentioned embodiments are only some embodiments of the embodiments of the present invention, but not all embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention. Those skilled in the art can make other levels of modifications and changes to the present invention. In this way, if these modifications and changes of the present invention fall within the scope of the claims of the present invention, the present invention is also intended to include these changes and changes.

Claims

1. A method for generating a construction engineering quality control rule set integrating a large model and a process data chain, characterized in that: include: Construct the process data chain data structure of each construction process of the building project; Construct the data structure of quality control rules for each construction process of the building project; Establish a large-scale building model, fine-tune the basic large-scale model through the professional corpus of the construction process, and input the process data chain data structure and quality control rule data structure of the construction process into the basic large-scale model, so that the basic large-scale model contains professional knowledge of the construction process and can understand the steps of the construction process and its quality control rules, thus forming a large-scale building model; Construct / modify the structured prompt words for each construction process according to the process data chain of each construction process and its corresponding quality control rule data structure; Gradually input the structured prompt words of each construction process into the large building model, and output the structured construction process quality control rule set through the large building model; The output results of the large building model are reviewed to determine whether the construction process quality control rules are complete. If not, the structured prompt words of the corresponding construction process are modified to output a new structured construction process quality control rule set through the large building model; if complete, the large building model outputs a structured construction process quality control rule set.

2. The method for generating a construction engineering quality control rule set integrating a large model and a process data chain according to claim 1 is characterized in that: The process data chain data structure of the construction process is: P = {step p i =(Step Name pn i , input data set Ip i ={ip ij }, output data set Op i ={op ij })}, where P is the process data chain data structure, referred to as process data chain; Integrate the attributes of the elements of input data and output data in the process data chain data structure to form a parameter table TPP; The columns of the parameter table TPP include parameter ID, data element and value type.

3. The method for generating a construction engineering quality control rule set integrating a large model and a process data chain according to claim 2 is characterized in that: The quality control rule data structure of the construction process includes a quality control rule table and a rule condition table; wherein the columns of the quality control rule table include rule ID, rule name, rule description, rule level, and step name; The columns of the rule condition table include condition ID, rule ID of the rule table, parameter ID of the parameter table, calculation symbol, threshold, logical relationship and parent condition ID.

4. The method for generating a construction engineering quality control rule set integrating a large model and a process data chain according to claim 3 is characterized in that: The method for establishing a large building model comprises: Select a basic model; Use the collected professional corpus of construction technology to fine-tune the basic large model; The definitions of professional terms extracted from quality control rules are input into the basic big model, so that the basic big model can understand the concepts of rules, rule levels, rule conditions, parameters, calculation symbols, and thresholds; Input the process data chain into the basic large model, so that the basic large model can understand the steps and relevant data elements of the construction process; form a large building model with professional knowledge of construction technology.

5. The method for generating a construction engineering quality control rule set integrating a large model and a process data chain according to claim 4 is characterized in that: The method for constructing / modifying structured prompt words for each construction process according to the process data chain of each construction process and its corresponding quality control rule data structure includes: Write role prompt words: pro={"You are an expert in [construction technology]"}, where [construction technology] is input by the user; Traverse each step p in P i ; Traverse each step p in P i The input dataset Ip i The ip of each element ij , construct the prompt word pp ij = "Please extract the quality requirements of the [data element] data input by [step name] and output it as structured data, including [data element], [calculation symbol], [threshold], [rule level], [constraint condition]", and store it in the prompt word set PPJ={ ppj k = (process step p i , prompt word pp ij )}; Traverse each step p in P i The output dataset Op i Each element op ij , construct the prompt word pp ij = "Please extract the [data element] data requirements of the [step name] output and output them as structured data, including [data element], [calculation symbol], [threshold], [rule level], [constraint condition]", and store them in the prompt word set PPJ={ppj k = (process step p i , prompt word pp ij )}.

6. The method for generating a construction engineering quality control rule set integrating a large model and a process data chain according to claim 5 is characterized in that: The method of gradually inputting the structured prompt words of each construction process into the building model and outputting the structured construction process quality control rule set through the building model includes: Traverse each element ppj in the prompt word set PPJ k = (process step p k , prompt word pp k ); The role definition word pro and the prompt word pp ij Input the large model of the building; The output results are analyzed by the large building model to obtain the output [data elements], [calculation symbols], [thresholds], [rule levels] and [constraints] information; Add a new record in the quality control rule table and store the [rule level] information and p k Step name pn i Go to the Rule Level column and determine [Rule ID]; According to the [data element], determine the corresponding [parameter ID] from the parameter table TPP; Add a new record to the rule condition table, store [Rule ID], [Parameter ID], [Calculation Symbol], [Threshold], [Constraint] in the corresponding columns, and determine [Condition ID]; Transform thresholds and constraints into structured data through building big models.

7. The method for generating a construction engineering quality control rule set integrating a large model and a process data chain according to claim 6 is characterized in that: The method of converting thresholds and constraints into structured data through a large building model includes: Traverse each record rtr in the rule condition table i ; Matching record rtr in parameter table TPP i The parameter ID corresponding to the [threshold] of the ; if there is no match, the threshold is considered to be a simple value; If rtr i If [Constraints] is not empty, a prompt word ppys for constraint conversion is generated: "Please convert [Constraints] into structured constraint rules. The output content includes [Data elements], [Calculation symbols], [Thresholds], and [Rule levels]"; Parse the output results of the building model, including [data elements], [calculation symbols], [thresholds], [rule levels] and other information; match the corresponding [parameter ID] from the parameter table TPP according to [data elements] and [attribute parameters]; add a record to the rule condition table and store rtr i [Rule ID], [Parameter ID], [Calculation symbol], [Threshold], "AND", rtr i The [Condition ID] information is divided into rule ID, parameter ID, calculation symbol, threshold, logical relationship, and parent condition ID; The large building model outputs a structured set of construction process quality control rules.

Images