Construction scheme auditing system based on AI large model

Abstract

The invention relates to the technical field of intelligent auditing, and discloses a construction scheme auditing system based on an AI large model, and the auditing system consists of a user interface module, a file analysis module, an AI auditing engine module, an auditing tracking module, a feedback and optimization module, a report generation module and a data storage module. Various file formats such as PDF, DOCX and XLSX are analyzed through the file analysis module, text, chart and table information is automatically extracted, the tedious time and energy of manual auditing are remarkably reduced, the compliance and accuracy of a construction scheme are automatically evaluated through a large-scale pre-training model, a user flexibly uploads various files according to requirements, and the construction efficiency is improved. According to the method, the file does not need to be converted into a specific format, optimization suggestions are automatically generated through user feedback and the auditing result, and auditing rules are adjusted in real time, so that the system can adapt to different auditing requirements and business changes, the flexibility is enhanced, and the beneficial effects that automatic auditing of the complex file is flexible and intelligent are achieved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent review technology, and specifically to a construction plan review system based on an AI big model. Background Art

[0002] The construction plan is the implementation plan of the project and is a document material that construction company technicians write in large quantities on a daily basis. In the traditional review process, the workload of technicians is large, the repetition is high, and the efficiency is low. Therefore, it is necessary to improve the review speed and quality and apply intelligent system management methods to the review of construction plans to improve the review speed and quality.

[0003] For example, a construction plan intelligent audit system and its audit method with publication number CN 118278891 A, the audit method, S1: upload the uploaded construction technical documents by category; S2: parse all types of documents in the same construction plan according to a unified parsing format; S3: import each content in the parsed title, paragraph, picture and table sequence into a rule audit engine; S4: set up an audit rule synonym platform on the interactive terminal; S5: set up a database, which has the advantages of high accuracy of professional terms, standardized audit rules, and high similarity between database stored documents and audit texts.

[0004] Although the above-mentioned technical solution solves the problems of slow audit speed and low audit quality, it does not apply machine learning technology, and the system cannot automatically process complex file contents, reducing the flexibility and intelligence of automatic audits. At the same time, the existing technology does not have an effective audit tracking and feedback mechanism to monitor decisions and results during the audit process, which will make it difficult to make continuous improvements and optimizations. Summary of the invention

[0005] Technical issues solved:

[0006] In view of the shortcomings of the existing technology, the present invention provides a construction plan review system based on an AI large model, which has the advantages of automated review of complex documents, flexibility and intelligence, and an audit tracking and feedback mechanism with continuous improvement and optimization effects, thereby solving the problems of the above-mentioned technologies.

[0007] Technical solution:

[0008] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a construction plan review system based on an AI big model, the review system consisting of a user interface module, a file parsing module, an AI review engine module, an audit tracking module, a feedback and optimization module, a report generation module and a data storage module;

[0009] The user interface module is used to provide a user interface for uploading files, supporting multiple file formats: PDF, DOCX and XLSX;

[0010] The file parsing module is used to parse the uploaded construction files, extract key information such as text, charts and tables, and convert the extracted information into a unified data format;

[0011] The AI ​​review engine module uses large-scale pre-trained models to automatically review construction plans, evaluate content compliance and accuracy, apply industry-specific knowledge, and understand complex technical terms and standards;

[0012] The audit tracking module is used to record each step of the audit process, generate audit logs, ensure process transparency, and support the traceability of audit results;

[0013] The feedback and optimization module is used to collect user feedback, analyze audit results, automatically generate optimization suggestions, adjust audit rules, and continuously improve the accuracy and efficiency of audits;

[0014] The report generation module is used to automatically generate an audit report, including audit results, suggested improvement measures and trend analysis;

[0015] The data storage module is used to store the operation values ​​generated in the above modules and the files uploaded by the user;

[0016] The file format expression defined in the user interface module is:

[0017] F=PDF,DOCX,XLSX

[0018] Where: F is the set of supported file formats;

[0019] The operation expression for user uploaded files is:

[0020] U:I→F

[0021] U is the operation of users uploading files;

[0022] I is the file upload interface.

[0023] Preferably, the file parsing module function expression is:

[0024] F=f1,f2,f3,…,fn,n≧1

[0025] Where: F is the uploaded construction file set, f1, f2, f3, ..., fn represent multiple files;

[0026] The file parsing module converts the data into a unified format expression:

[0027] D=P(F)=T(F),G(F),B(F)

[0028] Where: D is a unified data format; P(F) takes the uploaded construction file set F as input, processes it, extracts the corresponding information: text, charts, tables, and finally outputs a unified data format D; T(F) represents the text information set extracted from the file set F; G(F) represents the chart information set extracted from the file set F; B(F) represents the table information set extracted from the file set F.

[0029] Preferably, the AI ​​audit engine module is expressed as:

[0030] C=E(S,M,K,T)=(s,f p (s,M,K,T),f a (s,M,K,T))|s∈S

[0031] Among them: S is the set of uploaded construction plans; M is the large-scale pre-trained model; C is the set of audit results, including compliance and accuracy assessment results; K is the set of industry-specific knowledge; T is the set of technical terms and standards; s is the plan.

[0032] Preferably, the audit tracking module is expressed as:

[0033] A=a1,a2,…,an,n≥1

[0034] Where: A is the set of steps in the audit process;

[0035] The process expression of the audit trail module is:

[0036] L=T(A)

[0037] Where: L is the audit log set; T is the timestamp.

[0038] Preferably, the comprehensive expression in the audit tracking module is: L = (T (a i ),E(a i ),R(a i ))|a i ∈A

[0039] The set of all audit trail results is: S=L.

[0040] Preferably, the process expression of analysis and production suggestion in the feedback and optimization module is:

[0041] G=A(f i ,r i )|f i ∈F,r i ∈R

[0042] Among them: F is the user feedback set; R is the audit result set; G is the optimization suggestion set; A is the analysis function; f i represents the i-th user feedback, which belongs to the user feedback set F; r i Represents the i-th audit result, which belongs to the audit result set R.

[0043] Preferably, the process expression for updating the audit rules in the feedback and optimization module is:

[0044] R′=Adjust Rules (U,G)

[0045] Where: R′ is the updated audit rule set; U is the audit rule set; Adjust Rules It is used to adjust the audit rules according to the current audit rule set I and the generated optimization suggestion set G to form an updated rule set R′.

[0046] Preferably, the sum and average values ​​of the audit results calculated in the report generation module are expressed as follows:

[0047]

[0048] Among them: SS is the sum of the audit results;

[0049] The expression for collecting improvement measures in the report generation module is:

[0050] M={m 1 ,m 2 ,…,m m}

[0051] Where: m is the number of recommended improvement measures.

[0052] Preferably, the trend analysis expression in the report generation module is:

[0053]

[0054] Where: σ is the standard deviation of the audit results; is the average value of the audit results;

[0055] Trend analysis results, including mean and standard deviation:

[0056] Preferably, the audit system performs the following steps:

[0057] Step 1: The user uploads the file through the user interface module;

[0058] Step 2: The file parsing module parses and formats the uploaded file;

[0059] Step 3: The AI ​​audit engine module receives the parsed data and audits it to generate preliminary audit results;

[0060] Step 4: The audit tracking module records the audit process and the generated results;

[0061] Step 5: The feedback and optimization module collects user feedback and continuously improves the review rules and engine performance;

[0062] Step 6: After the audit is completed, the report generation module automatically generates an audit report and displays it on the user interface;

[0063] Step 7. All data and logs are stored in the data storage module for subsequent query and analysis.

[0064] Compared with the prior art, the present invention provides a construction plan review system based on an AI big model, which has the following beneficial effects:

[0065] 1. The present invention parses a variety of file formats through the file parsing module: PDF, DOCX, XLSX, automatically extracts text, charts and table information, significantly reduces the time and energy of tedious manual review, and uses a large-scale pre-trained model to automatically evaluate the compliance and accuracy of the construction plan, reducing the errors caused by manual judgment and improving the review efficiency. The system supports a variety of file formats, and users can flexibly upload various documents according to their needs without converting the files into a specific format. The feedback and optimization module automatically generates optimization suggestions based on user feedback and review results, and adjusts the review rules in real time, so that the system can adapt to different review requirements and business changes, enhancing flexibility, AI The audit engine utilizes industry-specific knowledge, complex technical terms and standards, and is able to understand and audit highly professional content. Intelligent audits go beyond surface inspections and can deeply analyze the compliance and technical effectiveness of content. By recording each step of the audit process and generating audit logs, it ensures the transparency of the process, which not only improves the credibility of intelligent audits, but also provides data support for subsequent improvements and backtracking. By collecting user feedback and analyzing audit results, it can automatically adjust and improve audit specifications, and continuously improve the accuracy and efficiency of audits. The feedback mechanism enables the system to maintain an intelligent development trend, achieving the beneficial effect of having the flexibility and intelligence to automate the audit of complex documents.

[0066] 2. The present invention records each step of the audit process through the audit tracking module and generates a detailed audit log, which ensures the transparency of the audit process and makes each audit action have a basis. All audit steps and results are recorded and can be traced back at any time to facilitate subsequent audit and analysis, thereby improving the reliability of the system. Through the audit log, problems, loopholes or bottlenecks in the audit process can be quickly identified to provide data support for subsequent improvements. The feedback and optimization module collects user feedback on the audit results, analyzes the feedback content and generates optimization suggestions. The user feedback is combined with the audit results, and the analysis function automatically generates optimization suggestions. The audit rules are dynamically adjusted according to these suggestions to ensure that the audit specifications are continuously optimized in practice to adapt to actual operations. By analyzing user feedback, the system can identify common problems and user needs, and then optimize the audit rules to make the audit process more in line with user needs, improve the accuracy and efficiency of the final audit, and achieve the beneficial effect of having an audit tracking and feedback mechanism with continuous improvement and optimization effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a schematic diagram of the system of the present invention; DETAILED DESCRIPTION

[0068] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] See also Figure 1 ,The construction plan review system based on AI big model,the review system consists of user interface module, file parsing module, AI review engine module, audit tracking module, feedback and optimization module, report generation module and data storage module;

[0070] The user interface module is used to provide a user interface for uploading files, supporting multiple file formats: PDF, DOCX and XLSX;

[0071] The file parsing module is used to parse the uploaded construction documents, extract key information such as text, charts and tables, and convert the extracted information into a unified data format;

[0072] The AI ​​review engine module uses large-scale pre-trained models to automatically review construction plans, evaluate content compliance and accuracy, apply industry-specific knowledge, and understand complex technical terms and standards;

[0073] The audit tracking module is used to record each step of the audit process, generate audit logs, ensure process transparency, and support the traceability of audit results;

[0074] The feedback and optimization module is used to collect user feedback, analyze audit results, automatically generate optimization suggestions, adjust audit rules, and continuously improve the accuracy and efficiency of audits;

[0075] The report generation module is used to automatically generate audit reports, including audit results, recommended improvement measures and trend analysis;

[0076] The data storage module is used to store the operation values ​​generated in the above modules and the files uploaded by users;

[0077] The file format expression defined in the user interface module is:

[0078] F=PDF,DOCX,XLSX

[0079] Where: F is the set of supported file formats;

[0080] The operation expression for user uploaded files is:

[0081] U:I→F

[0082] U is the operation of users uploading files;

[0083] I is the file upload interface.

[0084] The file parsing module parses a variety of file formats: PDF, DOCX, XLSX, automatically extracts text, charts and table information, significantly reduces the time and effort of tedious manual review, and uses large-scale pre-trained models to automatically evaluate the compliance and accuracy of construction plans, reducing errors caused by manual judgment and improving review efficiency. The system supports multiple file formats, and users can flexibly upload various documents according to their needs without converting files into specific formats. The feedback and optimization module automatically generates optimization suggestions based on user feedback and review results, and adjusts the review rules in real time, so that the system can adapt to different review needs and business changes, enhancing flexibility. The AI ​​review engine uses industry-specific knowledge, complex technical terms and standards to understand and review highly professional content.

[0085] Intelligent auditing is more than just a surface check. It can deeply analyze the compliance and technical effectiveness of the content. By recording each step of the audit process and generating audit logs, it ensures the transparency of the process, which not only improves the credibility of intelligent auditing, but also provides data support for subsequent improvements and backtracking. By collecting user feedback and analyzing audit results, it can automatically adjust and improve audit specifications, continuously improve the accuracy and efficiency of audits, and the feedback mechanism enables the system to maintain an intelligent development trend.

[0086] Specifically, the function expression of the file parsing module is:

[0087] F=f1,f2,f3,…,fn,n≧1

[0088] Where: F is the uploaded construction file set, f1, f2, f3, ..., fn represent multiple files;

[0089] The file parsing module converts the data into a unified format expression:

[0090] D=P(F)=T(F),G(F),B(F)

[0091] Where: D is a unified data format; P(F) takes the uploaded construction file set F as input, processes it, extracts the corresponding information: text, charts, tables, and finally outputs a unified data format D; T(F) represents the text information set extracted from the file set F; G(F) represents the chart information set extracted from the file set F; B(F) represents the table information set extracted from the file set F.

[0092] Specifically, the AI ​​audit engine module expression is:

[0093] C=E(S,M,K,T)=(s,f p (s,M,K,T),f a (s,M,K,T))|s∈S

[0094] Among them: S is the set of uploaded construction plans; M is the large-scale pre-trained model; C is the set of audit results, including compliance and accuracy assessment results; K is the set of industry-specific knowledge; T is the set of technical terms and standards; s is the plan.

[0095] Specifically, the audit trail module expression is:

[0096] A=a1,a2,…,an,n≥1

[0097] Where: A is the set of steps in the audit process;

[0098] The process expression of the audit trail module is:

[0099] L=T(A)

[0100] Where: L is the audit log set; T is the timestamp.

[0101] Specifically, the comprehensive expression in the audit tracking module is: L = (T (a i ),E(a i ),R(a i ))|a i ∈A

[0102] The set of all audit trail results is: S=L.

[0103] Specifically, the process expression of analysis and production suggestions in the feedback and optimization module is:

[0104] G=A(f i ,r i )|f i ∈F,r i ∈R

[0105] Among them: F is the user feedback set; R is the audit result set; G is the optimization suggestion set; A is the analysis function; f i represents the i-th user feedback, which belongs to the user feedback set F; r i Represents the i-th audit result, which belongs to the audit result set R.

[0106] Specifically, the process expression for updating the audit rules in the feedback and optimization module is:

[0107] R′=Adjust Rules (U,G)

[0108] Where: R′ is the updated audit rule set; U is the audit rule set; Adjust Rules It is used to adjust the audit rules according to the current audit rule set I and the generated optimization suggestion set G to form an updated rule set R′.

[0109] Specifically, the sum and average expressions for calculating the audit results in the report generation module are:

[0110]

[0111] Among them: SS is the sum of the audit results;

[0112] The expression for collecting improvement measures in the report generation module is:

[0113] M={m 1 ,m 2 ,…,m m}

[0114] Where: m is the number of recommended improvement measures.

[0115] Specifically, the trend analysis expression in the report generation module is:

[0116]

[0117] Where: σ is the standard deviation of the audit results; is the average value of the audit results;

[0118] Trend analysis results, including mean and standard deviation:

[0119] Specifically, the audit system execution steps are:

[0120] Step 1: The user uploads the file through the user interface module;

[0121] Step 2: The file parsing module parses and formats the uploaded file;

[0122] Step 3: The AI ​​audit engine module receives the parsed data and audits it to generate preliminary audit results;

[0123] Step 4: The audit tracking module records the audit process and the generated results;

[0124] Step 5: The feedback and optimization module collects user feedback and continuously improves the review rules and engine performance;

[0125] Step 6: After the audit is completed, the report generation module automatically generates an audit report and displays it on the user interface;

[0126] Step 7. All data and logs are stored in the data storage module for subsequent query and analysis.

[0127] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The construction plan review system based on AI big model is characterized by: The audit system consists of a user interface module, a file parsing module, an AI audit engine module, an audit tracking module, a feedback and optimization module, a report generation module and a data storage module; The user interface module is used to provide a user interface for uploading files, supporting multiple file formats: PDF, DOCX and XLSX; The file parsing module is used to parse the uploaded construction files, extract key information such as text, charts and tables, and convert the extracted information into a unified data format; The AI ​​review engine module uses large-scale pre-trained models to automatically review construction plans, evaluate content compliance and accuracy, apply industry-specific knowledge, and understand complex technical terms and standards; The audit tracking module is used to record each step of the audit process, generate audit logs, ensure process transparency, and support the traceability of audit results; The feedback and optimization module is used to collect user feedback, analyze audit results, automatically generate optimization suggestions, adjust audit rules, and continuously improve the accuracy and efficiency of audits; The report generation module is used to automatically generate an audit report, including audit results, suggested improvement measures and trend analysis; The data storage module is used to store the operation values ​​generated in the above modules and the files uploaded by the user; The file format expression defined in the user interface module is: F=PDF,DOCX,XLSX Where: F is the set of supported file formats; The operation expression for user uploaded files is: U:I→F U is the operation of users uploading files; I is the file upload interface.

2. The construction plan review system based on AI big model according to claim 1 is characterized by: The function expression of the file parsing module is: F=f1,f2,f3,…,fn,n≧1 Where: F is the uploaded construction file set, f1, f2, f3, ..., fn represent multiple files; The file parsing module converts the data into a unified format expression: D=P(F)=T(F),G(F),B(F) Where: D is a unified data format; P(F) takes the uploaded construction file set F as input, processes it, extracts the corresponding information: text, charts, tables, and finally outputs a unified data format D; T(F) represents the text information set extracted from the file set F; G(F) represents the chart information set extracted from the file set F; B(F) represents the table information set extracted from the file set F.

3. The construction plan review system based on AI big model according to claim 1 is characterized by: The AI ​​audit engine module expression is: C=E(S,M,K,T)=(s,f p (s,M,K,T),f a (s,M,K,T))|s∈S Among them: S is the set of uploaded construction plans; M is the large-scale pre-trained model; C is the set of audit results, including compliance and accuracy assessment results; K is the set of industry-specific knowledge; T is the set of technical terms and standards; s is the plan.

4. The construction plan review system based on AI big model according to claim 3 is characterized by: The audit trail module expression is: A=a1,a2,…,an,n≥1 Where: A is the set of steps in the audit process; The process expression of the audit trail module is: L=T(A) Where: L is the audit log set; T is the timestamp.

5. The construction plan review system based on AI big model according to claim 4 is characterized by: The comprehensive expression in the audit tracking module is: L = (T (a i ),E(a i ),R(a i ))|a i ∈A The set of all audit trail results is: S=L.

6. The construction plan review system based on AI big model according to claim 1 is characterized by: The process expression of analysis and production suggestions in the feedback and optimization module is: G=A(f i ,r i )|f i ∈F,r i ∈R Among them: F is the user feedback set; R is the audit result set; G is the optimization suggestion set; A is the analysis function; f i represents the i-th user feedback, which belongs to the user feedback set F; r i Represents the i-th audit result, which belongs to the audit result set R.

7. The construction plan review system based on AI big model according to claim 6 is characterized by: The process expression for updating the audit rules in the feedback and optimization module is: R′=Adjust Rules (U,G) Where: R′ is the updated audit rule set; U is the audit rule set; Adjust Rules It is used to adjust the audit rules according to the current audit rule set U and the generated optimization suggestion set G to form an updated rule set R′.

8. The construction plan review system based on AI big model according to claim 7 is characterized by: The sum and average values ​​of the audit results calculated in the report generation module are as follows: Among them: SS is the sum of the audit results; The expression for collecting improvement measures in the report generation module is: M={m1,m2,…,m m } Where: m is the number of recommended improvement measures.

9. The construction plan review system based on AI big model according to claim 1 is characterized by: The trend analysis expression in the report generation module is: Where: σ is the standard deviation of the audit results; is the average value of the audit results; Trend analysis results, including mean and standard deviation: T = {R, σ}.

10. The construction plan review system based on AI big model according to claim 1 is characterized by: The audit system performs the following steps: Step 1: The user uploads the file through the user interface module; Step 2: The file parsing module parses and formats the uploaded file; Step 3: The AI ​​audit engine module receives the parsed data and audits it to generate preliminary audit results; Step 4: The audit tracking module records the audit process and the generated results; Step 5: The feedback and optimization module collects user feedback and continuously improves the review rules and engine performance; Step 6: After the audit is completed, the report generation module automatically generates an audit report and displays it on the user interface; Step 7. All data and logs are stored in the data storage module for subsequent query and analysis.

Images