Highway engineering project electronic archive compliance verification method and system

Abstract

This invention discloses a method and system for verifying the compliance of electronic archives in highway engineering projects. The method includes: acquiring construction management data, IoT device status data, supervision system data, and original electronic archive data, and integrating and preprocessing them to generate an archive verification feature vector; acquiring highway engineering industry standards, parsing the standards to construct a standard compliance knowledge graph containing archive nodes, process logic, and parameter thresholds, and generating machine-recognizable verification rules based on this knowledge graph; using natural language processing technology to parse the project's electronic archives, extracting key feature values, and comparing these key feature values ​​with the verification rules to obtain a compliance score for the project's electronic archives; performing a cross-domain comparison between the archive verification feature vector and raw real-time data generated by an external IoT terminal to obtain the authenticity confidence level of the project's electronic archives; and generating a verification evaluation report based on the compliance score and authenticity confidence level.

Description

Technical Field

[0001] This invention relates to the field of document compliance verification technology, and in particular to a method and system for verifying the compliance of electronic documents in highway engineering projects. Background Technology

[0002] In the final acceptance process of highway engineering projects, verifying the compliance and authenticity of electronic archives is crucial. Existing verification methods mainly rely on manual sampling combined with the system's logical verification of basic archive attributes (such as file format, signing time, and completeness of required fields).

[0003] The methods described above can verify the formal compliance of electronic archives, i.e., "whether the document exists," but cannot delve into the accuracy of the archive content, i.e., "whether the content is correct." For example, the system cannot automatically identify whether the construction parameters recorded in the archive (such as pile number and strength value) match the original construction records or industry standards. Furthermore, because archive management systems, on-site IoT monitoring systems, and laboratory systems are often independent and their data is not interconnected, existing technologies struggle to perform cross-system verification of the facts recorded in the archives (such as when and where a certain construction was carried out). This makes it difficult to effectively identify acts such as supplementary archives, data manipulation, or false cross-regional inspections, resulting in a huge workload for manual review and difficulty in ensuring the quality of acceptance. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for verifying the compliance of electronic archives in highway engineering projects, so as to solve the technical problems of existing technologies that cannot automatically verify the content of electronic archives and that it is difficult to conduct penetrating verification of the authenticity of archives, and realize intelligent and automated in-depth auditing of the compliance of electronic archive content and the authenticity of facts.

[0005] To achieve the above objectives, a method for verifying the compliance of electronic archives in highway engineering projects is provided, which includes the following steps: Acquire construction management data, IoT device status data, supervision system data, and original electronic file data, and integrate and preprocess the acquired construction management data, IoT device status data, supervision system data, and original electronic file data to generate file verification feature vectors; Obtain highway engineering industry standards, parse the highway engineering industry standards to construct a standard compliance knowledge graph containing file nodes, process logic and parameter thresholds, and generate machine-recognizable verification rules based on the standard compliance knowledge graph; Natural language processing technology is used to parse the electronic project file, extract key feature values ​​from the electronic project file, and call the verification rules to compare the key feature values ​​to obtain the compliance score of the electronic project file. The document verification feature vector is compared with the raw real-time data generated by the external IoT terminal across domains to obtain the authenticity confidence level of the project's electronic document; A verification evaluation report is generated based on the compliance score and the confidence level of authenticity.

[0006] According to the aforementioned method for verifying the compliance of electronic archives for highway engineering projects, the process of acquiring construction management data, IoT device status data, supervision system data, and original electronic archive data, and then integrating and preprocessing them, specifically includes: The construction management data, the IoT device status data, the supervision system data, and the original electronic archive data are integrated in a time sequence to form the project electronic archive; Establish the association between the metadata, business process data, and physical entity data of the project's electronic archives; Based on the aforementioned relationship, the file verification feature vector is generated.

[0007] Based on the aforementioned method for verifying the compliance of electronic archives for highway engineering projects, highway engineering industry standards are analyzed to construct a standard compliance knowledge graph, specifically including: The highway engineering industry standard is semantically parsed to identify the hierarchical structure of sub-projects, procedures, inspection items and quality requirements, and numerical features are extracted from the text. Define file nodes, process nodes, and attribute nodes, and define the temporal relationships, inclusion relationships, or mutual exclusion relationships between nodes; Based on the hierarchical structure, the numerical features, and the node relationships, the standard compliance knowledge graph is constructed. According to the aforementioned method for verifying the compliance of electronic archives for highway engineering projects, machine-recognizable verification rules are generated based on the aforementioned standard compliance knowledge graph, specifically including: The logic in the standard compliance knowledge graph is transformed into machine-executable code logic to obtain the verification rules; Establish a mapping table between the verification rules and the file verification feature vectors; The invocation of the verification rule includes: automatically invoking the corresponding verification rule from the mapping table for comparison based on the category label of the project's electronic file.

[0008] According to the aforementioned method for verifying the compliance of electronic archives for highway engineering projects, the verification rules are invoked to compare the key feature values, including at least one of execution time conflict verification, spatial consistency verification, and physical index coupling verification.

[0009] According to the aforementioned method for verifying the compliance of electronic archives for highway engineering projects, the time conflict verification includes: Extract the document signing time, original record generation time, and physical equipment operation time trajectory from the project's electronic archives; Based on the verification rules, the preset time logic constraints are invoked; The extracted file signing time, original record generation time, and physical device operation time trajectory are compared with the time logic constraints to identify time logic contradictions. Specifically, the extracted file signing time, original record generation time, and physical device operation time trajectory are compared with the time logic constraints, including performing at least one of the following: performing reverse order contradiction judgment, performing false idle judgment, and performing time sequence offset calculation. When performing the reverse order contradiction determination, if the file signing time is earlier than the original record generation time, and / or the original record generation time is earlier than the operation start time in the physical equipment operation time trajectory, then a reverse order contradiction is determined to exist; When performing a false idling determination, if the physical equipment running time trajectory of the associated construction machinery is shown to be offline during the construction period recorded in the project's electronic file, and the offline duration exceeds a preset first threshold, then it is determined to be a false idling.

[0010] When performing time offset calculation, the time offset coefficient is calculated based on the original record generation time and the actual peak operation time of the physical device running time trajectory. If the time offset coefficient is greater than the preset deviation threshold, it is determined that there is a time logic conflict.

[0011] According to the aforementioned method for verifying the compliance of electronic archives for highway engineering projects, the spatial consistency verification includes: Extract the geographic coordinates of the archival records from the project's electronic archives; Obtain the set of mechanical trajectory points of the construction machinery associated with the electronic file of the project during the corresponding work process time period, and the set of trajectory points for supervision verification; Spatial matching is performed between the geographic coordinates of the archive records and the set of mechanical trajectory points and the set of supervisory verification trajectory points to identify trajectory deviations; Specifically, spatial matching is performed between the geographic coordinates of the archive records and the mechanical trajectory point set and the supervision and verification trajectory point set, including spatial position deviation determination and execution trajectory missing determination. When determining spatial position deviation, the Euclidean distance between the geographic coordinates of the archive record and the centroid coordinates of the mechanical trajectory point set is calculated. If the Euclidean distance is greater than the preset spatial tolerance radius, it is determined to be a spatial position deviation. When determining if a trajectory is missing, if the project's electronic file records key processes, but no associated set of mechanical trajectory points or the set of supervisory verification trajectory points are obtained within the project's geofence during the corresponding time period, then the trajectory is determined to be missing.

[0012] According to the aforementioned method for verifying the compliance of electronic archives for highway engineering projects, the physical index coupling verification includes: Extract key indicators for judgment from the experimental reports in the electronic archives of the project; Obtain the raw sensor sampling data corresponding to the experimental report; The key judgment indicators are compared with the physical feature values ​​calculated based on the original sampling data of the sensor to identify experimental errors.

[0013] According to the aforementioned method for verifying the compliance of electronic archives for highway engineering projects, obtaining the confidence level of the authenticity of the electronic archives specifically includes: Based on the document verification feature vector, obtain the corresponding physical evidence data from an external IoT terminal; A weighted similarity method is used to calculate the matching degree between the archival verification feature vector and the physical evidence data; Based on the matching degree, the confidence level of authenticity is obtained through a confidence calculation model; The authenticity confidence level is compared with a preset confidence level range, and the judgment result is output. The confidence level calculation model is as follows: in, The confidence level for the stated authenticity. For the first Weight coefficients for each verification dimension For the first The matching degree of each verification dimension. As a data source reliability correction factor, This represents the total number of verification dimensions.

[0014] This invention also provides a compliance verification system for electronic archives of highway engineering projects, comprising: The data integration and preprocessing module is used to acquire construction management data, IoT device status data, supervision system data and original electronic archive data, integrate and preprocess them, and generate archive verification feature vectors. The knowledge graph construction module is used to acquire highway engineering industry standards, parse the highway engineering industry standards to construct a standard compliance knowledge graph containing file nodes, process logic and parameter thresholds, and generate machine-recognizable verification rules based on the standard compliance knowledge graph. The content compliance audit module is used to parse the electronic files of the project using natural language processing technology, extract key feature values, and call the verification rules to compare the key feature values ​​to obtain a compliance score. The authenticity penetration verification module is used to perform cross-domain comparison between the archive verification feature vector and the raw real-time data generated by the external IoT terminal to obtain the authenticity confidence of the project's electronic archive. The report generation module is used to generate a verification evaluation report based on the compliance score and the confidence level of authenticity.

[0015] Beneficial Effects: By constructing a standard compliance knowledge graph and generating machine-recognizable verification rules, this method provides accurate and automated comparison criteria for auditing archival content, thus solving the problem that existing technologies cannot verify the correctness of archival content. Through cross-domain comparison of archival verification feature vectors with raw real-time data from external IoT terminals, it achieves penetrating verification of the authenticity of archival records, effectively identifying logical contradictions and false information. This method and system, by quantifying compliance scores and authenticity confidence levels and automatically generating verification evaluation reports, achieves intelligent and comprehensive auditing of the compliance and authenticity of electronic archives in highway engineering projects, significantly reducing manual verification costs and improving the efficiency of project completion and acceptance.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments; Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0018] This section will describe in detail specific embodiments of the present invention. Preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the drawings is to supplement the textual description with graphics, so that people can intuitively and vividly understand each technical feature and overall technical solution of the present invention, but they should not be construed as limiting the scope of protection of the present invention.

[0019] Reference Figure 1 The method for verifying the compliance of electronic archives of highway engineering projects includes steps S101 to S105.

[0020] Step S101: Data Integration and Preprocessing. Specifically, through a pre-defined data interface, construction progress data, construction quality data, and construction safety data (collectively referred to as construction management data) are obtained from the construction management system; through an IoT data acquisition module, equipment status data (IoT equipment status data) generated by sensors installed on construction machinery (such as pavers and road rollers) is obtained; supervision approval and verification data (supervision system data) is obtained from the supervision office system; and simultaneously, the original electronic archive data to be verified is obtained. The data processing unit (such as a processor) arranges and aligns these data from different sources and in different formats according to a unified timestamp, integrating them into a structured project electronic archive. Furthermore, the processor, based on entity recognition and relationship extraction algorithms, establishes the association relationships between the project electronic archive's metadata (such as archive number and type), business process data (such as approval flow nodes), and physical entity data at the construction site (such as machinery number and sensor ID), and generates a multi-dimensional archive verification feature vector based on these association relationships. This feature vector serves as the core data carrier for subsequent verification, realizing the link between static archive information and the dynamic physical world. The association relationships can be manually verified to ensure accuracy.

[0021] Step S102: Constructing the knowledge graph and verification rules. In this embodiment, step S102 specifically includes the following sub-steps.

[0022] Sub-step S1021 involves using a deep learning model (such as BERT) to perform semantic parsing on the acquired highway engineering industry standard text. The processor runs the model to identify the implicit hierarchical structure of "sub-project-process-inspection item-quality requirements" in the standard text, and automatically extracts specific numerical features from the text, such as the threshold "±10" in "allowable deviation of road surface thickness is ±10mm".

[0023] Sub-step S1022 defines the nodes and relationships of the knowledge graph. The processor creates archive nodes (such as "concrete strength test report"), process nodes (such as "beam and slab pouring"), and attribute nodes (such as "design strength C50") in memory or a graph database, and defines the temporal relationships (such as "reinforcement binding" must precede "concrete pouring"), inclusion relationships (such as "bridge engineering" includes "pile foundation construction"), or mutual exclusion relationships between these nodes. Based on the hierarchical structure and numerical features parsed from sub-step S1021, the processor automatically or semi-automatically constructs a standard compliance knowledge graph that includes archive nodes, process logic, and parameter thresholds.

[0024] In sub-step S1023, the processor transforms the logical relationships (such as timing constraints and numerical range constraints) defined in the standard compliance knowledge graph into machine-executable code logic (such as IF-THEN rules or functions) to obtain machine-recognizable verification rules.

[0025] In sub-step S1024, the processor establishes a mapping table between the verification rules and each dimension of the file verification feature vector generated in step S101. When it is necessary to verify a specific category of file, the system can automatically and accurately call the corresponding set of verification rules from the mapping table for comparison based on the category label of the file (such as "subgrade compaction record table"), thereby improving the automation and targeting of the audit.

[0026] Step S103: Content Compliance Audit. The processor uses natural language processing technologies (such as named entity recognition and keyword extraction) to parse the project's electronic archives, extracting key feature values ​​from the archive text, tables, or structured fields, such as station number (K100+500), strength grade (C30), and measured deviation (+5mm). Then, the processor calls the verification rules generated in step S102 and corresponding to the archive to automatically compare these key feature values ​​and calculate the archive's compliance score. The compliance score can be a percentage or a grade, reflecting the degree to which the archive content conforms to industry standards.

[0027] Specifically, content compliance audits may include one or more of the following: time conflict verification, spatial consistency verification, and physical indicator coupling verification.

[0028] Regarding time conflict verification: The processor extracts the document signing time, original record generation time, and physical equipment operation time trajectory from the project's electronic archives. Based on verification rules, the processor invokes preset time logic constraints, such as "the original record generation time must be within the equipment's operating time period." Then, the processor compares the extracted time information with the time logic constraints to identify time logic contradictions. The specific comparison process can be executed as one or a combination of the following judgment logics: Reversal Contradiction Detection: The processor compares the file signing time, the original record generation time, and the job start time in the physical device runtime timeline. If the file signing time is earlier than the original record generation time, and / or the original record generation time is earlier than the job start time, a reversal contradiction is determined, indicating that the file flow is illogical.

[0029] False Idle Running Detection: The processor checks the construction time period recorded in the project's electronic file. If, during this time period, the physical equipment running time trajectory corresponding to the associated construction machinery shows that the equipment is always offline (no data), and the offline duration exceeds a preset first threshold (e.g., 30 minutes), it is determined to be false idle running, indicating that the file may have fabricated construction activities.

[0030] Time offset calculation: The processor calculates the time offset coefficient γ based on the original record generation time Tr and the actual peak operation time Tp of the physical equipment's running time trajectory. The calculation formula is: in, β represents the total construction period recorded in the project's electronic archives, and β is a preset proportional coefficient (e.g., a value of 1). If the calculated time offset coefficient γ is greater than the preset deviation threshold (e.g., 0.2), a time logic conflict is determined to exist.

[0031] Regarding spatial consistency verification: The processor extracts the geographic coordinates (La, Ba) of the archive records from the project's electronic archives. Simultaneously, it obtains the continuous coordinate point set (machine trajectory point set) Pi of the construction machinery associated with the project's electronic archives during the corresponding work process's operation time period from the IoT positioning terminal, and obtains the supervisor's verification trajectory point set Pj from the supervisor's mobile terminal. The processor performs spatial matching between the archive record geographic coordinates and these trajectory point sets to identify trajectory deviations. The specific matching process can be executed as one or a combination of the following judgment logic: Euclidean distance calculation and determination: The processor calculates the Euclidean distance between the geographic coordinates (La, Ba) of the file record and the centroid coordinates (Lm, Bm) of the machine trajectory point set Pi. : If Euclidean distance If the spatial tolerance radius exceeds the preset radius R (e.g., 50 meters), it is determined to be a spatial position deviation. Remote inspection judgment: The processor maps all coordinate points in the machinery trajectory point set Pi to the project's digital alignment map. If the project's electronic file records construction in the first section (e.g., K10-K15), but Pi shows that all coordinate points of the associated construction machinery within the corresponding time period are located in the second section (e.g., K20-K25), it is determined to be a remote inspection and identified as a potentially false file.

[0032] Trajectory Missing Determination: If the project's electronic file records a key process (such as "bridge deck paving"), but the processor does not obtain any associated mechanical trajectory point set Pi and supervisor verification trajectory point set Pj within the project's preset geofence during the corresponding time period recorded in the file, it is determined to be a trajectory missing, indicating that there may be no corresponding activity on site.

[0033] Regarding the coupling verification of physical indicators: The processor utilizes natural language processing and document parsing technology to extract the key indicators for final judgment (such as "compressive strength 35.2 MPa") from the experimental reports (e.g., PDF format) attached to the project's electronic archives. Simultaneously, based on information such as the archive number, it retrieves the original sensor sampling data (i.e., the sequence of physical quantities changing over time) corresponding to the experimental report from the laboratory information management system. The processor compares the key indicators extracted from the report with the physical characteristic values ​​obtained through calculations (e.g., finding peak values, taking average values) based on the original sensor sampling data, calculating the difference between the two to identify whether the experimental error is within the allowable range of equipment accuracy or specifications. If the error exceeds the limit, the compliance score for that indicator is lowered.

[0034] Step S104: Authenticity Penetration Verification. The processor performs a cross-domain comparison between the document verification feature vector (representing the facts "claimed" by the document) formed in step S101 and the original physical evidence data acquired in real time from external IoT terminals (such as construction machinery controllers and laboratory sensors). Specifically, the processor extracts the corresponding physical evidence data from the IoT data warehouse based on the associated information (such as time and machinery number) in the document verification feature vector. Then, a weighted similarity method (such as weighted cosine similarity) is used to calculate the matching degree between the document verification feature vector and the physical evidence data in various dimensions (such as time, space, and physical quantity). Based on these matching degrees, a confidence level calculation model is used to obtain the authenticity confidence level of the project's electronic archives. One implementation of this model is as follows: in, For the confidence level of authenticity, The weight coefficient for the k-th verification dimension (such as time, space, physical indicators) can be preset by the importance of industry standards in step S102. The matching degree of the k-th verification dimension; This is a data source reliability correction factor used to account for the reliability differences of different IoT data sources; n is the total number of verification dimensions.

[0035] Obtain the confidence level of authenticity Then, the processor compares it with a preset confidence range. Compare and output the result: If If the electronic files of the project highly match the physical facts, then it is determined that the project's electronic files are highly consistent with the physical facts; if If the project's electronic archives show a data discrepancy, then it is determined that there is a data deviation. If the electronic file of the project is deemed to be a potentially falsified file, a risk warning will be triggered, and relevant personnel will be notified.

[0036] Step S105: Generate a verification and evaluation report. The report generation module (which can be implemented by the processor executing the corresponding program) performs a comprehensive analysis and formatting based on the compliance score obtained in step S103 and the authenticity confidence level and its judgment results obtained in step S104. The report automatically outputs a multi-dimensional verification and evaluation report containing an overview of the file, compliance analysis (including the verification results of each sub-item), authenticity assessment, identified specific risk points (such as "time sequence inconsistency" and "spatial location deviation exceeding the threshold"), risk level, and improvement suggestions. This report can be directly used for project acceptance decisions, significantly reducing the workload of manual review and report preparation.

[0037] This invention also provides a compliance verification system for electronic archives of highway engineering projects, comprising a data integration and preprocessing module, a knowledge graph construction module, a content compliance audit module, an authenticity penetration verification module, and a report generation module. The data integration and preprocessing module acquires construction management data, IoT device status data, supervision system data, and original electronic archive data, integrates and preprocesses them to generate archive verification feature vectors. The knowledge graph construction module acquires highway engineering industry standards, parses these standards to construct a standard compliance knowledge graph containing archive nodes, process logic, and parameter thresholds, and generates machine-recognizable verification rules based on the standard compliance knowledge graph. The content compliance audit module uses natural language processing technology to parse project electronic archives, extracts key feature values, and calls the verification rules to compare these key feature values ​​to obtain a compliance score. The authenticity penetration verification module performs a cross-domain comparison between the archive verification feature vectors and original real-time data generated by external IoT terminals to obtain the authenticity confidence level of the project electronic archives. The report generation module generates a verification evaluation report based on the compliance score and the authenticity confidence level.

[0038] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A method for verifying the compliance of electronic archives in highway engineering projects, characterized in that, Includes the following steps: Acquire construction management data, IoT device status data, supervision system data, and original electronic file data, and integrate and preprocess the acquired construction management data, IoT device status data, supervision system data, and original electronic file data to generate file verification feature vectors; Obtain highway engineering industry standards, parse the highway engineering industry standards to construct a standard compliance knowledge graph containing file nodes, process logic and parameter thresholds, and generate machine-recognizable verification rules based on the standard compliance knowledge graph; Natural language processing technology is used to parse the electronic project file, extract key feature values ​​from the electronic project file, and call the verification rules to compare the key feature values ​​to obtain the compliance score of the electronic project file. The document verification feature vector is compared with the raw real-time data generated by the external IoT terminal across domains to obtain the authenticity confidence level of the project's electronic document; A verification evaluation report is generated based on the compliance score and the confidence level of authenticity.

2. The method for verifying the compliance of electronic archives of highway engineering projects according to claim , characterized in that, The acquisition, integration, and preprocessing of construction management data, IoT device status data, supervision system data, and original electronic archive data specifically include: The construction management data, the IoT device status data, the supervision system data, and the original electronic archive data are integrated in a time sequence to form the project electronic archive; Establish the association between the metadata, business process data, and physical entity data of the project's electronic archives; Based on the aforementioned relationship, the file verification feature vector is generated.

3. The method for verifying the compliance of electronic archives of highway engineering projects according to claim 1, characterized in that, Analyzing highway engineering industry standards to construct a knowledge graph of standard compliance, specifically including: The highway engineering industry standard is semantically parsed to identify the hierarchical structure of sub-projects, procedures, inspection items and quality requirements, and numerical features are extracted from the text. Define file nodes, process nodes, and attribute nodes, and define the temporal relationships, inclusion relationships, or mutual exclusion relationships between nodes; Based on the hierarchical structure, the numerical features, and the node relationships, the standard compliance knowledge graph is constructed.

4. The method for verifying the compliance of electronic archives of highway engineering projects according to claim 1, characterized in that, Based on the aforementioned standard compliance knowledge graph, machine-recognizable verification rules are generated, specifically including: The logic in the standard compliance knowledge graph is transformed into machine-executable code logic to obtain the verification rules; Establish a mapping table between the verification rules and the file verification feature vectors; The invocation of the verification rule includes: automatically invoking the corresponding verification rule from the mapping table for comparison based on the category label of the project's electronic file.

5. The method for verifying the compliance of electronic archives of highway engineering projects according to claim 1, characterized in that, The verification rules are invoked to compare the key feature values, including at least one of time conflict verification, spatial consistency verification, and physical index coupling verification.

6. The method for verifying the compliance of electronic archives of highway engineering projects according to claim 5, characterized in that, The time conflict verification includes: Extract the document signing time, original record generation time, and physical equipment operation time trajectory from the project's electronic archives; Based on the verification rules, the preset time logic constraints are invoked; The extracted file signing time, original record generation time, and physical device operation time trajectory are compared with the time logic constraints to identify time logic contradictions. Specifically, the extracted file signing time, original record generation time, and physical device operation time trajectory are compared with the time logic constraints, including performing at least one of the following: performing reverse order contradiction judgment, performing false idle judgment, and performing time sequence offset calculation. When performing the reverse order contradiction determination, if the file signing time is earlier than the original record generation time, and / or the original record generation time is earlier than the operation start time in the physical equipment operation time trajectory, then a reverse order contradiction is determined to exist; When performing a false idling determination, if the physical equipment running time trajectory of the associated construction machinery shows that the equipment is offline during the construction period recorded in the project's electronic file, and the offline duration exceeds a preset first threshold, then it is determined to be a false idling. When performing time offset calculation, the time offset coefficient is calculated based on the original record generation time and the actual peak operation time of the physical device running time trajectory. If the time offset coefficient is greater than the preset deviation threshold, it is determined that there is a time logic conflict.

7. The method for verifying the compliance of electronic archives of highway engineering projects according to claim 5, characterized in that, The spatial consistency check includes: Extract the geographic coordinates of the archival records from the project's electronic archives; Obtain the set of mechanical trajectory points of the construction machinery associated with the electronic file of the project during the corresponding work process time period, and the set of trajectory points for supervision verification; Spatial matching is performed between the geographic coordinates of the archive records and the set of mechanical trajectory points and the set of supervisory verification trajectory points to identify trajectory deviations; Specifically, spatial matching is performed between the geographic coordinates of the archive records and the mechanical trajectory point set and the supervision and verification trajectory point set, including spatial position deviation determination and execution trajectory missing determination. When determining spatial position deviation, the Euclidean distance between the geographic coordinates of the archive record and the centroid coordinates of the mechanical trajectory point set is calculated. If the Euclidean distance is greater than the preset spatial tolerance radius, it is determined to be a spatial position deviation. When determining if a trajectory is missing, if the project's electronic file records key processes, but no associated set of mechanical trajectory points or the set of supervisory verification trajectory points are obtained within the project's geofence during the corresponding time period, then the trajectory is determined to be missing.

8. The method for verifying the compliance of electronic archives of highway engineering projects according to claim 5, characterized in that, The physical index coupling verification includes: Extract key indicators for judgment from the experimental reports in the electronic archives of the project; Obtain the raw sensor sampling data corresponding to the experimental report; The key judgment indicators are compared with the physical feature values ​​calculated based on the original sampling data of the sensor to identify experimental errors.

9. The method for verifying the compliance of electronic archives of highway engineering projects according to claim 1, characterized in that, The confidence level of the authenticity of the obtained electronic archives of the project specifically includes: Based on the document verification feature vector, obtain the corresponding physical evidence data from an external IoT terminal; A weighted similarity method is used to calculate the matching degree between the archival verification feature vector and the physical evidence data; Based on the matching degree, the confidence level of authenticity is obtained through a confidence calculation model; The authenticity confidence level is compared with a preset confidence level range, and the judgment result is output. The confidence level calculation model is as follows: in, The confidence level for the stated authenticity. For the first The weight coefficients of each verification dimension. For the first The matching degree of each verification dimension. As a data source reliability correction factor, This represents the total number of verification dimensions.

10. A compliance verification system for electronic archives of highway engineering projects, characterized in that, include: The data integration and preprocessing module is used to acquire construction management data, IoT device status data, supervision system data and original electronic archive data, integrate and preprocess them, and generate archive verification feature vectors. The knowledge graph construction module is used to acquire highway engineering industry standards, parse the highway engineering industry standards to construct a standard compliance knowledge graph containing file nodes, process logic and parameter thresholds, and generate machine-recognizable verification rules based on the standard compliance knowledge graph. The content compliance audit module is used to parse the electronic files of the project using natural language processing technology, extract key feature values, and call the verification rules to compare the key feature values ​​to obtain a compliance score. The authenticity penetration verification module is used to perform cross-domain comparison between the archive verification feature vector and the raw real-time data generated by the external IoT terminal to obtain the authenticity confidence of the project's electronic archive. The report generation module is used to generate a verification evaluation report based on the compliance score and the confidence level of authenticity.

Images