A method for intelligent extraction of engineering quantities in highway engineering

Abstract

This application relates to the technical field of information technology applications in construction engineering, and discloses a method for intelligently extracting engineering quantities for highway engineering applications. The method comprises: standardizing the engineering quantity table used in highway engineering design to form a standardized engineering quantity table; establishing a budget project reference template; establishing a connection relationship between the budget project reference template and the standardized engineering quantity table, wherein the connection relationship includes at least a semantic correspondence between project sections and engineering quantity table fields, logical matching rules between quota sub-items and engineering parameters, and an adaptive adjustment mechanism for differences in table formats; and when a new engineering quantity table is imported, automatically identifying and parsing the data fields and structure of the new engineering quantity table based on the connection relationship. The data in the new engineering quantity correspondence table is mapped to the corresponding project sections and quota sub-items of the budget project reference template through semantic parsing and logical matching rules, thereby generating a budget file that meets the requirements for cost compilation.

Description

Technical Field

[0001] The present application relates to the technical field of application of information technology in construction engineering, and in particular to a method for intelligently extracting engineering quantities for highway engineering. Background Art

[0002] In the field of budget preparation for highway projects, especially when there are a large number of structures such as bridges and tunnels, the traditional method relies on manual extraction of design quantity table data, matching quotas and entering them into cost software. For example, compiling the cost of 60 tunnels takes 10 working days, and there are problems such as low efficiency, strong format dependence, and accuracy risks. Although existing technologies have attempted to achieve partial automation by extracting data through spreadsheet rules, they can only adapt to fixed formats (such as Excel column order and fixed cell structure). When the design unit adjusts the table structure (such as merging cells and changing the column order), the program needs to be redeveloped, and its versatility is insufficient. At the same time, manual understanding of the correspondence between design parameters and quota sub-items is prone to ambiguity, leading to matching errors. The core defects of existing technologies are:

[0003] 1. Data conversion relies on manual intervention: The mapping process from the design quantity table to the budget document requires a lot of manual work, which cannot meet the needs of efficient comparison and selection when design schemes are adjusted (such as changes in tunnel surrounding rock grade);

[0004] 2. Poor format compatibility: Relying on a fixed table structure, it lacks the ability to adaptively parse differences in table formats;

[0005] 3. Inaccurate semantic matching: Parameter understanding and quota matching rely on manual experience and lack a standardized semantic parsing mechanism, resulting in insufficient accuracy.

[0006] Therefore, a method is urgently needed to solve at least one of the above problems. Summary of the Invention

[0007] This application provides a method for intelligently extracting engineering quantities for highway engineering projects, aiming to solve problems in the field of highway engineering budget preparation, especially when there are a large number of structures such as bridges and tunnels, such as the existing technology that data conversion relies on manual intervention, has poor format compatibility, and inaccurate semantic matching.

[0008] In a first aspect, an embodiment of the present application provides a method for intelligently extracting engineering quantities for a highway project, comprising:

[0009] Standardizing the bill of quantities in highway engineering design to form a standardized bill of quantities; wherein the standardization includes defining unified data fields, data structures and parameter rules;

[0010] Establishing a budget project reference template, wherein the budget project reference template includes the project section structure, corresponding quota sub-items and matching rules required for the budget of the highway project;

[0011] Establishing a connection relationship between the budget project reference template and the standardized bill of quantities, wherein the connection relationship includes at least a semantic correspondence between project sections and bill of quantities fields, a logical matching rule between quota sub-items and engineering parameters, and an adaptive adjustment mechanism for differences in table formats;

[0012] When a new bill of quantities is imported, the data fields and structure of the new bill of quantities are automatically identified and parsed based on the connection relationship. Through semantic analysis and logical matching rules, the data in the new bill of quantities corresponding table are mapped to the corresponding project sections and quota sub-items of the budget project reference template to generate a budget file that meets the cost compilation requirements.

[0013] In some embodiments, the standardization of the quantity table in highway engineering design to form a standardized quantity table includes: defining unified data field names, data types and field descriptions, and establishing a unique mapping relationship between fields and engineering objects; standardizing the hierarchical relationship of the data structure, including the list level, sub-project level and specific parameter level; formulating parameter rules, including standardization rules for quantity units, numerical precision constraint rules and parameter integrity verification rules; and forming a standardized quantity table based on the unique mapping relationship, hierarchical relationship and parameter rules.

[0014] In some embodiments, the establishment of a budget project reference template includes: constructing a hierarchical project section structure including project sections, sub-item sections, and quota items in accordance with the highway engineering budget preparation specifications; establishing a quota sub-item library, the quota sub-item library including the engineering parameter ranges, engineering quantity calculation rules, and price components corresponding to each quota; obtaining matching rules, the matching rules including the quota sub-item screening logic based on engineering parameter thresholds, and the correlation between engineering quantity calculation rules and design parameters; establishing a budget project reference template based on the hierarchical project section structure, the quota sub-item library, and the matching rules.

[0015] In some embodiments, establishing a connection relationship between the budget project reference template and the standardized engineering quantity table includes: constructing a highway engineering field terminology library, the highway engineering field terminology library stores the semantic mapping relationship between design parameter terms and budget project section terms, for eliminating ambiguity between different expressions; establishing a logical matching rule library, the logical matching rule library contains the correspondence between engineering parameters and quota sub-item selection conditions; the engineering parameters include at least bridge span and tunnel surrounding rock grade; designing a table format difference analysis algorithm, the table format difference analysis algorithm automatically adapts to structural changes such as column order changes and cell merging by identifying semantic tags of cell contents; establishing the connection relationship based on the highway engineering field terminology library, the logical matching rule library and the table format difference analysis algorithm.

[0016] In some embodiments, the data in the new engineering quantity correspondence table is mapped to the corresponding project section and quota sub-item of the budget project reference template through semantic analysis and logical matching rules, including: using the highway engineering field terminology library to semantically convert the parameter terms in the new engineering quantity correspondence table and match them to the standard terms of the project section; based on the logical matching rule library, conditional judgment is made on the engineering parameters corresponding to the new engineering quantity, and matching quota sub-items are automatically screened; consistency verification is performed on the mapped new engineering quantity corresponding data and the parameters required for the quota sub-items, and a parameter matching report is generated to complete the mapping of the data in the new engineering quantity correspondence table to the corresponding project section and quota sub-item of the budget project reference template.

[0017] In some embodiments, the semantic parsing process introduces knowledge graph technology to construct a relationship network between highway engineering design parameters and budget quotas, and solves the ambiguity problem of complex parameter combinations through graph reasoning algorithms to improve matching accuracy; the complex parameter combinations include quota matching for collaborative construction of multiple components.

[0018] In some embodiments, the establishment of a logic matching rule library includes: optimizing the matching relationship between engineering parameters and quota sub-items through a machine learning model based on training of historical engineering data, and the machine learning model can dynamically update the matching rules according to the design scheme adjustment; the design scheme adjustment includes changes in bridge span and tunnel surrounding rock grade.

[0019] In some embodiments, the automatic identification and parsing of the data fields and structure of the new engineering quantity table includes: scanning the cell contents of the new engineering quantity table through preset field semantic tags to match standardized data fields; based on the table structure tree algorithm, analyzing the row and column merging relationship and hierarchical indentation relationship of the table corresponding to the new engineering quantity table, generating a parsing model compatible with the standardized data structure, and completing the automatic identification and parsing of the data fields and structure of the new engineering quantity table.

[0020] In some embodiments, the method of scanning the cell contents of the new engineering quantity table through preset field semantic tags and matching standardized data fields includes: using natural language processing technology to perform semantic analysis on the table title and cell contents of the new engineering quantity table, identifying implicit engineering parameter information, and supplementing missing standardized data fields.

[0021] In some embodiments, the generation of a budget estimate file that meets the cost preparation requirements includes: importing the mapped project section structure, quota sub-items and engineering quantity parameters into the preset software; generating a new engineering quantity table corresponding to the project section list, quota application table and cost summary table in accordance with the cost document preparation specifications to obtain the budget estimate file; performing multi-dimensional verification on the generated budget estimate file, including engineering quantity balance verification, quota application compliance verification and price calculation logic verification.

[0022] The embodiment of the present application provides a method for intelligently extracting engineering quantity applications for highway projects. By constructing an automated mapping system of "standardized engineering quantity table-budget template-dynamic connection relationship", it breaks through the existing technology's reliance on fixed formats and manual operations, and realizes intelligent data conversion based on semantic analysis and logical matching. The method includes the following core steps: engineering quantity table standardization: defining unified data fields, structures and parameter rules to form a standardized engineering quantity table, eliminating the underlying barriers of data format differences; template construction: establishing a budget project reference template containing project section structure, quota sub-items and matching rules, and building a standardized framework for cost compilation; connection relationship establishment: constructing the semantic correspondence between project sections and engineering quantity table fields, the logical matching rules between quota sub-items and engineering parameters, and the adaptive adjustment mechanism for table format differences, opening up the mapping channel between design data and cost compilation; automatic generation: when importing a new engineering quantity table, automatically mapping data through semantic analysis and logical matching to generate a budget file that meets the requirements, without the need for manual entry and item-by-item matching throughout the process.

[0023] Through automated data conversion, the cost compilation time is shortened, the time consumption of manual operations is reduced, and the timeliness requirements of multiple scheme comparisons are met; through difference analysis and semantic tag recognition, the table structure changes such as column order changes and cell merging are automatically adapted, and there is no need to redevelop the program for the table formats of different design units, and the versatility is greatly improved; based on standardized templates and semantic parsing mechanisms, parameter matching errors caused by human misunderstanding are avoided, and the data consistency and compliance of the generated budget files are significantly improved, and the traceability is stronger; a semantic and structural mapping system of design parameters and cost item sections is constructed to realize the transformation from "human experience-driven" to "data rule-driven", providing a new intelligent solution for the compilation of highway project budgets.

[0024] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0026] Figure 1 This is a schematic flow chart of the steps of a method for intelligently extracting engineering quantities for highway engineering provided by an embodiment of the present application;

[0027] Figure 2 This is a schematic block diagram of a device for intelligently extracting engineering quantities for highway engineering projects provided in one embodiment of the present application;

[0028] Figure 3 This is a schematic block diagram of the structure of a computer device provided in one embodiment of the present application.

[0029] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0031] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0032] It should be understood that the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0033] It should be further understood that the term “and / or” used in this specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0034] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0035] In the field of budget preparation for highway projects, especially when there are a large number of structures such as bridges and tunnels, the traditional method relies on manual extraction of design quantity table data, matching quotas and entering them into cost software. For example, compiling the cost of 60 tunnels takes 10 working days, and there are problems such as low efficiency, strong format dependence, and accuracy risks. Although existing technologies have attempted to achieve partial automation by extracting data through spreadsheet rules, they can only adapt to fixed formats (such as Excel column order and fixed cell structure). When the design unit adjusts the table structure (such as merging cells and changing the column order), the program needs to be redeveloped, and its versatility is insufficient. At the same time, manual understanding of the correspondence between design parameters and quota sub-items is prone to ambiguity, leading to matching errors. The core defects of existing technologies are:

[0036] 1. Data conversion relies on manual intervention: The mapping process from the design quantity table to the budget document requires a lot of manual work, which cannot meet the needs of efficient comparison and selection when design schemes are adjusted (such as changes in tunnel surrounding rock grade);

[0037] 2. Poor format compatibility: Relying on a fixed table structure, it lacks the ability to adaptively parse differences in table formats;

[0038] 3. Inaccurate semantic matching: Parameter understanding and quota matching rely on manual experience and lack a standardized semantic parsing mechanism, resulting in insufficient accuracy.

[0039] Therefore, a method is urgently needed to solve at least one of the above problems.

[0040] To resolve the above issues, please refer to Figure 1 , Figure 1 This is a flowchart illustrating the steps of a method for intelligently extracting construction quantities for highway projects, provided in one embodiment of the present application. This method can be implemented using a computer device, which can be deployed on a single server or a server cluster. Alternatively, it can be deployed on a handheld terminal, laptop computer, wearable device, or robot.

[0041] It should be noted that any information involved in the embodiments of the present application complies with relevant policies and regulations and is carried out with the authorization and permission of relevant users, and will not infringe on the privacy of users.

[0042] like Figure 1 As shown, the method for intelligently extracting engineering quantities for highway engineering specifically includes steps S101 to S104:

[0043] Step S101: Standardize the bill of quantities in highway engineering design to form a standardized bill of quantities; wherein the standardization includes defining unified data fields, data structures and parameter rules.

[0044] Specifically, by defining unified data fields, data structures and parameter rules, the original engineering quantity tables with different formats in different design units are converted into standardized data carriers with unified structure and clear semantics, eliminating parsing obstacles caused by differences in table formats (such as column order, field naming, and inconsistent units), and laying the foundation for subsequent data mapping.

[0045] For example, data field standardization can be achieved by establishing a basic field library for highway engineering, covering core parameters such as bridge span, tunnel surrounding rock grade, concrete strength grade, and engineering units (e.g., cubic meters, linear meters). Each field is assigned a unique identifier (e.g., "bridge span" corresponds to the field code GJ-001), data type (numeric / text), and field description. Semantic normalization is performed on fields in the original table. For example, different expressions such as "tunnel lining thickness" and "lining thickness (tunnel)" are uniformly mapped to the "tunnel lining thickness" field in the basic field library.

[0046] Data structure standardization can be achieved by constructing a hierarchical data structure model, including a project layer (e.g., an entire highway project), a structure layer (e.g., bridges and tunnels), a sub-project layer (e.g., bridge pile foundations and tunnel linings), and a parameter layer (e.g., pile foundation diameter and lining concrete strength). The inclusion relationships and data association rules between each layer are clearly defined. The hierarchical attribution of table cells can be standardized, for example, by using merge cell markers or field prefixes (e.g., "tunnel-lining-concrete strength") to identify the sub-project to which a parameter belongs.

[0047] Standardization of parameter rules can be achieved by formulating engineering quantity unit conversion rules, such as converting "cubic" and "m³" into the standard unit "cubic meter"; setting numerical precision constraints (such as retaining 2 decimal places for engineering quantities) and integrity verification rules (such as key parameters cannot be empty), and automatically detecting whether the original table data complies with the rules through the program, and issuing early warnings for missing or erroneous data.

[0048] By unifying fields and structures, the original table's "format-driven" parsing is transformed into "semantics-driven" parsing. Subsequent steps no longer rely on fixed column order or cell positions, significantly improving the system's compatibility with diverse table designs. The standardized fields clearly correspond to the project sections of the budget template, avoiding matching errors caused by ambiguous field names and providing underlying data support for "linguistic consistency" in automated mapping.

[0049] Step S102: Establish a budget project reference template, wherein the budget project reference template includes the project section structure, corresponding quota sub-items and matching rules required for the budget of the highway project.

[0050] Specifically, by constructing a standardized template that conforms to the specifications for the preparation of highway project estimates and budgets, including the project section hierarchy structure, corresponding quota sub-items and matching rules, a "translation" framework from design parameters to cost results is formed, replacing the traditional manual empirical operation of looking up quotas.

[0051] The project section structure is constructed based on relevant documents to establish a hierarchical project section model, for example: project layer (A expressway): structure layer (bridge engineering); sub-project layer (pile foundation engineering); quota sub-items (drilling by rotary drilling rig, diameter within 1.5m); sub-project layer (superstructure) quota sub-items (prefabrication of prestressed concrete T-beams); structure layer (tunnel engineering) sub-project layer (lining engineering); quota sub-items (C30 concrete lining, surrounding rock grade III); each project section is associated with the quota sub-items it contains, the calculation rules for engineering quantities, and the cost calculation logic (such as the quota base price, the consumption of manpower, materials, and machinery).

[0052] The quota sub-item library is constructed by including all sub-items in the current quota standards, and noting the applicable engineering parameter range for each sub-item (such as the "rotary drilling rig drilling" sub-item is applicable to hole diameters of 0.8-2.5m), required input parameters (such as drilling depth, formation type) and output attributes (such as quota number, unit price, and labor-day consumption).

[0053] The matching rules are preset, such as by establishing parameter-quota association rules, for example, when "tunnel surrounding rock grade = Grade III" and "lining concrete strength = C30", it is automatically associated with the corresponding lining quota sub-item; setting priority rules, when multiple quota sub-items meet the conditions, the optimal sub-item is selected according to parameter accuracy (such as aperture exact matching takes precedence over range matching) or project properties (such as new projects take precedence over reconstruction projects).

[0054] Templates are used to convert industry standards into computable structured data, preventing manual misapplication of quotas due to misunderstandings and improving the standardization of results. Pre-set matching rules transform "manual quota search" into "computer-automated matching." Especially when design adjustments are made (such as changes to bridge spans), there's no need to manually search for quotas again; the rule library is directly triggered to update the matching results, significantly improving the efficiency of multiple-scheme comparisons.

[0055] Step S103: Establish a connection relationship between the budget project reference template and the standardized bill of quantities, wherein the connection relationship at least includes the semantic correspondence between the project section and the bill of quantities field, the logical matching rules between the quota sub-items and the engineering parameters, and the adaptive adjustment mechanism for the differences in the table formats.

[0056] Specifically, by building a semantic bridge and logical link between design data and cost templates, including the semantic correspondence between project sections and bill of quantities fields, the logical matching rules between quota sub-items and engineering parameters, and an adaptive mechanism to cope with changes in table formats, accurate conversion from "data language" to "cost language" is achieved.

[0057] Semantic correspondences are established, for example, by developing a term mapping library for highway engineering, storing bidirectional mappings between design parameter terms and budget item terms. For example, the design term "surrounding rock grade" is mapped to the cost term "tunnel surrounding rock grade (affecting lining quota matching)"; the cost term "pile foundation length" is mapped to the design term "drilling depth (excluding virtual piles)." Natural language processing (NLP) technology is used to semantically identify non-standard terms in tables, for example, automatically mapping "beam height" to "bridge beam height (affecting superstructure quota)."

[0058] Logical matching rules are constructed by establishing a parameter matching algorithm. For example, for the "bridge span" parameter, the following rules are set: when the span is ≤20m, the "simply supported beam" quota is matched; when 20m<span≤50m, the "continuous beam" quota is matched; the sub-items are dynamically filtered based on the span value and the applicable scope of the quota; a conditional expression engine is introduced to support customized complex matching logic (such as "when the surrounding rock grade = Grade IV and the burial depth is >50m, the reinforced lining quota is enabled").

[0059] Adaptive formatting mechanisms, such as those developed through table structure parsing algorithms, automatically locate data fields by identifying semantic labels within cell content (e.g., keywords like "surrounding rock grade" and "span") rather than fixed column positions. For example, regardless of whether "Tunnel Length" is listed in the third or fifth column of a table, as long as the cell content contains the keyword "Length," it will be matched to the standardized field "Tunnel Length." For merged cells, a hierarchical recursive parsing method is used to infer data ownership based on cross-row and column information within the cell. For example, if three rows are merged and the "Bridge Name" cell's value applies to all sub-project parameters in the three rows below it.

[0060] A terminology library and NLP technology ensure accurate understanding of design parameters, avoiding issues such as "misjudgment of surrounding rock grade" and "confusion of engineering units" caused by manual interpretation, and improving matching accuracy. Through semantic tagging and structural parsing algorithms, the system automatically adapts to over 90% of table format changes (such as adjusting column order and merging cells) without manual intervention or secondary program development.

[0061] Step S104: When a new bill of quantities is imported, the data fields and structure of the new bill of quantities are automatically identified and parsed based on the connection relationship, and the data in the new bill of quantities corresponding table are mapped to the corresponding project sections and quota sub-items of the budget project reference template through semantic analysis and logical matching rules to generate a budget file that meets the cost compilation requirements.

[0062] Specifically, based on the standardized system and connection relationships established in the early stage, the newly imported engineering quantity bills are automatically parsed, mapped and results generated. There is no need for manual data entry or manual matching of quotas throughout the process, realizing end-to-end intelligent conversion from "design data" to "cost documents".

[0063] Data identification and analysis, such as through field matching: by scanning the cell contents of the new table, using the terminology library to match standardized data fields, and generating an "original field-standard field" mapping table (such as automatically mapping the "concrete strength" in the table to the standard field "concrete strength grade"); structure analysis: using a table structure tree algorithm (such as the XML tree model) to analyze the row and column merging relationship and hierarchical indentation of the table, and generate an intermediate data model that is compatible with the standardized data structure, such as converting a nested table into a hierarchical JSON object.

[0064] Data mapping and verification, such as through project node mapping: according to the structure type (bridge / tunnel) and sub-project information of the intermediate data model, automatically locate the corresponding project node in the template (such as "tunnel lining" corresponds to the "tunnel engineering-lining engineering" node in the template); quota matching: call the logical matching rule library, based on the engineering parameters (such as "tunnel surrounding rock grade = Grade III, lining thickness = 50cm"), filter the matching quota sub-items, and automatically fill in the engineering quantity and unit price parameters; consistency verification: perform multi-dimensional checks on the mapping results, including whether the engineering quantity units match (such as "cubic meter" corresponds to the volume unit required by the quota), whether the parameters are within the applicable range of the quota (such as "pile foundation diameter 1.8m" is within the 1.5-2.0m range allowed by the corresponding sub-item).

[0065] For example, the generation and output of results can be connected with mainstream cost estimation software (such as Tongwang and Glodon) through the API interface, and the mapped project section structure, quota sub-items and parameters can be automatically imported in the format required by the software (such as Excel templates, XML data files); standardized result files can be generated, including sub-item project lists, quota application tables, cost summary tables, and attached data mapping logs (recording the source and matching basis of each parameter) to facilitate later traceability and verification.

[0066] Taking the compilation of construction costs for 60 tunnels as an example, the traditional method requires 10 working days. This step is reduced to 3 working days through fully automated processing, and there is no need for repeated manual verification, which improves efficiency by more than 70%. Through standardized verification and logical rule matching, manual input errors (such as incorrect digital entry and missed quota sub-items) are avoided, and no quantity extraction errors occurred in the actual measured cases. It completely gets rid of the traditional model of "manual search-manual entry-repeated verification", realizes the seamless conversion of "design data is cost data", and provides real-time support for the comparison of multiple schemes for highway projects.

[0067] In some embodiments, the standardization of the quantity table in highway engineering design to form a standardized quantity table includes: defining unified data field names, data types and field descriptions, and establishing a unique mapping relationship between fields and engineering objects; standardizing the hierarchical relationship of the data structure, including the list level, sub-project level and specific parameter level; formulating parameter rules, including standardization rules for quantity units, numerical precision constraint rules and parameter integrity verification rules; and forming a standardized quantity table based on the unique mapping relationship, hierarchical relationship and parameter rules.

[0068] Data field standardization, for example, involves establishing a "Highway Engineering Core Data Dictionary" that defines a unique identifier for each data field (e.g., "bridge span" corresponds to ID-GL-001), field name (uniformly designated as "bridge span"), data type (numeric, unit: meter), and field description (e.g., "horizontal distance between the centers of adjacent bridge piers"), ensuring a one-to-one mapping between each field and an engineering object (e.g., bridge, tunnel) (e.g., the "tunnel surrounding rock grade" field corresponds only to tunnel structures). A field mapping verification tool is developed to scan the original table fields. Through keyword matching (e.g., fields containing the word "span" are automatically associated with "bridge span") and manual calibration interfaces, non-standard fields (e.g., "beam span length") are forcibly mapped to standard fields, forming a "original field-standard field" mapping table.

[0069] The hierarchical specification of data structure can be achieved by defining a three-level data structure: list level: corresponding to the entire highway engineering project (such as "XX Expressway Project"); sub-project level: split according to the type of structure (such as "Bridge Engineering-Pile Foundation Sub-item" and "Tunnel Engineering-Lining Sub-item"); parameter level: specific engineering parameters (such as "Pile Foundation Diameter" and "Lining Concrete Strength"), each parameter is bound to a unique sub-project level node (such as "Pile Foundation Diameter" only belongs to "Bridge Engineering-Pile Foundation Sub-item").

[0070] Use XML or JSON format to build a structured model. When the original table is converted through the plug-in, the data must be nested according to the hierarchical relationship (for example, tunnel lining parameters must be included in the "Tunnel Engineering - Lining Sub-item" tag). Parameter rule formulation: Unit standardization: Establish a "Conversion Table of Engineering Quantity Units", such as "m 3 " is automatically converted to "cubic meters" and "t" is converted to "tons". Unsupported units (such as non-standard abbreviations) trigger error prompts; numerical precision constraints: engineering quantity values ​​are forced to retain 2 decimal places (such as "100" is automatically completed to "100.00"), and the allowable error range is ±0.01; integrity check: through preset rules (such as "bridge span" and "tunnel length" are required fields), tables with missing key parameters are rejected for import and the missing field positions are marked.

[0071] Unique mapping relationships avoid field ambiguity (for example, eliminating the confusion that "lining thickness" corresponds to both bridges and tunnels). The clear three-level structure improves the efficiency of subsequent template matching, eliminating the need for manual judgment of parameter attribution. Format errors and missing data are automatically filtered, reducing the time spent on traditional manual verification and laying the foundation for error-free data for automated processes.

[0072] In some embodiments, the establishment of a budget project reference template includes: constructing a hierarchical project section structure including project sections, sub-item sections, and quota items in accordance with the highway engineering budget preparation specifications; establishing a quota sub-item library, the quota sub-item library including the engineering parameter ranges, engineering quantity calculation rules, and price components corresponding to each quota; obtaining matching rules, the matching rules including the quota sub-item screening logic based on engineering parameter thresholds, and the correlation between engineering quantity calculation rules and design parameters; establishing a budget project reference template based on the hierarchical project section structure, the quota sub-item library, and the matching rules.

[0073] The hierarchical project section structure includes a four-level template structure: project sections, such as "Part I Building and Installation Engineering - Bridge Engineering"; sub-item sections, such as "Bridge Engineering - Pile Foundation Engineering"; quota items, such as "Pile Foundation Engineering - Rotary Drilling (within 1.5m diameter)"; and parameter items, such as "Drilling Depth" and "Soil Type" (required input parameters for quota items). Each hierarchical node is associated with a compilation specification (e.g., quota items correspond to clause P32-5-1-2 of the "Budget Quota"), creating a traceable basis for compliance.

[0074] The construction of quota sub-item database is as follows: if the current quota is entered through structured input, each sub-item includes: engineering parameter range: such as "drilling diameter 0.8-2.5m", "stratum type is sand and clay"; engineering quantity calculation rules: such as "drilling depth is calculated based on the designed pile length plus overdrilling depth"; price composition: labor, material, machine-hour consumption and base price (such as "labor day 10.5 man-days / 10m 3 ”).

[0075] Matching rule acquisition includes threshold screening logic: setting parameter interval matching rules, such as "bridge span ≤ 20m → simply supported beam quota, 20m < span ≤ 50m → continuous beam quota"; association relationship definition: binding design parameters (such as "tunnel surrounding rock grade") with engineering quantity calculation rules (such as "Grade IV surrounding rock lining thickness needs to be increased by 10cm") to form a chain association of "parameter → rule → quota".

[0076] The template strictly follows industry quota standards, and the generated cost documents meet audit requirements, reducing rework caused by quota application errors; clear parameter ranges and rules enable computers to replace manual work to complete most quota screening work, and the quota matching time for complex structures (such as high-pier bridges) is shortened from 30 minutes / item to 2 minutes / item; each quota item is associated with specification clauses, which facilitates rapid verification of compliance during audits and enhances the credibility of the output documents.

[0077] The establishment of the connection relationship between the budget project reference template and the standardized engineering quantity table includes: constructing a highway engineering field terminology library, the highway engineering field terminology library stores the semantic mapping relationship between design parameter terms and budget project section terms, and is used to eliminate ambiguity between different expressions; establishing a logical matching rule library, the logical matching rule library contains the correspondence between engineering parameters and quota sub-item selection conditions; the engineering parameters include at least bridge span and tunnel surrounding rock grade; designing a table format difference analysis algorithm, the table format difference analysis algorithm automatically adapts to structural changes such as column order changes and cell merging by identifying semantic tags of cell content; and establishing the connection relationship based on the highway engineering field terminology library, the logical matching rule library and the table format difference analysis algorithm.

[0078] A bidirectional mapping table is used to store the correspondence between design and cost terms. For example: Design term: surrounding rock grade; Cost term: tunnel surrounding rock grade; Note: Standard terms are used uniformly. Design term: concrete strength C30l; Cost term: concrete strength grade C30; Note: Abbreviations are converted to full names. Design term: pile foundation length; Cost term: borehole depth (excluding virtual piles); Note: Calculation rules clarify semantics. Custom term extension is supported, allowing users to add new term mappings (such as local standard-specific terms) through the interface.

[0079] The design of the logical matching rule library includes establishing rules for core parameters (bridge span, tunnel surrounding rock grade): bridge span: "span ≤ 10m → small bridge quota, 10m < span ≤ 30m → medium bridge quota, span > 30m → large bridge quota"; tunnel surrounding rock grade: "Grade I-II → ordinary lining quota, Grade III-IV → reinforced lining quota, Grade V-VI → special support quota", combined with burial depth parameters (such as "Grade IV surrounding rock and burial depth > 50m → enable higher grade support quota").

[0080] Table format difference analysis algorithm: Develop a semantic tag scanner to match keywords in cell content (for example, scan for words like "span" and "length" to locate structural parameter fields), ignoring the impact of column order and merged cells; use a dynamic programming algorithm to parse merged cells. For example, if a "bridge name" cell spans two columns, its value is automatically filled in the corresponding parameter rows of all unmerged cells below until the next merged cell is encountered.

[0081] The terminology library eliminates most ambiguities in expressions (such as "concrete" and "concrete" are unified as standard terms), and the error rate in parameter understanding is reduced from 15% of traditional methods to below 2%; the algorithm can recognize most non-standard tables (such as random column order, merged cells and nested cells), and can be directly imported without manual adjustment of the table, with improved compatibility compared to existing technologies; multi-parameter combination matching is achieved through the logical rule library (such as the linkage selection of quotas for surrounding rock grade + burial depth), solving complex scenarios that are difficult to handle quickly by traditional manual methods.

[0082] In some embodiments, the data in the new engineering quantity correspondence table is mapped to the corresponding project section and quota sub-item of the budget project reference template through semantic analysis and logical matching rules, including: using the highway engineering field terminology library to semantically convert the parameter terms in the new engineering quantity correspondence table and match them to the standard terms of the project section; based on the logical matching rule library, conditional judgment is made on the engineering parameters corresponding to the new engineering quantity, and matching quota sub-items are automatically screened; consistency verification is performed on the mapped new engineering quantity corresponding data and the parameters required for the quota sub-items, and a parameter matching report is generated to complete the mapping of the data in the new engineering quantity correspondence table to the corresponding project section and quota sub-item of the budget project reference template.

[0083] The semantic conversion process includes scanning the parameter terms in the new table and calling the term library for two-way conversion (such as converting "surrounding rock grade IV" to "tunnel surrounding rock grade IV"). If an unlisted term is encountered, the manual review interface is triggered and it is recommended to add it to the term library; a "Semantic Conversion Log" is generated to record the original term, converted term and mapping basis of each parameter (such as "based on record 23 in the term library").

[0084] Automatic screening of quota sub-items includes inputting engineering parameters (such as "bridge span 25m"), triggering the conditional judgment of "10m<span≤30m→medium bridge quota" in the rule library, and screening out all quota sub-items that meet the conditions; if there are multiple matching sub-items (such as different pile foundation types), they are sorted according to the preset priority (such as "priority for process specified in the design drawing" and "priority for the latest quota version"), and the sub-item with the highest priority is selected by default.

[0085] Consistency verification includes parameter range verification: checking whether the "tunnel lining thickness 0.3m" is within the 0.2-0.5m range allowed by the corresponding quota. If it exceeds the range, an early warning will be marked; unit consistency verification: ensuring that the "engineering quantity unit cubic meter" is consistent with the volume unit required by the quota. If there is inconsistency, unit conversion will be automatically triggered (such as "cubic" to "cubic meter"); generating a "Parameter Matching Report" including a list of parameters for successful / failed matching and the reasons (such as "pile foundation diameter 2.6m exceeds the 2.5m upper limit allowed by the quota").

[0086] Semantic conversion ensures the consistency of parameter "language" and template. Combined with conditional judgment rules, the accuracy of quota matching is greatly improved. The verification report clearly displays matching anomalies, and the manual review time is greatly shortened. The entire process from term conversion to verification is automated, eliminating the need for manual intervention in parameter understanding and rule judgment, avoiding matching deviations caused by subjective experience.

[0087] In some embodiments, the semantic parsing process introduces knowledge graph technology to construct a relationship network between highway engineering design parameters and budget quotas, and solves the ambiguity problem of complex parameter combinations through graph reasoning algorithms to improve matching accuracy; the complex parameter combinations include quota matching for collaborative construction of multiple components.

[0088] The construction of the knowledge graph includes defining entity types: engineering parameters (such as bridge span, surrounding rock grade), quota sub-items (such as "simply supported beam prefabrication"), structures (such as bridges and tunnels), and calculation rules (such as "pile foundation engineering quantity is calculated according to the designed pile length"); establishing a relationship network: the "bridge span" entity and the "simply supported beam prefabrication" sub-item entity are connected through the "applicable span range" relationship; the "tunnel surrounding rock grade" entity and the "lining quota" sub-item entity are connected through the "support strength requirement" relationship; and multi-component collaborative construction scenarios (such as "bridge pile foundation + abutment + pier") construct sub-graphs through "construction sequence" and "dependency".

[0089] The application of graph reasoning algorithms includes traversing the association relationships through graph reasoning for complex parameter combinations (such as "large-span bridges + deep-water pile foundations + high-intensity earthquake zones"), and giving priority to matching quota sub-items that simultaneously meet multiple entity relationships (such as "deep-water pile foundation quotas + additional sub-items for seismic-resistant structures"); using path search algorithms (such as SPARQL queries) to extract implicit rules from the knowledge graph (such as "when the pile foundation diameter is greater than 2m and the water depth is greater than 10m, the special machinery unit quota must be activated") to supplement the deficiencies of the preset rule library.

[0090] It solves the problem of multi-parameter linkage matching that is difficult to cover in traditional rule libraries (such as the combination scenario where tunnel lining needs to consider surrounding rock grade, burial depth, and concrete strength at the same time), and improves the matching coverage rate; clarifies the dependency logic between parameters through graph relationships (such as the inevitable association between "span" and "bridge type"), and avoids errors caused by single parameter matching (such as only matching by span and ignoring the bridge structure form); the graph can accumulate successful matching cases of historical projects to form an industry knowledge base, so that new projects can directly inherit historical experience and reduce the cost of developing duplicate rules.

[0091] In some embodiments, the establishment of a logic matching rule library includes: optimizing the matching relationship between engineering parameters and quota sub-items through a machine learning model based on training of historical engineering data, and the machine learning model can dynamically update the matching rules according to the design scheme adjustment; the design scheme adjustment includes changes in bridge span and tunnel surrounding rock grade.

[0092] Historical data training involves collecting design parameter and quota matching records from over 1,000 highway projects in the past, and then cleaning them to form a training data set (including samples such as "bridge span = 35m → matching continuous beam quota"); using algorithms such as random forest and gradient boosting tree to train the model, with the input being a combination of engineering parameters (span, concrete strength, surrounding rock grade, etc.), and the output being the optimal quota sub-item number.

[0093] Dynamic rule updates include automatically identifying parameter changes when a design plan is adjusted (such as a bridge span changing from 30m to 40m), calling the trained algorithm to recalculate matching sub-items, and comparing the results with the preset rule base. If the difference exceeds a threshold (such as 10%), a manual review process is triggered. Incremental learning is supported, and the matching results of new projects (after manual confirmation) are automatically added to the training data. The model is updated quarterly to ensure that the rule base adapts to new industry regulations and changes in design practices.

[0094] By learning from historical data, the model can capture implicit matching rules that are difficult to summarize manually (such as the habit of some regions to enable special quotas for specific span ranges), thereby improving matching efficiency; when the design plan is adjusted, the rule base responds in real time (such as updating the matching results within 30 seconds after the span is changed), without the need for manual modification of the rules, meeting the real-time requirements of multiple plan comparisons; traditional rule bases require manual updates on a regular basis (about once every six months), while machine learning models automatically absorb new data, reducing the workload of rule maintenance.

[0095] In some embodiments, the automatic identification and parsing of the data fields and structure of the new engineering quantity table includes: scanning the cell contents of the new engineering quantity table through preset field semantic tags to match standardized data fields; based on the table structure tree algorithm, analyzing the row and column merging relationship and hierarchical indentation relationship of the table corresponding to the new engineering quantity table, generating a parsing model compatible with the standardized data structure, and completing the automatic identification and parsing of the data fields and structure of the new engineering quantity table.

[0096] Field semantic label scanning includes preset semantic labels for 100+ core parameters (such as "span", "length", "strength", and "grade"), and performs keyword matching on the cell contents of the new table. For example, cells containing the "span" label are automatically associated with the "bridge span" or "culvert span" standard fields; fuzzy matching is supported (such as matching "beam length" to the "bridge length" field), and similarity is calculated using an edit distance algorithm (such as Levenshtein distance). Automatic mapping is triggered when the threshold exceeds 80%.

[0097] The table structure tree algorithm includes converting the table into a tree structure, with each cell as a node, recording row and column merge properties (such as rowspan=2 for merging across two rows), and hierarchical indentation (determined by cell spacing or prefix symbols, such as "├─" for child nodes); and generating a JSON parsing model compatible with the standardized data structure, for example:

[0098] { ;

[0099] "Project layer": "XX Expressway",

[0100] "Structure layer": [ ;

[0101] { ;

[0102] "Name": "XX Bridge", ;

[0103] "Sub-project layer": [;

[0104] { ;

[0105] "Name": "Pile Foundation Engineering", ;

[0106] "parameterlayer": { ;

[0107] "Pile diameter": "1.8m", ;

[0108] "Drilling Depth": "50m";

[0109] };

[0110] }; ] ;

[0112] }; ] ;

[0114] };

[0115] Regardless of how the table adjusts the column order or merges cells, the system can locate fields through semantic tags and supports direct import of most non-standard tables (traditional methods only support fixed formats). The tree parsing model ensures accurate restoration of hierarchical relationships, and reduces the parsing error rate of complex nested tables (such as nested sub-projects in multiple layers). The recognition and parsing process does not require manual pre-processing of tables, and directly reads Excel, PDF (OCR conversion required) and other formats, thereby improving data import efficiency.

[0116] In some embodiments, the method of scanning the cell contents of the new engineering quantity table through preset field semantic tags and matching standardized data fields includes: using natural language processing technology to perform semantic analysis on the table title and cell contents of the new engineering quantity table, identifying implicit engineering parameter information, and supplementing missing standardized data fields.

[0117] Semantic analysis and processing performs word segmentation and part-of-speech tagging (such as "pile foundation" as a noun and "diameter" as an attribute word) on the cell header row (such as "bridge engineering quantity table") and cell content, and combines the domain dictionary to identify implicit parameters. For example, there is no clear "bridge span" field in the table, but there are "the length of the first span of the left span is 30m" and "the length of the second span of the right span is 35m". The "span length" keyword is extracted through NLP, and the missing "bridge span" field is inferred and automatically completed.

[0118] Missing fields are completed using missing field inference rules: if relevant fields such as "number of spans" and "single span length" exist, the "total length of the bridge" field is completed by calculating (number of spans × single span length); for fields that cannot be automatically completed (such as missing key parameters), a "Missing Field Suggestion Table" is generated to prompt the user to complete the fields (such as "Tunnel engineering is detected but the 'surrounding rock grade' field is missing, please confirm").

[0119] By extracting implicit information, 20%-30% of missing fields can be supplemented to avoid process interruptions caused by missing fields (traditional methods require manual line-by-line checking). The system allows for the absence of non-critical fields in design tables, and automatically infers or prompts for completion through associated fields to adapt to the common problem of incomplete data in actual projects. The workload of manual pre-processing is reduced, the availability of non-standard tables is improved, and the user threshold is significantly lowered.

[0120] In some embodiments, the generation of a budget estimate file that meets the cost preparation requirements includes: importing the mapped project section structure, quota sub-items and engineering quantity parameters into the preset software; generating a new engineering quantity table corresponding to the project section list, quota application table and cost summary table in accordance with the cost document preparation specifications to obtain the budget estimate file; performing multi-dimensional verification on the generated budget estimate file, including engineering quantity balance verification, quota application compliance verification and price calculation logic verification.

[0121] The data import preset software develops API interfaces for mainstream cost management software (such as Tongwang and Glodon), automatically filling in the mapped project section structure, quota sub-items, and engineering quantity parameters according to the software data format (such as XML interface specifications and Excel templates), and supports batch import of 100+ structural object data without lag.

[0122] The document is generated by generating three core tables according to the Methods for Preparing Budgets for Highway Projects: Project Section List: Displays the hierarchical relationship between engineering structures (e.g., "Tunnel Engineering - Lining Engineering - Concrete Lining"); Quota Application Table: Lists the quota number, name, quantity and unit price of each sub-project (e.g., "Quota 5-1-2, rotary drilling rig drilling, quantity 100m 3 Cost Summary Sheet: Summarizes direct costs, indirect costs, profits, etc. by chapter and automatically calculates the total cost.

[0123] Multi-dimensional verification includes engineering quantity balance verification: checking whether the engineering quantity of the sub-project is equal to the sum of each sub-item (such as whether the "total length of the bridge pile foundation" is equal to the sum of the lengths of each pile). An error of more than 1% will trigger an early warning; quota application compliance verification: verifying whether the matching sub-item is within the validity period of the current quota, and whether the parameter range complies with the quota description (such as "pile foundation diameter of 2.6m uses the quota within 2.5m" is marked as an error); price calculation logic verification: verifying through the built-in formula "comprehensive unit price = labor, material and machine consumption × market price + management fee + profit" to ensure that there are no logical errors in the calculation.

[0124] The generated files comply with industry compilation standards and can be directly used for bidding or auditing after verification, reducing the time for format adjustment; multi-dimensional verification covers common errors (such as errors in total engineering quantities and application of expired quotas), and one-stop processing from data import to file generation and verification forms a complete automated chain from "inputting design tables to outputting deliverable cost documents", completely getting rid of the tedious process of manual compilation.

[0125] In some embodiments, this application designs a dynamic field mapping model that integrates engineering domain knowledge and uses an improved cosine similarity formula to achieve intelligent matching of non-standard fields to standard libraries, thereby solving the problems of semantic ambiguity and domain differences in traditional rule matching. The improved cosine similarity formula includes:

[0126] ;

[0127] Here, fi is a non-standard bill of quantities field to be standardized (such as "beam height", "pile foundation length", and "C30 concrete"). It may contain abbreviations of engineering parameters, local idioms, or non-standard expressions. It is a string type, usually 2-20 characters long, and supports numbers, unit symbols (such as "m" and "cm"), and engineering abbreviations (such as "concrete" for concrete).

[0128] Fj is a target field from a standard library (such as "Bridge Beam Height," "Concrete Pile Length," and "C30 Concrete") that complies with industry standards. It is a string type and comes from a pre-built standard field library.

[0129] vi, vj are domain-specific word embeddings for fi and Fj, generated through pre-training of the engineering domain BERT model to capture the semantic characteristics of the fields (e.g., the correlation between "pile foundation" and "cast-in-place pile"). vi, vj are 1024-dimensional real number vectors (consistent with the BERT hidden layer dimensions), with each dimension representing a semantic feature (e.g., dimension 300 corresponds to "structure type" and dimension 600 corresponds to "parameter attribute"). Training was performed using highway engineering tabular data, and engineering entity tags (e.g., [parameter name], [unit]) were added to the word embedding layer to strengthen the semantic distinction of domain terms (e.g., the vector distance between "span" and "span" is 30% smaller than that of the general model).

[0130] c(fi,Fj) is the number of times fi and Fj appear together in the same bill of quantities in historical data, reflecting frequent mapping practices within the field (e.g., "pile diameter" often corresponds to "pile foundation diameter"). The corresponding value is an integer in the range [0,10] (counts exceeding 10 are rounded to 10 to avoid the influence of extreme data). Each time a matching pair is manually confirmed (e.g., mapping "pier height" to "bridge pier height"), c(fi,Fj) += 1, forming a two-way optimization process of "data-driven + manual calibration."

[0131] β(di) adjusts the similarity weight based on the data type of fi to address matching difficulties caused by differences in units and symbols for numeric parameters (e.g., "30cm" vs. "0.3m," "C30" vs. "30# concrete"). If fi contains both a numeric value and a unit (e.g., "15m," "C50"), it is considered a numeric parameter and β=1.2 (increasing the matching priority, as numeric parameters are more critical to cost calculations). If it is a plain text description (e.g., "reinforcement mesh," "mortared stone"), it is considered a text parameter and β=1.0 (default weight). Regular expressions are used to detect the presence of numbers and unit symbols (e.g., m, cm, C, %).

[0132] vi vj is the dot product of the word vectors, measuring the semantic similarity between fi and Fj (larger values ​​indicate smaller vector angles and closer semantics). It is a real number in the range [−1024, 1024] (because the modulus of vectors approaches 1 after normalization, the actual value is more concentrated in the range [−500, 500]).

[0133] ∥vi∥∥vj∥ is the modulus-length product of the word vector, which is used to normalize the dot product and convert it into the denominator of the cosine similarity (range [0, 10242], and is actually close to 1 after layer normalization).

[0134] ϵ is a smoothing factor to avoid calculation errors caused by the denominator being 0 (for example, in extreme cases, all word vectors are 0). The corresponding value is a very small positive number 10 −8 (Safety values ​​are commonly used in engineering calculations and their impact on the results can be ignored).

[0135] The word vectors for the engineering domain are constructed by training a domain-specific BERT model using historical engineering quantity tables and introducing engineering entity labels (such as [parameter], [unit], and [structure]) in the word embedding layer, so that the vector distance between "pile foundation diameter" and "pile diameter" is less than 20% of that of the non-domain model.

[0136] Dynamic mapping rule generation involves calculating S(fi,Fj) for the input field fi and all Fj in the standard library, selecting the top three matching results with S>0.7, and adding unit normalization verification for numeric fields (for example, after "30cm" is automatically converted to "0.3m", the matching priority is increased by combining β(di)=1.2 in the formula).

[0137] The incremental learning mechanism includes manually reviewing and correcting matching pairs (such as mapping "pier spacing" to "bridge pier center distance"), updating the co-occurrence coefficient according to c(fi,Fj)+=1, and retraining the word vector bias item after every 500 corrections.

[0138] Through the domain co-occurrence coefficient α⋅c(fi,Fj), the correct matching rate of regional difference terms such as "platform cap" and "cap beam" is improved; using β(di), the unit / symbol expressions such as "concrete strength C30" and "No. 30 concrete" are intelligently normalized to improve matching efficiency; there is no need to manually write mapping tables, and the field adaptation time of new project types is shortened to.

[0139] See also Figure 2 , Figure 2 2 is a schematic diagram of the structure of an apparatus 200 for intelligently extracting engineering quantities for highway projects, provided in an embodiment of the present application. The apparatus 200 is configured to execute the steps of the methods for intelligently extracting engineering quantities for highway projects described in the aforementioned embodiments. The apparatus 200 can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, a laptop computer, a wearable device, or a robot.

[0140] like Figure 2 As shown, the device 200 for intelligently extracting engineering quantities for highway engineering includes:

[0141] The standard processing unit 201 is used to standardize the engineering quantity table in highway engineering design to form a standardized engineering quantity table; wherein the standardization processing includes defining unified data fields, data structures and parameter rules;

[0142] The template establishment unit 202 is used to establish a budget project reference template, wherein the budget project reference template includes the project section structure, corresponding quota sub-items and matching rules required for the budget of the highway project;

[0143] A connection establishing unit 203 is configured to establish a connection relationship between the budget project reference template and the standardized bill of quantities, wherein the connection relationship includes at least a semantic correspondence between project sections and bill of quantities fields, a logical matching rule between quota sub-items and project parameters, and an adaptive adjustment mechanism for differences in table formats;

[0144] The file generation unit 204 is used to automatically identify and parse the data fields and structure of the new engineering quantity table based on the connection relationship when importing a new engineering quantity table, and map the data in the new engineering quantity corresponding table to the corresponding project section and quota sub-item of the budget project reference template through semantic analysis and logical matching rules, so as to generate a budget file that meets the cost compilation requirements.

[0145] In some embodiments, the standardization of the quantity table in highway engineering design to form a standardized quantity table includes: defining unified data field names, data types and field descriptions, and establishing a unique mapping relationship between fields and engineering objects; standardizing the hierarchical relationship of the data structure, including the list level, sub-project level and specific parameter level; formulating parameter rules, including standardization rules for quantity units, numerical precision constraint rules and parameter integrity verification rules; and forming a standardized quantity table based on the unique mapping relationship, hierarchical relationship and parameter rules.

[0146] In some embodiments, the establishment of a budget project reference template includes: constructing a hierarchical project section structure including project sections, sub-item sections, and quota items in accordance with the highway engineering budget preparation specifications; establishing a quota sub-item library, the quota sub-item library including the engineering parameter ranges, engineering quantity calculation rules, and price components corresponding to each quota; obtaining matching rules, the matching rules including the quota sub-item screening logic based on engineering parameter thresholds, and the correlation between engineering quantity calculation rules and design parameters; establishing a budget project reference template based on the hierarchical project section structure, the quota sub-item library, and the matching rules.

[0147] In some embodiments, establishing a connection relationship between the budget project reference template and the standardized engineering quantity table includes: constructing a highway engineering field terminology library, the highway engineering field terminology library stores the semantic mapping relationship between design parameter terms and budget project section terms, for eliminating ambiguity between different expressions; establishing a logical matching rule library, the logical matching rule library contains the correspondence between engineering parameters and quota sub-item selection conditions; the engineering parameters include at least bridge span and tunnel surrounding rock grade; designing a table format difference analysis algorithm, the table format difference analysis algorithm automatically adapts to structural changes such as column order changes and cell merging by identifying semantic tags of cell contents; establishing the connection relationship based on the highway engineering field terminology library, the logical matching rule library and the table format difference analysis algorithm.

[0148] In some embodiments, the data in the new engineering quantity correspondence table is mapped to the corresponding project section and quota sub-item of the budget project reference template through semantic analysis and logical matching rules, including: using the highway engineering field terminology library to semantically convert the parameter terms in the new engineering quantity correspondence table and match them to the standard terms of the project section; based on the logical matching rule library, conditional judgment is made on the engineering parameters corresponding to the new engineering quantity, and matching quota sub-items are automatically screened; consistency verification is performed on the mapped new engineering quantity corresponding data and the parameters required for the quota sub-items, and a parameter matching report is generated to complete the mapping of the data in the new engineering quantity correspondence table to the corresponding project section and quota sub-item of the budget project reference template.

[0149] In some embodiments, the semantic parsing process introduces knowledge graph technology to construct a relationship network between highway engineering design parameters and budget quotas, and solves the ambiguity problem of complex parameter combinations through graph reasoning algorithms to improve matching accuracy; the complex parameter combinations include quota matching for collaborative construction of multiple components.

[0150] In some embodiments, the establishment of a logic matching rule library includes: optimizing the matching relationship between engineering parameters and quota sub-items through a machine learning model based on training of historical engineering data, and the machine learning model can dynamically update the matching rules according to the design scheme adjustment; the design scheme adjustment includes changes in bridge span and tunnel surrounding rock grade.

[0151] In some embodiments, the automatic identification and parsing of the data fields and structure of the new engineering quantity table includes: scanning the cell contents of the new engineering quantity table through preset field semantic tags to match standardized data fields; based on the table structure tree algorithm, analyzing the row and column merging relationship and hierarchical indentation relationship of the table corresponding to the new engineering quantity table, generating a parsing model compatible with the standardized data structure, and completing the automatic identification and parsing of the data fields and structure of the new engineering quantity table.

[0152] In some embodiments, the method of scanning the cell contents of the new engineering quantity table through preset field semantic tags and matching standardized data fields includes: using natural language processing technology to perform semantic analysis on the table title and cell contents of the new engineering quantity table, identifying implicit engineering parameter information, and supplementing missing standardized data fields.

[0153] In some embodiments, the generation of a budget estimate file that meets the cost preparation requirements includes: importing the mapped project section structure, quota sub-items and engineering quantity parameters into the preset software; generating a new engineering quantity table corresponding to the project section list, quota application table and cost summary table in accordance with the cost document preparation specifications to obtain the budget estimate file; performing multi-dimensional verification on the generated budget estimate file, including engineering quantity balance verification, quota application compliance verification and price calculation logic verification.

[0154] It should be noted that those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0155] The above-mentioned device can be realized in the form of a computer program. The computer program can be used in Figure 3 Runs on the computer equipment shown.

[0156] See also Figure 3 , Figure 3FIG. 1 is a schematic block diagram of a computer device in one embodiment. The computer device may be a server.

[0157] See Figure 3 The computer device includes a processor, a memory, and a network interface connected through a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0158] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, which, when executed, can cause a processor to execute any one of the methods for intelligently extracting engineering quantities for highway engineering.

[0159] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0160] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any method for intelligent extraction of engineering quantity applications in highway engineering.

[0161] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0162] It should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0163] In one embodiment, the processor is configured to execute a computer program stored in the memory to implement the following steps:

[0164] Standardizing the engineering quantity table in highway engineering design to form a standardized engineering quantity table; wherein the standardization includes defining unified data fields, data structures and parameter rules;

[0165] Establishing a budget project reference template, wherein the budget project reference template includes the project section structure, corresponding quota sub-items and matching rules required for the budget of the highway project;

[0166] Establishing a connection relationship between the budget project reference template and the standardized bill of quantities, wherein the connection relationship includes at least a semantic correspondence between project sections and bill of quantities fields, a logical matching rule between quota sub-items and engineering parameters, and an adaptive adjustment mechanism for differences in table formats;

[0167] When a new bill of quantities is imported, the data fields and structure of the new bill of quantities are automatically identified and parsed based on the connection relationship. Through semantic analysis and logical matching rules, the data in the new bill of quantities corresponding table are mapped to the corresponding project sections and quota sub-items of the budget project reference template to generate a budget file that meets the cost compilation requirements.

[0168] In some embodiments, the standardization of the quantity table in highway engineering design to form a standardized quantity table includes: defining unified data field names, data types and field descriptions, and establishing a unique mapping relationship between fields and engineering objects; standardizing the hierarchical relationship of the data structure, including the list level, sub-project level and specific parameter level; formulating parameter rules, including standardization rules for quantity units, numerical precision constraint rules and parameter integrity verification rules; and forming a standardized quantity table based on the unique mapping relationship, hierarchical relationship and parameter rules.

[0169] In some embodiments, the establishment of a budget project reference template includes: constructing a hierarchical project section structure including project sections, sub-item sections, and quota items in accordance with the highway engineering budget preparation specifications; establishing a quota sub-item library, the quota sub-item library including the engineering parameter ranges, engineering quantity calculation rules, and price components corresponding to each quota; obtaining matching rules, the matching rules including the quota sub-item screening logic based on engineering parameter thresholds, and the correlation between engineering quantity calculation rules and design parameters; establishing a budget project reference template based on the hierarchical project section structure, the quota sub-item library, and the matching rules.

[0170] In some embodiments, establishing a connection relationship between the budget project reference template and the standardized engineering quantity table includes: constructing a highway engineering field terminology library, the highway engineering field terminology library stores the semantic mapping relationship between design parameter terms and budget project section terms, for eliminating ambiguity between different expressions; establishing a logical matching rule library, the logical matching rule library contains the correspondence between engineering parameters and quota sub-item selection conditions; the engineering parameters include at least bridge span and tunnel surrounding rock grade; designing a table format difference analysis algorithm, the table format difference analysis algorithm automatically adapts to structural changes such as column order changes and cell merging by identifying semantic tags of cell contents; establishing the connection relationship based on the highway engineering field terminology library, the logical matching rule library and the table format difference analysis algorithm.

[0171] In some embodiments, the data in the new engineering quantity correspondence table is mapped to the corresponding project section and quota sub-item of the budget project reference template through semantic analysis and logical matching rules, including: using the highway engineering field terminology library to semantically convert the parameter terms in the new engineering quantity correspondence table and match them to the standard terms of the project section; based on the logical matching rule library, conditional judgment is made on the engineering parameters corresponding to the new engineering quantity, and matching quota sub-items are automatically screened; consistency verification is performed on the mapped new engineering quantity corresponding data and the parameters required for the quota sub-items, and a parameter matching report is generated to complete the mapping of the data in the new engineering quantity correspondence table to the corresponding project section and quota sub-item of the budget project reference template.

[0172] In some embodiments, the semantic parsing process introduces knowledge graph technology to construct a relationship network between highway engineering design parameters and budget quotas, and solves the ambiguity problem of complex parameter combinations through graph reasoning algorithms to improve matching accuracy; the complex parameter combinations include quota matching for collaborative construction of multiple components.

[0173] In some embodiments, the establishment of a logic matching rule library includes: optimizing the matching relationship between engineering parameters and quota sub-items through a machine learning model based on training of historical engineering data, and the machine learning model can dynamically update the matching rules according to the design scheme adjustment; the design scheme adjustment includes changes in bridge span and tunnel surrounding rock grade.

[0174] In some embodiments, the automatic identification and parsing of the data fields and structure of the new engineering quantity table includes: scanning the cell contents of the new engineering quantity table through preset field semantic tags to match standardized data fields; based on the table structure tree algorithm, analyzing the row and column merging relationship and hierarchical indentation relationship of the table corresponding to the new engineering quantity table, generating a parsing model compatible with the standardized data structure, and completing the automatic identification and parsing of the data fields and structure of the new engineering quantity table.

[0175] In some embodiments, the method of scanning the cell contents of the new engineering quantity table through preset field semantic tags and matching standardized data fields includes: using natural language processing technology to perform semantic analysis on the table title and cell contents of the new engineering quantity table, identifying implicit engineering parameter information, and supplementing missing standardized data fields.

[0176] In some embodiments, the generation of a budget estimate file that meets the cost preparation requirements includes: importing the mapped project section structure, quota sub-items and engineering quantity parameters into the preset software; generating a new engineering quantity table corresponding to the project section list, quota application table and cost summary table in accordance with the cost document preparation specifications to obtain the budget estimate file; performing multi-dimensional verification on the generated budget estimate file, including engineering quantity balance verification, quota application compliance verification and price calculation logic verification.

[0177] A computer-readable storage medium is also provided in an embodiment of the present application, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and the processor executes the program instructions to implement any method for intelligent extraction of engineering quantities for highway projects provided in an embodiment of the present application.

[0178] The computer-readable storage medium may be an internal storage unit of the computer device described in the aforementioned embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a SmartMedia Card (SMC), a Secure Digital (SD) card, a flash memory card, etc., equipped on the computer device.

[0179] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A method for intelligently extracting engineering quantities for highway engineering, characterized in that: The method comprises: Standardizing the engineering quantity table in highway engineering design to form a standardized engineering quantity table includes: defining unified data field names, data types, and field descriptions, and establishing a unique mapping relationship between fields and engineering objects; standardizing the hierarchical relationship of the data structure, including the list level, sub-project level, and specific parameter level; formulating parameter rules, including engineering quantity unit standardization rules, numerical precision constraint rules, and parameter integrity verification rules; forming a standardized engineering quantity table based on the unique mapping relationship, hierarchical relationship, and parameter rules; wherein the standardization includes defining unified data fields, data structures, and parameter rules; Establishing a budget project reference template, wherein the budget project reference template includes the project section structure, corresponding quota sub-items and matching rules required for the budget of the highway project; Establishing a connection relationship between the budget project reference template and the standardized bill of quantities, wherein the connection relationship includes at least a semantic correspondence between project sections and bill of quantities fields, a logical matching rule between quota sub-items and engineering parameters, and an adaptive adjustment mechanism for differences in table formats; When a new bill of quantities is imported, the data fields and structure of the new bill of quantities are automatically identified and parsed based on the connection relationship. Through semantic analysis and logical matching rules, the data in the new bill of quantities corresponding table are mapped to the corresponding project sections and quota sub-items of the budget project reference template to generate a budget file that meets the cost compilation requirements.

2. The method according to claim 1, characterized in that The reference template for establishing a budget project includes: According to the highway project budget compilation specifications, a hierarchical project section structure including project sections, sub-item sections, and quota items is constructed; Establishing a quota sub-item database, wherein the quota sub-item database includes the engineering parameter range, engineering quantity calculation rules and price components corresponding to each quota; Obtaining matching rules, wherein the matching rules include quota sub-item screening logic based on engineering parameter thresholds, and an association relationship between engineering quantity calculation rules and design parameters; A reference template for a budget project is established based on the hierarchical project section structure, quota sub-item library and matching rules.

3. The method according to claim 1, characterized in that The establishing of a connection relationship between the budget project reference template and the standardized bill of quantities includes: Constructing a highway engineering terminology database, wherein the highway engineering terminology database stores semantic mapping relationships between design parameter terms and budget item section terms, so as to eliminate ambiguity between different expressions; Establishing a logic matching rule library, wherein the logic matching rule library contains the corresponding relationship between engineering parameters and quota sub-item selection conditions; the engineering parameters include at least the bridge span and the tunnel surrounding rock grade; Designing a table format difference analysis algorithm that automatically adapts to structural changes such as column order changes and cell merging by identifying semantic tags of cell content; The connection relationship is established based on the highway engineering field terminology library, the logic matching rule library and the table format difference analysis algorithm.

4. The method according to claim 3, characterized in that The mapping of the data in the new engineering quantity corresponding table to the corresponding project sections and quota sub-items of the budget project reference template through semantic analysis and logical matching rules includes: Using the highway engineering field terminology library, semantic conversion is performed on parameter terms in the new engineering quantity correspondence table to match them with standard terms in the project section; Based on the logic matching rule library, conditional judgment is performed on the engineering parameters corresponding to the new engineering quantity, and matching quota sub-items are automatically screened; Perform consistency check on the new engineering quantity corresponding data after mapping and the parameters required for the quota sub-items, generate a parameter matching report, and complete the mapping of the data in the new engineering quantity corresponding table to the corresponding project sections and quota sub-items of the budget project reference template.

5. The method according to claim 4, characterized in that The semantic parsing process introduces knowledge graph technology to construct a relationship network between highway engineering design parameters and budget quotas, and resolves the ambiguity of complex parameter combinations through graph reasoning algorithms to improve matching accuracy. The complex parameter combination includes quota matching for collaborative construction of multiple components.

6. The method according to claim 3, characterized in that The establishing of a logic matching rule base includes: Based on training of historical engineering data, the matching relationship between engineering parameters and quota sub-items is optimized through a machine learning model. The machine learning model can dynamically update the matching rules according to the design scheme adjustment; the design scheme adjustment includes changes in bridge span and tunnel surrounding rock grade.

7. The method according to claim 1, characterized in that The automatic identification and parsing of the data fields and structure of the new bill of quantities includes: Scan the cell contents of the new engineering quantity table using preset field semantic tags to match standardized data fields; Based on the table structure tree algorithm, the row and column merging relationship and hierarchical indentation relationship of the new engineering quantity table corresponding to the table are analyzed, and a parsing model compatible with the standardized data structure is generated to complete the automatic identification and parsing of the data fields and structure of the new engineering quantity table.

8. The method according to claim 7, characterized in that Scanning the cell contents of the new engineering quantity table using preset field semantic tags to match standardized data fields includes: Natural language processing technology is used to perform semantic analysis on the table titles and cell contents of the new engineering quantity table to identify implicit engineering parameter information and supplement missing standardized data fields.

9. The method according to claim 1, characterized in that The generation of a budget estimate document that meets the cost compilation requirements includes: Import the mapped project section structure, quota sub-items and engineering quantity parameters into the preset software; According to the cost document compilation specifications, generate a new bill of quantities corresponding to the project section list, quota application table and cost summary table to obtain the budget document; The generated budget estimate documents are verified in multiple dimensions, including quantity balance verification, quota application compliance verification and price calculation logic verification.

Images