A method for auxiliary review of the integrity of power distribution network feasibility study projects

By performing multimodal parsing and structured processing of power distribution network feasibility study project documents, a review rule base was constructed and cross-document data consistency verification was performed. This solved the problems of low efficiency and inconsistent standards in the review of power distribution network feasibility study projects, realized an automated and standardized review process, and improved the quality and scientific nature of the review.

CN123118031APending Publication Date: 2026-11-17GUANGDONG POWER GRID CO LTD DONGGUAN POWER SUPPLY BUREAU
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611447850.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-17
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The current review efficiency of power distribution network feasibility study projects is low, the review standards are inconsistent, the cross-document consistency verification is weak, and there is a lack of quantitative evaluation methods, which makes it difficult to guarantee the quality of the review.

Method used

Multimodal parsing and structured processing technologies are used to uniformly process heterogeneous files, build a review rule base, and perform automated review of chapters, fields and formats. A three-layer matching framework of logical association analysis and semantic-element-evaluation is used to perform cross-file data consistency verification and problem matching evaluation.

Benefits of technology

It has enabled automated and standardized review of power distribution network feasibility study projects, improved review efficiency and accuracy, reduced human subjectivity, enhanced the comprehensiveness and scientific nature of the review, and ensured the scientific nature of investment decisions and the safety of project implementation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123118031A_ABST
    Figure CN123118031A_ABST
Patent Text Reader

Abstract

This invention relates to the field of distribution network planning technology and discloses a method for assisting in the integrity review of distribution network feasibility study projects. By using multimodal parsing and structured processing of distribution network feasibility study project file packages, it achieves unified processing and automatic review of heterogeneous files, significantly improving efficiency. Based on natural language processing, it extracts review specifications and constructs a rule base, transforming institutional rules into executable machine logic and unifying review standards. Dependency parsing is used to perform logical association analysis on cross-file data, automatically comparing key data and solving the problem of omissions in manual comparison. A three-layer matching framework of semantics, elements, and evaluation is constructed to perform semantic parsing, element extraction, and dynamic weighted evaluation on the proposed problem and alternative projects, achieving quantitative evaluation of problem-project matching. A review report containing problem location, regulatory basis, and modification suggestions is generated, providing direct decision support and comprehensively improving the efficiency, objectivity, and scientific nature of distribution network feasibility study reviews, ensuring scientific investment decisions and engineering safety.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of power distribution network planning technology, and in particular to a method for assisting in the review of the integrity of power distribution network feasibility study projects. Background Technology

[0002] The feasibility study project for the distribution network (hereinafter referred to as the "distribution network feasibility study project") is a crucial preliminary stage in the planning and construction of the distribution network. The quality of its feasibility study report and supporting documents directly affects the scientific nature of investment decisions and the safety of project implementation. According to the relevant management regulations of power grid companies, the distribution network feasibility study project document package typically includes various heterogeneous documents such as the project proposal, feasibility study report (Word document), investment estimate and equipment and material list (Excel spreadsheet), relevant approval documents and drawing specifications (PDF file).

[0003] Currently, the review of feasibility studies for power distribution network projects mainly relies on manual work, which has the following shortcomings: (1) Low review efficiency: Feasibility study documents are long and numerous. Reviewers need to check the chapter settings, field filling and format specifications page by page, which is a huge workload. Moreover, the review standards vary from person to person, which is highly subjective and inconsistent.

[0004] (2) Difficulty in obtaining review rules: The review norms are mostly institutional documents described in natural language. The rules are scattered and varied in expression, making it difficult to directly convert them into executable machine review logic.

[0005] (3) Weak cross-document consistency verification: Key data such as project scale, investment amount, and equipment quantity need to be consistent in different documents (such as feasibility study report, estimation table, and inventory list). Manual comparison is very easy to miss, while existing text review tools are mostly for single documents and lack the ability to analyze cross-document logical relationships.

[0006] (4) Problem-project matching assessment lacks quantitative means: In the feasibility study review, it is necessary to determine the degree of matching between the problem to be solved by the project and the alternative solutions (such as new construction, renovation, expansion and other alternative projects). Existing methods mostly rely on expert experience for qualitative judgment and lack a systematic and quantifiable evaluation framework.

[0007] Therefore, there is an urgent need for an automated, standardized, and intelligent method to assist in the review of the integrity of power distribution network feasibility study projects, so as to improve the review efficiency and quality. Summary of the Invention

[0008] This invention provides a method for assisting in the review of the integrity of power distribution network feasibility study projects, in order to solve the problems existing in the prior art.

[0009] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for assisting in the integrity review of a power distribution network feasibility study project, comprising the following steps: S1. Obtain the power distribution network feasibility study project file package to be reviewed, perform multimodal parsing and structuring on the heterogeneous files in the file package to obtain structured text in a unified format. The heterogeneous files include Word documents, Excel spreadsheets and PDF files. S2. Based on natural language processing technology, extract rules from the feasibility study review specification text of the power distribution network and construct a review rule base; S3. Perform a completeness review on the structured text based on the review rule base. The completeness review includes chapter completeness review, field completeness review, and format conformity review. S4. Use dependency resolution technology to perform logical association analysis on cross-file data in the structured text and verify the consistency of cross-file data; S5. Construct a three-layer matching framework of semantics-element-evaluation, perform semantic parsing, element extraction and dynamic weighted evaluation on the problem to be solved by the feasibility study project and the alternative projects, and calculate the matching degree between the problem and the alternative projects. S6. Based on the results of integrity review, consistency verification and matching assessment, generate a review report, which includes problem identification, standard basis and modification suggestions.

[0010] Furthermore, in the method for assisting in the integrity review of the power distribution network feasibility study project, step S1 includes: S11. Obtain the project feasibility study file package for the distribution network to be reviewed; S12. Decompress the file package and identify the file types of heterogeneous files within the file package; S13. For PDF files, determine whether they are scanned documents. If they are scanned documents, use OCR technology to recognize text; otherwise, extract the text directly. For Word documents, perform document structure analysis to extract chapter titles, paragraph levels, and table row and column structures. For Excel tables, perform header recognition and key field location. S14. The extracted content is uniformly converted into structured text containing location tags, which are used to identify the file, chapter and field to which the content belongs.

[0011] Furthermore, in the method for assisting in the integrity review of the power distribution network feasibility study project, step S2 includes: S21. Perform word segmentation, part-of-speech tagging, named entity recognition, and dependency parsing on the feasibility study review specification text of the power distribution network. S22. Identify the review objects, review elements, threshold conditions and qualification criteria in the power distribution network feasibility study review specification text, and extract them into review rules in the form of object-element-condition-criteria quadruples; S23. Store the review rules in the review rule base and establish a reference mapping relationship between the review rules and the normative clauses.

[0012] Furthermore, in the method for assisting in the integrity review of the power distribution network feasibility study project, in step S3, the chapter integrity review is as follows: comparing the chapter structure of the feasibility study report with the standard chapter template to identify missing and redundant chapters; The field integrity review is as follows: verifying whether the essential fields are complete. The essential fields include a method for auxiliary review of the integrity of a power distribution network feasibility study project, construction location, voltage level, construction scale, total investment and sub-item investment. The format compliance review involves checking whether the data format of the fields conforms to the specifications, including numerical units, numbering formats, and date formats.

[0013] Furthermore, in the method for assisting in the integrity review of the power distribution network feasibility study project, step S4 includes: S41. Extract cross-file data fields from the structured text, and construct a cross-file data dependency graph with data fields as nodes and logical relationships between fields as edges. The logical relationships include summation relationships, equality relationships, range constraint relationships, and reference relationships. S42. Verify the values ​​of the corresponding fields along the dependency edges of the data dependency graph to identify data contradictions; wherein the verification includes at least: matching verification of total investment and sum of sub-investments, matching verification of equipment and material list and investment estimate, and matching verification of engineering quantity and drawing parameters.

[0014] Furthermore, in the method for assisting in the integrity review of the power distribution network feasibility study project, in step S5, the semantic-element-evaluation three-layer matching framework includes: Semantic layer: Semantic parsing is performed on the problem description text and the project description text respectively to generate semantic vectors and calculate semantic similarity; Element layer: Extract key elements from the problem description text and project description text. The key elements include region, voltage level, capacity, investment and time limit. Match the corresponding key elements one by one and calculate the element matching degree. Evaluation layer: Using the semantic similarity, the element matching degree, and the planning scene features as state features, the evaluation weights of the semantic layer and the element layer are dynamically optimized using a reinforcement learning model, and the comprehensive matching degree between the problem and the alternative options is output.

[0015] Furthermore, in the method for assisting in the integrity review of the power distribution network feasibility study project, the step of dynamically optimizing the evaluation weights of the semantic layer and the element layer using a reinforcement learning model includes: The evaluation weights are used as actions, and the consistency between the evaluation results and the conclusions of manual review is used as a reward signal. The evaluation weights are updated iteratively through strategies. The evaluation weights are also adaptively adjusted according to the planning scenario type, which includes new scenarios, renovation scenarios, and expansion scenarios.

[0016] Furthermore, in the method for assisting in the integrity review of the power distribution network feasibility study project, in step S6, the review report classifies and sorts the problems found according to their severity, and marks the corresponding normative clauses for each problem; the review report can be exported as a PDF or Excel file.

[0017] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the auxiliary review method for the integrity of the power distribution network feasibility study project as provided in the first aspect above.

[0018] Thirdly, the present invention provides a computer-readable storage medium having computer-executable instructions stored thereon, the computer-executable instructions being executed by a computer processor to implement the auxiliary review method for the integrity of power distribution network feasibility study projects as provided in the first aspect above.

[0019] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a method for assisting in the integrity review of power distribution network feasibility study projects. By performing multimodal parsing and structured processing on the project file package, it achieves unified processing and automated machine review of various heterogeneous files such as Word, Excel, and PDF, avoiding manual page-by-page verification and significantly improving review efficiency. Based on natural language processing technology, it extracts rules from the review specification text and constructs a rule base, transforming scattered, naturally language-described institutional rules into executable machine review logic. This achieves unified and standardized review standards, eliminating the subjectivity of human review, which varies from person to person and has inconsistent standards. Finally, it utilizes dependency parsing technology to perform logical association analysis on cross-file data. It can automatically compare the consistency of key data across different documents, effectively solving the problem of easy omissions in manual comparison and enhancing the comprehensiveness and accuracy of the review. Furthermore, by constructing a three-layer matching framework of semantics-element-evaluation, it performs semantic parsing, element extraction, and dynamic weighted evaluation of the proposed problem and alternative projects, realizing a systematic and quantifiable evaluation of the degree of matching between the problem and the project, reducing the reliance on expert experience in the evaluation. The final review report includes problem identification, normative basis, and modification suggestions, providing reviewers with directly usable decision support, and improving the efficiency, objectivity, and scientific nature of the review of power distribution network feasibility study projects, ensuring the scientific nature of investment decisions and the safety of project implementation. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating a method for assisting in the review of the integrity of a power distribution network feasibility study project, provided in Embodiment 1 of the present invention. Figure 2 This is a schematic diagram of the structure of a computer device provided in Embodiment 2 of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Example 1

[0024] Please refer to Figure 1 This is a flowchart illustrating a method for assisting in the integrity review of a power distribution network feasibility study project, provided in Embodiment 1 of the present invention. This method can be implemented using software and / or hardware. The method specifically includes the following steps: S1. Obtain the power distribution network feasibility study project file package to be reviewed, perform multimodal parsing and structuring on the heterogeneous files in the file package to obtain structured text in a unified format. The heterogeneous files include Word documents, Excel spreadsheets and PDF files. It should be noted that the heterogeneous files refer to various files in the power distribution network feasibility study project file package that have different file formats, data organization forms, and content presentation forms, specifically including Word documents, Excel spreadsheets, and PDF files.

[0025] Preferably, in this embodiment, step S1 is further refined into the following sub-steps: S11. Obtain the project feasibility study file package for the distribution network to be reviewed; Specifically, the power distribution network feasibility study project file package submitted by the project applicant and organized in compressed or other packaged forms can be obtained through file upload, directory scanning or interface call. The file package usually includes supporting documents such as project proposal, feasibility study report, investment estimate and equipment and material list.

[0026] S12. Decompress the file package and identify the file types of heterogeneous files within the file package; Specifically, files within a file package can be identified and categorized as Word documents, Excel spreadsheets, PDF files, or other file types based on file extensions, file header identification information, or a combination thereof, so that appropriate parsing strategies can be invoked for different file types.

[0027] S13. For PDF files, determine whether they are scanned documents. If they are scanned documents, use OCR technology to recognize text; otherwise, extract the text directly. For Word documents, perform document structure analysis to extract chapter titles, paragraph levels, and table row and column structures. For Excel tables, perform header recognition and key field location. Specifically, for PDF files, the first step is to determine whether they are scanned documents. This determination can be made, but is not limited to, detecting whether the PDF file contains a text layer, counting the number and density of extractable characters, and analyzing the image occupancy of each page. If the PDF file is determined to be a scanned document, Optical Character Recognition (OCR) technology is used to recognize the text in the scanned image to extract its content. If the PDF file is determined not to be a scanned document (i.e., an electronic file containing an extractable text layer), its text content is directly extracted. This method effectively ensures compatibility between PDF materials from both paper-based scanned archives and electronically generated documents, guaranteeing the text usability of approval documents, drawings, and other materials.

[0028] For Word documents, document structure analysis is performed. While extracting the main text content, the document's outline structure is also analyzed. Specifically, this includes extracting chapter titles, paragraph hierarchy, and the row and column structure of embedded tables, thereby fully preserving the organizational structure information of the feasibility study report and providing a structural foundation for subsequent chapter integrity review.

[0029] For Excel spreadsheets, header identification and key field location are performed. Specifically, by analyzing the text features of the first row and first column of the table, as well as cell merging, the location and hierarchical relationship of the header can be identified. Based on this, the column positions of key fields (such as equipment name, specifications, quantity, unit price, amount, etc.) in business tables such as investment estimates and equipment and material lists can be located, thereby accurately extracting structured data from the tables.

[0030] S14. The extracted content is uniformly converted into structured text containing location tags, which are used to identify the file, chapter and field to which the content belongs.

[0031] For example, a multi-level tag format of filename-chapter path-field name can be used to label each extracted content, so that the original file and specific location of the problem can be quickly traced back based on the location tags during the subsequent review process. At the same time, it is convenient to establish field-level correspondence between different files based on location tags.

[0032] S2. Based on natural language processing technology, extract rules from the feasibility study review specification text of the power distribution network and construct a review rule base; Preferably, in this embodiment, step S2 is further refined into the following sub-steps: S21. Perform word segmentation, part-of-speech tagging, named entity recognition, and dependency parsing on the feasibility study review specification text of the power distribution network. Among them, word segmentation and part-of-speech tagging are used to divide the standard text into word units and label their grammatical attributes; named entity recognition is used to identify business entities such as equipment names, voltage levels, project names, amounts, and dates involved in the standard text; dependency parsing is used to reveal the syntactic dependency relationships between words in the standard text, thereby understanding the logical semantics expressed by the standard clauses.

[0033] S22. Identify the review objects, review elements, threshold conditions and qualification criteria in the power distribution network feasibility study review specification text, and extract them into review rules in the form of object-element-condition-criteria quadruples; The review object refers to the document or field to which the rule applies (such as a feasibility study report or investment estimate table); the review element refers to the specific content items that need to be verified (such as construction scale or total investment); the threshold condition refers to the numerical boundary at which the rule is triggered or judged (such as investment amount accurate to ten thousand yuan, date format YYYY-MM-DD); and the qualification criterion refers to the standard for judging whether the review result meets the requirements (such as "should include", "must not be missing", "must be consistent", etc.). By expressing the normative clauses in a structured form using a four-tuple, the institutional rules described in natural language can be transformed into machine-executable review logic.

[0034] S23. Store the review rules in the review rule base and establish a reference mapping relationship between the review rules and the normative clauses.

[0035] The reference mapping relationship records the name, clause number, and original text of the normative document corresponding to each review rule. This allows for the automatic association and reference of the normative basis when a rule is triggered during subsequent reviews, providing support for the normative basis annotation in the review report. Furthermore, when the review norms are revised, the corresponding rules in the rule base can be updated and maintained to ensure the timeliness and accuracy of the review basis.

[0036] S3. Perform a completeness review on the structured text based on the review rule base. The completeness review includes chapter completeness review, field completeness review, and format conformity review. Specifically, the chapter integrity review involves comparing the chapter structure of the feasibility study report with a pre-configured standard chapter template. The standard chapter template is pre-set based on the compilation procedures or standardization outlines for power distribution network feasibility study reports, and includes standard chapter names and their hierarchical order. During the comparison, missing chapters and redundant chapters exceeding the standard template are identified through text matching and hierarchical structure matching of chapter titles, and a list of missing chapters and a list of redundant chapters are output. It should be noted that for non-essential chapters, conditional exemptions can be granted based on criteria in the rule base to avoid misjudging chapters that are unnecessary based on the actual project situation as missing.

[0037] The field integrity review involves verifying whether all required fields in the feasibility study document are complete. These required fields include, but are not limited to, key fields such as project name, construction location, voltage level, construction scale, total investment, and sub-investment. During the review, the field values ​​corresponding to each required field are retrieved from the structured text based on location tags. If a field value is empty, filled with a placeholder (such as " / ", "none", etc.), or clearly does not meet the minimum length requirement, the field is determined to be missing, and the missing field and its document and section location are recorded.

[0038] The format compliance review involves verifying whether the data format of each field conforms to the specifications, specifically including numerical units (e.g., whether the monetary unit is ten thousand yuan, whether the engineering quantity unit is meters / unit, etc.), numbering formats (e.g., whether the coding rules for project numbers and equipment numbers conform to the prescribed number of digits and segmentation methods), and date formats (e.g., whether the date uses the prescribed year-month-day format). Format verification can be achieved by combining regular expression matching, numerical range checks, and unit dictionary comparisons.

[0039] S4. Use dependency resolution technology to perform logical association analysis on cross-file data in the structured text and verify the consistency of cross-file data; Preferably, in this embodiment, step S4 is further refined into the following sub-steps: S41. Extract cross-file data fields from the structured text, and construct a cross-file data dependency graph with data fields as nodes and logical relationships between fields as edges. The logical relationships include summation relationships, equality relationships, range constraint relationships, and reference relationships. Among these, the relationships include: summation (e.g., total investment equals the sum of all sub-investments); equality (e.g., the quantity of equipment recorded in the feasibility study report should match the quantity in the equipment and materials list); range constraints (e.g., a certain sub-investment should be less than the total investment); and referencing (e.g., the value of a certain field references a corresponding field in another file). When constructing the dependency graph, nodes expressing the same business meaning in different files are mapped based on location labels and field semantics. Dependency edges between nodes are established based on predefined rules in the rule base or logical relationships extracted from the standard text, thus forming a field-level association network covering multiple documents such as the feasibility study report, investment estimation table, equipment and materials list, and drawing specifications.

[0040] S42. Verify the values ​​of the corresponding fields along the dependency edges of the data dependency graph to identify data contradictions; wherein the verification includes at least: matching verification of total investment and sum of sub-investments, matching verification of equipment and material list and investment estimate, and matching verification of engineering quantity and drawing parameters.

[0041] Specifically, for summation edges, the sum of the values ​​of the relevant sub-items is calculated and compared with the sum of the values ​​of the summary field; for equality edges, the values ​​of the fields at both ends are directly compared to see if they are consistent; for range constraint edges, the field values ​​are checked to see if they meet the constraint conditions; for reference edges, the values ​​of the referenced field and the referenced field are checked to see if they are consistent. If the comparison result exceeds the preset error tolerance (for example, a reasonable rounding tolerance can be set for monetary fields), then a data contradiction is determined, and the contradictory field and its dependency path are recorded. The verification includes at least: matching verification of total investment and the sum of sub-items (between feasibility study report and estimation table), matching verification of equipment and material list and investment estimate (between inventory table and estimation table), and matching verification of quantities and drawing parameters (between report text and drawing instructions). Through the above verification, cross-file data consistency checks that are difficult to cover with traditional manual comparisons can be transformed into an automated dependency graph traversal verification process, significantly improving the coverage and accuracy of consistency checks.

[0042] S5. Construct a three-layer matching framework of semantics-element-evaluation, perform semantic parsing, element extraction and dynamic weighted evaluation on the problem to be solved by the feasibility study project and the alternative projects, and calculate the matching degree between the problem and the alternative projects. In this embodiment, the semantic-feature-evaluation three-layer matching framework includes, from bottom to top, a semantic layer, a feature layer, and an evaluation layer: Semantic layer: The problem description text and the project description text are input into a pre-trained language model for semantic parsing, generating their respective semantic vector representations. The semantic similarity between the two is then calculated based on these semantic vectors, thus measuring the degree of association between the problem and the candidate projects at the semantic expression level. The advantage of the semantic layer is its ability to capture deep semantic connections between the texts; even if the problem description and project description use different wording, their semantic matching relationship can still be identified.

[0043] Element Layer: Key elements are extracted from both the problem description text and the project description text. These key elements include, but are not limited to, elements reflecting the characteristics of the problem and the project, such as region, voltage level, capacity, investment, and time limit. Then, the key elements of the corresponding categories in the problem and candidate projects are matched item by item, and the element matching degree is calculated based on the degree of agreement of the element values. The advantage of the element layer is that it provides interpretable and traceable matching criteria, allowing reviewers to intuitively view the matching status of each element item.

[0044] Evaluation layer: The semantic similarity output by the semantic layer, the element matching degree output by the element layer, and the planning scene features are used together as state features. The evaluation weights of the semantic layer and the element layer are dynamically optimized using a reinforcement learning model. Based on the optimized weights, the semantic similarity and element matching degree are weighted and fused to output the comprehensive matching degree between the problem and the candidate items.

[0045] Preferably, the dynamic optimization of the evaluation weights of the semantic layer and the element layer using a reinforcement learning model includes: using evaluation weights as actions and the consistency between the matching evaluation results and the conclusions of human review as reward signals, continuously updating the evaluation weights through strategy iteration, so that the model gradually improves the consistency between the matching degree evaluation results and expert judgments during the continuous learning process; simultaneously, adaptively adjusting the evaluation weights according to the planning scenario type, which includes new construction scenarios, renovation scenarios, and expansion scenarios. For example, in a new construction scenario, the weights of elements such as capacity and region in the element layer can be appropriately increased, and in a renovation scenario, the weights of the semantic layer can be appropriately increased to capture the semantic associations of existing equipment. Through the dynamic weighting mechanism, the matching degree evaluation can adapt to the differences in evaluation focus under different planning scenarios, improving the flexibility and accuracy of the matching evaluation.

[0046] S6. Based on the results of integrity review, consistency verification and matching assessment, generate a review report, which includes problem identification, standard basis and modification suggestions.

[0047] The problem identification is based on the location label, which clearly indicates the document, chapter, and field where the problem is located; the standard basis is based on the reference mapping relationship between the review rules and the standard clauses, which indicates the standard clause number and the original text of the clause that the problem violates or refers to; the modification suggestions provide targeted modification guidance based on the problem type (such as "supplementing the missing construction scale field", "unifying the total investment amount in the feasibility study report and the estimation table", etc.).

[0048] Preferably, in step S6, the review report categorizes and sorts the various issues discovered during the review process according to their severity. For example, issues can be divided into different levels such as "serious," "general," and "noticeable," and displayed in descending order of severity so that reviewers can prioritize addressing critical issues. Simultaneously, each issue is accompanied by its corresponding regulatory clause, ensuring that the review conclusions are verifiable. Furthermore, the review report supports exporting to PDF or Excel format to meet the archiving and transfer needs of different review stages.

[0049] Although this invention makes extensive use of terms such as structured text and natural language processing, the possibility of using other terms is not excluded. These terms are used merely for the convenience of describing and explaining the essence of this invention; interpreting them as any additional limitation would contradict the spirit of this invention.

[0050] This invention provides a method for assisting in the integrity review of power distribution network feasibility study projects. By performing multimodal parsing and structured processing on the project file package, it achieves unified processing and automated machine review of various heterogeneous files such as Word, Excel, and PDF, avoiding manual page-by-page verification and significantly improving review efficiency. Based on natural language processing technology, it extracts rules from the review specification text and constructs a rule base, transforming scattered, naturally language-described institutional rules into executable machine review logic, achieving unified and standardized review standards and eliminating the subjectivity of human review, which varies from person to person and has inconsistent standards. Finally, it utilizes dependency parsing technology to perform logical association analysis on cross-file data. It can automatically compare the consistency of key data across different documents, effectively solving the problem of omissions that are easily missed in manual comparison and enhancing the comprehensiveness and accuracy of the review. Furthermore, by constructing a three-layer matching framework of semantics-element-evaluation, it performs semantic parsing, element extraction, and dynamic weighted evaluation on the proposed problem and alternative projects, realizing a systematic and quantifiable evaluation of the degree of matching between the problem and the project, reducing the reliance on expert experience in the evaluation. The final review report includes problem identification, normative basis, and modification suggestions, providing reviewers with directly usable decision support, and improving the efficiency, objectivity, and scientific nature of the review of power distribution network feasibility study projects, ensuring the scientific nature of investment decisions and the safety of project implementation.

[0051] Example 2

[0052] Figure 2 This is a schematic diagram of the structure of a computer device provided in Embodiment 2 of the present invention. Figure 2 A block diagram of an exemplary computer device 12 suitable for implementing embodiments of the present invention is shown. Figure 2 The computer device 12 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0053] like Figure 2 As shown, the computer device 12 is represented in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0054] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0055] Computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0056] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 2 Not shown; usually referred to as a "hard drive"). Although Figure 2Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0057] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of the present invention.

[0058] Computer device 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), and with one or more devices that enable a user to interact with the computer device 12, and / or with any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 22. Furthermore, computer device 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of computer device 12 via bus 18. It should be understood that, although... Figure 2 As not shown, it can be used in conjunction with computer device 12 with other hardware and / or software modules, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0059] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the auxiliary review method for the integrity of power distribution network feasibility study projects provided in the embodiments of the present invention.

[0060] Example 3

[0061] Embodiment 3 of the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the method for auxiliary review of the integrity of power distribution network feasibility study projects as provided in all embodiments of the present invention.

[0062] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0063] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0064] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0065] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0066] Finally, it should be noted that although the above embodiments have been described in the description and drawings of this invention, this should not limit the scope of patent protection of this invention. Any technical solutions that are based on the essential concept of this invention, utilize the content described in the description and drawings of this invention to make equivalent structural or procedural substitutions or modifications, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this invention.

Claims

1. A method for auxiliary review of the integrity of a power distribution network feasibility study project, characterized in that, Includes the following steps: S1. Obtain the power distribution network feasibility study project file package to be reviewed, perform multimodal parsing and structuring on the heterogeneous files in the file package to obtain structured text in a unified format. The heterogeneous files include Word documents, Excel spreadsheets and PDF files. S2. Based on natural language processing technology, extract rules from the feasibility study review specification text of the power distribution network and construct a review rule base; S3. Perform a completeness review on the structured text based on the review rule base. The completeness review includes chapter completeness review, field completeness review, and format conformity review. S4. Use dependency resolution technology to perform logical association analysis on cross-file data in the structured text and verify the consistency of cross-file data; S5. Construct a three-layer matching framework of semantics-element-evaluation, perform semantic parsing, element extraction and dynamic weighted evaluation on the problem to be solved by the feasibility study project and the alternative projects, and calculate the matching degree between the problem and the alternative projects. S6. Based on the results of integrity review, consistency verification and matching assessment, generate a review report, which includes problem identification, standard basis and modification suggestions.

2. The method for auxiliary review of the integrity of power distribution network feasibility study projects according to claim 1, characterized in that, S1 includes: S11. Obtain the project feasibility study file package for the distribution network to be reviewed; S12. Decompress the file package and identify the file types of heterogeneous files within the file package; S13. For PDF files, determine whether they are scanned documents. If they are scanned documents, use OCR technology to recognize text; otherwise, extract the text directly. For Word documents, perform document structure analysis to extract chapter titles, paragraph levels, and table row and column structures. For Excel tables, perform header recognition and key field location. S14. The extracted content is uniformly converted into structured text containing location tags, which are used to identify the file, chapter and field to which the content belongs.

3. The method for auxiliary review of the integrity of power distribution network feasibility study projects according to claim 1, characterized in that, S2 includes: S21. Perform word segmentation, part-of-speech tagging, named entity recognition, and dependency parsing on the feasibility study review specification text of the power distribution network. S22. Identify the review objects, review elements, threshold conditions and qualification criteria in the power distribution network feasibility study review specification text, and extract them into review rules in the form of object-element-condition-criteria quadruples; S23. Store the review rules in the review rule base and establish a reference mapping relationship between the review rules and the normative clauses.

4. The method for auxiliary review of the integrity of power distribution network feasibility study projects according to claim 1, characterized in that, In S3, the chapter integrity review is as follows: the chapter structure of the feasibility study report is compared with the standard chapter template to identify missing and redundant chapters; The field integrity review is as follows: verifying whether the essential fields are complete. The essential fields include a method for auxiliary review of the integrity of a power distribution network feasibility study project, construction location, voltage level, construction scale, total investment and sub-item investment. The format compliance review involves checking whether the data format of the fields conforms to the specifications, including numerical units, numbering formats, and date formats.

5. The method for auxiliary review of the integrity of power distribution network feasibility study projects according to claim 1, characterized in that, S4 includes: S41. Extract cross-file data fields from the structured text, and construct a cross-file data dependency graph with data fields as nodes and logical relationships between fields as edges. The logical relationships include summation relationships, equality relationships, range constraint relationships, and reference relationships. S42. Verify the values ​​of the corresponding fields along the dependency edges of the data dependency graph to identify data contradictions; wherein the verification includes at least: matching verification of total investment and sum of sub-investments, matching verification of equipment and material list and investment estimate, and matching verification of engineering quantity and drawing parameters.

6. The method for auxiliary review of the integrity of power distribution network feasibility study projects according to claim 1, characterized in that, In S5, the semantic-feature-evaluation three-layer matching framework includes: Semantic layer: Semantic parsing is performed on the problem description text and the project description text respectively to generate semantic vectors and calculate semantic similarity; Element layer: Extract key elements from the problem description text and project description text. The key elements include region, voltage level, capacity, investment and time limit. Match the corresponding key elements one by one and calculate the element matching degree. Evaluation layer: Using the semantic similarity, the element matching degree, and the planning scene features as state features, the evaluation weights of the semantic layer and the element layer are dynamically optimized using a reinforcement learning model, and the comprehensive matching degree between the problem and the alternative options is output.

7. The method for auxiliary review of the integrity of power distribution network feasibility study projects according to claim 6, characterized in that, The method of dynamically optimizing the evaluation weights of the semantic layer and the feature layer using a reinforcement learning model includes: The evaluation weights are used as actions, and the consistency between the evaluation results and the conclusions of manual review is used as a reward signal. The evaluation weights are updated iteratively through strategies. The evaluation weights are also adaptively adjusted according to the planning scenario type, which includes new scenarios, renovation scenarios, and expansion scenarios.

8. The method for auxiliary review of the integrity of power distribution network feasibility study projects according to claim 1, characterized in that, In step S6, the review report categorizes and sorts the identified issues according to their severity, and indicates the relevant regulatory clauses for each issue; the review report can be exported as a PDF or Excel file.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the auxiliary review method for the integrity of the power distribution network feasibility study project as described in any one of claims 1-8.

10. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that, The computer-executable instructions are executed by a computer processor to implement the auxiliary review method for the integrity of the distribution network feasibility study project as described in any one of claims 1-8.