A grid infrastructure project feasibility study report structured design method and system

Abstract

The application relates to the technical field of power grid infrastructure projects. A power grid infrastructure project feasibility study report structured design method and system comprises obtaining file data of power grid infrastructure related management regulations in a historical file library to obtain a structured text data set; performing structured information extraction on the text data set to obtain a text structure framework set; performing key review element extraction of feasibility study content review rules on the text data set to obtain a key review element set; establishing logical association for each key review element, each node and each text in the text data set to obtain a mapping relationship M of key review element IDs, node IDs and text IDs; forming a feasibility study report structure diagram according to the mapping relationship M to create a power grid infrastructure project feasibility study report template. The output standardized template provides a solid and unified data basis for the management, integration and deep analysis of power grid data, thereby directly supporting the scientificity and accuracy of investment decisions.

Description

Technical Field

[0001] This application relates to the technical field of power grid infrastructure projects, and more specifically, to a structured design method and system for feasibility study reports of power grid infrastructure projects. Background Technology

[0002] Feasibility study reports are crucial in the power grid sector, providing a scientific basis and technical support for investment decisions on power grid projects. They not only reflect the project's economic benefits, technical feasibility, and environmental impact, but also its compliance, risk assessment, and implementation plan. A structured feasibility study report ensures that all key information is systematically considered and presented, helping decision-makers fully understand the project and make informed investment decisions. The necessity of structuring lies in its ability to improve readability and logic, making complex information easier to understand and track, and ensuring that the report's writing logic conforms to relevant project evaluation rules. Furthermore, standardized feasibility study reports provide a solid foundation for the governance and subsequent application of power grid data, facilitating data integration and analysis.

[0003] Currently, in the power grid sector, due to its complexity and interdisciplinary nature, the structuring and standardization of feasibility study reports have not yet met the needs of project evaluation and decision-making. Therefore, there is an urgent need for a structured design method and system for power grid infrastructure project feasibility study reports to address these existing problems. Summary of the Invention

[0004] The main purpose of this application is to provide a structured design method and system for feasibility study reports of power grid infrastructure projects, so as to solve the problems of non-standardization, lack of standardization and poor structure in existing feasibility study reports in the power grid field.

[0005] To achieve the above objectives, the first aspect of this application proposes a structured design method for feasibility study reports of power grid infrastructure projects, including:

[0006] In the historical document library, document data related to power grid infrastructure management regulations are obtained to obtain a structured text data set, wherein the document data includes at least policy documents, industry standards and enterprise specifications for power grid infrastructure projects;

[0007] The text data set is subjected to structured information extraction to obtain a set of text structure frameworks, wherein the set of structure frameworks includes at least a node relationship diagram formed by title name, title level and association relationship, and each node is composed of multiple key review elements;

[0008] The key review elements of the feasibility study content review rules are extracted from the text data set to obtain a key review element set, wherein the key review element set includes a list of review elements corresponding to the nodes.

[0009] For each key review element in the text data set, each node in the structural framework set, and each text in the text data set, a key review element ID, a node ID, and a text ID are created respectively. Logical associations are established based on semantic analysis or a rule engine to obtain a mapping relationship M between key review element ID, node ID, and text ID. Each mapping relationship includes a triple relationship of key review element-node-text.

[0010] Based on the mapping relationship M, a feasibility study report structure diagram is formed, and a template for a feasibility study report of a power grid infrastructure project is created.

[0011] In some feasible methods, the step of obtaining a structured text data set by acquiring document data related to power grid infrastructure management regulations from a historical document library includes:

[0012] Searching the historical document database yields policy documents, industry standards, and corporate specifications related to power grid infrastructure management, forming document data.

[0013] Using a file parsing algorithm, the text content of the file data is extracted and the data is cleaned to obtain intermediate text data, wherein the intermediate text data contains clean text content.

[0014] Based on the natural paragraphs and chapter boundaries of the intermediate text data, text structuring processing of logical blocks is performed to obtain a text data set.

[0015] In some feasible methods, the step of extracting structured information from the text data set to obtain a set of text structure frameworks includes:

[0016] Construct a structured rule matching library, wherein the structured rule matching library includes at least font feature information, indentation information, heading numbering pattern, positional relationship and level identifier, for identifying heading levels in text;

[0017] The structured rule matching library is matched with the text data set, the title name and title level are identified through feature parsing, and the association relationship is constructed based on the title level relationship to obtain a node relationship graph, wherein each node in the node relationship graph corresponds to multiple key review elements;

[0018] The nodes and relationships in the node relationship graph are extracted to obtain the set of text structure frameworks.

[0019] In some feasible methods, the step of matching the structured rule matching library with the text data set, identifying the title name and title level through feature parsing, and constructing an association relationship based on the title level relationship to obtain a node relationship graph includes:

[0020] By using conflict resolution methods, nodes in different documents, such as policy documents, industry standards, and corporate specifications of power grid infrastructure projects, are processed to obtain a node relationship diagram after conflict resolution.

[0021] In some feasible methods, the step of extracting key review elements from the text data set according to the feasibility study content review rules to obtain a key review element set includes:

[0022] Using regular expressions and a list of semantic trigger words, the feasibility study content review rules are extracted from the text data set to obtain the feasibility study content review rules.

[0023] The key review elements in the feasibility study content review rules are extracted to obtain a set of key review elements.

[0024] In some feasible methods, the step of extracting key review elements from the feasibility study content review rules to obtain a set of key review elements includes:

[0025] The feasibility study content review rules were analyzed using jieba word segmentation to convert the data into a word sequence.

[0026] The TF-IDF algorithm is used to identify the key elements of the term sequence, thereby obtaining a set of key review elements.

[0027] In some feasible methods, the step of creating a key review element ID, node ID, and text ID for each key review element in the text data set, each node in the structural framework set, and each text in the text data set, and establishing logical associations based on semantic analysis or a rule engine to obtain the mapping relationship M between the key review element ID, node ID, and text ID includes:

[0028] For each key review element in the text data set, each node in the structural framework set, and each text in the text data set, a corresponding key review element ID, node ID, and text ID are created respectively;

[0029] Using semantic similarity calculation methods or predefined rule engines, logical associations are formed between elements and nodes, and text source analysis is used to logically associate elements and texts, as well as the attribution of nodes and texts, thus forming logical associations between nodes and texts. This yields the mapping relationships between element IDs and node IDs, element IDs and text IDs, and node IDs and text IDs.

[0030] Based on the mapping relationship between the element ID and the node ID, the mapping relationship between the element ID and the text ID, and the mapping relationship between the node ID and the text ID, a mapping relationship M is obtained that links the key review element ID, node ID, and text ID.

[0031] In some feasible methods, the step of forming a feasibility study report structure diagram and creating a feasibility study report template for a power grid infrastructure project based on the mapping relationship M includes:

[0032] Based on the mapping relationship M, the element titles of the nodes and the element content corresponding to the nodes are established in units of text, forming the structure diagram of the feasibility study report;

[0033] Load the structure diagram of the feasibility study report into the template to create a feasibility study report template for power grid infrastructure projects.

[0034] In some feasible methods, after the step of forming a feasibility study report structure diagram based on the mapping relationship M and creating a feasibility study report template for the power grid infrastructure project, the following steps are included:

[0035] The current file library is periodically polled to obtain the current file data, and the mapping relationship M is updated based on the current file data to obtain an updated feasibility study report structure diagram;

[0036] The updated feasibility study report structure diagram is used to update the template for the created power grid infrastructure project feasibility study report.

[0037] Secondly, this application provides a structured design system for feasibility study reports of power grid infrastructure projects, applied to the aforementioned structured design method for feasibility study reports of power grid infrastructure projects, including:

[0038] The acquisition unit is used to acquire document data related to power grid infrastructure management regulations from the historical document library to obtain a structured text data set, wherein the document data includes at least policy documents, industry standards and enterprise specifications for power grid infrastructure projects;

[0039] A structured unit is used to extract structured information from the text data set to obtain a set of text structure frameworks, wherein the set of structure frameworks includes at least a node relationship diagram formed by title name, title level and association relationship, and each node is composed of multiple key review elements;

[0040] The extraction unit is used to extract key review elements of the feasibility study content review rules from the text data set to obtain a key review element set, wherein the key review element set includes a list of review elements corresponding to the nodes.

[0041] The logical association unit is used to create a key review element ID, node ID, and text ID for each key review element in the text data set, each node in the structural framework set, and each text in the text data set, respectively, and establish logical associations based on semantic analysis or a rule engine to obtain a mapping relationship M between the key review element ID, node ID, and text ID, wherein each mapping relationship includes a triple relationship of key review element-node-text.

[0042] The result unit is used to generate a feasibility study report structure diagram and create a feasibility study report template for power grid infrastructure projects based on the mapping relationship M.

[0043] The technical solutions provided by the embodiments of this application may include the following beneficial effects:

[0044] This application presents a structured design method for feasibility study reports of power grid infrastructure projects. By systematically extracting and integrating standardized feasibility study report templates from multi-level management regulations, it addresses the challenges of insufficient structure and standardization in feasibility study reports. This method first transforms scattered, unstructured text data into a structural framework with clear hierarchical relationships and a set of key review elements. Then, by constructing precise mapping relationships between elements, nodes, and text, it ensures that the report template comprehensively covers all compliance requirements in content and strictly conforms to project evaluation rules in logic. This not only significantly improves the readability and logical coherence of the report, making complex information easier to track and understand, but more importantly, the standardized templates it produces provide a solid and unified data foundation for the governance, integration, and in-depth analysis of power grid data, thereby directly supporting the scientific and accurate nature of investment decisions. Attached Figure Description

[0045] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application. In the drawings:

[0046] Figure 1A flowchart illustrating a structured design method for a feasibility study report of a power grid infrastructure project, provided in this application.

[0047] Figure 2 This application provides a structural element diagram of a feasibility study report for a power grid infrastructure project, which is part of a structured design method for such reports. Detailed Implementation

[0048] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0049] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0050] In this application, the terms "upper," "lower," "left," "right," "front," "rear," "top," "bottom," "inner," "outer," "middle," "vertical," "horizontal," "lateral," and "longitudinal" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are primarily for the purpose of better describing this application and its embodiments, and are not intended to limit the indicated device, element, or component to having a specific orientation, or to be constructed and operated in a specific orientation.

[0051] Furthermore, in addition to indicating location or positional relationship, some of the aforementioned terms may also have other meanings. For example, the term "above" may also be used in some cases to indicate a certain dependency or connection relationship. Those skilled in the art can understand the specific meaning of these terms in this application based on the specific circumstances.

[0052] Furthermore, the terms "installation," "setup," "equipped with," "connection," "linked," and "socketing" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral structure; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or an internal connection between two devices, components, or parts. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0053] Text structuring, also known as semantic text processing or information extraction, refers to the process of transforming disordered text data into ordered, understandable information. Through text structuring, various information elements in project feasibility study documents, such as technical solutions, economic benefits, social benefits, and environmental impacts, can be classified, organized, and labeled, facilitating subsequent data element interaction and rule-based correlation analysis. Firstly, regarding data element interaction, after structuring, the feasibility study documents form ordered data stored in a database. Enterprises can then inventory their data resources and establish a data resource catalog. Subsequently, based on specific production needs, they can aggregate, organize, and process this data to form data products, which are then incorporated into the production process to fully release their value and form data assets. Data elements are interactively utilized in this process. Secondly, regarding rule-based correlation analysis, structured information is correlated with rules using analytical methods. This allows for analysis of the differences between the rules shown in the current feasibility study report and the actual rule documents, providing decision support for the subsequent structured design of the feasibility study report.

[0054] In China, scholars widely utilize various technologies for report structure design, including but not limited to XML, ontology, and machine learning, with applications across multiple fields, but the primary research focus is on the medical field. For example: 1. Following the DICOM SR standard, using XML as the information exchange method, and through parsing and converting XML files, visualization is achieved using Web services, realizing the application of DICOM structured reports in PACS workstations based on a B / S architecture. 2. Based on ontology concepts and management report terminology, combined with the Clinical Document Structure Standard (CDA), 1093 primary liver cancer CT imaging reports were randomly selected, analyzed, and organized to form a terminology set for primary liver cancer imaging reports. This terminology set was then mapped to the CDA to establish a standard for the clinical document structure of primary liver cancer imaging reports. 3. By describing and processing the structural characteristics of technical documents during the development of a single model of aviation products, a three-level XML file structured storage method was proposed. By parsing and outputting the file content node by node, a standard-compliant structured template can be formed. 4. A part-of-speech dictionary for thyroid ultrasound reports was constructed using methods such as sentence segmentation, cluster analysis, and named entity recognition to obtain the part-of-speech tags of the reports. Then, dependency parsing was performed on the short sentences to obtain syntactic relations. Combining the part-of-speech tags and syntactic relations, a structured template tree for thyroid ultrasound reports was automatically constructed, achieving the structuring of thyroid ultrasound reports. 5. The machine learning model Bi-LSTM + CRF was used to extract text feature labels for mammography medical image reports, and mapping rules were designed to achieve the structuring of image reports.

[0055] Abroad, scholars' research on structured reports mainly focuses on the medical field. For example, based on guidelines for standardized interpretation and post-processing of cardiovascular magnetic resonance images and CMR reporting guidelines, and combining German experience with the actual situation in China, a structured CMR report template has been developed to improve report quality. It mainly includes three parts: equipment and methods, findings (structure and function, tissue characteristics), summary, and conclusions.

[0056] In summary, while there has been some research progress in the structured design of documents and reports both domestically and internationally, most of it has been concentrated in the medical field, and research on the structured design of project feasibility study reports (project feasibility study reports) is still in its infancy.

[0057] Current technologies for structuring reports often rely on automatic / semi-automatic / non-automatic extraction and transformation of existing unstructured data into structured information. This approach, rather than starting from the actual needs of the data to achieve a more structured design for the report itself, can lead to structured reports that fail to fully meet established regulations and requirements.

[0058] In summary, the existing technical solutions still have the following drawbacks:

[0059] 1) The lack of in-depth analysis on data governance and data application makes it difficult for the structured content to fully reflect actual needs, resulting in deficiencies in the standardization and practicality of the report's structure.

[0060] 2) Some existing technologies still rely on the original report format and are limited to one or more data transformations, but their generation logic has not changed fundamentally, which makes it difficult to achieve the goal of data standardization.

[0061] To address the aforementioned shortcomings, this application aims to propose a structured design method for feasibility study reports of power grid infrastructure projects. Primarily targeting power grid infrastructure projects, this method enables the structured design of feasibility study reports, meeting multi-level management regulations while considering actual project decision-making needs, and promoting the standardization, efficient governance, and application of unstructured power grid data.

[0062] This application proposes a structured design method for feasibility study reports of power grid infrastructure projects. This method is mainly aimed at power grid infrastructure projects. By combining the concept of unstructured data metadata management, it adopts an element-based approach to realize the structured design of feasibility study reports. [3] By comprehensively analyzing the existing multi-level report structure regulations and the corresponding report content review knowledge and rules, the basic structure of the feasibility report and the expression of key elements are determined. At the same time, in order to effectively deal with subsequent data management issues, this invention takes into account the updates of policies and rules, and establishes a flexible adjustment mechanism so that the structural elements and expression of the feasibility study report can meet the latest industry dynamics and regulatory requirements.

[0063] like Figure 1 As shown, in a first aspect, this application provides a structured design method for feasibility study reports of power grid infrastructure projects, the method comprising:

[0064] S100 retrieves document data related to power grid infrastructure management regulations from the historical document library, resulting in a structured text data set.

[0065] The document data includes at least policy documents, industry standards, and corporate specifications for power grid infrastructure projects.

[0066] The purpose of step S100 is to extract clean text content with a preliminary logical structure from scattered, multi-format original management regulations documents, forming a unified and well-organized text data set, laying a solid data foundation for subsequent structuring and element extraction.

[0067] Specifically, obtaining a structured collection of text data may include the following steps:

[0068] S101. Search the historical document database to obtain policy documents, industry standards, and enterprise specifications related to power grid infrastructure management, forming document data.

[0069] Specifically, the historical document database is searched based on predefined keywords (such as "feasibility study report", "power grid infrastructure", "compilation specifications", "review points") and document categories (such as "policy documents", "industry standards", "enterprise specifications"). Next, all retrieved documents are collected; these documents may contain multiple formats, such as PDF and Word.

[0070] It should be noted that the historical document library can refer to documents related to power grid infrastructure up to the present.

[0071] S102, using a file parsing algorithm, extract the text content from the file data and perform data cleaning to obtain intermediate text data.

[0072] The intermediate text data contains clean text content.

[0073] Specifically, different parsing technologies are used for different file formats. For example, for PDF files, a PDF parsing library is used to distinguish between "text-based PDFs" and "scanned image-based PDFs." For text-based PDFs, characters and their position information are directly extracted; for scanned PDFs, optical character recognition (OCR) is performed first to convert the image into text. For Word documents, the corresponding document processing library is used to directly parse their document object model (DOM) and extract text content such as titles and paragraphs, as well as their style information.

[0074] After extracting the text content, the next step is data cleaning, which involves cleaning the extracted raw text to obtain "clean" text content. Cleaning rules may include: removing irrelevant elements: deleting headers, footers, page numbers, watermarks, and other information unrelated to the main text; correcting recognition errors: automatically correcting character errors generated during the OCR process; and standardizing formatting: standardizing inconsistent spaces, line breaks, etc. After data cleaning, intermediate text data is obtained.

[0075] S103, based on the natural paragraphs and chapter boundaries of the intermediate text data, perform text structuring processing on the logical blocks to obtain a text data set.

[0076] Specifically, after obtaining the intermediate text data in the aforementioned S102 step, the boundaries of natural paragraphs and chapters are identified.

[0077] For example:

[0078] For paragraph boundary recognition, continuous text is segmented into natural paragraphs based on visual and formatting features such as line breaks and indentation.

[0079] For chapter boundary recognition, chapter boundaries can be identified by analyzing patterns in the text, for example:

[0080] Title pattern matching: Identifies text lines that match specific number or letter numbering patterns (such as "Chapter 1", "1.1", "(I)", etc.).

[0081] Formatting feature aids: Combine formatting cues (such as font size and bolding) retained or reanalyzed from S102 to confirm the heading level.

[0082] Keyword triggering: Identify key words such as "General Provisions" and "Appendix" to divide large logical blocks.

[0083] Based on the identified boundaries, the text is organized into hierarchical logical blocks. For example, all text under a first-level heading (including its subordinate second-level headings and paragraphs) is grouped into a single logical block, forming a small unit. This results in a structured collection of text data, where the text is no longer unordered but organized into structured data with clear chapter and paragraph relationships. Each logical block is labeled with its hierarchy and belonging relationships.

[0084] Through the above three steps, this application transforms the policy documents, industry standards, and enterprise specifications related to power grid infrastructure management from their original unstructured form into a structured form, so that they can be extracted and used in subsequent steps as a basis for those steps.

[0085] S200, perform structured information extraction on the text data set to obtain a set of text structure frameworks.

[0086] The structural framework set includes at least a node relationship diagram formed by title name, title level and association relationship, and each node consists of multiple key review elements.

[0087] The purpose of step S200 is to further transform the preliminary structured text data set obtained in S100 into a precise, machine-readable hierarchical structure (i.e., a set of structural frameworks).

[0088] Specifically, obtaining the structural framework set of the text may include the following steps:

[0089] S201, Build a structured rule matching library.

[0090] The structured rule matching library includes at least font feature information, indentation information, heading numbering pattern, positional relationship and level identifier, which are used to identify heading levels in the text.

[0091] Specifically, based on the format analysis of a large number of management regulation documents, a structured rule matching library is constructed to identify heading levels. The structured rule matching library contains a series of judgment rules; for example, the judgment rules are as follows:

[0092] Font characteristics: Headings typically use bold, larger fonts or specific font families.

[0093] Indentation information: Headings at different levels often have different left indentation amounts; the higher the level, the smaller the indentation may be.

[0094] Title numbering pattern: Define common numbering sequences, such as "Chapter X", "XY", "(X)", "X.)", etc., and assign them possible levels.

[0095] Positional relationship: The title is usually located at the beginning of a line, and the content of the following paragraph has a clear visual connection with it.

[0096] Level identifiers: Combine the above features to clearly define what constitutes a "first-level heading", "second-level heading", etc.

[0097] The importance of features formed by each rule is not equal. Therefore, a preset weight is assigned to each rule. Using these weights, confidence is calculated during the judgment process. This is a multi-feature weighted fusion process, the goal of which is to calculate the total score for a title judged to be at a certain level. In other words, the scores of all features are weighted and summed according to their respective weights to obtain a total score, i.e., the comprehensive confidence score. A judgment threshold is then set; if the score is greater than the threshold, it is considered true; otherwise, it is considered false.

[0098] S202, the structured rule matching library is matched with the text data set, the title name and title level are identified through feature parsing, and the association relationship is constructed based on the title level relationship to obtain the node relationship graph.

[0099] In the node relationship diagram, each node corresponds to multiple key review elements.

[0100] Specifically, each logical block (such as a paragraph) in the text data set is traversed. For each logical block, its text content, formatting features, and location information are extracted and compared with a rule matching library. Through a weighted scoring mechanism, it is determined whether the logical block is a heading and its specific level is determined (e.g., it is determined to be a "second-level heading").

[0101] Node creation: For each identified title, a corresponding node is created in the graph. Node attributes must include at least: node ID, title name, and title level.

[0102] Relationship Building: Based on the heading level and order of appearance, establish the relationships between nodes to form a node relationship diagram. Parent-Child Relationship: All lower-level headings that follow a higher-level heading and precede the next sibling or higher-level heading are child nodes of that higher-level heading. For example, the "Chapter 1" node is the parent node of the immediately following "1.1", "1.2", etc. Sequence Relationship: Headings at the same level have a sequential order.

[0103] Furthermore, step S202 includes:

[0104] By using conflict resolution methods, nodes in different documents, such as policy documents, industry standards, and corporate specifications of power grid infrastructure projects, are processed to obtain a node relationship diagram after conflict resolution.

[0105] Specifically, in steps S201 and S202, the system extracts the respective report structures (i.e., "node relationship diagrams") from multiple documents such as national policies, industry standards, and enterprise specifications. However, these independently extracted structures may conflict, mainly in the following ways:

[0106] Node hierarchy conflict: The same content topic is specified at different levels in different documents.

[0107] For example, in national policies, "environmental impact assessment" may be required to be a separate first-level chapter (Chapter 1); while in a certain corporate standard, it may be classified as a second-level chapter (such as Section 1.5).

[0108] Node name conflict: The same content topic has different names in different files.

[0109] For example, policy documents may refer to it as "social stability risk analysis," while industry standards may call it "social stability assessment."

[0110] Node necessity conflict: A certain chapter is mandatory in a high-level file, but may be optional or missing in a low-level file.

[0111] If these conflicts are not addressed and nodes are merged directly, the resulting structure diagram will be logically chaotic and inconsistent, failing to meet compliance requirements at all levels.

[0112] In view of this, the conflict resolution method is used to handle the issue, and its working principle is as follows:

[0113] 1. By comparing node relationship graphs from different files, and calculating the semantic similarity of node names (e.g., determining whether "risk analysis" and "risk assessment" refer to the same thing), the system automatically identifies which nodes are "corresponding nodes" (i.e., describing the same content topic) but have different attributes (such as hierarchy and name). These differences are marked as "conflict points".

[0114] 2. For detected conflicts, the system makes decisions based on a preset priority strategy. Core strategies typically include:

[0115] The authority priority strategy stipulates that higher-level documents have higher authority than lower-level documents. The priority order is typically: national policies > industry standards > corporate specifications. When a conflict occurs between node levels, the system will adopt the level specified by the more authoritative document.

[0116] The inclusive strategy involves retaining nodes that are not present in higher-level files and do not conflict with them, to ensure report completeness. For necessary conflicts, if a higher-level file requires a node, it must be retained; if only the lower-level file contains it, it can be retained as a supplement.

[0117] The naming standardization strategy prioritizes the name in the more authoritative file or selects the most standard name from a predefined standardized terminology library when there is a conflict in node names.

[0118] 3. After applying the above strategy, nodes with the same semantics will be merged and given unified, adjudicated attributes (such as unified hierarchy and name). All unique nodes that do not conflict will be retained. The final output is a single node relationship graph that resolves all conflicts. This relationship graph satisfies the requirements of all levels to the greatest extent and is the basis for generating the final result.

[0119] In summary, when different documents contain conflicting descriptions of the hierarchy or names of the same chapter, a preset strategy is used to resolve the conflict, ensuring that the final generated node relationship diagram is logically consistent.

[0120] S203, extract the nodes and relationships in the node relationship graph to obtain the text structure framework set.

[0121] Specifically, after constructing the node relationship graph in step S202, the node relationship graph undergoes format conversion of its nodes and relationships. This transforms the graph data structure into a serialized or exported set of structural frameworks that can be read and processed by subsequent processes. This set of textual structural frameworks is essentially a data representation of the node relationship graph, containing a list of all nodes and relationships.

[0122] In summary, the S200 steps utilize a rule base to transform visual and formatting features into explicit hierarchical logic, constructing a clear and accurate feasibility study report structure. This structure serves as a bridge connecting the original regulations and the final standardized report, ensuring the compliance and systematic nature of the report structure.

[0123] S300, extract the key review elements of the feasibility study content review rules from the text data set to obtain a key review element set.

[0124] The set of key review elements includes a list of review elements corresponding to each node.

[0125] Specifically, obtaining the set of key review elements may include the following steps:

[0126] S301, using regular expressions and a list of semantic trigger words, extract the feasibility study content review rules from the text data set to obtain the feasibility study content review rules.

[0127] Specifically, for the preprocessed text data set, all clauses expressing feasibility study review requirements are identified from the text data, forming a structured list of "feasibility study content review rules." This is equivalent to building a review rule knowledge base, providing a basis for subsequent element extraction.

[0128] Furthermore, regular expression matching: using a predefined pattern library (such as r"should contain .*", r"must be argued .*") (the pattern library is manually compiled based on domain knowledge (such as power grid standard documents) or learned through historical data), it accurately matches mandatory expressions with fixed structures in the text (e.g., "the report should contain technical solution comparison"). Regular expressions excel at handling formatted text patterns, ensuring the accuracy of rule extraction.

[0129] Semantic trigger word list assistance: Simultaneously, a predefined semantic trigger word library (containing words such as "analysis," "assessment," "calculation," and "verification") is used to identify sentences expressing review intent through text. This expands the recall scope, capturing clauses with non-fixed patterns but semantic relevance (e.g., "environmental impact assessment required").

[0130] The combination of these two approaches ensures both efficient and comprehensive extraction of rule clauses. The output is a structured set of "feasibility study content review rules," with each rule clearly describing a review requirement (such as "the report must provide detailed investment estimates").

[0131] S302 extracts the key review elements from the feasibility study content review rules to obtain a set of key review elements.

[0132] Extract the most core and operable key elements from each review rule to form a "set of key review elements". These elements will serve as specific content points in the feasibility study report template to ensure that the report meets the review requirements.

[0133] Furthermore, step S302 includes:

[0134] S3021, Use jieba segmentation to segment and transform the feasibility study content review rules to obtain a sequence of terms.

[0135] Specifically, Chinese text is a continuous sequence of characters and needs to be segmented into independent lexical units (terms) for computer processing. Jieba segmentation is an efficient Chinese word segmentation tool. It is based on a dictionary and a statistical model to transform regular text into a discrete sequence of terms. For example, the rule "The report should conduct a technical and economic comparison of multiple options" is segmented into the term sequence: ["report", "should", "conduct", "multiple options", "technical and economic", "comparison"]. Furthermore, during the segmentation process, a custom dictionary in the power grid field (such as "N-1 verification", "load-to-capacity ratio") will be loaded to ensure that professional terms are accurately segmented and avoid incorrect splitting.

[0136] S3022, Use the TF-IDF algorithm to identify the key elements of the term sequence to obtain a set of key review elements.

[0137] TF-IDF (Term Frequency - Inverse Document Frequency) is a statistical method used to evaluate the importance of a term in a text collection. Its principle is: The more frequently a term appears in a single rule (high term frequency), and the rarer it appears in the entire rule set (high inverse document frequency), the more representative the term is of the core content of the rule and should be identified as a key review element. Furthermore, calculate the TF-IDF weights of all terms and select the term with the highest weight as the key element. For example, in the rule set, "technical and economic comparison" may have both high TF and IDF and be extracted as an element, while the common word "of" has a low weight and is filtered. Finally, through normalization (such as merging synonyms, using a synonym dictionary or entity linking technology to unify the expression), a standardized set of key review elements is formed.

[0138] In step S300, the transformation from text to structured knowledge is achieved: S301 constructs a review rule library using pattern matching and semantic analysis, and S302 refines key elements through segmentation and statistical feature extraction. The entire process requires little or no manual intervention, improving the automation level and accuracy of the feasibility study report compilation, and providing a data basis for the standardized review of power grid infrastructure projects. The set of key review elements will be used for the subsequent construction of the mapping relationship M and the generation of the report template.

[0139] In one embodiment, it is known that not every node has key review elements. However, if a node without key review elements continues to appear in subsequent steps, it will not affect the final creation of the power grid infrastructure project feasibility study report template, but will increase the amount of computation. Therefore, it is necessary to remove nodes without key review elements, specifically including the following:

[0140] Obtain the constructed node relationship graph and the set of key review elements;

[0141] Based on the correspondence between key review elements and nodes in the key review element set, nodes without key review elements are filtered out, resulting in nodes to be processed.

[0142] The nodes to be processed are then processed and restructured to obtain a node relationship graph containing only the key review element set. The methods for processing the nodes to be processed include:

[0143] Delete the leaf nodes in the node list that do not contain key review elements in the node relationship diagram;

[0144] For non-leaf nodes in the node list that do not have key review elements in the node relationship diagram, if all child nodes are invalid, remove the entire branch containing the node; if there are valid child nodes in the non-leaf node, retain the non-leaf node as a structural placeholder.

[0145] During the structural reorganization process, the node hierarchy is recalibrated, and the parent-child node relationships are updated. For example, a node marked as "invalid" (i.e., lacking key review elements) may have one or more "valid" child nodes (i.e., child nodes have associated key review elements). The handling strategy is to retain this invalid node as a "structural placeholder."

[0146] A node is marked as "invalid," and all its child nodes, and even all descendant nodes under the entire branch, are also invalid (i.e., the entire branch has no key review elements). Remove this invalid node and its entire invalid branch from the structure diagram. This method completes the structural reorganization, forming a more concise node relationship diagram for use in subsequent steps.

[0147] S400, for each key review element in the text data set, each node in the structural framework set, and each text in the text data set, a key review element ID, a node ID, and a text ID are created respectively, and logical associations are established based on semantic analysis or a rule engine to obtain the mapping relationship M of key review element ID, node ID, and text ID.

[0148] Each of the mapping relationships includes a triplet relationship of key review element-node-text.

[0149] Specifically, obtaining the mapping relationship M between key review element IDs, node IDs, and text IDs may include the following steps:

[0150] S401, for each key review element in the text data set, each node in the structural framework set, and each text in the text data set, a key review element ID, a node ID, and a text ID are created respectively.

[0151] Specifically, each entity is assigned a globally unique identifier (ID) for accurate subsequent association.

[0152] It should be noted that "text" refers to policy documents, industry standards, and corporate specifications for power grid infrastructure projects. This represents a progressively decreasing granularity between text, nodes, and elements. From text to nodes, the framework is established; from text to elements, content is filled in; and nodes and elements are mapped and integrated.

[0153] S402, using semantic similarity calculation methods or predefined rule engines, logically associate elements and nodes, use text source analysis to logically associate elements and text, and determine the attribution of nodes and text to form logical associations between nodes and text, thereby obtaining the mapping relationship between the element ID and the node ID, the mapping relationship between the element ID and the text ID, and the mapping relationship between the node ID and the text ID.

[0154] Specifically, establishing logical relationships between entities through multi-dimensional analysis is divided into three sub-tasks:

[0155] The association between elements and nodes is achieved using semantic similarity calculation or a rule engine. Element names (e.g., "Technical Solution Comparison") and node names (e.g., "Chapter 3 Technical Solution") are converted into word vectors (using models such as Word2Vec), and cosine similarity is calculated. If the similarity exceeds a preset threshold (e.g., 0.8), an association is established. The rule engine uses predefined mapping rules (e.g., "All economic elements must be associated with the 'Economic Benefit Analysis' node") to directly bind elements through pattern matching. This results in a set of mapping relationships between element IDs and node IDs.

[0156] The association between elements and text is based on text source analysis. Through keyword matching or contextual analysis, the source text of an element is determined. For example, the element "N-1 verification" might originate from a text paragraph (the content of which is "The report requires N-1 security verification"). This records the element's location within the text, establishing a direct association.

[0157] The association between nodes and text is based on node attribution analysis. The node structure itself originates from management regulations (e.g., the node "1.1 Necessity of Construction" corresponds to a certain text clause). The system directly establishes attribution relationships by parsing the structural framework source in step S200, obtaining a set of mapping relationships between node IDs and text IDs.

[0158] S403, based on the mapping relationship between the element ID and the node ID, the mapping relationship between the element ID and the text ID, and the mapping relationship between the node ID and the text ID, a mapping relationship M relating the key review element ID, node ID, and text ID is obtained.

[0159] This step integrates discrete mapping relationships into a unified triple mapping relationship M. The construction of mapping relationship M achieves a closed loop from data identification to logical association. S401 ensures the traceability of entities, S402 establishes precise associations through semantics and rules, and S403 integrates them into an operable triple mapping. This process not only improves the efficiency of data management but also ensures that the feasibility study report template fully covers the review rules in terms of content and strictly follows the hierarchical relationship in terms of structure, providing core data support for the standardized review of power grid infrastructure projects.

[0160] S500, Based on the mapping relationship M, form a feasibility study report structure diagram and create a feasibility study report template for power grid infrastructure projects.

[0161] Specifically, creating a feasibility study report template for a power grid infrastructure project may include the following steps:

[0162] S501, Based on the mapping relationship M, establish the element title of the node and the element content corresponding to the node in text units to form the feasibility study report structure diagram.

[0163] Specifically, the node IDs and their relationships in the mapping relationship M are parsed to restore the complete tree hierarchy. For each node, the corresponding key review elements are attached according to the mapping in M. The element content is stored in the form of structured fields, including element name, filling requirements, example instructions, etc.

[0164] By using the associations in M, the source text number is labeled for each node or element, enhancing the compliance traceability of the template.

[0165] S502, Load the structure diagram of the feasibility study report into the template to create a feasibility study report template for power grid infrastructure projects.

[0166] Choose the target template format based on your actual needs, such as a Microsoft Word template (.docx), XML Schema, or Markdown template. Word templates are easy for humans to edit, while XML templates are easy for machines to parse.

[0167] Convert the hierarchical relationship of nodes into the title style of the template (e.g., set the first-level title to the "Heading 1" style and the second-level title to the "Heading 2" style).

[0168] Under each heading, a placeholder is generated based on the content of the bound element. For example:

[0169] Insert formatted prompt text (such as "

Please fill in the technical solution comparison process here

[0170] Insert data-based elements into standardized tables (such as investment estimate templates).

[0171] Automatically add source text citations for elements to the template footer or footer. The final result is a standardized feasibility study report template for power grid infrastructure projects (e.g., a .docx file), which may include:

[0172] Preset chapter title structure;

[0173] Embedded element filling prompts and standardized forms;

[0174] Source of compliance information is clearly marked.

[0175] In one embodiment, after step S500, the method further includes:

[0176] S601, periodically poll the current file library to obtain the current file data, and update the mapping relationship M based on the current file data to obtain an updated feasibility study report structure diagram.

[0177] S602, using the updated feasibility study report structure diagram, updates the created power grid infrastructure project feasibility study report template.

[0178] Specifically, the current file library (a database or file system that stores the latest management regulations) is scanned at a preset cycle (such as monthly), and only the files that have changed are processed, avoiding full calculations and improving efficiency.

[0179] The modified document undergoes processes S100 to S400 (text structuring, rule extraction, and feature extraction) to generate a new local mapping relationship M. This new local mapping relationship M is then merged with the existing mapping relationship M, resolving conflicts: if the requirements of the old and new rules for the same node conflict (e.g., changes in node level or feature content), an authoritative priority strategy (national policy > industry standard > enterprise specification) is used for automatic adjudication. Related updates: if features or nodes extracted from the new document are semantically similar to existing content (calculated using word vector cosine similarity), the old data is overwritten; if they are entirely new content, they are appended to M.

[0180] Next, based on the updated mapping relationship M, step S501 of S500 (node-level reconstruction and element binding) is re-executed to generate a feasibility study report structure diagram that is synchronized with the latest policy requirements. Then, the feasibility study report structure diagram required by the latest policy is compared with the old feasibility study report structure diagram to identify the differences, and template updates are triggered only for the differences to reduce unnecessary global reconstruction.

[0181] Example

[0182] A structured design method for feasibility study reports of power grid infrastructure projects includes the following steps:

[0183] 1. Obtain project feasibility studies and related management regulations at the national, industry, enterprise, and special project levels related to power grid infrastructure construction;

[0184] 2. Process and analyze the unstructured data obtained in step 1), filter the relevant provisions for the feasibility study report structure outline, align the report structure nodes in each provision, and obtain the basic report structure that meets each provision;

[0185] ① Structure Acquisition: Using PDF structure parsing technology, extract the headings at each level specified in the report outline to determine the basic structural framework of the feasibility study report (see Table 1 for algorithm examples).

[0186]

[0187] Table 1

[0188] ② Structure Alignment: Take the titles at each level in the feasibility study report structure framework set(F) obtained in ① as nodes N, and the hierarchical relationship between titles as node association relationship R, store them in the Neo4j graph database, construct the power grid infrastructure feasibility study report structure network graph G = (N,R), and perform node alignment to obtain the basic structure of the feasibility study report that meets the requirements of each level.

[0189] 3. Screening 1) the relevant review rules for the feasibility study content of the power grid infrastructure project, and using jieba word segmentation and TF-IDF algorithm to identify and extract the key elements that each structure should contain;

[0190] Jieba is a popular Python library for Chinese text segmentation, known for its ease of use, flexibility, and high efficiency. It supports multiple segmentation modes, allows for custom dictionaries, and can perform keyword extraction and part-of-speech tagging, enabling professional segmentation processing in the power grid field.

[0191] The Term Frequency-Inverse Document Frequency (TF-IDF) algorithm is a statistical method widely used in information retrieval and text mining to evaluate the importance of a word in a document within a document set or corpus. It is also a widely used text feature extraction method. The TF-IDF algorithm calculates the importance of a word in a document using two main components: Term Frequency (TF) and Inverse Document Frequency (IDF). The basic principle of the TF-IDF algorithm is as follows:

[0192] ①Term frequency (TF) represents the frequency with which a term (keyword) appears in a document. This number is usually standardized to prevent it from being biased towards long documents. The formula is:

[0193] (1)

[0194] ② Inverse Document Frequency (IDF) represents the general importance of a term. The fewer documents containing a term, the larger the IDF, indicating that the term has good category discrimination ability and contributes more to the term's weight; conversely, the more documents containing a term, the smaller the contribution. The formula is:

[0195] (2)

[0196] ③TF-IDF is the product of TF and IDF to obtain the importance score of a term in a document.

[0197] By comprehensively applying the above methods and conducting manual screening, we obtain a set (E) of key review elements for each structure. i This enables the element-based representation of all levels of the feasibility study report for power grid infrastructure projects;

[0198] 4. Analyze the execution paths of 2) and 3), clarify and construct the mapping relationship between the element set, feasibility study report structure and relevant regulations, and lay the foundation for the operation of the subsequent adjustment mechanism;

[0199] 5. Establish a structured adjustment mechanism for power grid infrastructure feasibility study reports, synchronize updates to power grid infrastructure project management regulations at all levels, and achieve a structured design of power grid infrastructure feasibility study reports that meets the requirements of regulations and review at all levels (see Table 2 for algorithm examples). The results are as follows: Figure 2 .

[0200]

[0201] Table 2

[0202] In summary, the structured design method for feasibility study reports of power grid infrastructure projects proposed in this application has the following beneficial effects:

[0203] This application aims to standardize and normalize feasibility study data, enhance the efficiency and accuracy of data management, reduce redundant information, and improve review efficiency. Compared with existing technologies, the main advantage of this application is "structured design of feasibility study reports oriented towards management and review needs."

[0204] In the power grid sector, power grid projects in the reserve phase require rigorous feasibility study reviews to examine their compliance and feasibility. In this process, the feasibility study report is a crucial basis for the review process and represents key data in the feasibility study phase; therefore, the management requirements and review standards are extremely stringent.

[0205] Therefore, this application, through its structured design, ensures that the report content comprehensively covers the regulations at all levels, improves review efficiency, reduces compliance risks, and effectively supports the scientific decision-making and efficient implementation of power grid infrastructure projects.

[0206] Secondly, this application provides a structured design system for feasibility study reports of power grid infrastructure projects, applied to the aforementioned structured design method for feasibility study reports of power grid infrastructure projects, including:

[0207] The acquisition unit is used to acquire document data related to power grid infrastructure management regulations from the historical document library to obtain a structured text data set, wherein the document data includes at least policy documents, industry standards and enterprise specifications for power grid infrastructure projects;

[0208] A structured unit is used to extract structured information from the text data set to obtain a set of text structure frameworks, wherein the set of structure frameworks includes at least a node relationship diagram formed by title name, title level and association relationship, and each node is composed of multiple key review elements;

[0209] The extraction unit is used to extract key review elements of the feasibility study content review rules from the text data set to obtain a key review element set, wherein the key review element set includes a list of review elements corresponding to the nodes.

[0210] The logical association unit is used to create a key review element ID, node ID, and text ID for each key review element in the text data set, each node in the structural framework set, and each text in the text data set, respectively, and establish logical associations based on semantic analysis or a rule engine to obtain a mapping relationship M between the key review element ID, node ID, and text ID, wherein each mapping relationship includes a triple relationship of key review element-node-text.

[0211] The result unit is used to generate a feasibility study report structure diagram and create a feasibility study report template for power grid infrastructure projects based on the mapping relationship M.

[0212] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0213] Obviously, those skilled in the art should understand that the various units or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps into a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0214] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A structured design method for feasibility study reports of power grid infrastructure projects, characterized in that, include: Retrieve relevant management regulations for power grid infrastructure from the historical document library to obtain a structured text data set, wherein the document data includes at least policy documents, industry standards and enterprise specifications for power grid infrastructure projects; The text data set is subjected to structured information extraction to obtain a set of text structure frameworks, wherein the set of structure frameworks includes at least a node relationship diagram formed by title name, title level and association relationship, and each node is composed of multiple key review elements; The key review elements of the feasibility study content review rules are extracted from the text data set to obtain a key review element set, wherein the key review element set includes a list of review elements corresponding to the nodes. For each key review element in the text data set, each node in the structural framework set, and each text in the text data set, a key review element ID, a node ID, and a text ID are created respectively. Logical associations are established based on semantic analysis or a rule engine to obtain a mapping relationship M between key review element ID, node ID, and text ID. Each mapping relationship includes a triple relationship of key review element-node-text. Based on the mapping relationship M, a feasibility study report structure diagram is formed, and a template for a feasibility study report of a power grid infrastructure project is created.

2. The structured design method for feasibility study reports of power grid infrastructure projects as described in claim 1, characterized in that, The step of obtaining a structured text data set by retrieving relevant management regulations for power grid infrastructure from a historical document library includes: Searching the historical document database yields policy documents, industry standards, and corporate specifications related to power grid infrastructure management, forming document data. Using a file parsing algorithm, the text content of the file data is extracted and the data is cleaned to obtain intermediate text data, wherein the intermediate text data contains clean text content. Based on the natural paragraphs and chapter boundaries of the intermediate text data, text structuring processing of logical blocks is performed to obtain a text data set.

3. The structured design method for feasibility study reports of power grid infrastructure projects as described in claim 1, characterized in that, The step of extracting structured information from the text data set to obtain a set of text structure frameworks includes: Construct a structured rule matching library, wherein the structured rule matching library includes at least font feature information, indentation information, heading numbering pattern, positional relationship and level identifier, for identifying heading levels in text; The structured rule matching library is matched with the text data set, the title name and title level are identified through feature parsing, and the association relationship is constructed based on the title level relationship to obtain a node relationship graph, wherein each node in the node relationship graph corresponds to multiple key review elements; The nodes and relationships in the node relationship graph are extracted to obtain the set of text structure frameworks.

4. The structured design method for feasibility study reports of power grid infrastructure projects as described in claim 3, characterized in that, The step of matching the structured rule matching library with the text data set, identifying the title name and title level through feature parsing, and constructing an association relationship based on the title level relationship to obtain a node relationship graph includes: By using conflict resolution methods, nodes in different documents, such as policy documents, industry standards, and corporate specifications of power grid infrastructure projects, are processed to obtain a node relationship diagram after conflict resolution.

5. The structured design method for feasibility study reports of power grid infrastructure projects as described in claim 1, characterized in that, The step of extracting key review elements from the text data set according to the feasibility study content review rules to obtain a key review element set includes: Using regular expressions and a list of semantic trigger words, the feasibility study content review rules are extracted from the text data set to obtain the feasibility study content review rules. The key review elements in the feasibility study content review rules are extracted to obtain a set of key review elements.

6. The structured design method for feasibility study reports of power grid infrastructure projects as described in claim 5, characterized in that, The step of extracting key review elements from the feasibility study content review rules to obtain a set of key review elements includes: The feasibility study content review rules were analyzed using jieba word segmentation to convert the data into a word sequence. The TF-IDF algorithm is used to identify the key elements of the term sequence, thereby obtaining a set of key review elements.

7. The structured design method for feasibility study reports of power grid infrastructure projects as described in claim 1, characterized in that, The steps of creating a key review element ID, node ID, and text ID for each key review element in the text data set, each node in the structural framework set, and each text in the text data set, and establishing logical relationships based on semantic analysis or a rule engine to obtain the mapping relationship M between the key review element ID, node ID, and text ID include: For each key review element in the text data set, each node in the structural framework set, and each text in the text data set, a corresponding key review element ID, node ID, and text ID are created respectively; Using semantic similarity calculation methods or predefined rule engines, logical associations are formed between elements and nodes, and text source analysis is used to logically associate elements and texts, as well as the attribution of nodes and texts, thus forming logical associations between nodes and texts. This yields the mapping relationships between element IDs and node IDs, element IDs and text IDs, and node IDs and text IDs. Based on the mapping relationship between the element ID and the node ID, the mapping relationship between the element ID and the text ID, and the mapping relationship between the node ID and the text ID, a mapping relationship M is obtained that links the key review element ID, node ID, and text ID.

8. The structured design method for feasibility study reports of power grid infrastructure projects as described in claim 1, characterized in that, The steps of forming a feasibility study report structure diagram and creating a feasibility study report template for power grid infrastructure projects based on the mapping relationship M include: Based on the mapping relationship M, the element titles of the nodes and the element content corresponding to the nodes are established in units of text, forming the structure diagram of the feasibility study report; Load the structure diagram of the feasibility study report into the template to create a feasibility study report template for power grid infrastructure projects.

9. The structured design method for feasibility study reports of power grid infrastructure projects as described in claim 1, characterized in that, After the steps of forming a feasibility study report structure diagram based on the mapping relationship M and creating a feasibility study report template for power grid infrastructure projects, the following steps are included: The current file library is periodically polled to obtain the current file data, and the mapping relationship M is updated based on the current file data to obtain an updated feasibility study report structure diagram; The updated feasibility study report structure diagram is used to update the template for the created power grid infrastructure project feasibility study report.

10. A structured design system for feasibility study reports of power grid infrastructure projects, characterized in that, The structured design method for feasibility study reports of power grid infrastructure projects applied to any one of claims 1-9 includes: The acquisition unit is used to acquire document data related to power grid infrastructure management regulations from the historical document library to obtain a structured text data set, wherein the document data includes at least policy documents, industry standards and enterprise specifications for power grid infrastructure projects; A structured unit is used to extract structured information from the text data set to obtain a set of text structure frameworks, wherein the set of structure frameworks includes at least a node relationship diagram formed by title name, title level and association relationship, and each node is composed of multiple key review elements; The extraction unit is used to extract key review elements of the feasibility study content review rules from the text data set to obtain a key review element set, wherein the key review element set includes a list of review elements corresponding to the nodes. The logical association unit is used to create a key review element ID, node ID, and text ID for each key review element in the text data set, each node in the structural framework set, and each text in the text data set, respectively, and establish logical associations based on semantic analysis or a rule engine to obtain a mapping relationship M between the key review element ID, node ID, and text ID, wherein each mapping relationship includes a triple relationship of key review element-node-text. The result unit is used to generate a feasibility study report structure diagram and create a feasibility study report template for power grid infrastructure projects based on the mapping relationship M.

Images