Intelligent management method and system for in-service inspection outline of nuclear power plant

Abstract

The invention discloses an intelligent management method and system for an in-service inspection outline of a nuclear power plant. The method comprises the following steps: S0, collecting and establishing an outline knowledge base related to an in-service inspection outline of a nuclear power plant; s1, generating a nuclear power plant in-service inspection outline according to the outline knowledge base; s2, performing intelligent review on the nuclear power plant in-service inspection outline according to the outline knowledge base, generating a review report, and generating a question list according to the review report; the step of intelligently evaluating the in-service inspection outline of the nuclear power plant and generating the evaluation report comprises the substeps of obtaining a reference in-service inspection outline, intelligently comparing the reference in-service inspection outline with the in-service inspection outline of the nuclear power plant, and labeling difference contents to generate the evaluation report. By implementing the technical scheme of the invention, differentiated contents can be labeled, the review report is intelligently generated, and the management efficiency of the in-service inspection outline is improved.

Description

Technical Field

[0001] This invention relates to the field of in-service inspection of nuclear power plants, and more particularly to an intelligent management method and system for in-service inspection outlines of nuclear power plants. Background Technology

[0002] The in-service inspection outline for nuclear power plants is a crucial technical document for ensuring nuclear safety. Work directly related to this outline includes its drafting and review. Drafting is undertaken by the nuclear power plant and its supporting units, while review is conducted by nuclear safety regulatory authorities and their supporting units. Currently, the drafting and review of the in-service inspection outline primarily rely on manual processes. Review requires meticulous word-by-word verification, and the creation of a problem list addressing errors, questions, or suggestions. This problem list must be sent to the power plant for response or clarification. Upon receiving the plant's response, the corresponding sections of the outline must be reviewed based on their feedback. This process is time-consuming, inefficient, and of inconsistent quality. Furthermore, the in-service inspection outline, problem lists, and response documents are independent and isolated, lacking effective integration and correlation. This necessitates repeated searching and organization of relevant information when developing downstream documents (such as the unit in-service inspection plan outline), increasing workload, reducing efficiency, and increasing the risk of errors due to inconsistencies. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an intelligent management method and system for the in-service inspection outline of nuclear power plants.

[0004] The technical solution adopted by this invention to solve its technical problem is: to construct an intelligent management method for the in-service inspection outline of a nuclear power plant, the method comprising the following steps: S0. Collect and establish a knowledge base related to the in-service inspection outline of the nuclear power plant; S1. Generate the nuclear power plant in-service inspection outline based on the outline knowledge base; S2. Perform intelligent review of the nuclear power plant in-service inspection outline based on the outline knowledge base, generate an review report, and generate a problem list based on the review report; the intelligent review of the nuclear power plant in-service inspection outline and generation of the review report includes: obtaining a benchmark in-service inspection outline, performing intelligent comparison with the nuclear power plant in-service inspection outline, marking the differences, and generating the review report.

[0005] Furthermore, in step S0, information related to the in-service inspection outline of the nuclear power plant is collected, classified, and organized to form the outline knowledge base. The outline knowledge base includes: drawings and design manuals of nuclear power plant mechanical equipment, laws, regulations, standards, and regulatory requirements.

[0006] Furthermore, in step S2, the intelligent review also includes: performing format checks, content completeness checks, and / or compliance checks on the in-service inspection outline.

[0007] Furthermore, in step S2, the intelligent comparison with the nuclear power plant in-service inspection outline includes: establishing a database of equipment correspondences for multiple power plants, and performing intelligent comparison based on the database of equipment correspondences.

[0008] Furthermore, the differences include formal differences and essential differences; in step S2, the labeling of the differences includes: classifying and labeling the differences according to their degree of impact on nuclear safety.

[0009] Furthermore, after step S2, the method further includes: obtaining the power plant's response document to the question form, updating the outline knowledge base and / or the corresponding nuclear power plant in-service inspection outline based on the response document, and recording the data before and after the update.

[0010] Furthermore, the method also includes: obtaining historical response files from multiple power plants to historical problem forms, performing intelligent analysis on the multiple historical problem forms and the historical response files, and establishing a common problem response knowledge base.

[0011] Further, step S1 includes: A preset template for the nuclear power plant in-service inspection outline is generated based on the outline specifications; The corresponding content is extracted from the outline knowledge base using natural language processing technology and filled into the preset template to generate the nuclear power plant in-service inspection outline.

[0012] Furthermore, in step S0, when the content of the outline knowledge base changes, the corresponding content of the outline knowledge base is updated. In step S1, the following is also included: when the content of the outline knowledge base changes, a self-assessment is performed on the nuclear power plant in-service inspection outline, and the associated data is updated based on the results of the self-assessment.

[0013] This invention also constructs an intelligent management system for the in-service inspection outline of nuclear power plants, the system comprising: The knowledge base module is used to collect and establish a knowledge base related to the in-service inspection outline of the nuclear power plant. The outline generation module is used to generate the in-service inspection outline of the nuclear power plant based on the outline knowledge base. The intelligent review module is used to conduct an intelligent review of the nuclear power plant in-service inspection outline based on the outline knowledge base, generate an review report, and generate a problem list based on the review report. The intelligent review of the nuclear power plant in-service inspection outline and the generation of the review report include: obtaining a benchmark in-service inspection outline, conducting an intelligent comparison with the nuclear power plant in-service inspection outline, marking the differences, and generating the review report.

[0014] Implementing this invention has the following beneficial effects: By integrating information related to multiple power plants and in-service inspection outlines to form an outline knowledge base, information silos can be broken down, and information integration and sharing can be achieved; generating in-service inspection outlines based on the integrated outline knowledge base can improve writing efficiency and ensure outline quality; during the review process, intelligent comparison with the benchmark in-service inspection outline is performed, and the differences in the in-service inspection outline are marked, generating a review report for reviewers to refer to, eliminating the need for word-by-word review of the entire outline document, thereby improving review efficiency; intelligent review reduces human error and improves the quality and stability of in-service inspection outlines. Attached Figure Description

[0015] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 This is a flowchart of an intelligent management method for in-service inspection outlines of nuclear power plants, as described in one embodiment of the present invention. Figure 2 This is a schematic diagram of a template for an inspection item table of an in-service inspection outline for nuclear power plants, according to one embodiment of the present invention. Detailed Implementation

[0016] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart of an intelligent management method for in-service inspection outlines of nuclear power plants according to one embodiment of the present invention. This embodiment includes the following steps: S0. Collect and establish a knowledge base of outlines related to the in-service inspection outlines of nuclear power plants; In this step, it should be understood that the in-service inspection outline for nuclear power plants needs to be generated based on relevant documents such as laws, regulations, standards, and regulatory requirements, and is formulated according to the actual equipment and design of the corresponding power plant. During the outline review, reviewers also need to refer to the relevant information and regulatory requirements of the corresponding power plant for the review. The outline knowledge base includes the documents and information required for the in-service inspection outline. The data collection process can be achieved by manually uploading relevant information files to form the outline knowledge base, or by the outline knowledge base automatically retrieving relevant information.

[0018] In some optional embodiments, in step S0, information related to the in-service inspection outline of nuclear power plants is collected, categorized, and organized to form an outline knowledge base. The outline knowledge base may include: drawings of nuclear power plant mechanical equipment, design manuals, laws, regulations, standards, and regulatory requirements. Specifically, regulatory requirements are requirement documents issued by nuclear safety regulatory authorities, including safety review requirements. In step S0, the collected information can be categorized according to information type and power plant based on usage needs, such as categorizing by information type into legal and regulatory databases, standard databases, equipment information databases, and regulatory requirement databases. For outlines from different power plants, some content is the same, such as laws and regulations, which can be grouped into a common knowledge base. Some content may differ, such as equipment drawings and regulatory requirements. This differentiated content can be categorized by different power plants, with differentiated content from the same power plant grouped into a differentiated knowledge base. The relevant information for each power plant can be maintained by the respective power plant itself. It should be understood that the outline knowledge base is not limited to the above content and can be updated, added, or deleted according to actual needs. By establishing an outline knowledge base that distinguishes between public and differentiated knowledge bases, a complete and highly applicable outline knowledge base was created for each power plant.

[0019] S1. Generate an in-service inspection outline for nuclear power plants based on the outline knowledge base; In this step, an in-service inspection outline for the nuclear power plant is generated based on the outline specification requirements document in the outline knowledge base and relevant information about the corresponding power plant. In some optional embodiments, step S1 may include: generating a preset template of the in-service inspection outline for the nuclear power plant according to the outline specification; and extracting corresponding content from the outline knowledge base using natural language processing technology to fill the preset template, thereby generating the in-service inspection outline for the nuclear power plant. Specifically, the outline specification requires that the in-service inspection outline for the nuclear power plant must conform to the standardized format and content requirements stipulated by the nuclear safety regulatory authority. The format requirements require that the components and section divisions of the in-service inspection outline conform to the fixed format requirements issued by the nuclear safety regulatory authority; the content requirements require that the full text of the in-service inspection outline should include content such as the main text, Appendix 1, Appendix 2, etc. A preset template can be generated based on the above outline specifications. Then, using natural language processing techniques such as LSTM and Transformer deep learning models, corresponding information is extracted from the outline knowledge base and populated into the preset template to generate the nuclear power plant in-service inspection outline. The format is then adjusted to conform to the required style, including font, font size, and layout. The nuclear power plant in-service inspection outline can include text, tables, and flowcharts, with the content format determined according to the outline specifications. For example, for an inspection table for a project in the nuclear power plant in-service inspection outline, the required fields of the table are already determined according to the outline specifications, such as... Figure 2As shown, including but not limited to the fields shown in the figure, the required content for each field is extracted and generated from the knowledge base. For example, the field describing the inspected part is determined according to the specifications and drawings and design manuals of nuclear power plant mechanical equipment. It can be extracted from the corresponding power plant specifications, mechanical equipment drawings and design manuals in the outline knowledge base using natural language processing technology. Further, combined with other specifications, information such as the inspection quantity and scope, inspection method, and access method corresponding to the inspected part is extracted and filled into the project inspection table. The source of the content is marked, such as the drawing number, design manual chapter, specification / standard chapter, regulatory requirement clause, etc., from which the field content originates. The table style is adjusted according to the specifications. In this embodiment, the in-service inspection outline of a nuclear power plant is automatically generated using natural language processing technology, eliminating the need for manual review of numerous documents, reducing the workload and time of manual writing, improving writing efficiency, and ensuring that the generated content meets the outline requirements, thus improving the quality of the in-service inspection outline.

[0020] In some optional embodiments, when the content of the outline knowledge base changes in step S0, the corresponding content of the outline knowledge base is updated. Step S1 further includes: when the content of the outline knowledge base changes, a self-assessment of the nuclear power plant in-service inspection outline is performed, and the associated data is updated based on the self-assessment results. Specifically, when the source of the content of documents such as standards and laws and regulations in the outline knowledge base changes, or when the equipment of the nuclear power plant changes (e.g., increases or decreases in quantity), or when a major anomaly occurs during the inspection of the nuclear power plant, the outline knowledge base needs to be updated to ensure its accuracy. In step S1, the knowledge base content can be monitored periodically or detected as needed to determine whether changes have occurred, and a self-assessment of the generated nuclear power plant in-service inspection outline can be performed based on the changes in the knowledge base content. This includes comparing the knowledge base content and the nuclear power plant in-service inspection outline, outputting a difference report, and updating the nuclear power plant in-service inspection outline. When the content in the nuclear power plant in-service inspection outline changes, its corresponding external data may also change. A corresponding relationship data table can be maintained, and the associated data can be further updated based on the self-assessment results and the corresponding relationship data table. For example, when the inspection content, including scope, methods, and cycle, changes in the nuclear power plant's in-service inspection outline, the corresponding human and equipment resources required to perform the inspection will also change. The information in the corresponding relational data table is retrieved, and the human and equipment resources corresponding to the changed content in the outline are updated. Through outline self-assessment and related data updates, the outline content can be updated in a timely manner, ensuring its accuracy. Simultaneously, related data is updated based on changes in the outline, forming a closed-loop update process.

[0021] S2. Intelligently review the in-service inspection outline of the nuclear power plant based on the outline knowledge base and generate an review report, and generate a question sheet based on the review report; the intelligent review of the in-service inspection outline of the nuclear power plant and the generation of a review report include: obtaining the benchmark in-service inspection outline, intelligently comparing it with the in-service inspection outline of the nuclear power plant, marking the differences, and generating a review report.

[0022] In this step, the nuclear power plant's in-service inspection outline can be either automatically generated using natural language processing technology based on the outline knowledge base, or it can be manually generated. It should be noted that for automatically generated in-service inspection outlines, if the corresponding knowledge base content changes after generation but before review, and the in-service inspection outline has not been updated, discrepancies will exist. Therefore, intelligent review based on the latest outline knowledge base content is required during the review process. The benchmark in-service inspection outline can be either automatically generated and reviewed, or manually generated and reviewed. The benchmark in-service inspection outline serves as a standard compliance outline, which other power plants can use as a reference during its development and review. It should be understood that the in-service inspection outlines of different power plants will vary due to differences in equipment and regulatory requirements, but there are also many common parts, such as laws and regulations, and identical equipment. In this embodiment, during intelligent review, a benchmark in-service inspection outline is obtained and intelligently compared with the in-service inspection outline of the nuclear power plant to be reviewed, which is automatically generated or manually generated. The differences are marked, and finally, an review report is automatically generated. The review report can provide an intuitive comparison of differences. Reviewers do not need to check word by word, and common parts can be exempted from review. Only the differences need to be focused on, which improves review efficiency and reduces repetitive work and human error.

[0023] In some optional embodiments, step S2, the intelligent review further includes: performing format checks, content completeness checks, and / or compliance checks on the in-service inspection outline. Specifically, intelligent review can be performed using machine learning algorithms. The format check verifies whether the format of the nuclear power plant's in-service inspection outline meets the requirements of the nuclear safety regulatory authority, including font, font size, and layout. The content completeness check verifies whether the nuclear power plant's in-service inspection outline covers all necessary content, such as laws, regulations, standards, and equipment information. The compliance check verifies whether the content of the nuclear power plant's in-service inspection outline meets the review requirements of the nuclear safety regulatory authority and whether there are any violations or non-compliance issues.

[0024] In some optional embodiments, in step S2, the intelligent comparison with the nuclear power plant's in-service inspection outline includes: establishing a database of equipment correspondences among multiple power plants, and performing intelligent comparison based on the database. Specifically, the most significant differences between the in-service inspection outlines of different power plants stem from equipment differences. Different power plants may name the same equipment differently, but in essence, they are the same. In this embodiment, a database of equipment correspondences between power plants can be established. For example, for the same weld, the benchmark in-service inspection outline might name it weld A0, power plant A's weld A1, and power plant B's weld A2. Weld A0, weld A1, and weld A2 are then mapped to establish a database of equipment correspondences. When performing intelligent comparison, for power plant A's in-service inspection outline, weld A1 is considered equivalent to weld A0 in the benchmark in-service inspection outline; for power plant B's in-service inspection outline, weld A2 is considered equivalent to weld A0 in the benchmark in-service inspection outline, without any difference marking, or only marked as a different name. By establishing device correspondences, the reliability and accuracy of labeled content can be further improved, thereby increasing review efficiency.

[0025] In some optional embodiments, the differences include formal differences and essential differences. In step S2, labeling the differences includes classifying and labeling them according to their impact on nuclear safety. Specifically, formal differences may include differences in descriptive language, such as adding or removing words without substantial impact, equivalent expressions like "every 12 months" and "every year," different units but the same substantive meaning, etc., which can be interpreted with the assistance of artificial intelligence. Essential differences may include differences in equipment design, differences in inspection requirements, etc., such as different equipment structures, the need to add or remove an inspection of a weld, adding or reducing inspection steps, changing the inspection frequency in the inspection requirements from "annual" to "quarterly," etc. Semantic training is used to identify and distinguish between formal and essential differences, and the differences are further classified and labeled according to their impact on nuclear safety. In this embodiment, a preliminary judgment can be made based on the equipment or system. Since different equipment or systems have varying degrees of impact on nuclear safety, different importance coefficients or risk weight coefficients can be assigned to them. In nuclear power plants, equipment and systems have corresponding nuclear safety levels. For example, equipment or systems with a nuclear safety level of 1 are assigned an importance coefficient of 1, while those with a nuclear safety level of 3 are assigned an importance coefficient of 0.6. Secondly, the identified differences are further classified according to their type. For example, differences involving missing or newly added items are classified as the first category; differences in inspection, testing, or monitoring requirements due to different methods or inconsistent cycles are classified as the second category; and differences involving only text descriptions, document formats, or inconsistent expressions are classified as the third category. A graded labeling is then created by combining the equipment or system coefficients and the type of difference. This grading of the differences can serve as a review reference, focusing on high-level differences to avoid overlooking important equipment differences and improve the reliability and efficiency of the review. It should be understood that the above grading and classification can be adjusted according to actual needs and is not limited to the content mentioned in this embodiment.

[0026] In some optional embodiments, after step S2, the method further includes: obtaining the power plant's response document to the problem form; updating the outline knowledge base and / or the corresponding nuclear power plant in-service inspection outline based on the response document; and recording the data before and after the update. Specifically, the power plant's response to the problem form may include modifications to the content of the in-service inspection outline. For example, if a piece of equipment requires an increased inspection frequency due to an anomaly, the power plant's response may be to change the inspection frequency to monthly. In this embodiment, after obtaining the response document, the corresponding in-service inspection outline is updated based on the modifications to the content of the response document. If the update involves changes to the outline knowledge base, such as design drawings, the outline knowledge base is updated, and the data before and after the update is recorded for traceability and comparison.

[0027] In some optional embodiments, the method further includes: acquiring historical response files from multiple power plants to historical problem forms, performing intelligent analysis on the multiple historical problem forms and historical response files, and establishing a common problem response knowledge base. Specifically, some common problems exist in the power plant response files. In this embodiment, machine learning algorithms are used to intelligently analyze multiple historical problem forms and historical response files, sorting out the common problems and their corresponding responses to establish a common problem response knowledge base. This can provide intelligent recommendations for subsequent responses from other power plants, helping to quickly resolve similar problems. After updating the outline knowledge base and / or the corresponding nuclear power plant in-service inspection outline based on the response files, and recording the data before and after the update, the updated data can also be associated with the common problem response knowledge base in this embodiment, and intelligent analysis can be used to generate suggestions for modifying the responses.

[0028] The intelligent management system for nuclear power plant in-service inspection outlines of the present invention can be used to execute the intelligent management method for nuclear power plant in-service inspection outlines described in the above embodiments. The intelligent management system for nuclear power plant in-service inspection outlines in this embodiment includes: The knowledge base module is used to collect and build a knowledge base related to the in-service inspection outline of nuclear power plants; The outline generation module is used to generate an in-service inspection outline for nuclear power plants based on the outline knowledge base. The intelligent review module is used to intelligently review the in-service inspection outline of nuclear power plants based on the outline knowledge base, generate a review report, and generate a problem list based on the review report. The intelligent review of the in-service inspection outline of nuclear power plants and the generation of a review report include: obtaining a benchmark in-service inspection outline, intelligently comparing it with the in-service inspection outline of nuclear power plants, marking the differences, and generating a review report.

[0029] It is understood that the above embodiments only illustrate preferred embodiments of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can freely combine the above technical features without departing from the concept of the present invention, and can also make several modifications and improvements, all of which fall within the protection scope of the present invention. Therefore, all equivalent transformations and modifications made with respect to the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. A method for intelligent management of in-service inspection outlines for nuclear power plants, characterized in that, include: S0. Collect and establish a knowledge base related to the in-service inspection outline of the nuclear power plant; S1. Generate the nuclear power plant in-service inspection outline based on the outline knowledge base; S2. Perform intelligent review of the nuclear power plant in-service inspection outline based on the outline knowledge base, generate an review report, and generate a question sheet based on the review report; The intelligent review and generation of a review report for the in-service inspection outline of the nuclear power plant includes: obtaining a benchmark in-service inspection outline, intelligently comparing it with the in-service inspection outline of the nuclear power plant, marking the differences, and generating the review report.

2. The intelligent management method for the in-service inspection outline of nuclear power plants according to claim 1, characterized in that, In step S0, information related to the in-service inspection outline of the nuclear power plant is collected, classified, and organized to form the outline knowledge base. The outline knowledge base includes: drawings and design manuals of nuclear power plant mechanical equipment, laws, regulations, standards, and regulatory requirements.

3. The intelligent management method for the in-service inspection outline of nuclear power plants according to claim 1, characterized in that, In step S2, the intelligent review also includes: performing format checks, content integrity checks, and / or compliance checks on the in-service inspection outline.

4. The intelligent management method for the in-service inspection outline of nuclear power plants according to claim 1, characterized in that, In step S2, the intelligent comparison with the nuclear power plant in-service inspection outline includes: establishing a database of equipment correspondences for multiple power plants, and performing intelligent comparison based on the database of equipment correspondences.

5. The intelligent management method for the in-service inspection outline of nuclear power plants according to claim 4, characterized in that, The differences include formal differences and essential differences; in step S2, the labeling of the differences includes: classifying and labeling the differences according to their degree of impact on nuclear safety.

6. The intelligent management method for the in-service inspection outline of nuclear power plants according to claim 1, characterized in that, The process after step S2 further includes: obtaining the power plant's response document to the question form, updating the outline knowledge base and / or the corresponding nuclear power plant in-service inspection outline based on the response document, and recording the data before and after the update.

7. The intelligent management method for the in-service inspection outline of nuclear power plants according to claim 1, characterized in that, The method further includes: obtaining historical response files from multiple power plants to historical problem forms, performing intelligent analysis on the multiple historical problem forms and the historical response files, and establishing a common problem response knowledge base.

8. The intelligent management method for the in-service inspection outline of nuclear power plants according to claim 1, characterized in that, Step S1 includes: A preset template for the nuclear power plant in-service inspection outline is generated based on the outline specifications; The corresponding content is extracted from the outline knowledge base using natural language processing technology and filled into the preset template to generate the nuclear power plant in-service inspection outline.

9. The intelligent management method for the in-service inspection outline of nuclear power plants according to claim 8, characterized in that, In step S0, when the content of the outline knowledge base changes, the corresponding content of the outline knowledge base is updated. In step S1, the following is also included: when the content of the outline knowledge base changes, a self-assessment is performed on the nuclear power plant in-service inspection outline, and the associated data is updated based on the results of the self-assessment.

10. An intelligent management system for in-service inspection outlines of nuclear power plants, characterized in that, include: The knowledge base module is used to collect and establish a knowledge base related to the in-service inspection outline of the nuclear power plant. The outline generation module is used to generate the in-service inspection outline of the nuclear power plant based on the outline knowledge base. The intelligent review module is used to conduct intelligent review of the nuclear power plant in-service inspection outline based on the outline knowledge base, generate a review report, and generate a question sheet based on the review report; The intelligent review and generation of a review report for the in-service inspection outline of the nuclear power plant includes: obtaining a benchmark in-service inspection outline, intelligently comparing it with the in-service inspection outline of the nuclear power plant, marking the differences, and generating the review report.

Images