Nuclear power equipment health intelligent management system and method based on digital twinning
By integrating digital twin technology and large language models, the problem of incomplete equipment status information has been solved, enabling accurate judgment of equipment health status and intelligent maintenance solutions, thereby improving the reliability and maintenance efficiency of nuclear power equipment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SANMEN NUCLEAR POWER CO LTD
- Filing Date
- 2025-11-24
- Publication Date
- 2026-04-24
AI Technical Summary
Traditional equipment maintenance suffers from incomplete equipment status information and a lack of accurate judgment, resulting in low equipment reliability, high failure frequency, low maintenance efficiency, high costs, and a lack of effective equipment health management and maintenance solutions.
A smart health management system for nuclear power equipment based on digital twins is adopted. Through the integration of data storage, governance, graph display and application layer, it realizes equipment status visualization and fault prediction. Combined with knowledge graph and large language model, it provides intelligent maintenance solutions.
It enables comprehensive management of equipment operating environment data, accurate identification of fault causes, reduction of equipment failure probability, improvement of maintenance efficiency, reduction of manual workload, reduction of costs, and shortening of fault diagnosis time.
Smart Images
Figure CN121920987A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the fields of nuclear power plant equipment maintenance management, knowledge graph, large language model, artificial intelligence, three-dimensional model and digital twin technology, and specifically relates to a smart health management system and method for nuclear power equipment based on digital twin. Background Technology
[0002] Safe, reliable, stable, and economical operation is the lifeline of nuclear power plants. my country's nuclear power industry has shifted from construction to technical services, with equipment maintenance and health management being core components. For enterprises, the health of equipment and excellent maintenance performance play a decisive role in their economic interests. Preventive maintenance and upkeep are crucial for ensuring the lifeline of the power plant. Over 30 years of operation and maintenance in China, each nuclear power plant has established sound experience feedback mechanisms and accumulated rich experience. Nuclear power plant equipment maintenance is a comprehensive activity characterized by concentrated tasks, complex logic, wide scope, numerous coordination factors, and high management requirements.
[0003] Traditional equipment maintenance suffers from incomplete information regarding equipment status, faults, documents, spare parts, and tools due to differences in personnel and systems involved in equipment management, operation, and maintenance. This results in a lack of supervision over equipment health, absence of preventative maintenance, failure to adopt good practices (experience feedback), and incomplete risk assessments, leading to problems such as low equipment reliability, high failure frequency, low maintenance efficiency, high maintenance costs, long equipment downtime, and short equipment lifespan. The main issues include: incomplete management of equipment operating environment data, making it impossible to accurately determine whether equipment is malfunctioning; fragmented management of individual equipment performance indicators, failing to comprehensively reflect equipment health status; lack of management of data on sensitive equipment components, making it impossible to accurately determine the location of equipment failures; lack of systematic management of historical equipment data, making it impossible to predict failures from data fluctuations; lack of physical simulations of equipment maintenance plans, failing to ensure efficient and successful maintenance activities; inability to display equipment documents, manpower, spare parts, tools, and environmental status in real time; lack of effective management of historical case data, failing to implement timely and accurate maintenance plans; lack of customized equipment simulation exercises, failing to conduct comprehensive pre-testing of equipment; lack of systematic management of historical equipment knowledge, failing to leverage data value to create value; and lack of effective explicit management of maintenance experience, failing to make tacit knowledge explicit. Summary of the Invention
[0004] The purpose of this invention is to provide a smart health management system and method for nuclear power equipment based on digital twins. This system can predict the health status of equipment through fluctuations in twin data, accurately determine the cause and location of equipment failures, visualize the operating status of nuclear power plant equipment, recommend equipment failures and historical cases, provide intelligent recommendations and equipment maintenance solutions, associate equipment maintenance-related data with the characteristics of status changes to facilitate decision-making during maintenance windows, and enable rapid formulation and comparison of multiple equipment maintenance solutions through visualized digital model operation.
[0005] Technical solution to achieve the purpose of this invention:
[0006] A smart health management system for nuclear power equipment based on digital twins, the system comprising:
[0007] Data storage layer: Used to store device data, image data, equipment operation data, technical personnel data, document data, spare parts data, and environmental data, forming a twin data repository;
[0008] Data governance layer: Used to perform data governance on the data in the twin data repository through metadata management, data lineage management, data cleaning, and data format conversion technologies, generating governed data that can be accurately identified and correctly calculated by computers;
[0009] Data Graph Display Layer: Based on the data processed by the data governance layer, knowledge graph technology and large language model technology are applied to construct 3D models of equipment and equipment operating environment, display the relationships between data, customize data analysis results, and 3D vectorized data of equipment.
[0010] The data storage layer includes: a device physical data module, a device file data module, a device environment data module, a device failure case module, a device experience feedback module, and a tricks and best practices module.
[0011] The data governance includes standard dictionaries and data training and fusion for various types of data in the basic data storage layer of power plant equipment; vectorization is adopted for physical entities that are directional and measurable; multi-source data is cleaned and preprocessed for training and application of knowledge graphs and large language models; for individual ambiguous data, power plant experts can make judgments and corrections online to achieve unambiguous expression of data; the data governance layer includes a standard dictionary module, an encoding management module, a vector data module, and an expert confirmation module.
[0012] The data graph display layer utilizes knowledge graph technology, large language model technology, 3D modeling technology, and semantic association technology to map the processed data into the 3D model of the equipment, constructing a static 3D model of the equipment and a 3D model of the equipment's operating environment. It displays the relationships between data, customized data analysis results, and 3D vectorized data of the equipment. Applying machine learning algorithms, it automatically extracts feature data, intelligently identifies the current equipment status and changes in the external environment, and determines the cause of entity alarms. Functions include alarms triggered by power changes, equipment misoperation, or changes in external factors. It simulates the operation of one or more identical devices under the same environment using digital twins. By establishing a multi-parameter correlation model for vibration, temperature, pressure, and flow, it achieves single-dimensional, definitive threshold detection for individual devices. It enables comprehensive monitoring of multi-dimensional dynamic thresholds, improving anomaly identification and reducing the probability of equipment failure. The data graph display layer includes: 3D models, individual demonstration models, multi-device linkage demonstration models, and faulty equipment operation simulation models.
[0013] The system also includes a data application layer: based on historical data and real-time data or configuration data, it uses AI to simulate equipment operation scenarios, analyzes equipment failure points, and forms feasible optimal maintenance solutions by configuring historical failure maintenance plans, feedback on equipment failure troubleshooting experience, operating parameters of upstream and downstream equipment, and equipment maintenance environment parameters; the data application layer includes: equipment digital model operation module, digital recommendation solution module, solution digital exercise and testing module, parameter configuration adjustment module, and optimal solution recommendation module.
[0014] The system also includes a data value mining layer: used to display the number of operating key equipment in the power plant, the number of high-risk equipment failures, medium-risk equipment and equipment under repair, changes in equipment-related documents, tools, spare parts, and health clock data based on the virtual operation of the twin data of the data application layer, to detect the health trend of the equipment, and to adjust the inventory of spare parts and tools; the data value mining layer includes: a key equipment sensitive component trend module, a high-risk equipment failure status module, a medium-risk data status module, an equipment under inspection data trend module, and a technical personnel data status module.
[0015] The system also includes a data search engine layer: used for simple and advanced data retrieval, and intelligent push based on user profiles; the data search engine layer includes: a retrieval module, an intelligent push module, a login and authorization module, a publishing and navigation module, and a system maintenance module.
[0016] The system also includes a basic capability layer for user management, organizational structure management, permission management, log management, audit management, and security management. The basic capability layer includes: user management module, organizational structure management module, permission management module, log management, audit log module, and security management module.
[0017] A method for intelligent health management of nuclear power equipment based on digital twins, the method comprising:
[0018] Step 1: Data acquisition and storage to form a twin data repository;
[0019] Step 2: Data governance, generating processed data that can be accurately identified and correctly calculated by computers;
[0020] Step 3: Data visualization, constructing a 3D model of the equipment and a 3D model of the equipment's operating environment, displaying the relationships between data, customized data analysis results, and 3D vectorized data of the equipment;
[0021] Step 4: Data application. Based on historical data, real-time data, or AI configuration data, the system uses AI to simulate equipment operation scenarios, analyzes equipment failure points, optimizes solutions, and forms the best feasible maintenance plan.
[0022] The method also includes step 5, data value mining, which displays the status of sensitive components of key equipment in the power plant, the fault status of high-risk equipment, the status of medium-risk equipment, the status of equipment under inspection, and the data status of technical personnel.
[0023] The beneficial technical effects of this invention are as follows:
[0024] This invention provides a digital twin-based intelligent health management system for nuclear power equipment, which enables comprehensive management of equipment operating environment data and prediction of equipment health status; centralized management of equipment data and accurate identification of equipment failure causes; mapping of physical entities of nuclear power plants to virtual scenes for visualized equipment operation; management of power plant equipment component data for accurate location of equipment failures; systematic management of historical equipment data to predict equipment failures from data fluctuations; correlation and intelligent recommendation of equipment failures with historical cases; recommendation of maintenance plans for faulty equipment; visualized drills of equipment maintenance plans to mitigate risks; correlation of equipment maintenance-related data with status to aid in decision-making during maintenance windows; rapid formulation and comparison of multiple equipment maintenance plans; solid foundation for digital transformation of equipment health management through equipment data management; automatic report generation to reduce manual workload and labor costs; combination of intelligent and expert fault diagnosis to effectively shorten engineers' troubleshooting time; maintenance decisions based on actual equipment operating status to reduce over-maintenance and under-maintenance; and reduction of spare parts inventory. Attached Figure Description
[0025] Figure 1 This invention provides an architecture diagram of a smart health management system for nuclear power equipment based on digital twins.
[0026] Figure 2The present invention provides a flowchart of a method for intelligent health management of nuclear power equipment based on digital twins. Detailed Implementation
[0027] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.
[0028] In this specification, unless the context clearly indicates otherwise, the term "nuclear equipment" refers to an entity capable of independently performing a specific function within a nuclear power plant system during operation. The term "sensitive component" refers to the smallest replaceable unit within a nuclear equipment, typically a component, part, or auxiliary support loop. Failure of this unit can lead to failure of the equipment it belongs to, resulting in reactor shutdown, power outage, reduced power output, or significant power fluctuations. The term "large language model" is an artificial intelligence model based on deep learning, trained on large amounts of text data to learn and generate natural language text. Its characteristics include large-scale parameters, deep learning architecture, pre-training capabilities, fine-tuning flexibility, contextual understanding, multi-task learning, and high computational resource requirements.
[0029] This invention provides a digital twin-based intelligent health management system for nuclear power equipment. Utilizing digital twin technology, knowledge graphs, and large language model fusion technology, it constructs a simulated operation model of nuclear power equipment. Through virtual equipment operation, it intelligently analyzes equipment status and maintenance data, performs semantic understanding and knowledge reasoning, and transforms the data into intuitive equipment status indicators. It fully leverages historical and real-time twin data from power plant equipment, mapping relevant equipment data to the platform to form a simulated virtual equipment and its operating conditions. Equipment health is verified through fluctuations in actual data operation. When a fault occurs, the platform automatically issues an alarm and displays the fault source. Combining historical fault case libraries and expert diagnoses, it generates maintenance plans and allows for different data (element) configurations to form maintenance plans and drills, continuously improving and generating intelligent maintenance solutions. The platform can also digitally demonstrate maintenance plans, flexibly expanding element configurations to form new plans. In digital simulation, it shortens fault diagnosis time, improves equipment maintenance efficiency, reduces equipment fault troubleshooting risks, and achieves dynamic management of equipment health to guide nuclear power plant equipment maintenance performance.
[0030] This invention provides a method for constructing a smart health management system for nuclear power equipment based on digital twins, as follows: data acquisition → data classification → data fusion → 3D model drawing (knowledge graph drawing) → digital dynamic model creation → fault detection and determination → smart data configuration → solution recommendation → solution AI drill → solution adjustment → solution AI drill → solution improvement → solution implementation → process and result data are acquired.
[0031] Specifically, this invention provides a smart health management system for nuclear power equipment based on digital twins, such as... Figure 1As shown, it includes:
[0032] Data storage layer: This layer stores equipment data, image data, equipment operation data, technician data, document data, spare parts data, and environmental data, forming a twin data repository for the nuclear power plant management system and management tools. The data storage layer stores equipment data, image data, equipment operation data, technician data, document data, spare parts data, and environmental data through input, import, and online mapping. The data stored in the data storage layer includes data, files, graphics, audio-visual materials, and other formats.
[0033] The data storage layer includes: equipment physical data module, equipment file data module, equipment environment data module, equipment failure case module, equipment experience feedback module, and tips and best practices module.
[0034] Data Governance Layer: This layer utilizes knowledge graphs and large language models to govern data in the twin data repository, generating governed data that can be accurately identified and correctly processed by computers. By integrating knowledge graphs and large language models, semantic networks are used to accurately express the relationships between equipment entities and their associated equipment, constructing complex equipment scenario knowledge chains and providing explicit semantic support for the large model. The large language model is then used to correct and optimize the knowledge graph. Through enhanced training, semantic understanding capabilities are improved, and knowledge injection enables high-precision prediction tasks, mitigating the problem of knowledge sparsity. By coordinating knowledge graph and large language model technologies, a standard dictionary and data training and fusion functions have been established for various types of data in the power plant equipment basic data storage layer. This multi-source data, after cleaning and preprocessing, is used for training and application of the knowledge graph and large language model. For ambiguous data, power plant experts can perform online judgment and correction to achieve unambiguous data expression.
[0035] The data governance layer is a functional layer that enables power plant equipment data to be accurately identified and correctly processed by computers. Data governance includes standard dictionaries and data training and fusion for various types of data in the basic data storage layer of power plant equipment; vectorization for physical entities with direction and measurement; cleaning and preprocessing of multi-source data for training and application of knowledge graphs and large language models; and online judgment and correction for ambiguous data to achieve unambiguous expression of the data.
[0036] The data governance layer includes a standard dictionary module, an encoding management module, a vector data module, and an expert confirmation module.
[0037] Data Graph Display Layer: Based on the data processed by the data governance layer, knowledge graph technology and large language model technology are applied to construct 3D models of equipment and equipment operating environment, realize the correlation between data, display the relationships between data, customize data analysis results, and display 3D vectorized data of equipment.
[0038] The data visualization layer utilizes 3D equipment modeling and semantic association, leveraging knowledge graph and large language model technologies to map processed data onto the 3D models of the equipment. This constructs 3D models of the equipment and their operating environment, showcasing relationships between data, customized data analysis results, and vectorized 3D data of the equipment. Applying machine learning algorithms and AI technology, it automatically extracts feature data, intelligently identifies current equipment status and changes in the external environment, and determines the causes of entity alarms. These alarm causes include power changes, equipment misoperation, or changes in external factors. Through digital twin simulation of single or multiple identical devices under identical environments, and by establishing multi-parameter correlation models such as vibration, temperature, pressure, and flow, it achieves single-dimensional, definitive threshold detection for individual devices, as well as comprehensive monitoring of multi-dimensional dynamic thresholds, improving anomaly identification and reducing the probability of equipment failure. Deviation data is highlighted to draw user attention.
[0039] The data visualization layer includes: 3D model, single-unit demonstration model, multi-device linkage demonstration model, and faulty equipment operation simulation model.
[0040] Data Application Layer: Based on historical and real-time data or configuration data, AI simulates equipment operation scenarios, analyzes equipment failure points, and forms feasible optimal maintenance plans by configuring historical failure maintenance plans, equipment failure troubleshooting experience feedback, upstream and downstream equipment operating parameters, equipment maintenance environment and other parameters and conditions. These plans include elements such as window period, maintenance human resources, documents, tools, spare parts, required time, and risk points, providing real-time data support for equipment health assessment and equipment failure diagnosis.
[0041] The data application layer is responsible for the formation and improvement of equipment maintenance and repair plans. By modeling entities and relationships, it constructs a semantic network for nuclear power plants, which becomes a large model and an important source of knowledge.
[0042] The data application layer establishes connections between equipment, faults, historical best practices, and environmental data to help intelligently generate equipment maintenance plans, conduct simulations, and recommend improvement measures based on the simulations. This enhances the perfection of equipment maintenance plans and extends equipment lifespan. Specifically, the platform can use historical and real-time data, AI-configured conditions, and AI-simulated equipment operation scenarios to analyze equipment fault points. By configuring parameters and conditions such as historical fault maintenance plans, equipment fault troubleshooting experience feedback, upstream and downstream equipment operating parameters, and the equipment maintenance environment, it forms a feasible optimal maintenance plan. This plan includes elements such as window periods, maintenance human resources, documents, tools, spare parts, required time, and risk points, providing real-time data support for equipment health assessment and equipment fault diagnosis. Equipment managers can also adjust the conditions in the recommended plan to form a new plan, conduct digital simulations again, and repeat this process to develop the optimal plan.
[0043] The data application layer includes: equipment digital model operation module, digital recommendation solution module, solution digital exercise and testing module, parameter configuration adjustment module, and best solution recommendation module.
[0044] Data Value Mining Layer: Based on the virtual operation of twin data from the data application layer, it displays changes in data such as the number of operating key equipment in the power plant, high-risk equipment failures, medium-risk equipment and equipment under repair, equipment-related documents, tools, spare parts, and health clocks. It detects equipment health trends, adjusts inventory of spare parts and tools, and achieves lean management.
[0045] The data value mining layer includes: a trend module for sensitive components of key equipment, a fault status module for high-risk equipment, a data status module for medium-risk data, a data trend module for equipment under inspection, and a data status module for technical personnel.
[0046] Data search engine layer: used for simple and advanced data retrieval, and intelligent push based on user profiles.
[0047] The data search engine layer includes: a retrieval module, an intelligent push module, a login and authorization module, a publishing and navigation module, and a system maintenance module.
[0048] Basic capability layer: used for user management, organizational structure management, access control, log management, audit management, security management, etc.
[0049] The basic capability layer includes: user management module, organizational structure management module, permission management module, log management, audit log module, and security management module.
[0050] This invention also provides a method for intelligent health management of nuclear power equipment based on digital twins, such as... Figure 2 As shown, it includes:
[0051] Step 1: Data acquisition and storage to form a twin data repository;
[0052] Data such as equipment data, image data, equipment operation data, technical personnel data, document data, spare parts data, and environmental data are stored through input, import, and online mapping.
[0053] Step 2: Data governance, generating processed data that can be accurately identified and correctly calculated by computers;
[0054] By using knowledge graphs and large language models, data governance is performed on the data in the twin data repository to generate governed data that can be accurately identified and correctly processed by computers.
[0055] Step 3: Data visualization, constructing a static 3D model of the equipment and a 3D model of the equipment's operating environment, displaying the relationships between data, customized data analysis results, and 3D vectorized data of the equipment;
[0056] Based on the data after governance by the data governance layer, knowledge graph technology and large language model technology are applied to construct a static 3D model of the equipment and a 3D model of the equipment's operating environment, realize the correlation between data, display the relationships between data, customize data analysis results, and generate 3D vectorized data of the equipment.
[0057] Step 4: Data application. Based on historical data and real-time data or AI configuration data, the system uses AI to simulate equipment operation scenarios, analyzes equipment failure points, optimizes solutions, and forms the best feasible maintenance plan.
[0058] Based on historical and real-time data or AI-configured data, the system simulates equipment operation scenarios using AI, analyzes equipment failure points, and forms feasible optimal maintenance plans by configuring historical failure repair plans, feedback on equipment troubleshooting experience, operating parameters of upstream and downstream equipment, and equipment maintenance environment parameters and conditions. These plans include factors such as window periods, maintenance human resources, documents, tools, spare parts, required time, and risk points, providing real-time data support for equipment health assessment and equipment failure diagnosis.
[0059] Step 5: Data value mining, displaying the status of sensitive components of key equipment in the power plant, the fault status of high-risk equipment, the status of medium-risk equipment, the status of equipment under inspection, and the data status of technical personnel;
[0060] Based on the virtual operation of twin data in the data application layer, the system displays changes in data such as the number of operating key equipment in the power plant, high-risk equipment failures, medium-risk equipment and equipment under repair, equipment-related documents, tools, spare parts, and health clocks. It can detect the health trend of equipment, adjust inventory of spare parts and tools, and achieve lean management.
[0061] The present invention has been described in detail above with reference to the accompanying drawings and embodiments. However, the present invention is not limited to the above embodiments, and various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention. All contents not described in detail in the present invention can be derived from existing technologies.
Claims
1. A smart health management system for nuclear power equipment based on digital twins, characterized in that, The system includes: Data storage layer: Used to store device data, image data, equipment operation data, technical personnel data, document data, spare parts data, and environmental data, forming a twin data repository; Data governance layer: Used to perform data governance on the data in the twin data repository through metadata management, data lineage management, data cleaning, and data format conversion technologies, generating governed data that can be accurately identified and correctly calculated by computers; Data Graph Display Layer: Based on the data processed by the data governance layer, knowledge graph technology and large language model technology are applied to construct 3D models of equipment and equipment operating environment, display the relationships between data, customize data analysis results, and 3D vectorized data of equipment.
2. The intelligent health management system for nuclear power equipment based on digital twins according to claim 1, characterized in that, The data storage layer includes: a device physical data module, a device file data module, a device environment data module, a device failure case module, a device experience feedback module, and a tricks and best practices module.
3. The intelligent health management system for nuclear power equipment based on digital twins according to claim 1, characterized in that, The data governance includes standard dictionaries and data training and fusion for various types of data in the basic data storage layer of power plant equipment; vectorization is adopted for physical entities that are directional and measurable; multi-source data is cleaned and preprocessed for training and application of knowledge graphs and large language models; for individual ambiguous data, power plant experts can make judgments and corrections online to achieve unambiguous expression of data; the data governance layer includes a standard dictionary module, an encoding management module, a vector data module, and an expert confirmation module.
4. The intelligent health management system for nuclear power equipment based on digital twins according to claim 1, characterized in that, The data graph display layer utilizes knowledge graph technology, large language model technology, 3D modeling technology, semantic association technology, etc., to map the processed data into the 3D model of the equipment, construct the static 3D model of the equipment and the 3D model of the equipment operating environment, and display the relationship between data, customized data analysis results, and 3D vectorized data of the equipment. Applying machine learning algorithms, the system automatically extracts feature data, intelligently identifies the current equipment status and changes in the external environment, and determines the cause of entity alarms. Functions include: alarms triggered by power changes, equipment malfunctions, or changes in external factors; simulation of single or multiple identical devices under the same environment using digital twins; establishment of multi-parameter correlation models for vibration, temperature, pressure, and flow to achieve single-dimensional, definitive threshold detection for individual devices; comprehensive monitoring of multi-dimensional dynamic thresholds to improve anomaly identification and reduce the probability of equipment failure; and a data visualization layer including: 3D models, single-device demonstration models, multi-device linkage demonstration models, and faulty equipment operation simulation models.
5. The intelligent health management system for nuclear power equipment based on digital twins according to claim 1, characterized in that, The system also includes a data application layer: based on historical data and real-time data or configuration data, it uses AI to simulate equipment operation scenarios, analyzes equipment failure points, and forms a feasible optimal maintenance plan by configuring historical failure maintenance plans, feedback on equipment failure troubleshooting experience, operating parameters of upstream and downstream equipment, and equipment maintenance environment parameters. The data application layer includes: equipment digital model operation module, digital recommendation solution module, solution digital exercise and testing module, parameter configuration adjustment module, and best solution recommendation module.
6. The intelligent health management system for nuclear power equipment based on digital twins according to claim 1, characterized in that, The system also includes a data value mining layer: based on the virtual operation of the twin data of the data application layer, it displays the number of operating key equipment in the power plant, the number of high-risk equipment failures, medium-risk equipment and equipment under repair, changes in equipment-related documents, tools, spare parts, and health clock data, detects the health trend of the equipment, and adjusts the inventory of spare parts and tools. The data value mining layer includes: a trend module for sensitive components of key equipment, a fault status module for high-risk equipment, a data status module for medium-risk data, a data trend module for equipment under inspection, and a data status module for technical personnel.
7. A smart health management system for nuclear power equipment based on digital twins according to claim 1, characterized in that, The system also includes a data search engine layer: used for simple and advanced data retrieval, and intelligent push based on user profiles; The data search engine layer includes: a retrieval module, an intelligent push module, a login and authorization module, a publishing and navigation module, and a system maintenance module.
8. The intelligent health management system for nuclear power equipment based on digital twins according to claim 1, characterized in that, The system also includes a basic capability layer for user management, organizational structure management, permission management, log management, audit management, and security management. The basic capability layer includes: user management module, organizational structure management module, permission management module, log management, audit log module, and security management module.
9. A method for intelligent health management of nuclear power equipment based on digital twins, characterized in that, The method includes: Step 1: Data acquisition and storage to form a twin data repository; Step 2: Data governance, generating processed data that can be accurately identified and correctly calculated by computers; Step 3: Data visualization, constructing a 3D model of the equipment and a 3D model of the equipment's operating environment, displaying the relationships between data, customized data analysis results, and 3D vectorized data of the equipment; Step 4: Data application. Based on historical data, real-time data, or AI configuration data, the system uses AI to simulate equipment operation scenarios, analyzes equipment failure points, optimizes solutions, and forms the best feasible maintenance plan.
10. A method for intelligent health management of nuclear power equipment based on digital twins according to claim 9, characterized in that, The method also includes step 5, data value mining, which displays the status of sensitive components of key equipment in the power plant, the fault status of high-risk equipment, the status of medium-risk equipment, the status of equipment under inspection, and the data status of technical personnel.