A nuclear power plant generator set overhaul rehearsal system

Abstract

The present invention discloses a nuclear power plant generator set overhaul rehearsal system, which relates to the field of nuclear power plant maintenance technology. The system includes a three-dimensional digital twin module, which constructs a three-dimensional digital twin model based on the design drawings and pipeline layout of the nuclear power plant generator set and its surrounding environment; an artificial intelligence simulation engine module, which integrates a physical engine and process rules based on the three-dimensional digital twin model and real-time data to simulate the entire overhaul process and intelligently adjust the initial overhaul plan through a machine learning algorithm; and an artificial intelligence decision module, which uses deep learning technology to comprehensively evaluate and compare the performance of each optimized overhaul plan in multiple dimensions, including construction period, cost, and radiation dose, and provides optimal decision recommendations. By organically integrating the three-dimensional digital twin model, artificial intelligence simulation engine, virtual reality interaction system, and blockchain security module, the present invention achieves digital, intelligent, and visual management of nuclear power plant overhaul operations.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear power plant maintenance, in particular to a nuclear power plant generator set overhaul rehearsal system. Background Art

[0002] Nuclear power plant generators are crucial infrastructure for national energy security, and their operational safety and economic efficiency are paramount. To ensure continuous, stable, and safe operation, they typically undergo an overhaul every 18-24 months, comprehensively inspecting and maintaining all key systems and equipment. Overhauls typically last two to three months, involve tens of thousands of specific processes, and involve thousands of personnel, making them a complex, systematic project. A well-designed and detailed overhaul plan is crucial for ensuring overhaul quality, controlling overhaul cycles, and minimizing radiation exposure.

[0003] Traditional overhaul work plan design relies primarily on the experience of on-site managers, manually developing work processes and timelines based on two-dimensional nuclear power plant design drawings and limited on-site data. This approach not only lacks realistic simulation capabilities but also struggles to fully assess multi-dimensional constraints, including time, space, environment, and personnel. This often leads to process conflicts, schedule delays, operational violations, and excessive radiation doses, adversely impacting the successful completion of overhaul work. Furthermore, existing work plan design lacks information technology, and data traceability and protection measures across the entire overhaul process are insufficient to ensure data integrity. Summary of the Invention

[0004] In view of the problems in the existing design of nuclear power plant generator unit overhaul operation plans, such as poor real simulation capabilities, difficulty in comprehensively evaluating multi-dimensional constraints, low efficiency in operation process planning, and inability to ensure the authenticity and integrity of full-process data, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to simulate the entire overhaul process with high precision, automatically generate and optimize alternative plans in multiple dimensions, visualize interaction and manual intervention, comprehensively evaluate each plan and make the optimal decision, and ensure that the data of the entire process is traceable and tamper-proof.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides a nuclear power plant generator set overhaul rehearsal system, which includes a three-dimensional digital twin module, which uses CAD and BIM technology to construct a three-dimensional digital twin model based on the design drawings and pipeline layout of the nuclear power plant generator set and the surrounding environment; a data acquisition module, which obtains the working status data of the nuclear power plant on-site equipment in real time through a variety of sensors, and provides real input parameters for the three-dimensional digital twin model; an artificial intelligence simulation engine module, which integrates the physical engine and process rules based on the three-dimensional digital twin model and real-time data, simulates the entire overhaul process, and intelligently adjusts the initial overhaul plan through a machine learning algorithm; a virtual reality interaction module, equipped with virtual reality equipment, enables operators to observe and interactively control the overhaul simulation process in a virtual environment; an artificial intelligence decision-making module, which uses deep learning technology to comprehensively evaluate and compare the performance of each optimized overhaul plan in multiple dimensions, and gives the best decision recommendation, the multiple dimensions including construction period, cost, and radiation dose; a blockchain security module, which uses blockchain technology to ensure the traceability and tamper-proofing of data throughout the overhaul process, and automatically supervises work compliance through smart contracts.

[0008] As a preferred solution of the nuclear power plant generator set overhaul rehearsal system described in the present invention, the following steps are performed: obtaining two-dimensional CAD design drawings and existing BIM model data of the nuclear power plant generator set and its surrounding environment, and classifying and organizing the CAD drawings and BIM model data; establishing a unified three-dimensional rectangular coordinate system with reference to the overall layout of the nuclear power plant; performing three-dimensional parametric modeling of the building structure and pipeline layout of the nuclear power plant in the BIM modeling software based on the existing BIM model data; importing the two-dimensional CAD drawings into the BIM modeling software, performing three-dimensional solid modeling of each system component and equipment according to the two-dimensional CAD drawings, and integrating the three-dimensional solid model with the BIM model; using a model checking tool to detect whether the integrated three-dimensional model has errors, and repairing the three-dimensional model based on the detection results; performing secondary simplification on non-core areas of the three-dimensional model that are not related to simulation analysis, and refining the core areas related to simulation analysis; integrating the geometric information of the three-dimensional model with non-geometric information including parameter information and material properties of system components and equipment to form a three-dimensional digital twin model; and publishing the three-dimensional digital twin model to a simulation analysis platform and integrating it with an artificial intelligence simulation engine module, a virtual reality interaction module, and a data acquisition module.

[0009] As a preferred solution of the nuclear power plant generator set overhaul rehearsal system described in the present invention, the following steps are included: repairing the three-dimensional model according to the detection results: if a surface overlap error is detected, running the repair geometry function of the modeling software to automatically repair the overlapped surface; if a spatial position conflict is detected between the pipeline and the equipment model, adjusting the pipeline route in the pipeline professional model to eliminate the interference; if it is detected that a certain equipment model is missing an auxiliary bracket component, the bracket structure dimension data is obtained through on-site measurement and the corresponding component is supplemented in the modeling; if it is detected that a certain area model is missing a surveillance camera, and the design document has clear requirements, the modeling is supplemented according to the camera model and position; and the repaired three-dimensional model is comprehensively checked using a model inspection tool to ensure that there are no residual errors after the repair.

[0010] As a preferred embodiment of the nuclear power plant generator unit overhaul rehearsal system of the present invention, the artificial intelligence simulation engine module is constructed as follows: obtaining three-dimensional digital twin model data from a three-dimensional digital twin module and obtaining real-time collected on-site equipment operating status data from a data acquisition module; constructing a virtual nuclear power plant environment model in a physics engine based on the acquired data, and applying the on-site equipment operating status data as physical simulation input parameters; converting standardized overhaul operation process rules into an executable rule set and importing them into the constructed virtual environment model; the physics engine drives the simulation execution and obtains simulation data generated during the simulation in real time; determining whether the acquired simulation data violates the constraints in the executable rule set, and if so, sending a prompt signal to the virtual reality interaction module; an operator interactively observes and controls the violation simulation process through a virtual reality device, manually intervenes and adjusts the violation link to ensure that it complies with the operating standards; recording the values ​​of all key indicators in the adjusted simulation process; a machine learning algorithm automatically identifies bottlenecks and risk points in the simulation process based on the recorded data, and adjusts the operation parameters through an optimization algorithm to generate multiple optimized alternative overhaul plans; and performing multiple rounds of evaluation simulation on each optimized alternative plan to determine its comprehensive performance in terms of construction period, cost, and radiation dose.

[0011] As a preferred solution of the nuclear power plant generator set overhaul rehearsal system described in the present invention, the steps of constructing the virtual reality interaction module are as follows: when the simulation process violates the constraints in the executable rule set, the artificial intelligence simulation engine module sends a violation prompt signal to the virtual reality interaction module; according to the violation prompt signal, the virtual reality interaction module locates the violation link through one of the following technical means or a combination thereof: responding to the operator's voice command, automatically switching the perspective of the virtual reality scene so that the violation area is located in the center of the field of view; detecting the operator's gestures, controlling the virtual scene to switch the perspective and zoom ratio until the violation link is clearly seen; using virtual reality three-dimensional visualization technology, the operator can fully The operator analyzes the specific causes of the violations based on the constraints in the executable rule set and proposes adjustment plans based on the causes of the violations. The operator simulates and verifies the adjustment plans in the virtual scene to check whether other constraints in the operating rules are violated. If the adjustment plan is verified, the operator confirms the execution of the adjustment through voice or gestures in the virtual interface, and the system sends the adjustment instructions to the simulation engine. The simulation engine re-executes according to the adjustments, and the operator continues to observe. If new violations are found, the adjustment process is repeated until there are no violations in the entire process. The virtual reality system automatically records every adjustment decision and observation result of the operator, and feeds this data back to the simulation engine module.

[0012] As a preferred embodiment of the nuclear power plant generator unit overhaul rehearsal system of the present invention, the system proposes an adjustment plan for the cause of the violation, including: if it is detected that the operation time has exceeded the specified critical time node, the cause is analyzed in combination with the time constraint requirements. If the analysis shows that the actual start time is delayed due to a delay in the preparatory work phase, the personnel and materials for the next process are deployed in advance. Once the current operation is completed, the next operation is seamlessly connected to catch up with the progress; if on-site personnel are missing during the equipment relocation operation, the system analyzes the staffing requirements. If the analysis shows that the actual operating staff is insufficient and violates the staffing standards for the corresponding type of work, the required crane operators and protective personnel are deployed from other areas, and the existing personnel are used to carry out preparatory work without radiation risk, and then officially start the work after the staffing is replenished; during the chemical decontamination operation, if the ventilation system in the operation area is detected to be not operating normally, the cause is analyzed in combination with the environmental constraint requirements. If the analysis shows that the ventilation parameters at the operation site do not meet the standards, posing a safety hazard of chemical leakage, the operation is immediately suspended, the ventilation system fault is first checked, and additional monitoring posts are deployed in vacant areas upwind. Operations are resumed when conditions are met.

[0013] As a preferred solution of the nuclear power plant generator set overhaul rehearsal system described in the present invention, the steps of constructing the artificial intelligence decision-making module are as follows: obtaining multiple optimized alternative overhaul plans from the artificial intelligence simulation engine module; randomly dividing the historical overhaul plan data into two parts, a training set and a test set, according to a preset ratio; standardizing the multi-dimensional data in the training set and the test set, and converting data of different dimensions into the same numerical range; constructing a multi-input and multi-output decision evaluation model based on deep learning technology, with the input being the performance data of each alternative plan in different dimensions, and the output being the comprehensive score of each plan; using the data in the training set, the decision evaluation model is trained through the back propagation algorithm to enable it to learn different The weight of the impact of dimensional data on the final decision; evaluate the trained decision evaluation model on the test set, and analyze its generalization ability on unseen data based on the evaluation indicators; if the test effect is not ideal, optimize the decision evaluation model by adjusting the network structure, optimizing hyperparameters, and increasing training data; input the multi-dimensional data of each optimized alternative plan into the trained decision evaluation model, and the model outputs a comprehensive score for each plan; sort all alternative plans according to the comprehensive score output by the model, and select the one with the highest score as the optimal overhaul plan given by the artificial intelligence decision module; after the actual execution of the output optimal plan, record its actual performance, compare it with the model prediction, and add this part of new data to the historical data set.

[0014] On the second aspect, an embodiment of the present invention provides a method for rehearsing the overhaul of a nuclear power plant generator set, which includes using a three-dimensional digital twin module to construct a three-dimensional digital twin model based on the design drawings and pipeline layout of the nuclear power plant generator set and the surrounding environment using CAD and BIM technology; through the data acquisition module, various sensors are used to obtain the working status data of the nuclear power plant equipment in real time, providing real input parameters for the three-dimensional digital twin model; in the artificial intelligence simulation engine module, the physical engine and process rules are integrated to perform a virtual simulation of the entire overhaul process based on the three-dimensional digital twin model and real-time data; the simulation results are analyzed using a machine learning algorithm, and the initial overhaul plan is intelligently adjusted and optimized; through the virtual reality interaction module, the operator observes and interactively controls the overhaul simulation process in a virtual environment; the artificial intelligence decision-making module uses deep learning technology to comprehensively evaluate and compare the performance of each optimized overhaul plan from multiple dimensions, and give the best decision recommendation; the blockchain security module uses blockchain technology to ensure the traceability and tamper-proofness of data throughout the overhaul process, and automatically supervises work compliance through smart contracts.

[0015] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the nuclear power plant generator set overhaul rehearsal system as described in the first aspect of the present invention are implemented.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the nuclear power plant generator set overhaul rehearsal system as described in the first aspect of the present invention are implemented.

[0017] The beneficial effects of the present invention are as follows: By organically integrating a three-dimensional digital twin model, an artificial intelligence simulation engine, a virtual reality interactive system, and a blockchain security module, the present invention achieves digital, intelligent, and visual management of nuclear power plant overhaul operations. This can significantly improve the accuracy, efficiency, and safety of overhaul operations, shorten construction periods, reduce labor and operating costs, mitigate radiation risks, and protect the health of workers. At the same time, blockchain technology is used to ensure data security and traceability throughout the entire process, and deep learning models are used to scientifically evaluate and support decision-making for alternative plans. This comprehensively enhances the intelligence and modernization of overhaul operations, injecting new impetus into the high-quality development of the nuclear power industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 Module connection diagram of the rehearsal system for overhaul of nuclear power plant generator sets.

[0020] Figure 2 Construct a flow chart for the 3D digital twin module of the nuclear power plant generator unit overhaul rehearsal system.

[0021] Figure 3 Computer equipment diagram of a rehearsal system for overhaul of nuclear power plant generator units. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Example 1

[0026] Reference Figures 1 to 3 , which is the first embodiment of the present invention, provides a nuclear power plant generator set overhaul rehearsal system, comprising:

[0027] The data acquisition module obtains the working status data of the nuclear power plant's on-site equipment in real time through a variety of sensors, providing real input parameters for the three-dimensional digital twin model.

[0028] The three-dimensional digital twin module uses CAD and BIM technologies to build a three-dimensional digital twin model based on the design drawings and pipeline layout of the nuclear power plant's generator units and surrounding environment.

[0029] Specifically, the steps for constructing the 3D digital twin module are as follows: 2D CAD design drawings and existing BIM model data of the nuclear power plant's generator units and surrounding environment are obtained, and the CAD drawings and BIM model data are classified and organized; a unified 3D rectangular coordinate system is established with reference to the overall layout of the nuclear power plant; 3D parametric modeling of the nuclear power plant's building structure and pipeline layout is performed in the BIM modeling software based on the existing BIM model data; 2D CAD drawings are imported into the BIM modeling software, and 3D solid modeling of each system component and equipment is performed based on the 2D CAD drawings, and the 3D solid model is integrated with the BIM model; a model checking tool is used to detect whether the integrated 3D model contains errors and the 3D model is repaired based on the detection results; non-core areas of the 3D model that are not related to simulation analysis are simplified to reduce the computational burden, while core areas related to simulation analysis are refined to improve simulation accuracy; the geometric information of the 3D model is integrated with non-geometric information such as parameter information and material properties of system components and equipment to form a 3D digital twin model; the 3D digital twin model is published to the simulation analysis platform and integrated with the artificial intelligence simulation engine module, virtual reality interaction module, and data acquisition module.

[0030] Preferably, a model checking tool is used to detect whether there are any errors in the integrated three-dimensional model, and the three-dimensional model is repaired according to the detection results, including the following steps: if a face overlap error is detected, the repair geometry function of the modeling software is run to automatically repair the overlapped surface; if a spatial position conflict is detected between the pipeline and the equipment model, the pipeline route is adjusted in the pipeline professional model to eliminate the interference; if a certain equipment model is detected to be missing an auxiliary bracket component, the bracket structure dimension data is obtained by on-site measurement, and the corresponding components are supplemented in the modeling; if a certain area model is detected to be missing a surveillance camera, and the design document has clear requirements, the modeling is supplemented according to the camera model and position; the repaired three-dimensional model is comprehensively checked using the model checking tool to ensure that there are no residual errors after the repair.

[0031] The artificial intelligence simulation engine module, based on the three-dimensional digital twin model and real-time data, integrates the physical engine and process rules to simulate the entire overhaul process, and intelligently adjusts the initial overhaul plan through machine learning algorithms.

[0032] Specifically, the AI ​​simulation engine module operates as follows: It obtains 3D digital twin model data from the 3D digital twin module and real-time on-site equipment operating status data from the data acquisition module. Based on the acquired data, a virtual nuclear power plant environment model is constructed in the physics engine, and the on-site equipment operating status data is used as input parameters for the physical simulation. Standardized overhaul operation process rules are converted into an executable rule set and imported into the constructed virtual environment model. The physics engine drives the simulation execution and acquires the simulation data generated during the simulation in real time. It determines whether the acquired simulation data violates the constraints in the executable rule set and, if so, sends a prompt signal to the virtual reality interaction module. The operator uses the virtual reality device to interactively observe and control the violation simulation process and manually intervene to adjust the violation link to ensure that it complies with the operating standards. The values ​​of all key indicators in the adjusted simulation process are recorded, including construction period, manpower, energy consumption, and radiation dose. The machine learning algorithm automatically identifies bottlenecks and risk points in the simulation process based on the recorded data, and adjusts the operation parameters through the optimization algorithm to generate multiple optimized alternative overhaul plans. Multiple rounds of evaluation simulation are performed on each optimized alternative plan to determine its comprehensive performance in terms of construction period, cost, and radiation dose.

[0033] Preferably, the executable rule set includes multi-dimensional constraints including operation sequence, time requirements, staffing, and environmental conditions.

[0034] The virtual reality interaction module, equipped with virtual reality equipment, enables operators to observe and interactively control the overhaul simulation process in a virtual environment.

[0035] Specifically, when a simulation violates a constraint in the executable rule set, the AI ​​simulation engine module sends a violation alert signal to the VR interaction module. This signal is displayed as a red warning sign on the VR headset's display interface, accompanied by an audio prompt. Based on the violation alert signal, the VR interaction module locates the offending link through one or a combination of the following technical means: responding to the operator's voice commands to automatically switch the perspective of the VR scene so that the violation area is centered in the field of view; detecting the operator's gestures, such as pressing buttons on the VR controller or making specific gestures, to control the virtual scene's perspective switching and zoom ratio changes until the offending link is clearly visible. Using virtual reality three-dimensional visualization technology, operators can observe the detailed information of the violation in all directions, including but not limited to the three-dimensional spatial coordinate location of the violation, the time of the violation, the list of staff present at the time, the equipment number involved, etc.; the operator analyzes the specific cause of the violation in combination with the constraint requirements of the executable rule set, and proposes an adjustment plan for the cause of the violation; the operator simulates and verifies the adjustment plan in the virtual scene to check whether it violates other constraints in the operating rules; if the adjustment plan is verified, the operator confirms the execution of this adjustment in the virtual interface through voice or gesture, and the system sends the adjustment instruction to the simulation engine; the simulation engine re-executes according to the adjustment, and the operator continues to observe. If new violations are found, the above steps are repeated until there are no violations in the entire process; the virtual reality system automatically records every adjustment decision and observation result of the operator, and feeds this data back to the simulation engine module.

[0036] Preferably, the adjustment plan proposed for the cause of the violation includes but is not limited to the following: if it is detected that the operation time has exceeded the specified critical time node, the cause is analyzed in combination with the time constraint requirements. If the analysis finds that the actual start time is delayed due to the delay in the preparatory operation link, the personnel and materials for the next process are deployed in advance. Once the current operation is completed, the next operation is seamlessly connected to catch up with the progress; if it is found that there are missing on-site personnel during the equipment relocation and placement operation, the personnel deployment requirements are analyzed. If the analysis finds that the actual operating personnel are insufficient and violates the staffing standards of the corresponding types of work, the required crane operators and protective personnel are deployed from other areas, and the personnel already on site carry out preparatory operations without radiation risks, and then formally execute the operation after the personnel are replenished; during the chemical pollution reduction operation, if it is detected that the ventilation system of the operation area is not operating normally, the cause is analyzed in combination with the environmental constraint requirements. If the analysis finds that the ventilation parameters of the operation site do not meet the standards and there is a safety hazard of chemical leakage, the operation is immediately suspended, the ventilation system fault is first checked, and monitoring posts are added to the vacant area upwind. The operation is resumed when the conditions are met.

[0037] The artificial intelligence decision-making module uses deep learning technology to comprehensively evaluate and compare the performance of each optimized overhaul plan in multiple dimensions, including construction period, cost, and radiation dose, and provides the best decision-making recommendations.

[0038] Specifically, multiple optimized alternative overhaul plans are obtained from the artificial intelligence simulation engine module, including simulation performance data of multiple dimensions; the historical overhaul plan data are randomly divided into two parts, training set and test set, according to the preset ratio; the multi-dimensional data in the training set and the test set are standardized, and the data of different dimensions are converted into the same numerical range; based on deep learning technology, a multi-input and multi-output decision evaluation model is constructed, with the input being the performance data of each alternative plan in different dimensions, and the output being the comprehensive score of each plan; the decision evaluation model is trained by the back propagation algorithm using the data in the training set, so that it learns the influence weight of data of different dimensions on the final decision; the training set is trained on the test set. A good decision evaluation model is evaluated, and its generalization ability on unseen data is analyzed based on the evaluation indicators; if the test effect is not ideal, the decision evaluation model is optimized by adjusting the network structure, optimizing hyperparameters, and increasing training data; the multi-dimensional data of each optimized alternative plan is input into the trained decision evaluation model, and the model outputs a comprehensive score for each plan; all alternative plans are sorted according to the comprehensive score output by the model, and one or more with the highest score (if there are multiple plans with similar scores) are selected as the optimal overhaul plan given by the artificial intelligence decision module; after the output of the optimal plan is actually executed, its actual performance is recorded and compared with the model prediction, and this part of the new data is added to the historical data set.

[0039] It should be noted that a deep neural network model consisting of multiple fully connected layers is used. The number of nodes in the input layer is equal to the number of dimensions considered. The hidden layer is appropriately designed according to the amount of data and model complexity. The output layer gives a score between 0 and 1 for each node.

[0040] The blockchain security module uses blockchain technology to ensure data traceability and tamper-proofing throughout the entire overhaul process, and automatically monitors work compliance through smart contracts.

[0041] Furthermore, this embodiment also provides a method for rehearsing the overhaul of a nuclear power plant generator set, including using a three-dimensional digital twin module to construct a three-dimensional digital twin model based on the design drawings and pipeline layout of the nuclear power plant generator set and the surrounding environment using CAD and BIM technology; through the data acquisition module, various sensors are used to obtain the working status data of the nuclear power plant equipment in real time, providing real input parameters for the three-dimensional digital twin model; in the artificial intelligence simulation engine module, the physical engine and process rules are integrated to perform a virtual simulation of the entire overhaul process based on the three-dimensional digital twin model and real-time data; the simulation results are analyzed using a machine learning algorithm, and the initial overhaul plan is intelligently adjusted and optimized; through the virtual reality interaction module, the operator observes and interactively controls the overhaul simulation process in a virtual environment; the artificial intelligence decision-making module uses deep learning technology to comprehensively evaluate and compare the performance of each optimized overhaul plan from multiple dimensions, and give the best decision recommendation; the blockchain security module uses blockchain technology to ensure the traceability and tamper-proofness of data throughout the overhaul process, and automatically supervises work compliance through smart contracts.

[0042] This embodiment also provides a computer device suitable for the nuclear power plant generator set overhaul rehearsal system, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the nuclear power plant generator set overhaul rehearsal system proposed in the above embodiment.

[0043] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0044] This embodiment further provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the nuclear power plant generator set overhaul rehearsal system proposed in the above embodiment.

[0045] In summary, this invention achieves digital, intelligent, and visual management of nuclear power plant overhaul operations by organically integrating a three-dimensional digital twin model, an artificial intelligence simulation engine, a virtual reality interactive system, and a blockchain security module. This can significantly improve the accuracy, efficiency, and safety of overhaul operations, shorten construction periods, reduce labor and operating costs, mitigate radiation risks, and protect the health of workers. Furthermore, blockchain technology is used to ensure data security and traceability throughout the entire process, and deep learning models are used to scientifically evaluate and support decision-making for alternative solutions. This comprehensively enhances the intelligence and modernization of overhaul operations, injecting new impetus into the high-quality development of the nuclear power industry.

[0046] Example 2

[0047] Reference Figures 1 to 3 , which is the second embodiment of the present invention, provides a nuclear power plant generator set overhaul rehearsal system. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0048] First, the system collected input data, including 5,628 CAD design drawings covering the nuclear power plant's 2,500 square kilometers, as well as existing BIM model data for buildings and pipelines, totaling 17,326 model instances. A unified 3D rectangular coordinate system was then established within the BIM software to model 27 major structures, including the reactor building and turbine house. Parametric technology was used to create a 3D model of the 230-kilometer main circulation pipeline, encompassing 34,427 pipeline nodes. Next, the system imported the CAD drawings and created 3D solid models for 27,500 pieces of equipment, such as the reactor pressure vessel and steam generator.

[0049] After the model was built, it was inspected and repaired. The system detected 105 surface overlap errors, which were corrected using the automatic repair function. The inspection also found 63 spatial conflicts between pipelines and equipment, requiring manual adjustment of pipeline routes to eliminate interference. To balance computing performance and accuracy, the system refined the mesh in the reactor core area and simplified non-core auxiliary buildings. Finally, non-geometric data such as equipment parameters and material information were integrated into the model, and a complete three-dimensional digital twin model was released, with a model volume of 23.7GB.

[0050] At a real nuclear power plant site, a large number of monitoring devices are deployed to collect real-time data. 1,200 high-precision temperature sensors are installed in the reactor core area, 1,840 pressure transmitters are placed along the main pipelines, and 960 radiation dose monitors are deployed in critical areas. The real-time data collected by these sensors is fed into the 3D model at a throughput of up to 200MB / second, serving as the real-world parameters for the physics simulation. The physics engine creates 22 million computational meshes and imports 375 executable rules, which serve as the foundation for a preliminary simulation of the entire overhaul process.

[0051] During the initial simulation, 83 violations were detected. Operators used virtual reality equipment to observe the locations of the violations, analyze the causes, and propose adjustments. After verification in the virtual environment, the adjustments were confirmed and implemented. The simulation engine then re-executed the simulation based on the adjustments until no violations occurred. Based on the simulated data, a machine learning algorithm conducted analysis and optimization, ultimately generating eight alternative overhaul plans. The deep learning decision evaluation model was then trained on 120,000 historical data points, achieving a generalization MSE loss of 0.037 on a test set of 30,000 items.

[0052] Furthermore, multi-dimensional performance data from the eight alternatives was fed into a decision-making evaluation model, which performed a comprehensive evaluation and ultimately output the two optimal options with the shortest renovation period and lowest total cost. Furthermore, the blockchain security module automatically generates new blocks every five seconds, ensuring that all data throughout the entire process is untraceable and tamper-proof. The smart contract-based monitoring system detected 12 constraint violations, issued an alarm, and forced the suspension of the offending simulation to prevent the incorrect approval of the work plan.

[0053] Preferably, the comparison indicators of the present invention and the traditional method are shown in Table 1.

[0054] Table 1 Comparative indexes of the present invention and the traditional method

[0055] index The present invention Traditional methods Simulation realism high Low Solution optimization capabilities Automatic intelligent optimization Limited manual optimization Evaluation Dimensions Multi-dimensional comprehensive assessment Single dimension assessment Human-computer interaction capabilities Virtual reality intuitive interaction Poor human-computer interaction Operational compliance Smart contract automatic supervision Manual supervision Simulation efficiency high Low Data security Blockchain No guarantee Model maintenance low cost High costs

[0056] Specifically, the method of the present invention is significantly superior to traditional overhaul rehearsal methods in many key indicators such as simulation realism, solution optimization, multi-dimensional evaluation, human-computer interaction, operation compliance supervision, simulation efficiency, data security and model maintenance.

[0057] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A nuclear power plant generator overhaul rehearsal system, characterized by: include, The 3D digital twin module uses CAD and BIM technologies to build a 3D digital twin model based on the design drawings and pipeline layout of the nuclear power plant's generator units and surrounding environment; The data acquisition module uses a variety of sensors to obtain real-time operating status data of nuclear power plant equipment, providing real input parameters for the three-dimensional digital twin model; An artificial intelligence simulation engine module, based on the three-dimensional digital twin model and real-time data, integrates a physics engine and process rules to simulate the entire overhaul process and intelligently adjust the initial overhaul plan through machine learning algorithms; A virtual reality interaction module, equipped with virtual reality equipment, enables operators to observe and interactively control the overhaul simulation process in a virtual environment; The AI ​​decision-making module uses deep learning technology to comprehensively evaluate and compare the performance of each optimized overhaul plan across multiple dimensions, including construction period, cost, and radiation dose, and provides optimal decision recommendations. The blockchain security module uses blockchain technology to ensure data traceability and tamper-proofing throughout the overhaul process, and automatically monitors work compliance through smart contracts; The steps for constructing the virtual reality interaction module are as follows: When the simulation process violates the constraints in the executable rule set, the artificial intelligence simulation engine module sends a violation prompt signal to the virtual reality interaction module; Based on the violation prompt signal, the VR interaction module locates the violation link through one or a combination of the following technical means: Responding to the operator's voice commands, the system automatically switches the perspective of the virtual reality scene so that the area where the violation occurred is in the center of the field of view; Detect the operator's gestures and control the virtual scene to switch perspectives and change the zoom ratio until the violation is clearly visible; Using virtual reality 3D visualization technology, operators can observe all-round details of violations. Based on the constraints in the executable rule set, operators can analyze the specific causes of violations and propose adjustment plans based on the causes of violations. Operators simulate and verify adjustment plans in virtual scenarios to check whether other constraints in the operating rules are violated; If the adjustment plan is verified, the operator confirms the execution of the adjustment through voice or gestures on the virtual interface, and the system sends the adjustment instruction to the simulation engine; The simulation engine re-executes based on the adjustments, and the operator continues to observe. If new violations are found, the adjustment process is repeated until there are no violations in the entire process; The virtual reality system automatically records every adjustment decision and observation of the operator and feeds this data back to the simulation engine module.

2. The nuclear power plant generator set overhaul rehearsal system according to claim 1, characterized in that: The steps for constructing the three-dimensional digital twin module are as follows: Obtain 2D CAD design drawings and existing BIM model data of the nuclear power plant's generator units and surrounding environment, and classify and organize the CAD drawings and BIM model data; Taking the overall layout of the nuclear power plant as a reference, a unified three-dimensional rectangular coordinate system is established; In BIM modeling software, three-dimensional parametric modeling of the nuclear power plant's building structure and pipeline layout is performed based on existing BIM model data; Import 2D CAD drawings into BIM modeling software, perform 3D solid modeling of each system component and equipment based on the 2D CAD drawings, and integrate the 3D solid model with the BIM model; Use model checking tools to detect whether there are errors in the integrated 3D model and repair the 3D model based on the detection results; Secondary simplification of non-core areas irrelevant to simulation analysis in the 3D model, while refining the core areas relevant to simulation analysis; Integrate the geometric information of the 3D model with non-geometric information including parameter information and material properties of system components and equipment to form a 3D digital twin model; The three-dimensional digital twin model is published to the simulation analysis platform and integrated with the artificial intelligence simulation engine module, virtual reality interaction module, and data acquisition module.

3. The nuclear power plant generator set overhaul rehearsal system according to claim 2, characterized in that: Repairing the three-dimensional model according to the detection results includes the following steps: If a face overlap error is detected, run the repair geometry function of the modeling software to automatically repair the overlapped faces; If a spatial position conflict is detected between the pipeline and the equipment model, the pipeline route will be adjusted in the pipeline professional model to eliminate the interference; If it is detected that a certain equipment model is missing auxiliary bracket components, the bracket structure dimension data will be obtained through on-site measurement and the corresponding components will be supplemented in the model; If it is detected that a surveillance camera is missing from the model of a certain area, and the design document has clear requirements, the model will be supplemented based on the camera model and location; Use model checking tools to conduct a comprehensive check on the repaired 3D model to ensure that there are no residual errors after the repair.

4. The nuclear power plant generator set overhaul rehearsal system according to claim 1, characterized in that: The steps for constructing the artificial intelligence simulation engine module are as follows: Acquire 3D digital twin model data from the 3D digital twin module, and acquire real-time on-site equipment working status data from the data acquisition module; Based on the acquired data, a virtual nuclear power plant environment model is constructed in the physical engine, and the on-site equipment working status data is used as the physical simulation input parameters; Convert the standardized overhaul operation process rules into executable rule sets and import them into the constructed virtual environment model; The physics engine drives the simulation and obtains the simulation data generated during the simulation in real time; Determine whether the obtained simulation data violates the constraints in the executable rule set, and if so, send a prompt signal to the virtual reality interaction module; Operators use virtual reality equipment to interactively observe and control the violation simulation process, and manually intervene and adjust the violation links to make them comply with operating standards; Record all key indicator values ​​during the adjusted simulation process; Based on the recorded data, the machine learning algorithm automatically identifies bottlenecks and risk points in the simulation process, and adjusts the operating parameters through the optimization algorithm to generate multiple optimized alternative overhaul plans; Multiple rounds of evaluation simulation are performed on each optimized alternative plan to determine its comprehensive performance in terms of construction period, cost, and radiation dose.

5. The nuclear power plant generator set overhaul rehearsal system according to claim 1, characterized in that: The proposed adjustment plan for the reasons for the violation includes: If it is detected that the operation time has exceeded the specified critical time node, the cause will be analyzed in combination with the time constraint requirements. If the analysis finds that the actual start time is delayed due to the delay in the preparation phase, the personnel and materials for the next process will be deployed in advance. Once the current operation is completed, the next operation will be seamlessly connected to catch up with the progress; If on-site personnel are missing during equipment relocation and installation, an analysis will be conducted based on staffing requirements. If the analysis reveals that the actual number of workers is insufficient and violates the staffing standards for the corresponding types of work, the required crane operators and safety personnel will be deployed from other areas. The personnel already on site will carry out preparatory work without radiation risks, and the formal implementation will be carried out after the personnel are replenished. During chemical pollution reduction operations, if it is detected that the ventilation system in the operating area is not operating normally, the cause will be analyzed in combination with environmental constraints. If the analysis finds that the ventilation parameters at the operating site do not meet the standards and there is a safety hazard of chemical leakage, the operation will be suspended immediately, the ventilation system fault will be checked first, and additional monitoring posts will be deployed in the vacant areas upwind. Operations will not continue until the conditions are met.

6. The nuclear power plant generator set overhaul rehearsal system according to claim 1, characterized in that: The steps for constructing the artificial intelligence decision-making module are as follows: Obtain multiple optimized alternative overhaul plans from the artificial intelligence simulation engine module; According to the preset ratio, the historical overhaul plan data is randomly divided into two parts: training set and test set; Standardize the multi-dimensional data in the training and test sets to convert data of different dimensions into the same numerical range; Based on deep learning technology, a multi-input and multi-output decision-making evaluation model is constructed. The input is the performance data of each alternative plan in different dimensions, and the output is the comprehensive score of each plan. Using the data in the training set, the decision evaluation model is trained through the back propagation algorithm to learn the influence weights of data of different dimensions on the final decision; Evaluate the trained decision evaluation model on the test set and analyze its generalization ability on unseen data based on the evaluation indicators; If the test results are not ideal, the decision evaluation model will be optimized by adjusting the network structure, optimizing hyperparameters, and increasing training data; The multi-dimensional data of each optimized alternative plan is input into the trained decision evaluation model, and the model outputs the comprehensive score of each plan; All alternatives are ranked according to the comprehensive scores output by the model, and the one with the highest score is selected as the optimal overhaul plan given by the artificial intelligence decision-making module; After the optimal solution is actually executed, its actual performance is recorded and compared with the model prediction, and this new data is added to the historical data set.

7. A nuclear power plant generator set overhaul rehearsal method, based on the nuclear power plant generator set overhaul rehearsal system according to any one of claims 1 to 6, characterized in that: Also includes, Using the 3D digital twin module, a 3D digital twin model is constructed using CAD and BIM technologies based on the design drawings and pipeline layout of the nuclear power plant generator units and surrounding environment; Through the data acquisition module, various sensors are used to obtain the working status data of nuclear power plant equipment in real time, providing real input parameters for the three-dimensional digital twin model; In the artificial intelligence simulation engine module, the physical engine and process rules are integrated to conduct a virtual simulation of the entire overhaul process based on a three-dimensional digital twin model and real-time data; Use machine learning algorithms to analyze simulation results and intelligently adjust and optimize the initial overhaul plan; Through the virtual reality interaction module, operators observe and interactively control the overhaul simulation process in a virtual environment; The AI ​​decision-making module uses deep learning technology to comprehensively evaluate and compare the performance of each optimized overhaul plan from multiple dimensions and provide the best decision recommendations; The blockchain security module uses blockchain technology to ensure data traceability and tamper-proofing throughout the entire overhaul process, and automatically monitors work compliance through smart contracts.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the nuclear power plant generator set overhaul rehearsal system according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the nuclear power plant generator set overhaul rehearsal system according to any one of claims 1 to 6 are implemented.

Images