Method for analyzing switching state of nuclear power plant and computer readable storage medium

Abstract

The application discloses a nuclear power equipment switching state analysis method and a computer readable storage medium, and the method comprises the following steps: collecting data of each nuclear power equipment; identifying characteristic data related to the health state of the nuclear power equipment in the data of each nuclear power equipment; dividing modules according to the types of the nuclear power equipment, and constructing a respective model for each module, wherein the model is used to analyze the switching state of the corresponding nuclear power equipment according to the characteristic data related to the health state of the nuclear power equipment. The application modularizes the switching state analysis module of the nuclear power equipment through modular modeling, each module focuses on the switching state analysis of the corresponding nuclear power equipment, improves the flexibility, maintainability and scalability of the model, can meet the needs of future equipment upgrading and function expansion, can comprehensively cover various nuclear power equipment in a nuclear power station, helps management personnel to identify potential problems in advance, reduces the probability of shutdown and downtime, and significantly improves the intelligent level and safety of nuclear power equipment management.

Description

Technical Field

[0001] This invention relates to the field of nuclear power equipment management technology, and in particular to a method for analyzing the switching status of nuclear power equipment and a computer-readable storage medium. Background Technology

[0002] With the rapid development of the nuclear power industry, the operational safety and reliability of nuclear power equipment have become core issues in nuclear power plant management. The on / off state of nuclear power equipment directly affects its operational efficiency and safety, thus requiring monitoring and precise analysis. However, existing technologies for analyzing the on / off state of nuclear power equipment suffer from the following major problems: Nuclear power plants have a large number of complex equipment types, making it difficult for traditional methods to comprehensively cover all equipment; data sources are diverse and their formats are complex, limiting the data acquisition and preprocessing capabilities of traditional methods and making it difficult to guarantee data reliability and consistency; traditional models typically use lengthy formulas, making it difficult to accurately reflect the nonlinear relationships of equipment on / off states, and their complex structures make maintenance and upgrades difficult; and the low accuracy of error assessment makes it difficult to provide accurate basis for on / off state analysis, affecting the reliability and safety of equipment management. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to address at least one deficiency of the related technologies mentioned in the background: traditional nuclear power equipment switch status analysis is difficult to fully cover all equipment, and the model structure is complex, making maintenance and upgrading difficult. The present invention provides a method for analyzing the switch status of nuclear power equipment and a computer-readable storage medium.

[0004] The technical solution adopted by this invention to solve its technical problem is: to construct a method for analyzing the switching state of nuclear power equipment, comprising the following steps: Collect data from various nuclear power equipment; Identify characteristic data related to the health status of each nuclear power plant from the data of each nuclear power plant; The modules are divided according to the type of nuclear power equipment. Each module constructs its own model, which is used to analyze the on / off state of the corresponding nuclear power equipment based on the characteristic data related to the health status of the nuclear power equipment.

[0005] In some embodiments, the switch state includes a basic state and an abnormal state; The basic status includes at least one of the following: on / run, off / stop, testing, and under maintenance; The abnormal state includes at least one of the following states: unexpected disconnection, startup failure, instruction response delay, instruction refusal to execute, and instruction sticking / invalidity.

[0006] In some embodiments, each of the modules constructs its own model, including: The model is constructed by training on historical data using machine learning or deep learning algorithms, enabling the model to autonomously learn the nonlinear mapping relationship between the characteristic data related to the health status of nuclear power equipment and the switching status under different operating conditions.

[0007] In some embodiments, characteristics refer to key parameters, metrics, and attributes used to describe, evaluate, and predict the switching state of a device, including at least one of operating parameters, performance metrics, condition monitoring data, maintenance and historical records, environmental and operating condition data, and switching states and control signals.

[0008] In some embodiments, analyzing the on / off state of the corresponding nuclear power equipment based on characteristic data related to the health status of the nuclear power equipment includes: If the characteristic data related to the health status of the nuclear power equipment are under normal operating conditions, then the analysis of the corresponding nuclear power equipment's on / off state indicates it is in operation; or, If the characteristic data related to the health status of the nuclear power equipment shows that the nuclear power equipment is not started or has been shut down, then the analysis of the corresponding switch status of the nuclear power equipment indicates that it is stopped, under testing, or under maintenance; or, If any abnormal value appears in the characteristic data related to the health status of the nuclear power equipment, it will be mapped to the specific abnormal state through a preset rule.

[0009] In some embodiments, the method further includes: Historical data is continuously used to train and evaluate the models of each module, and optimization algorithms are used to dynamically adjust the parameters of the models in order to continuously optimize and update the analytical performance of the models.

[0010] In some embodiments, the error assessment method includes: defining the assessment objective, selecting multiple indicators for comprehensive assessment, dynamically adjusting error weights, performing phased error analysis, and visualizing error analysis; The optimization algorithm includes: defining the optimization objective, selecting a suitable optimization algorithm, parameter initialization, iterative optimization, error feedback mechanism, verification and updating, setting termination conditions, and model updating and deployment.

[0011] In some embodiments, the method further includes: Based on the operating specifications and historical data of nuclear power equipment, the safety range and alarm threshold of the characteristic data related to the health status of nuclear power equipment are set, and the thresholds are dynamically adjusted according to the dynamic characteristics of the equipment to adapt to different operating conditions. During the judgment process, each characteristic data related to the health status of nuclear power equipment is independently judged by threshold, and the changing trends and interrelationships of multiple characteristic data related to the health status of nuclear power equipment over time are comprehensively analyzed by statistical analysis or machine learning models to identify anomalies.

[0012] In some embodiments, the method further includes: When analyzing the changing trends of the characteristic data related to the health status of nuclear power equipment over time, the analysis also combines the historical data and switch status sequences to analyze potential fault risks, predict potential fault risks, and trigger preventive maintenance alarms in advance.

[0013] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for analyzing the switching state of nuclear power equipment as described in any of the preceding claims.

[0014] By implementing this invention, the following beneficial effects are achieved: This invention modularizes the analysis of the switching status of nuclear power equipment through modular modeling. Each module focuses on the analysis of the switching status of the corresponding nuclear power equipment, which improves the flexibility, maintainability, and scalability of the model. It can meet the needs of future equipment upgrades and functional expansion, and at the same time, it can comprehensively cover various nuclear power equipment in nuclear power plants. It helps managers to identify potential problems in advance, reduce the probability of reactor shutdown and downtime, significantly improve the intelligence level and safety of nuclear power equipment management, and provide reliable technical support for the safe and stable operation of nuclear power plants. Attached Figure Description

[0015] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings: Figure 1 A flowchart of an embodiment of the method for analyzing the switching status of nuclear power equipment according to the present invention is shown. Detailed Implementation

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

[0017] It should be noted that the flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0018] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0019] The following can be one, two, three or any number.

[0020] like Figure 1 As shown, some embodiments of the present invention disclose a method for analyzing the switching state of nuclear power equipment, including the following steps: Collect data from various nuclear power equipment; Identify characteristic data related to the health status of each nuclear power plant from the data of each nuclear power plant; The modules are divided according to the type of nuclear power equipment. Each module builds its own model, which is used to analyze the on / off status of the corresponding nuclear power equipment based on characteristic data related to the health status of the nuclear power equipment.

[0021] The switch status includes a basic status and an abnormal status. The basic status includes at least one of the following: on / run, off / stop, testing, and maintenance. The abnormal status includes at least one of the following: unexpected disconnection, startup failure, instruction response delay, instruction refusal to execute, and instruction sticking / invalidity.

[0022] Specifically, the basic state is the stable operating mode exhibited by nuclear power equipment during normal operation. Among them, "on / operating" indicates that the equipment is performing its functions normally according to the preset program; "off / stopped" indicates that the equipment is in a non-operating state and no operation is being performed; "under testing" indicates that the equipment is undergoing performance testing and commissioning; and "under maintenance" indicates that the equipment is undergoing planned maintenance or fault repair.

[0023] Abnormal states reflect unexpected situations that occur during equipment operation. Unexpected disconnections may be caused by circuit faults or external interference; startup failures may be related to abnormalities in the power system or control signals; delayed command response can affect the timeliness of equipment operation; refusal to execute commands may stem from control system malfunctions; and command sticking / invalidity can cause the equipment to continuously perform erroneous operations or fail to respond to valid commands. Accurate identification of these abnormal states is crucial for ensuring the safe and stable operation of nuclear power equipment.

[0024] This invention modularizes the analysis of the switching status of nuclear power equipment through modular modeling. Each module focuses on the analysis of the switching status of the corresponding nuclear power equipment, which improves the flexibility, maintainability, and scalability of the model. It can meet the needs of future equipment upgrades and functional expansion, and at the same time, it can comprehensively cover various nuclear power equipment in nuclear power plants. It helps managers to identify potential problems in advance, reduce the probability of reactor shutdown and downtime, significantly improve the intelligence level and safety of nuclear power equipment management, and provide reliable technical support for the safe and stable operation of nuclear power plants.

[0025] In some embodiments, the analysis of the on / off status of nuclear power equipment relies on a large amount of data from diverse sources with complex formats. To address the challenges in data processing, data requirements must be clearly defined during data collection. Therefore, data is collected from various nuclear power equipment, specifically including: Data is collected from each nuclear power plant based on at least one preset acquisition standard, including data type, acquisition frequency, and accuracy. For example, data types include at least one of real-time operating data, historical maintenance records, fault data, and external environmental data to ensure coverage of operating parameters throughout the entire lifecycle of the nuclear power plant, thereby comprehensively capturing potential factors affecting the on / off status. The acquisition frequency is either real-time or timed. The accuracy must meet the requirements of subsequent model analysis to ensure that the collected data accurately reflects the characteristics of the nuclear power plant.

[0026] In some embodiments, to ensure data reliability and consistency, the method further includes: The collected data undergoes cleaning, standardization, and missing value processing. Cleaning involves removing noise and outliers; standardization involves unifying the data format, units, and time base; and missing value processing involves filling in missing values.

[0027] In some embodiments, to support the storage and rapid access of large-scale data while ensuring data security, the method further includes: After collection, the data is integrated and stored in the corresponding databases according to data type and access frequency. For example, frequently accessed real-time running data is stored in an in-memory database to improve response speed, while infrequently accessed historical maintenance records, fault data, and external environment data are stored in relational databases or distributed file systems to balance storage cost and access efficiency.

[0028] In some embodiments, each module builds its own model, specifically including: Machine learning or deep learning algorithms are employed to build models by training on historical data. These models can autonomously learn the nonlinear mapping relationships between characteristic data related to the health status of nuclear power equipment and its on / off states under different operating conditions. Examples include nonlinear mappings such as sudden drops in reactor coolant pressure corresponding to unexpected valve disconnection and abnormal control rod positions corresponding to startup failures. Simultaneously, a cross-validation mechanism is introduced to continuously optimize the model's performance, ensuring the accuracy and reliability of the analysis results and providing strong support for real-time monitoring and fault early warning of nuclear power equipment on / off states.

[0029] In some embodiments, characteristics refer to key parameters, metrics, and attributes used to describe, evaluate, and predict the switching state of a device, including at least one of operating parameters, performance metrics, condition monitoring data, maintenance and historical records, environmental and operating condition data, and switching states and control signals.

[0030] If the nuclear power equipment is a reactor, it is divided into reactor modules. A reactor module construction model is used, and the characteristic data related to the reactor's health status includes at least one of the following: operating parameters, condition monitoring data, and maintenance and historical records. Operating parameters include at least one of the following: coolant inlet temperature, coolant outlet temperature, coolant pressure, coolant flow rate, reactor power level (thermal power), control rod position / insertion depth, neutron flux density, and core pressure drop. Condition monitoring data includes at least one of the following: pressure vessel wall temperature, in-core component vibration signals, and coolant radioactivity (used to determine fuel cladding damage). Maintenance and historical records include at least one of the following: current fuel cycle duration, cumulative full-power operating hours, and the date and results of the last major overhaul (e.g., pressure vessel non-destructive testing report).

[0031] If the nuclear power equipment is a steam generator, it is divided into a steam generator module. The steam generator module construction model includes at least one of the following characteristic data related to the health status of the steam generator: operating parameters, performance indicators, condition monitoring data, and maintenance and historical records. Operating parameters include at least one of the following: primary side (tube side) inlet temperature, pressure, and flow rate; secondary side (shell side) water level; steam outlet pressure, temperature, and flow rate; feedwater temperature, pressure, and flow rate; and heat transfer tube outer wall temperature. Performance indicators include at least one of the following: terminal temperature difference (difference between saturation temperature and feedwater temperature) and heat transfer efficiency. Condition monitoring data includes at least one of the following: tube sheet temperature, U-tube vibration monitoring, and wastewater quality analysis (detecting minor leaks in heat transfer tubes). Maintenance and historical records include at least one of the following: the last heat transfer tube eddy current test time and result, cumulative operating hours, and historical tube blockage records (number and location).

[0032] If the nuclear power equipment type is a cooling pump (or main pump / circulating pump), it is divided into a cooling pump module. The cooling pump module model includes at least one of the following characteristic data related to the cooling pump's health status: operating parameters, condition monitoring data, maintenance and historical records, and switch status and control signals. Operating parameters include at least one of the following: motor current / voltage / power, pump inlet / outlet pressure (head), pump flow rate, and pump speed. Condition monitoring data includes at least one of the following: bearing temperature, bearing vibration (velocity, acceleration, and displacement), shaft displacement, lubricating oil temperature, pressure, and cleanliness, and sealing water pressure, flow rate, and temperature. Maintenance and historical records include at least one of the following: cumulative operating hours, last overhaul date, replaced components (e.g., bearings, mechanical seals), and historical vibration trend analysis records. Switch status and control signals include at least one of the following: pump start / stop status signals and speed setpoint.

[0033] It should be noted that the reactor, steam generator, and cooling pump mentioned above are merely examples; other nuclear power equipment, such as diesel generator sets (i.e., emergency power supplies), can also be used. The health status characteristics of diesel generator sets include at least one of the following: operating parameters, condition monitoring data, and maintenance and historical records. Operating parameters include at least one of the following: engine speed, oil pressure, water temperature, exhaust temperature, and output voltage / frequency. Condition monitoring data includes at least one of the following: start-up battery voltage, vibration, and lubricating oil analysis. Maintenance and historical records include at least one of the following: last start-up test time, success rate, and cumulative number of starts.

[0034] For example, for valves (such as safety valves and control valves), the characteristic data related to the valve's health status include at least one of the following: operating parameters, condition monitoring data, and maintenance and historical records. Operating parameters include at least one of valve opening degree, differential pressure across the valve, and medium temperature. Condition monitoring data includes at least one of actuator motor current / torque (reflecting jamming) and valve stem displacement. Maintenance and historical records include at least one of the following: last operation test time, leakage rate test results, and maintenance cycle.

[0035] For example, for a transformer, characteristic data related to the transformer's health status includes at least one of the following: operating parameters, condition monitoring data, and maintenance and historical records. Operating parameters include at least one of load current, winding temperature, and oil temperature. Condition monitoring data includes at least one of dissolved gas analysis and partial discharge monitoring. Maintenance and historical records include at least one of the following: the last oil testing date and preventative test results.

[0036] In some embodiments, the on / off state of the corresponding nuclear power equipment is analyzed based on characteristic data related to the health status of the nuclear power equipment, specifically including: If the characteristic data related to the health status of the nuclear power equipment are under normal operating conditions (e.g., coolant flow rate within the set range or control rod position normal), then the analysis of the corresponding nuclear power equipment's on / off status is "operating"; or, If the characteristic data related to the health status of nuclear power equipment shows that the equipment is not started or has been shut down, then the analysis of the corresponding nuclear power equipment's on / off status is as follows: stopped, under testing, or under maintenance; or, If abnormal values ​​appear in characteristic data related to the health status of nuclear power equipment, they will be mapped to specific abnormal states through preset rules.

[0037] For example, if the coolant flow rate in the reactor is below the threshold and / or the control rod position is abnormal, it is determined that the relevant regulating valves of the reactor are at risk of jamming or failure, and then the opening and closing status of the regulating valves of the reactor is analyzed to determine whether the command response is delayed or the command is refused to be executed.

[0038] For example, if the steam outlet pressure on the secondary side of the steam generator fluctuates abnormally and the temperature distribution on the outer wall of the heat transfer tube is uneven, then the opening and closing status of the relevant steam traps or regulating valves of the steam generator should be analyzed as startup failure, accidental disconnection, or command sticking / invalidity.

[0039] For example, if the motor current of the cooling pump is overloaded and the pressure difference between the inlet and outlet is abnormally reduced, the switching status of the cooling pump motor can be analyzed as command sticking / invalid.

[0040] In some embodiments, the method further includes: Based on the analysis results of each module, the risk of the switch status is visualized in at least one of the forms of rating, level and probability, and an automatic alarm is triggered when the risk is higher than the warning value.

[0041] In some embodiments, the method further includes: The analysis results are fed back to the terminal so that managers can take emergency measures.

[0042] In some embodiments, the modules are interconnected to ensure seamless integration between them, facilitating data transfer and result aggregation.

[0043] In some embodiments, the method further includes: By continuously using historical data to train and evaluate the models of each module, and by dynamically adjusting the parameters of the models in conjunction with optimization algorithms, the analytical performance of the models is continuously optimized and updated. This further improves the sensitivity and accuracy of reactor switching status analysis, ensuring that potential problems can be detected in a timely manner even under complex operating conditions, and providing continuous assurance for the safe and stable operation of the reactor.

[0044] Error assessment and model optimization are key steps in improving the accuracy and stability of equipment switching status analysis. By improving error assessment methods and introducing dynamic optimization algorithms, the accuracy and reliability of status analysis can be enhanced. This intelligent optimization capability enables the system to maintain efficient operation even in complex environments.

[0045] The specific methods for error assessment include: defining the assessment objectives, selecting multiple indicators for comprehensive assessment, dynamically adjusting error weights, conducting phased error analysis, and visualizing error analysis.

[0046] For example, clearly defining the evaluation objectives includes: defining the core indicators the model aims to predict and the dimensions to be evaluated. Core indicators might include response time error of switching actions and / or accuracy of state recognition. Evaluation dimensions would cover model performance under different operating conditions, such as stability during normal operation, adaptability to sudden load changes, and robustness under aging conditions. By combining these core indicators with evaluation dimensions, a clear direction and specific measurement standards can be provided for subsequent error assessment, ensuring a more targeted and effective evaluation process.

[0047] Choosing a multi-indicator comprehensive evaluation aims to avoid the one-sidedness of a single indicator. Specifically, this includes: measuring the overall deviation by combining mean squared error (MSE) and reflecting the level of absolute error and accuracy by using mean absolute error (MAE), thus preventing deviations in the direction of model optimization.

[0048] Among them, mean squared error effectively reflects the squared deviation between the predicted and the true values, and is suitable for measuring the cumulative effect of overall error. Mean absolute error, on the other hand, can intuitively show the average magnitude of the error, avoiding the excessive influence of extreme values ​​on the evaluation results.

[0049] Dynamically adjusting error weights specifically involves quantifying the cost of different error types. For example, the weight of "false positives" can be set higher than that of "false negatives," thus making the model more aligned with the industry's safety-first requirements. This differentiated weight allocation mechanism guides the model to focus more on error types that have a greater impact on system safety during training.

[0050] This also includes assigning dynamically changing weights to errors in different scenarios, allowing the evaluation system to better adapt to the complex and ever-changing environmental conditions in actual operation. For example, different weights are set for the same type of error during the equipment startup phase and the stable operation phase, to ensure that the evaluation results always match the current operating status and core requirements.

[0051] Phased error analysis specifically includes: decomposing and evaluating model errors according to operational phases (such as startup, power operation, and shutdown) to ensure that the model remains reliable throughout the entire operational cycle.

[0052] Specifically, during the startup phase, the focus is on the model's adaptive error to initial parameters, analyzing whether the prediction deviation caused by the system's cold start is within acceptable limits. During the power operation phase, the focus is on error stability under continuous load changes, assessing the model's prediction accuracy degradation under long-term high load conditions. During the shutdown phase, special monitoring is needed on the error response speed during sudden changes in key parameters to ensure the model can promptly capture rapid system state transitions. This phased, refined analysis not only identifies the sources of error at different stages but also provides clear directions for phased model optimization. For example, optimizing the initial parameter calibration algorithm for errors during the startup phase and improving the model's dynamic tracking capability during the power operation phase can systematically improve the model's overall performance throughout its entire lifecycle.

[0053] Visual error analysis specifically includes using charts to present abstract error data intuitively, which helps engineers quickly locate error patterns.

[0054] The optimization algorithm specifically includes: defining the optimization objective, selecting a suitable optimization algorithm, parameter initialization, iterative optimization, error feedback mechanism, verification and updating, setting termination conditions, and model updating and deployment.

[0055] Define the optimization objective, which specifically includes: transforming the optimization objective into a quantifiable mathematical function, such as a loss function.

[0056] Choosing a suitable optimization algorithm specifically includes selecting the most effective search strategy based on the characteristics of the model and the shape of the loss function.

[0057] Parameter initialization specifically includes setting a reasonable starting point for the model, rather than starting from zero. This allows control over the range of initial parameters, preventing gradient explosion or vanishing problems in the early stages of training.

[0058] Iterative optimization specifically includes: repeatedly calculating the loss function and updating parameters with a preset step size (such as a small step size) in order to make the model output closer to the true value.

[0059] The error feedback mechanism specifically includes: evaluating the model performance on an independent validation dataset after each iteration or training cycle, and feeding the calculated error back to the optimization algorithm to guide the optimization direction.

[0060] The verification and update process specifically includes: after obtaining the verification error, determining the validity of the parameter update. If the verification error decreases significantly, the parameter update is accepted.

[0061] Termination conditions are set, specifically including: to avoid infinite loops and to determine the best stopping time, it is necessary to preset clear termination conditions, including: the loss value converges (no longer decreases significantly) and / or the preset maximum number of iterations is reached.

[0062] Model updates and deployments specifically include: once the termination conditions are met, the parameters of the finally validated model are fixed, packaged into a new model version, and used to replace the old version.

[0063] In some embodiments, the method further includes: Based on the operating specifications and historical data of nuclear power equipment, set safe ranges and alarm thresholds for characteristic data related to the health status of nuclear power equipment, and dynamically adjust the thresholds according to the dynamic characteristics of the equipment to adapt to different operating conditions. During the assessment process, each characteristic data related to the health status of nuclear power equipment is subject to an independent threshold assessment. Statistical analysis or machine learning models are used to comprehensively analyze the changing trends and interrelationships of multiple characteristic data related to the health status of nuclear power equipment over time to identify anomalies.

[0064] This approach allows for a more comprehensive capture of potential anomalies in the switching status of nuclear power equipment, reducing the possibility of misjudgments based on single-characteristic data. Simultaneously, combined with a dynamic threshold adjustment mechanism, it maintains monitoring accuracy across different operational phases, providing a more reliable data foundation for subsequent switching status analysis and ensuring stable operation and timely early warning of nuclear power equipment under complex conditions.

[0065] In some embodiments, the method further includes: When analyzing the changing trends of the characteristic data related to the health status of nuclear power equipment over time, historical data and switch status sequences are also combined to analyze potential fault risks, predict potential fault risks, and trigger preventive maintenance alarms in advance.

[0066] This allows potential problems to be detected in time before obvious signs of equipment failure appear, avoiding unplanned shutdowns caused by sudden failures and significantly improving the safety and reliability of nuclear power equipment operation.

[0067] In some embodiments, the method further includes: Once an alarm is triggered, it is prioritized and categorized according to its severity and urgency. Management personnel are notified through multiple methods, and detailed information on the alarm cause, current equipment status, and handling recommendations are provided. Abnormal states of nuclear power equipment (such as unexpected disconnection or startup failure) are treated as one of the highest priority alarms.

[0068] In some embodiments, the method further includes: By combining historical records of device on / off status and alarm records, the alarm judgment logic is dynamically adjusted to optimize system performance and reduce the possibility of false alarms and missed alarms.

[0069] In some embodiments, the method further includes: Record the trigger time, data value, handling measures and results of each alarm, analyze the alarm data regularly, further optimize the alarm judgment logic, and ensure the safe and stable operation of nuclear power equipment.

[0070] This invention, through alarm analysis and intelligent decision support, can comprehensively consider the historical records of equipment switching status and alarm records to generate more accurate alarm information and provide scientific maintenance or operation suggestions. This not only improves the accuracy of alarms but also reduces false alarms and missed alarms, providing stronger technical support for the safe operation of nuclear power plants.

[0071] In summary, this invention constructs an efficient, flexible, and safe nuclear power equipment switch status analysis system through the collaborative work of four core modules: modular modeling, high-precision data acquisition and processing, error assessment and system optimization, and alarm comprehensive decision-making. This reduces the need for manual intervention, lowers labor costs, and improves work efficiency, significantly enhancing the intelligence level and safety of nuclear power equipment management and providing reliable technical support for the safe and stable operation of nuclear power plants.

[0072] Some embodiments of the present invention also disclose a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method for analyzing the switching state of nuclear power equipment as described in any of the above embodiments, and will not be repeated here.

[0073] It is understood that the above embodiments only illustrate some implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can freely combine the above embodiments or technical features without departing from the concept of the present invention, and can also make several modifications and improvements, all of which fall within the protection scope of the present invention. That is, the embodiments described "in some embodiments" can be freely combined with any of the preceding and following embodiments. Therefore, all equivalent transformations and modifications made within the scope of the claims of the present invention should be covered by the claims of the present invention.

Claims

1. A method for analyzing the switching status of nuclear power equipment, characterized in that, Includes the following steps: Collect data from various nuclear power equipment; Identify characteristic data related to the health status of each nuclear power plant from the data of each nuclear power plant; The modules are divided according to the type of nuclear power equipment. Each module constructs its own model, which is used to analyze the on / off state of the corresponding nuclear power equipment based on the characteristic data related to the health status of the nuclear power equipment.

2. The method of claim 1, wherein, The switch states include basic states and abnormal states; The basic status includes at least one of the following: on / run, off / stop, testing, and under maintenance; The abnormal state includes at least one of the following states: unexpected disconnection, startup failure, instruction response delay, instruction refusal to execute, and instruction sticking / invalidity.

3. The method of claim 1, wherein, Each of the modules constructs its own model, including: The model is constructed by training on historical data using machine learning or deep learning algorithms, enabling the model to autonomously learn the nonlinear mapping relationship between the characteristic data related to the health status of nuclear power equipment and the switching status under different operating conditions.

4. The method of claim 1, wherein, Characteristics refer to key parameters, indicators, and attributes used to describe, evaluate, and predict the switching status of equipment, including at least one of operating parameters, performance indicators, condition monitoring data, maintenance and historical records, environmental and operating condition data, and switching status and control signals.

5. The method of claim 2, wherein, The analysis of the on / off status of corresponding nuclear power equipment based on characteristic data related to the health status of the nuclear power equipment includes: If the characteristic data related to the health status of the nuclear power equipment are under normal operating conditions, then the analysis of the corresponding nuclear power equipment's on / off state indicates it is in operation; or, If the characteristic data related to the health status of the nuclear power equipment shows that the nuclear power equipment is not started or has been shut down, then the analysis of the corresponding switch status of the nuclear power equipment indicates that it is stopped, under testing, or under maintenance; or, If any abnormal value appears in the characteristic data related to the health status of the nuclear power equipment, it will be mapped to the specific abnormal state through a preset rule.

6. The method of claim 1, wherein, The method further includes: Historical data is continuously used to train and evaluate the models of each module, and optimization algorithms are used to dynamically adjust the parameters of the models in order to continuously optimize and update the analytical performance of the models.

7. The method of claim 6, wherein the method further comprises: The error assessment method includes: defining the assessment objective, selecting multiple indicators for comprehensive assessment, dynamically adjusting error weights, performing phased error analysis, and visualizing error analysis. The optimization algorithm includes: defining the optimization objective, selecting a suitable optimization algorithm, parameter initialization, iterative optimization, error feedback mechanism, verification and updating, setting termination conditions, and model updating and deployment.

8. The method of claim 1, wherein, The method further includes: Based on the operating specifications and historical data of nuclear power equipment, the safety range and alarm threshold of the characteristic data related to the health status of nuclear power equipment are set, and the thresholds are dynamically adjusted according to the dynamic characteristics of the equipment to adapt to different operating conditions. During the judgment process, each characteristic data related to the health status of nuclear power equipment is independently judged by threshold, and the changing trends and interrelationships of multiple characteristic data related to the health status of nuclear power equipment over time are comprehensively analyzed through statistical analysis or machine learning models to identify anomalies.

9. The method of claim 8, wherein, The method further includes: When analyzing the changing trends of the characteristic data related to the health status of nuclear power equipment over time, the analysis also combines the historical data and switch status sequences to analyze potential fault risks, predict potential fault risks, and trigger preventive maintenance alarms in advance.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for analyzing the switching state of nuclear power equipment as described in any one of claims 1-9.

Images