A method and device for detecting abnormalities in a nuclear power plant containment

Abstract

The present application relates to nuclear power safety protection technical field, especially to a kind of nuclear power plant box abnormality detection method and device.The method is: determining the ideal influence factor of the box;Preliminary determine the influence factor of the box;Determine the final influence factor, generate corresponding training data set;Establish reconstruction model and regression model, and train;Obtain the influence factor data of current box, input the influence factor data into reconstruction model and regression model to obtain two groups of prediction data of the box;Obtain the actual operation data of current box, compare actual operation data with the prediction data to determine whether the box is abnormal.The device includes: influence factor preliminary determination module;Potential influence factor determination module;Training set processing module;Reconstruction model training module;Regression model training module;Prediction module;Abnormality determination module.The present application is accurate, and early warning ability is strong.

Description

Technical Field

[0001] This invention relates to the field of nuclear power safety protection technology, and in particular to a method and device for detecting anomalies in nuclear power plant enclosures. Background Technology

[0002] During normal operation, nuclear power plants may experience unexpected protective actions, such as load shedding, shutdown, or reactor shutdown, due to reasons such as faulty or malfunctioning process equipment, instrumentation and control system failure, power grid failure, fire, insufficient personnel skills, or human error.

[0003] Nuclear reactor enclosures or tanks play a crucial role in nuclear power plants. Most enclosures serve only a storage function, such as common storage tanks; others also function as heat exchangers, such as pressurizers. If an enclosure malfunctions, it can lead to unexpected system shutdowns or outages, causing significant economic losses. Therefore, timely detection and resolution of enclosure malfunctions are of paramount importance.

[0004] The detection of abnormalities in liquid level, pressure, or temperature signals during the operation of the current tank mainly relies on alarm information from the DCS system. Alarms are only triggered when a single measurement or a combination of multiple measurements reaches the corresponding threshold. There is a lack of trend analysis, resulting in poor anomaly warning capabilities. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method and device for detecting abnormalities in nuclear power plant enclosures, with strong early warning capabilities.

[0006] This invention provides a method for detecting anomalies in nuclear power plant enclosures, comprising the following steps:

[0007] Step S100: Establish the mass conservation equation for the box, and determine the influence factor of the ideal of the box based on the mass conservation equation;

[0008] Step S200: Based on the actual deployment of the instrument and business knowledge, preliminarily determine the influencing factors of the box;

[0009] Step S300: Obtain several historical working data of the box, use data analysis methods to analyze the correlation between potential influencing factors and historical working data, so as to determine the final influencing factor of the box, and generate a corresponding training dataset based on the historical working data and the final determined influencing factor.

[0010] Step S400: Establish a reconstruction model and train the reconstruction model using the training dataset to obtain a fully trained box reconstruction model and warning thresholds for each influencing factor.

[0011] Step S500: Establish a regression model and train the regression model using the training dataset to obtain a fully trained box regression model and warning thresholds for the regression output parameters;

[0012] Step S600: Obtain the current impact factor data of the box, and input the current impact factor data into the reconstruction model and regression model to obtain two sets of prediction data for the box;

[0013] Step S700: Obtain the actual operating data of the current enclosure, and compare the actual operating data with the predicted data to determine whether the enclosure is abnormal.

[0014] In one specific embodiment of the present invention, the historical operating data includes at least historical liquid level difference data and its corresponding influencing factor data. The influencing factor data includes the inlet flow rate and temperature of the tank, the outlet flow rate and temperature of the tank, the pressure at the top of the tank, the liquid temperature, and the gas phase pressure.

[0015] In a specific embodiment of the present invention, in establishing the reconstruction model and the regression model, considering the influence of temperature and pressure on density, the inlet mass of the box is calculated by the inlet volumetric flow rate, the inlet liquid density, and the inlet liquid pressure. The outlet mass of the box is calculated by the outlet volumetric flow rate, the outlet liquid density, and the outlet liquid pressure. The inlet liquid density and the inlet liquid pressure are related to the inlet liquid temperature.

[0016] In one specific embodiment of the present invention, the reconstruction model employs principal component analysis (PCA) reconstruction, autoencoder (AE), and variational autoencoder (VAE) reconstruction algorithms.

[0017] In one specific embodiment of the present invention, the regression model is one of the following: linear regression model, multinomial regression, decision tree regression model, support vector machine regression model, K-nearest neighbor regression model, random forest regression model, Adaboost regression model, gradient boosting random forest regression model, bagging regression model, and Extra TREE regression model.

[0018] In a specific embodiment of the present invention, step S700 specifically comprises:

[0019] Obtain the actual working data of the current box, and compare the actual working data with the predicted working data to obtain the residual between the actual working data and the predicted working data;

[0020] The abnormal alarm threshold for working data is obtained based on the training results. The residual is compared with the abnormal alarm threshold for working data. Based on the comparison result, it is determined whether the box has an abnormality.

[0021] When an abnormality is detected in the enclosure, a corresponding alarm message is issued. The alarm message includes the specific details of the abnormality and the cause of the abnormality.

[0022] In one specific embodiment of the present invention, it further includes: when it is determined that the cause of the cabinet abnormality alarm is a false alarm or a missed alarm, obtaining the latest historical data and updating the cabinet abnormality detection model using the latest historical data.

[0023] This invention provides a nuclear power plant enclosure anomaly detection device, comprising:

[0024] The preliminary determination module for influencing factors is used to establish the mass conservation equation of the box and determine the influencing factors of the box based on the mass conservation equation.

[0025] The potential impact factor determination module is used to initially determine the impact factors of the box by combining the actual deployment of the instrument with business knowledge.

[0026] The training set processing module is used to acquire several historical working data of the enclosure, use data analysis methods to analyze the correlation between potential influencing factors and historical working data, in order to determine the final influencing factor of the heat exchanger, and generate the corresponding training dataset based on the historical working data and the final determined influencing factor.

[0027] The reconstruction model training module is used to build a reconstruction model and train the reconstruction model using the training dataset to obtain a fully trained box reconstruction model and warning thresholds for each influencing factor.

[0028] The regression model training module is used to build a regression model and train the regression model using the training dataset to obtain a fully trained box regression model and warning thresholds for the regression output parameters.

[0029] The prediction module is used to obtain the current impact factor data of the box, and input the current impact factor data into the reconstruction model and regression model for reconstruction to obtain two sets of prediction data of the box.

[0030] The anomaly detection module is used to obtain the actual operating data of the current enclosure, compare the actual operating data with the predicted data, and determine whether the enclosure has an anomaly.

[0031] This invention provides an electronic device, comprising: a processor and a memory;

[0032] The memory stores computer programs that can be executed by the processor;

[0033] When the processor executes the computer program, it implements the steps in the nuclear power plant enclosure anomaly detection method.

[0034] The present invention provides a computer-readable storage medium storing one or more programs, which can be executed by one or more processors to implement the steps in the nuclear power plant enclosure anomaly detection method.

[0035] Compared with existing technologies, the present invention provides a method and apparatus for detecting anomalies in nuclear power plant tanks. By acquiring the influencing factors of tank anomalies and using reconstruction and regression algorithms to construct an anomaly detection model for the tanks, the present invention enables the direct prediction of liquid level changes in the tanks through the anomaly detection model. By comparing the prediction with the actual liquid level changes, the present invention can quickly and accurately determine whether an anomaly has occurred in the tanks. Attached Figure Description

[0036] Figure 1 A flowchart illustrating the method for detecting anomalies in nuclear power plant enclosures;

[0037] Figure 2 A schematic diagram showing the functional modules of a nuclear power plant enclosure anomaly detection device;

[0038] Figure 3 A schematic diagram showing the hardware structure of an electronic device;

[0039] In the diagram, 10 is an electronic device, 110 is a processor, and 120 is a memory. Detailed Implementation

[0040] To further understand the present invention, embodiments of the present invention are described below in conjunction with examples. However, it should be understood that these descriptions are only for further illustrating the features and advantages of the present invention, and not for limiting the present invention.

[0041] An embodiment of the present invention discloses a method for detecting anomalies in a nuclear power plant enclosure, such as... Figure 1 As shown, it includes the following steps:

[0042] S100: Establish the mass conservation equation of the enclosure based on the instrument layout and business knowledge, and determine the influence factors of the enclosure based on the mass conservation equation;

[0043] S200: Establish the mass conservation equation of the box, and determine the influence factor of the ideal of the box based on the mass conservation equation;

[0044] S300: Acquire several historical working data of the box, use data analysis methods to analyze the correlation between potential influencing factors and historical working data, in order to determine the final influencing factor of the box, and generate a corresponding training dataset based on the historical working data and the final determined influencing factor.

[0045] S400: Establish a reconstruction model, and train the reconstruction model using the training dataset to obtain a fully trained box reconstruction model and warning thresholds for each influencing factor.

[0046] S500: Establish a regression model, and train the regression model using the training dataset to obtain a fully trained box regression model and warning thresholds for the regression output parameters;

[0047] S600: Obtain the current impact factor data of the box, and input the current impact factor data into the box reconstruction model and regression model to obtain two sets of predicted data for the box;

[0048] S700: Obtain the actual operating data of the current enclosure, compare the actual operating data with the predicted data, and determine whether the enclosure is abnormal.

[0049] In this embodiment, firstly, the mass conservation equation of the box is established, and the influencing factors of the box are determined based on the mass conservation equation; then, several historical working data of the box are acquired, and its influence shadow is finally determined and the required features are processed based on data analysis methods, and a training set is established; a reconstruction model is established, and the regression model is trained using the training dataset to obtain a fully trained box reconstruction model and warning thresholds for each influencing factor; a regression model is established, and the regression model is trained using the training dataset to obtain a fully trained box regression model and warning thresholds for the regression output parameters; then, the current influencing factor data of the box is acquired, and the current influencing factor data is input into the reconstruction model and the regression model to obtain the predicted working data of the box; finally, the current actual working data of the box is acquired, and the actual working data is compared with the predicted working data to determine whether the box has an anomaly. This invention obtains the influencing factors of tank anomalies and uses regression and reconstruction algorithms to construct two anomaly detection models for tanks. The anomaly detection models predict the liquid level and temperature changes of the tanks, and then compare them with the actual liquid level and temperature changes to quickly and accurately determine whether the tanks have anomalies.

[0050] In some embodiments, the mass conservation equation is used to reflect the change in the liquid level of the tank. Ideally, the increase in the liquid level of the tank is related to the mass change on both sides of the tank. Therefore, in order to obtain the liquid level change of the tank, the present invention can establish a mass conservation equation, and then obtain the ideal influencing factor of the liquid level change of the tank by converting the formula. Then, combined with the actual situation of instrument deployment and business knowledge, the influencing factor of the tank is initially determined.

[0051] In some embodiments, in step S300, the anomaly detection model is built based on knowledge and experience. According to the operating experience of nuclear power plants, historical operating data of the tank is obtained. This historical operating data includes at least historical liquid level difference data and its corresponding influencing factor data. Optionally, it includes the tank's inlet flow rate and temperature, the tank's outlet flow rate and temperature, the tank's top pressure (necessary for sealed tanks), liquid temperature, and gas phase pressure; some tanks also need to consider the flow rate, temperature, and pressure on the cooling side, or electric heating, etc. It should be noted that some tanks need to be divided into multiple operating conditions. Through data analysis methods, the influencing factors of the tank under each operating condition are finally determined, and the features to be processed are extracted, thereby forming a training dataset for that operating condition.

[0052] In this embodiment, according to the law of conservation of mass, in the absence of leakage, the entire contents of the box satisfy the law of conservation of mass, that is... Therefore, in this embodiment of the invention, the regression model is modeled based on the law of mass conservation, deltalevel = function(Min1, Min2, Mout1, level), that is, the change of liquid level in the tank is obtained by the mass difference between the two sides of the tank.

[0053] In some embodiments, under the premise that the density change is not significant, the mass is only related to the flow rate. However, in reality, temperature and pressure will oscillate, which will cause the density to change. Therefore, in the modeling process of this embodiment, the influence of temperature and pressure on density needs to be considered. Optionally, the inlet mass of the box is calculated by the inlet volume flow rate, the inlet liquid density, and the inlet liquid pressure. The outlet mass of the box is calculated by the outlet volume flow rate, the outlet liquid density, and the outlet liquid pressure. The inlet liquid density and the inlet liquid pressure are related to the inlet liquid temperature.

[0054] In some embodiments, during step S400, since the power plant may undergo maintenance, the data may change after the maintenance, which may lead to false alarms in the model prediction. Therefore, in order to avoid this situation, the impact factor data is reconstructed. The reconstructed data can best represent the characteristics of the impact factors and thus avoid false alarms. Moreover, by reconstructing the model to reconstruct the impact factor data, when the impact factor data of the tank deviates during actual operation, it can be used as a basis for judging the tank abnormality.

[0055] Optionally, the reconstruction model employs an autoencoder (AE), a variational autoencoder (VAE), PCA reconstruction, or the TransFormer algorithm.

[0056] In this embodiment, three reconstruction algorithms can be selected to build the reconstruction model according to actual needs. Specifically:

[0057] a. Autoencoder (AE): This is an unsupervised learning model. In machine learning and deep learning, its main goal is to learn how to represent data from the input space as a compact representation in the latent variable space and recover it back to the original input space as accurately as possible.

[0058] b. PCA Reconstruction Algorithm: Principal Component Analysis (PCA) is a dimensionality reduction method widely used in statistics, machine learning, and data analysis;

[0059] c. Variational Autoencoder (VAE): This is an unsupervised learning method used to generate and understand the latent structure of data. It combines the concepts of autoencoders (AEs) and Bayesian inference, enabling it to compress and reconstruct input data, and also generate new data samples. VAEs are particularly adept at discovering low-dimensional representations from high-dimensional data, making them widely applicable in many fields such as image generation, semantic embedding, and anomaly detection.

[0060] In some embodiments, in step S400, the regression model is a predictive modeling technique that studies the relationship between the dependent variable (target) and the independent variable (predictor) for predictive analysis, time series modeling, and discovering causal relationships between variables. Optionally, the regression model can be one of the following: linear regression model, decision tree regression model, support vector machine regression model, K-nearest neighbor regression model, random forest regression model, Adaboost regression model, gradient boosting random forest regression model, bagging regression model, and Extra TREE regression model.

[0061] In some embodiments, the input factors of the regression model are influencing factor data, and the output of the regression model is the liquid level difference.

[0062] In this embodiment, after obtaining a fully trained box anomaly detection model, the current box's influence factor data is acquired. After inputting the current box's influence factor data into the box anomaly detection model, the predicted liquid level difference can be quickly obtained.

[0063] In some embodiments, step S600 specifically includes:

[0064] Obtain the actual working data of the current box, and compare the actual working data with the predicted working data to obtain the residual between the actual working data and the predicted working data;

[0065] The alarm threshold for abnormal working data is obtained based on the training results. The residual is compared with the alarm threshold for abnormal working data. Based on the comparison result, it is determined whether the enclosure has an abnormality.

[0066] In this embodiment, a preset liquid level difference alarm threshold is defined as the maximum value of the error between the actual value and the predicted value. Within this error range, it indicates that there is no abnormality in the tank. Therefore, after obtaining the predicted liquid level difference through the model, the predicted liquid level difference is compared with the actual liquid level difference to obtain the difference between the two. Then, this difference is compared with the preset liquid level difference alarm threshold. If it exceeds the liquid level difference alarm threshold, it indicates that there is an abnormality in the tank; otherwise, it indicates that there is no abnormality in the tank.

[0067] In some embodiments, the method further includes:

[0068] When an abnormality is detected in the enclosure, a corresponding alarm message will be issued.

[0069] In this embodiment, when an abnormality is detected in the enclosure, an alarm message is issued. The alarm message includes the specific details of the abnormality and the cause of the abnormality, so that staff can respond to the alarm message quickly.

[0070] In some embodiments, the method further includes:

[0071] When the cause of the enclosure abnormality alarm is determined to be a false alarm or a missed alarm, the latest historical data is obtained and the enclosure abnormality detection model is updated using the latest historical data.

[0072] In this embodiment, the above models are all enclosure anomaly detection models built based on historical operating data, and the accuracy of the models is constrained by the data. Since historical operating data of the unit cannot cover all normal or abnormal states, anomaly detection models developed based on historical operating data will produce false alarms or missed alarms when encountering new normal or abnormal states. If the previous model is used for detection at this time, it will lead to an increase in the false alarm rate or missed alarm rate. Therefore, this invention adds an online update process, uses historical data for verification, and adjusts it according to the new data.

[0073] Specifically, the alarm information is first analyzed manually. If it is determined to be a false alarm, it indicates that there may be a new abnormal pattern that does not exist in the historical data. In this case, such samples need to be added to the training sample, and the model needs to be retrained and updated based on these samples. If it is determined to be a missed alarm, it indicates that there are cases that should have been identified by the existing nuclear power important parameter anomaly identification system, but were not actually identified. In this case, the features of the missed alarm are extracted and an anomaly feature library is established. Subsequently, new training is carried out based on the abnormal sample data in the anomaly feature library to obtain an updated tank anomaly detection model.

[0074] Another embodiment of the present invention provides a nuclear power plant enclosure anomaly detection device; please refer to [link to relevant documentation]. Figure 2 The anomaly detection device for the nuclear power plant enclosure includes:

[0075] The influence factor determination module is used to establish the mass conservation equation of the box and determine the influence factor of the box based on the mass conservation equation.

[0076] The potential impact factor determination module is used to initially determine the impact factors of the box by combining the actual deployment of the instrument with business knowledge.

[0077] The training set processing module is used to acquire several historical working data of the enclosure, and uses data analysis methods to analyze the correlation between potential influencing factors and historical working data in order to determine the final influencing factors of the heat exchanger. Based on the historical working data and the final determined influencing factors, a corresponding training dataset is generated.

[0078] The model training module is used to establish a reconstruction model and train the reconstruction model using the training dataset to obtain a fully trained box reconstruction model and warning thresholds for each influencing factor.

[0079] The regression model training module is used to build a regression model. The regression model is trained using the training dataset to obtain a fully trained box regression model and warning thresholds for the regression output parameters.

[0080] The prediction module is used to obtain the current impact factor data of the box, and input the current impact factor data into the reconstruction model and regression model for reconstruction to obtain two sets of prediction data of the box;

[0081] The anomaly detection module is used to obtain the actual operating data of the current enclosure, compare the actual operating data with the predicted data, and determine whether the enclosure has an anomaly.

[0082] It should be noted that the module referred to in this invention refers to a series of computer program instruction segments that can perform specific functions. It is more suitable than a program for describing the execution process of nuclear power plant enclosure anomaly detection. For the specific implementation of each module, please refer to the corresponding method embodiments above, which will not be repeated here.

[0083] In some embodiments, the inlet flow rate and temperature of the influencing factor data box, the outlet flow rate and temperature of the box, the pressure at the top of the box, the liquid temperature, and the gas phase pressure are all included.

[0084] In some embodiments, the inlet mass of the housing is calculated using the inlet volumetric flow rate, inlet liquid density, and inlet liquid pressure, and the outlet mass of the housing is calculated using the outlet volumetric flow rate, outlet liquid density, and outlet liquid pressure, wherein the inlet liquid density and inlet liquid pressure are related to the inlet liquid temperature.

[0085] In some embodiments, the reconstruction model employs reconstruction algorithms such as principal component analysis (PCA), autoencoder (AE), and variational autoencoder (VAE).

[0086] In some embodiments, the regression model is one of the following: linear regression model, multinomial regression, decision tree regression model, support vector machine regression model, K-nearest neighbor regression model, random forest regression model, Adaboost regression model, gradient boosting random forest regression model, bagging regression model, and Extra TREE regression model.

[0087] In some embodiments, the anomaly detection module is specifically used for:

[0088] Obtain the actual working data of the current box, and compare the actual working data with the predicted working data to obtain the residual between the actual working data and the predicted working data;

[0089] The alarm threshold for abnormal working data is obtained based on the training results. The residual is compared with the alarm threshold for abnormal working data. Based on the comparison result, it is determined whether the enclosure has an abnormality.

[0090] In some embodiments, the apparatus further includes:

[0091] The online update module is used to obtain the latest historical data and update the enclosure anomaly detection model when the cause of the enclosure anomaly alarm is determined to be a false alarm or a missed alarm.

[0092] Another embodiment of the present invention provides an electronic device, such as... Figure 3 As shown, the electronic device 10 includes:

[0093] One or more processors and memory 120, Figure 3 The following description uses a processor 110 as an example. The processor 110 and the memory 120 can be connected via a bus or other means. Figure 3 Taking the example of a connection between China and Israel via a bus.

[0094] Processor 110 is used to perform various control logics of electronic device 10. It can be a general-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), microcontroller, ARM (Acorn RISC Machine) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of these components. Furthermore, processor 110 can also be any conventional processor, microprocessor, or state machine. Processor 110 can also be implemented as a combination of computing devices, such as a combination of DSP and microprocessor, multiple microprocessors, one or more microprocessors combined with DSP and / or any other such configuration.

[0095] The memory 120, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions corresponding to the nuclear power plant enclosure anomaly detection method in this embodiment of the invention. The processor 110 executes various functional applications and data processing of the electronic device 10 by running the non-volatile software programs, instructions, and units stored in the memory 120, thereby implementing the nuclear power plant enclosure anomaly detection method in the above method embodiment.

[0096] The memory 120 may include a program storage area and a data storage area. The program storage area may store applications required for the operating platform and at least one function; the data storage area may store data created based on the use of the electronic device 10. Furthermore, the memory 120 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 120 may optionally include memory remotely located relative to the processor 110, and these remote memories may be connected to the electronic device 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0097] One or more units are stored in memory 120. When executed by one or more processors 110, they perform the nuclear power plant enclosure anomaly detection method in any of the above method embodiments, for example, performing the above-described... Figure 1 The method steps S100 to S600.

[0098] Another embodiment of the present invention provides a computer-readable storage medium storing computer-executable instructions that are executed by one or more processors, for example, to perform the above-described instructions. Figure 1 The method steps S100 to S600.

[0099] In summary, the present invention discloses a method, device, electronic equipment, and storage medium for detecting anomalies in a nuclear power plant enclosure. First, a mass conservation equation for the enclosure is established, and the influencing factors of the enclosure are determined based on this equation. Then, several segments of historical operating data for the enclosure are acquired, and the influencing shadows are ultimately determined and the required features are processed using data analysis methods, and a training set is established. A reconstruction model is established, and the regression model is trained using the training dataset to obtain a fully trained enclosure reconstruction model and warning thresholds for each influencing factor. A regression model is established, and the regression model is trained using the training dataset to obtain a fully trained enclosure regression model and warning thresholds for the regression output parameters. Next, the current influencing factor data of the enclosure is acquired, and this data is input into the reconstruction model and regression model to obtain predicted operating data for the enclosure. Finally, the current actual operating data of the enclosure is acquired, and the actual operating data is compared with the predicted operating data to determine whether an anomaly has occurred in the enclosure. This invention obtains the influencing factors of tank anomalies, and uses reconstruction and regression algorithms to construct an anomaly detection model for the tank. Then, the liquid level change of the tank can be predicted directly through the anomaly detection model. After comparing it with the actual liquid level change, it can quickly and accurately determine whether the tank has an anomaly.

[0100] The above description of the embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. It should be noted that those skilled in the art can make several improvements and modifications to the present invention without departing from the principles of the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

[0101] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for detecting anomalies in a nuclear power plant enclosure, characterized in that, Includes the following steps: Step S100: Establish the mass conservation equation for the box, and determine the influence factor of the ideal of the box based on the mass conservation equation; Step S200: Based on the actual deployment of the instruments and business knowledge, preliminarily determine the influencing factors of the enclosure; Step S300: Obtain several historical working data of the box, use data analysis methods to analyze the correlation between potential influencing factors and historical working data, so as to determine the final influencing factor of the box, and generate a corresponding training dataset based on the historical working data and the final determined influencing factor. Step S400: Establish a reconstruction model and train the reconstruction model using the training dataset to obtain a fully trained box reconstruction model and warning thresholds for each influencing factor. Step S500: Establish a regression model and train the regression model using the training dataset to obtain a fully trained box regression model and warning thresholds for the regression output parameters; In the process of establishing the reconstruction model and the regression model, the influence of temperature and pressure on density is considered. The inlet mass of the box is obtained by calculating the inlet volume flow rate, inlet liquid density, and inlet liquid pressure. The outlet mass of the box is obtained by calculating the outlet volume flow rate, outlet liquid density, and outlet liquid pressure. The inlet liquid density and inlet liquid pressure are related to the inlet liquid temperature. Step S600: Obtain the current impact factor data of the box, and input the current impact factor data into the reconstruction model and regression model to obtain two sets of prediction data for the box; Step S700: Obtain the actual operating data of the current enclosure, and compare the actual operating data with the predicted data to determine whether the enclosure is abnormal.

2. The method for detecting anomalies in a nuclear power plant enclosure according to claim 1, characterized in that, The historical operating data includes at least historical liquid level difference data and its corresponding influencing factor data, including the inlet flow rate and temperature of the tank, the outlet flow rate and temperature of the tank, the pressure at the top of the tank, the liquid temperature, and the gas phase pressure.

3. The method for detecting anomalies in a nuclear power plant enclosure according to claim 1, characterized in that, The reconstruction model employs principal component analysis (PCA) reconstruction, autoencoder (AE), and variational autoencoder (VAE) reconstruction algorithms.

4. The method for detecting anomalies in a nuclear power plant enclosure according to claim 1, characterized in that, The regression model is one of the following: linear regression, multinomial regression, decision tree regression, support vector machine regression, K-nearest neighbor regression, random forest regression, Adaboost regression, gradient boosting random forest regression, bagging regression, or extra tree regression.

5. The method for detecting anomalies in a nuclear power plant enclosure according to claim 1, characterized in that, The specific steps of S700 are as follows: Obtain the actual working data of the current housing, and compare the actual working data with the predicted data to obtain the residual between the actual working data and the predicted data; The abnormal alarm threshold for working data is obtained based on the training results. The residual is compared with the abnormal alarm threshold for working data. Based on the comparison result, it is determined whether the box has an abnormality. When an abnormality is detected in the enclosure, a corresponding alarm message is issued, which includes the specific details of the abnormality and the cause of the abnormality.

6. The method for detecting anomalies in a nuclear power plant enclosure according to claim 1, characterized in that, Also includes: When the cause of the enclosure abnormality alarm is determined to be a false alarm or a missed alarm, the latest historical data is obtained and the enclosure abnormality detection model is updated using the latest historical data.

7. A nuclear power plant enclosure anomaly detection device, characterized in that, include: The preliminary determination module for influencing factors is used to establish the mass conservation equation of the box and determine the influencing factors of the box based on the mass conservation equation. The potential impact factor determination module is used to preliminarily determine the impact factors of the enclosure by combining the actual deployment of the instrument with business knowledge; The training set processing module is used to acquire several historical working data of the enclosure, use data analysis methods to analyze the correlation between potential influencing factors and historical working data, in order to determine the final influencing factors of the heat exchanger, and generate the corresponding training dataset based on the historical working data and the final determined influencing factors. The reconstruction model training module is used to build a reconstruction model and train the reconstruction model using the training dataset to obtain a fully trained box reconstruction model and warning thresholds for each influencing factor. The regression model training module is used to build a regression model and train the regression model using the training dataset to obtain a fully trained box regression model and warning thresholds for the regression output parameters. In the process of establishing the reconstruction model and the regression model, the influence of temperature and pressure on density is considered. The inlet mass of the box is obtained by calculating the inlet volume flow rate, inlet liquid density, and inlet liquid pressure. The outlet mass of the box is obtained by calculating the outlet volume flow rate, outlet liquid density, and outlet liquid pressure. The inlet liquid density and inlet liquid pressure are related to the inlet liquid temperature. The prediction module is used to obtain the current impact factor data of the box, and input the current impact factor data into the reconstruction model and regression model for reconstruction to obtain two sets of prediction data for the box. The anomaly detection module is used to obtain the actual operating data of the current enclosure, compare the actual operating data with the predicted data, and determine whether the enclosure has an anomaly.

8. An electronic device, characterized in that, include: Processor and memory; The memory stores computer programs that can be executed by the processor; When the processor executes the computer program, it implements the steps in the nuclear power plant enclosure anomaly detection method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which can be executed by one or more processors to implement the steps in the nuclear power plant enclosure anomaly detection method according to any one of claims 1 to 6.

Images