Intelligent analysis system and evaluation method for nuclear power plant containment safety monitoring

Abstract

An intelligent analysis system and evaluation method for nuclear power plant containment safety monitoring, relating to the technical field of nuclear power plant monitoring. The intelligent evaluation method for nuclear power plant containment safety monitoring comprises: step S101, constructing a sensing front-end layer (S1) so as to acquire containment monitoring data by means of the sensing front-end layer (S1); step S102, screening, analyzing, and denoising the containment monitoring data by means of an information acquisition layer (S2); step S103, performing deep mining and intelligent analysis on the containment monitoring data in a data acquisition unit by means of an intelligent identification layer (S3) to obtain an analysis result; and step S104, performing comprehensive intelligent containment analysis and evaluation on the analysis result by means of a comprehensive evaluation layer (S4). The intelligent analysis system and evaluation method for nuclear power plant containment safety monitoring ensures safe operation of nuclear power plants, and optimizes monitoring efficiency and costs.

Description

Intelligent Analysis System and Evaluation Method for Nuclear Power Plant Containment Safety Monitoring

[0001] Cross-references to related applications

[0002] This application claims the benefit of Chinese patent application CN202411919284.4, filed on December 24, 2024, the contents of which are incorporated herein by reference. Technical Field

[0003] This application relates to the field of nuclear power plant monitoring technology, and in particular to an intelligent analysis system and evaluation method for monitoring the safety of the containment of a nuclear power plant. Background Technology

[0004] The containment integrity test at a nuclear power plant aims to verify the structural stability of the containment under external pressure or impact, and is one of the most important tests in a nuclear power plant. By simulating pressure conditions under extreme operating conditions, the test evaluates the pressure-bearing capacity and structural integrity of the containment, thereby ensuring that it can remain intact and continue to function in the event of a LOCA (Local Occurrence-Altered Accident).

[0005] Currently, the overall testing content of containment systems in nuclear power plants both domestically and internationally includes strain, temperature, displacement, prestress, and appearance inspection data. Each monitoring data point is collected or tested independently, resulting in scattered data that lacks unified integration and analysis, making it difficult to comprehensively assess the performance of the containment. Some monitoring data also rely on the experience and subjective judgment of inspectors, leading to insufficient data volume to support subsequent life extension assessments of nuclear power plant containment systems.

[0006] Existing technologies typically address the issues of manual data reading and poor data reliability by optimizing the data reading and analysis methods of existing nuclear power plant containment pressure test intensity monitoring systems. However, with the increase in third-generation and fourth-generation nuclear power plant reactor types and the need for nuclear power plant life extension assessments, monitoring requirements and indicators are becoming increasingly stringent, and the data volume requirements are growing, making it impossible to meet the monitoring requirements of unsuitable areas. Existing technologies are unable to meet these monitoring requirements. Summary of the Invention

[0007] The purpose of this application is to provide a method, apparatus, equipment, and storage medium for intelligent evaluation of the safety monitoring of the containment of a nuclear power plant.

[0008] To achieve the above objectives, this application provides the following solution:

[0009] Firstly, this application provides an intelligent evaluation method for the safety monitoring of the containment structure of a nuclear power plant, including:

[0010] Construct a sensing front-end layer to acquire containment monitoring data through the sensing front-end layer;

[0011] The containment monitoring data is filtered, analyzed, and noise-reduced through the information acquisition layer;

[0012] The intelligent recognition layer is used to perform in-depth mining and intelligent analysis on the containment monitoring data in the data acquisition instrument to obtain analysis results;

[0013] The comprehensive evaluation layer performs a full intelligent analysis and evaluation of the containment structure based on the analysis results.

[0014] Optionally, the step of constructing a sensing front-end layer to acquire containment monitoring data through the sensing front-end layer includes:

[0015] Front-end sensors are installed to collect data on the containment vessel's strain, temperature, displacement, prestress, and appearance.

[0016] Optionally, the step of filtering, analyzing, and reducing noise in the containment monitoring data through the information acquisition layer includes:

[0017] The data collected by the sensing front-end layer is transmitted to the data acquisition instrument using data and communication interfaces, and the data is initially screened, preprocessed and efficiently stored using distributed database technology.

[0018] Optionally, after the step of using the intelligent recognition layer to perform in-depth mining and intelligent analysis of the containment monitoring data in the data acquisition instrument to obtain the analysis results, the method further includes:

[0019] Construct a multi-level monitoring network to achieve real-time and accurate monitoring and trend prediction of the containment status, as well as intelligent identification and immediate response to abnormal data.

[0020] Optionally, the step of using the intelligent recognition layer to perform in-depth mining and intelligent analysis of the containment monitoring data in the data acquisition instrument to obtain analysis results includes:

[0021] Based on the containment monitoring data, the actual elastic modulus and Poisson's ratio of the containment concrete are calculated, and an accurate data prediction model is constructed.

[0022] The analysis results are obtained through in-depth mining and intelligent analysis using the data prediction model.

[0023] Optionally, the step of performing comprehensive intelligent analysis and evaluation of the containment structure based on the analysis results through a comprehensive evaluation layer includes:

[0024] The analysis results are used to perform intelligent damage diagnosis and refined analysis using a state assessment model based on deep learning algorithms, and the layout of monitoring points and data acquisition strategies are optimized based on the amount of monitoring data and assessment requirements.

[0025] Optionally, the step of using a state assessment model based on a deep learning algorithm to perform intelligent damage diagnosis and refined analysis of the analysis results includes:

[0026] A state assessment model based on a CNN neural network algorithm is used to identify the characteristics of the containment vessel under different damage states and locate the damage location.

[0027] Secondly, this application provides an intelligent analysis system for monitoring the safety of a nuclear power plant containment structure, comprising:

[0028] The perception front-end layer is used to acquire containment monitoring data;

[0029] The information acquisition layer is used to transmit the containment monitoring data to the data acquisition instrument.

[0030] The intelligent identification layer is used for in-depth mining and intelligent analysis of monitoring data;

[0031] The comprehensive assessment layer is used for comprehensive monitoring, assessment, and optimization of the containment structure.

[0032] Construct a sensing front-end layer to acquire containment monitoring data through the sensing front-end layer;

[0033] The containment monitoring data is filtered, analyzed, and noise-reduced through the information acquisition layer;

[0034] The intelligent recognition layer is used to perform in-depth mining and intelligent analysis on the containment monitoring data in the data acquisition instrument to obtain analysis results;

[0035] The comprehensive evaluation layer performs a full intelligent analysis and evaluation of the containment structure based on the analysis results.

[0036] Optionally, the sensing front-end layer is further configured to:

[0037] Front-end sensors are installed to collect data on the containment vessel's strain, temperature, displacement, prestress, and appearance.

[0038] Optionally, the information acquisition layer is further configured to:

[0039] The data collected by the sensing front-end layer is transmitted to the data acquisition instrument using data and communication interfaces, and the data is initially screened, preprocessed and efficiently stored using distributed database technology.

[0040] Optionally, the intelligent recognition layer is further used for:

[0041] Construct a multi-level monitoring network to achieve real-time and accurate monitoring and trend prediction of the containment status, as well as intelligent identification and immediate response to abnormal data.

[0042] Optionally, the intelligent recognition layer is further used for:

[0043] Based on the containment monitoring data, the actual elastic modulus and Poisson's ratio of the containment concrete are calculated, and an accurate data prediction model is constructed.

[0044] The analysis results are obtained through in-depth mining and intelligent analysis using the data prediction model.

[0045] Optionally, the comprehensive evaluation layer is further used for:

[0046] The analysis results are used to perform intelligent damage diagnosis and refined analysis using a state assessment model based on deep learning algorithms, and the layout of monitoring points and data acquisition strategies are optimized based on the amount of monitoring data and assessment requirements.

[0047] Optionally, the comprehensive evaluation layer is further used for:

[0048] A state assessment model based on a CNN neural network algorithm is used to identify the characteristics of the containment vessel under different damage states and locate the damage location.

[0049] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the intelligent evaluation method for nuclear power plant containment safety monitoring as described above.

[0050] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the intelligent evaluation method for nuclear power plant containment safety monitoring described in any of the above-mentioned methods.

[0051] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the intelligent evaluation method for nuclear power plant containment safety monitoring described above.

[0052] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0053] This application provides a method, device, equipment, and storage medium for intelligent evaluation of nuclear power plant containment safety monitoring. By constructing a sensing front-end layer, an information acquisition layer, an intelligent identification layer, and a comprehensive evaluation layer, the intelligent analysis system for nuclear power plant containment monitoring achieves a closed-loop process from data acquisition to in-depth analysis and then to comprehensive evaluation. First, the sensing front-end layer ensures comprehensive and high-precision containment status monitoring, providing an accurate data foundation for the system. Subsequently, the information acquisition layer improves data processing efficiency through efficient data aggregation and processing, enabling real-time monitoring. Next, the intelligent identification layer utilizes advanced data mining and intelligent analysis technologies to significantly enhance anomaly identification and early warning capabilities, enabling timely detection of potential risks. Finally, the comprehensive evaluation layer combines historical data and real-time status to conduct a comprehensive evaluation and optimization of the containment, not only improving the operational reliability of the containment but also providing a scientific basis for life extension assessment and aging management. The entire system, through layer-by-layer progressive data processing and analysis, not only ensures the safe operation of the nuclear power plant but also optimizes monitoring efficiency and cost, demonstrating the significant application value of intelligent monitoring technology in the nuclear power field. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 is a flowchart illustrating an intelligent evaluation method for monitoring the safety of a nuclear power plant containment, according to an embodiment of this application.

[0056] Figure 2 is a schematic diagram of the structure of an intelligent evaluation system for monitoring the safety of a nuclear power plant containment according to an embodiment of this application;

[0057] Figure 3 is a schematic diagram of the wiring of the displacement measurement system of the information acquisition layer provided in an embodiment of this application;

[0058] Figure 4 is a schematic diagram of a strain data acquisition system provided in an embodiment of this application;

[0059] Figure 5 is a schematic diagram of a containment structure damage identification approach based on a neural network algorithm provided in an embodiment of this application;

[0060] Figure 6 is a schematic diagram of neural network construction and training provided in an embodiment of this application;

[0061] Figure 7 is a comparison chart of theoretical values ​​and experimental results of a model provided in an embodiment of this application;

[0062] Figure 8 is a comparison between the finite element secondary development analysis results and the numerical solution of the prototype algorithm provided in an embodiment of this application;

[0063] Figure 9 is a schematic diagram of the functional modules of an intelligent analysis system for monitoring the safety of a nuclear power plant containment vessel according to an embodiment of this application;

[0064] Figure 10 is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0065] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0066] As shown in Figure 1, some embodiments of this application provide an intelligent evaluation method for the safety monitoring of the containment of a nuclear power plant. In these embodiments, the method includes steps 101 to 104. Wherein:

[0067] Referring to Figure 2, the intelligent evaluation system for nuclear power plant containment safety monitoring provided in this embodiment includes: a sensing front-end layer S1, an information acquisition layer S2, an intelligent identification layer S3, and a comprehensive evaluation layer S4. Each layer will be explained in detail below in conjunction with steps 101 to 104.

[0068] Step 101: Construct a sensing front-end layer to acquire containment monitoring data through the sensing front-end layer.

[0069] In this embodiment, the sensing front-end layer is the very front end of the monitoring system, primarily composed of various sensors and detection devices deployed at key locations both inside and outside the nuclear power plant containment. These sensors can monitor the containment's structural state (e.g., stress, strain, temperature, humidity), environmental parameters (e.g., radiation levels, gas concentrations), and potential anomalies (e.g., leaks, vibrations) in real time. Through high-precision sensors and advanced detection technologies, the sensing front-end layer can accurately and in real-time acquire monitoring data from the containment, providing a foundation for subsequent data processing and analysis.

[0070] Specifically, the sensing front-end layer mainly includes the input of test data and detection results collected by front-end sensors. This layer is a key link in information acquisition and preliminary processing. It is mainly used for data collection on the containment structure's strain, temperature, displacement, prestress, and appearance. Among them, the overall displacement test of the containment structure includes the horizontal radial displacement of the containment shell, the vertical displacement of the containment dome, and the deflection of the dome relative to the ring beam. The containment structure strain test includes the strain at the dome center, the ring beam at the junction of the ring beam and the upper dome, the junction of the ring beam and the shell, and the junction of the middle shell and the bottom plate; containment outer surface temperature measurement; steel tendon force measurement; appearance inspection including inspection of concrete crack observation areas, inspection of the surface condition of the containment concrete, inspection of the epoxy resin coating on the inner surface of the containment structure, inspection of the anchorage of prestressed steel bars; and temperature field measurement.

[0071] Step 102: The containment monitoring data is filtered, analyzed, and noise-reduced through the information acquisition layer.

[0072] In this embodiment, the information acquisition layer is responsible for collecting, organizing, and transmitting the data acquired by the sensing front-end layer. This layer typically includes devices such as data acquisition instruments and communication networks. The data acquisition instrument can receive signals from various sensors and convert them into digital data for storage and transmission. The communication network ensures that this data can be reliably and quickly transmitted to subsequent data processing and analysis stages. Through the information acquisition layer, the monitoring data of the containment structure is effectively aggregated, facilitating subsequent in-depth analysis and intelligent analysis. Referring to Figure 3, a schematic diagram of the displacement measurement system connection of the information acquisition layer is shown.

[0073] Specifically, the information acquisition layer seamlessly connects to and transmits the diverse monitoring data captured by the sensing front-end layer to the data acquisition device via a wired Ethernet communication interface and high-precision USB and HDMI digital interfaces. As the core component of the terminal system, the data acquisition device utilizes distributed database and data storage technologies to achieve preliminary screening, preprocessing, and efficient storage of massive amounts of data.

[0074] The displacement test data is obtained using an FT11 type DC-LVDT displacement sensor and an FT7311J type modulation / demodulation device. During use, the 4-pin connector of the FT11 is connected to the 4-pin socket of the FT7311J. The input excitation voltage and output signal (voltage) of the displacement sensor are achieved through a 5-pin socket. Field wiring and installation are completed using a converter consisting of a 5-pin connector and a board-type audio connector. Signal acquisition and excitation are accomplished using a 60800 type JB connector and a 66810 type CDF connector. The other end of the multi-core wire in the JB box is connected to the IMP and DC regulated power supply via a 5-core shielded extension cable according to its function. All DC-LVDTs in the test share a single 5A DC regulated power supply in parallel (see Figure 4).

[0075] Strain test data is monitored and connected to a junction box via strain gauges. Several junction boxes are then connected to a main junction box via conduits. In the main junction box, each strain gauge is simultaneously connected to two data acquisition boxes. Each data acquisition box is an automated data acquisition system connected to a host computer via a modem, enabling automatic data acquisition, storage, and display (see Figure 5).

[0076] Visual inspection of accessible areas can be performed using instruments such as bulge analyzers and 3D scanners to conduct a comprehensive visual inspection of the concrete surface of the containment vessel wall and dome; visual inspection of inaccessible areas can be performed using a wall-climbing robot.

[0077] Temperature field measurement can be performed using two methods: infrared thermal imagers and temperature sensors. The main purpose of infrared thermal imager temperature field testing is to detect whether there are any temperature anomalies in the entire containment structure; while temperature sensor test data is mainly used to correct for temperature issues in direct test data of containment structural displacement and strain measurement points.

[0078] Step 103: The intelligent identification layer is used to perform in-depth mining and intelligent analysis on the containment monitoring data in the data acquisition instrument to obtain the analysis results.

[0079] In this embodiment, the intelligent identification layer is the core component of the monitoring system. It utilizes advanced data processing technologies and algorithms to perform in-depth mining and intelligent analysis of the collected data. This layer can identify key information such as abnormal patterns and trend changes in the data, and interpret and predict this information through technologies such as machine learning and artificial intelligence. Through the intelligent identification layer, the monitoring system can more accurately understand the operational status of the containment structure, promptly detect potential safety hazards, and predict their development trends.

[0080] Specifically, the intelligent identification layer deeply mines and intelligently analyzes the diverse data from the information acquisition layer. It constructs a broad and hierarchical data monitoring network, building a multi-layered monitoring system including sensor networks, data processing centers, and remote monitoring platforms. This enables comprehensive and high-precision acquisition of environmental parameters inside and outside the containment, capturing and integrating data from different sources in real time to ensure data comprehensiveness and accuracy. This layer can dynamically monitor the data interaction process, promptly detecting and correcting errors and anomalies in data transmission. By integrating theoretical analysis data, historical data, and real-time monitoring data, this layer can construct accurate data prediction models, achieving precise real-time monitoring and trend prediction of the status of containments of the same reactor type and pressure. This layer can accurately identify potential risk points and anomaly patterns, setting red-line indicators based on the limits of different monitoring data, and immediately triggering an early warning mechanism when these limits are exceeded.

[0081] The core element of the intelligent identification layer is to calculate the Poisson's ratio and elastic modulus at different elevations based on the measured strain at various elevation measurement points of the containment shell using the diversified monitoring data of the information acquisition layer, and take the average value as the overall elastic modulus and Poisson's ratio of the containment structure.

[0082] First, the actual elastic modulus and Poisson's ratio of the containment concrete are calculated, and then the analysis and calculation are performed according to the plane stress state based on the following simplified model.

[0083] Where, σ H σ represents the circumferential stress (MPa); V The vertical stress is represented by MPa; P represents the test pressure (MPa); r represents the inner radius of the containment vessel (mm); e represents the wall thickness (mm); and l represents the containment vessel wall height (mm).

[0084] According to the stress-strain relationship under plane stress state, the following formula is used:

[0085] From formulas (3) and (4), we can derive:

[0086] Where, ε H ε represents the circumferential strain of the cylinder wall (με); V E represents the vertical strain of the cylinder wall (με); E represents the elastic modulus; and μ represents Poisson's ratio.

[0087] Substituting the strain values ​​of each measuring point in the information acquisition layer under the highest pressure into formula (7), the Poisson's ratio μ of each measuring point is calculated:

[0088] Among them, E H This indicates the elastic modulus (MPa) calculated from the circumferential strain data.

[0089] Among them, E V The elastic modulus (MPa) is calculated from the vertical strain data. The calculated elastic modulus at each elevation measuring point is obtained from formulas (8) and (9). Based on the measured strain at each elevation measuring point of the containment shell, the Poisson's ratio and elastic modulus at different elevations can be calculated. The average value is taken as the overall elastic modulus and Poisson's ratio of the containment structure.

[0090] Step 104: Perform a comprehensive intelligent analysis and evaluation of the containment structure based on the analysis results through the comprehensive evaluation layer.

[0091] In this embodiment, the comprehensive assessment layer is a layer that provides a comprehensive assessment and optimization suggestions for the overall state of the containment based on the analysis results provided by the intelligent identification layer. This layer comprehensively considers the historical data, current state, and future predictions of the containment to conduct a comprehensive assessment of its integrity, stability, and safety. Based on the assessment results, the comprehensive assessment layer proposes corresponding optimization suggestions, such as maintenance plans and improvement measures, to ensure the long-term safe operation of the containment. Through the comprehensive assessment layer, the monitoring system not only provides real-time monitoring of the containment's state but also provides strong support for the safety management and optimization decision-making of the nuclear power plant.

[0092] Specifically, the core assessment model of the comprehensive assessment layer is based on the existing monitoring data of the structure, to establish an optimized model to identify the material parameters of the concrete model, and to establish a benchmark model for the analysis of the containment structure. Among these, establishing a long-term prestress loss model is the key to the integrity evaluation of the containment structure.

[0093] Based on adaptive Latin square sampling technology, a Kriging model is used to establish a time-varying equivalent model for calculating monitoring points. Using existing monitoring data considering temperature compensation, the following optimization problem is constructed to identify the parameters of the concrete shrinkage and creep model. The trust region method is used to solve the optimization problem, forming a baseline model for the finite element analysis of the containment structure. An inversion method is used to establish an evaluation method for the loss level of other bonded prestresses in the containment structure, and a method for quantitative analysis of integrity uncertainty is proposed. Based on this, a real-time accurate evaluation system for containment integrity is developed.

[0094] st x l ≤x≤x u (10)

[0095] Where: x is the material parameter to be determined, x l and x u , respectively, represent the upper and lower limits of x; m is the total number of discrete time points; w i T represents the weight value at time i; i f(x) represents the calculated prestress value at the monitoring point at time i. It is determined through calculation using the time-varying equivalence model obtained from the time-varying equivalence model of the monitoring point. f(x) is the optimization objective function; T i T (x) represents the measured target of the prestress at the monitoring point at time i.

[0096] The comprehensive assessment layer constructs a multi-dimensional, multi-level assessment index system, combines it with the analysis results of the intelligent identification layer, and utilizes a state assessment model driven by deep learning algorithms to conduct intelligent damage diagnosis and refined analysis. This process not only relies on real-time data provided by the containment perception front-end layer but also trains the model using a CNN neural network algorithm, enabling it to accurately identify the characteristics of the containment under different damage states, including internal and external damage such as cracks, corrosion, pitting, and fissures. When damage occurs to the containment, the model can accurately identify potential damage or anomalies in the system, issue timely warnings, accurately locate the damage position, and conduct refined analysis, thereby comprehensively reflecting the overall performance and safety status of the system.

[0097] Referring to Figure 6, a schematic diagram of neural network construction and training provided in an embodiment of this application is shown.

[0098] For damage identification of containment structures, this application adopts a complete process, including data collection and preprocessing, feature extraction and selection, CNN neural network model construction and preliminary training, model verification and optimization (K-fold cross-validation), damage identification and system monitoring.

[0099] In the data preprocessing stage, outliers and noise are removed by data cleaning, data normalization ensures that the data are on the same scale, as shown in formula (11), and data augmentation improves the generalization ability of the model, laying a solid foundation for subsequent model training.

[0100] Where X is the original data, X min and X max These are the minimum and maximum values ​​of the data, X. norm It is the normalized data.

[0101] The basic structure of a CNN model includes an input layer, convolutional layers, pooling layers, fully connected layers, and an output layer. The input layer receives preprocessed image data; the convolutional layers use kernels to extract local features from the image, which are crucial for identifying damage such as cracks, corrosion, and pitting on the surface of the containment structure; the pooling layers reduce the dimensionality of the feature maps through downsampling; the fully connected layers synthesize the extracted features and introduce nonlinear transformations through activation functions to enhance the model's expressive power; finally, the output layer produces the identification result, namely the type and location of damage to the containment structure.

[0102] The training process of a CNN model includes forward propagation and back propagation. Forward propagation involves processing the input data layer by layer through the model to obtain the output result; back propagation involves updating the model's weights and biases based on the error between the output result and the true label using the gradient descent algorithm. The loss function of a CNN model is usually the cross-entropy loss function, and its formula (12) is as follows:

[0103] Where N is the number of samples, y i It is the true label of the i-th sample. is the model's prediction result for the i-th sample, and L is the cross-entropy loss value.

[0104] The basic architecture of a Convolutional Neural Network (CNN) model consists of an input layer, convolutional layers, pooling layers, fully connected layers, and an output layer. The input layer receives raw data or preprocessed feature parameters, providing the foundation for subsequent processing. The convolutional layers perform convolution operations through a series of kernels, aiming to extract local features from the input data. The pooling layers reduce the spatial dimensionality of the feature maps through downsampling operations, typically using methods such as max pooling or average pooling to reduce computational complexity while preserving key features. These key features are particularly important in identifying external damage within containment structures, such as cracks, rust, and pitting on the containment surface.

[0105] The convolutional layer extracts local features from the input data by performing convolution operations through a series of convolutional kernels. During the convolution process, the convolutional kernel slides across the input data and generates a feature matrix by calculating the dot product with the local regions of the data. This feature matrix can capture various subtle features of the surface damage of the containment vessel, and its formula (13) is as follows:

[0106] Where F(x,y) is the value of the output feature map at (x,y); I(x+m,y+n) is the pixel value of the output image at position (x+m,y+n); K(m,n) is the weight of the convolution kernel at position (m,n); m and n are the offsets of the convolution kernel K relative to its center. For a convolution kernel of size (2a+1×2b+1), the ranges of m and n are [-a,a] and [-b,b], respectively, where a and b are positive integers that determine the height and width of the convolution kernel.

[0107] After the convolutional layer, a pooling layer is usually added to reduce the spatial dimension of the feature map. If the pooling kernel size is (m×n), the pooling operation can be expressed as follows (14):

[0108] The average of all values ​​in the feature region is calculated as the pooled result for that region, and the formula becomes:

[0109] In formulas (14) and (15), F(x,y) is the value of the output feature map at position (x,y); it means that the largest element value is selected as the output within the region covered by the pooling kernel. is the normalization coefficient of average pooling, where m and n are the height and width of the pooling window, respectively; This means that the largest element value is selected as the output within the region covered by the pooling kernel. This indicates that all elements within the pooling window are traversed and accumulated. I represents the input feature map. (i, j) represents the relative position index within the pooling window, used to traverse each element within the window, where the values ​​of i and j range from [0, m) and [0, n), respectively.

[0110] The fully connected layer, located in the latter half of the network, is responsible for synthesizing the features extracted by the preceding layers and introducing nonlinear transformations through activation functions to enhance the model's expressive power. These features are then mapped to a higher-dimensional space to support classification tasks, such as distinguishing the types of damage to the containment structure (cracks, corrosion, pitting, etc.). Finally, the output layer generates the identification result, i.e., the damage status of the containment structure's interior and exterior, based on the task requirements.

[0111] The training process of a CNN involves two phases: forward propagation and back propagation. In the forward propagation phase, the input data is processed sequentially through each layer until an output is produced. In the back propagation phase, based on the error between the output and the true label, optimization algorithms such as gradient descent are used to adjust the model's weights and bias parameters to minimize the loss function. In CNNs, the cross-entropy loss function is often used as a standard to measure the difference between the predicted and true values.

[0112] For a specific dataset (such as the S1 dataset containing images of external damage inside a containment vessel), it can be transformed into a form suitable for CNN processing through a series of preprocessing steps. These include cropping, scaling, and normalizing the images, and labeling damaged and healthy data into different categories. These processed image samples can then be fed into a CNN model for training and testing.

[0113] Model optimization, based on model validation, involves adjusting the model's parameters and structure to improve recognition accuracy and computational efficiency. K-fold cross-validation divides the dataset into K subsets. Each time, K-1 subsets are selected as the training set, and the remaining subset is used as the validation set. This process is repeated K times to obtain K validation results, and the average value is taken as the final performance evaluation of the model. In containment structure damage identification, damage identification results can be obtained through threshold judgment, cluster analysis, and other methods. System monitoring requires the establishment of a real-time data acquisition, transmission, and processing system to ensure data accuracy and timeliness.

[0114] In the application of containment structure damage identification, CNN models can analyze damage results through methods such as threshold judgment and cluster analysis. Simultaneously, to ensure data accuracy and real-time performance, a real-time data acquisition, transmission, and processing system needs to be established. By continuously optimizing the vibration signal acquisition method and sample partitioning strategy, the recognition efficiency and accuracy of the CNN model can be further improved, providing more reliable technical support for containment maintenance and safety monitoring. Throughout the CNN learning process, the model iteratively calculates to continuously narrow the difference between the validation value and the true value, demonstrating strong applicability to damage identification datasets. Referring to Figure 7, which shows a comparison between the model's theoretical values ​​and experimental results, and to Figure 8, which shows a comparison between the finite element secondary development analysis results and the prototype algorithm's numerical solution, it is evident that the intelligent evaluation method for nuclear power plant containment safety monitoring provided in this application embodiment can effectively and accurately detect nuclear power plant containment structures.

[0115] As nuclear power plant operating time increases, inversion analysis of monitoring data can reveal potential defects or deficiencies within the system. Supplementary monitoring points are added to areas where structural aging damage approaches limits, guiding the optimized layout of these points. The comprehensive assessment layer utilizes monitoring data and the predictive capabilities of neural network models to invert the internal state of the containment. Data inversion provides a more accurate understanding of the actual operating state and damage status of the containment. Based on the inversion results, the arrangement and number of monitoring points are optimized. Reducing unnecessary monitoring points lowers monitoring costs, while increasing monitoring points in critical areas improves accuracy and reliability. Using monitoring data and the predictive capabilities of neural network models, the structural state of the containment is reconstructed. This structural state reconstruction provides a more intuitive understanding of the overall structure and damage status of the containment, offering strong support for subsequent aging maintenance.

[0116] This application's embodiment constructs a sensing front-end layer, an information acquisition layer, an intelligent identification layer, and a comprehensive evaluation layer. This intelligent analysis system for nuclear power plant containment monitoring achieves a closed-loop process from data acquisition to in-depth analysis and comprehensive evaluation. First, the sensing front-end layer ensures comprehensive and high-precision containment status monitoring, providing an accurate data foundation for the system. Subsequently, the information acquisition layer improves data processing efficiency through efficient data aggregation and processing, enabling real-time monitoring. Next, the intelligent identification layer utilizes advanced data mining and intelligent analysis technologies to significantly enhance anomaly identification and early warning capabilities, enabling timely detection of potential risks. Finally, the comprehensive evaluation layer combines historical data with real-time status to conduct a comprehensive evaluation and optimization of the containment, not only improving the operational reliability of the containment but also providing a scientific basis for life extension assessments and aging management. The entire system, through layer-by-layer progressive data processing and analysis, not only ensures the safe operation of the nuclear power plant but also optimizes monitoring efficiency and cost, demonstrating the significant application value of intelligent monitoring technology in the nuclear power field.

[0117] Optionally, step 101 includes: setting up front-end sensors to collect data on the strain, temperature, displacement, prestress, and appearance of the containment.

[0118] Optionally, step 102 includes: transmitting the data collected by the sensing front-end layer to the data acquisition instrument using a data interface and a communication interface, and performing preliminary screening, preprocessing and efficient storage of the data using distributed database technology.

[0119] Optionally, after step 103, the method further includes: constructing a multi-level monitoring network to achieve real-time accurate monitoring and trend prediction of the containment status, as well as intelligent identification and immediate response to abnormal data.

[0120] Optionally, step 103 includes:

[0121] Step 1031: Based on the containment monitoring data, calculate the actual elastic modulus and Poisson's ratio of the containment concrete to construct an accurate data prediction model.

[0122] Step 1032: The data prediction model is used to perform in-depth mining and intelligent analysis to obtain the analysis results.

[0123] Optionally, step 104 includes: using a state assessment model based on a deep learning algorithm to perform intelligent damage diagnosis and refined analysis on the analysis results, and optimizing the layout of monitoring points and data acquisition strategies based on the amount of monitoring data and assessment requirements.

[0124] Optionally, step 104 includes: employing a state assessment model based on a CNN neural network algorithm to identify the characteristics of the containment under different damage states and locate the damage location.

[0125] Based on the same inventive concept, this application also provides an intelligent analysis system for nuclear power plant containment safety monitoring, used to implement the intelligent evaluation method for nuclear power plant containment safety monitoring described above. The solution provided by this device is similar to the implementation scheme described in the above method; therefore, the specific limitations of one or more embodiments of the intelligent analysis system for nuclear power plant containment safety monitoring provided below can be found in the limitations of the intelligent evaluation method for nuclear power plant containment safety monitoring described above, and will not be repeated here.

[0126] In one exemplary embodiment, as shown in FIG9, a nuclear power plant containment safety monitoring intelligent analysis system 20 is provided, comprising:

[0127] The sensing front-end layer 201 is used to acquire containment monitoring data;

[0128] Information acquisition layer 202 is used to aggregate the containment monitoring data to the data acquisition instrument;

[0129] The intelligent recognition layer 203 is used for in-depth mining and intelligent analysis of the aggregated data;

[0130] The comprehensive assessment layer 204 is used for comprehensive monitoring, assessment, and optimization of the containment.

[0131] A sensing front-end layer 201 is constructed to acquire containment monitoring data through the sensing front-end layer 201;

[0132] The containment monitoring data is aggregated to the data acquisition instrument via the information acquisition layer 202;

[0133] The intelligent identification layer 203 is used to perform in-depth mining and intelligent analysis on the containment monitoring data in the data acquisition instrument to obtain analysis results.

[0134] The analysis results are comprehensively monitored, evaluated, and optimized using the comprehensive evaluation layer 204.

[0135] Optionally, the sensing front-end layer 201 is further configured to:

[0136] Front-end sensors are installed to collect data on the containment vessel's strain, temperature, displacement, prestress, and appearance.

[0137] Optionally, the information acquisition layer 202 is further configured to:

[0138] The data collected by the sensing front-end layer 201 is transmitted to the data acquisition instrument using data and communication interfaces, and the data is initially screened, preprocessed and efficiently stored using distributed database technology.

[0139] Optionally, the intelligent recognition layer 203 is further used for:

[0140] Construct a multi-level monitoring network to achieve real-time and accurate monitoring and trend prediction of the containment status, as well as intelligent identification and immediate response to abnormal data.

[0141] Optionally, the intelligent recognition layer 203 is further used for:

[0142] Based on the containment monitoring data, the actual elastic modulus and Poisson's ratio of the containment concrete are calculated, and an accurate data prediction model is constructed.

[0143] The analysis results are obtained through in-depth mining and intelligent analysis using the data prediction model.

[0144] Optionally, the comprehensive evaluation layer 204 is further configured to:

[0145] The analysis results are used to perform intelligent damage diagnosis and refined analysis using a state assessment model based on deep learning algorithms, and the layout of monitoring points and data acquisition strategies are optimized based on the amount of monitoring data and assessment requirements.

[0146] Optionally, the comprehensive evaluation layer 204 is further configured to:

[0147] A state assessment model based on a CNN neural network algorithm is used to identify the characteristics of the containment vessel under different damage states and locate the damage location.

[0148] This application's embodiment constructs a sensing front-end layer, an information acquisition layer, an intelligent identification layer, and a comprehensive evaluation layer. This intelligent analysis system for nuclear power plant containment monitoring achieves a closed-loop process from data acquisition to in-depth analysis and comprehensive evaluation. First, the sensing front-end layer ensures comprehensive and high-precision containment status monitoring, providing an accurate data foundation for the system. Subsequently, the information acquisition layer improves data processing efficiency through efficient data aggregation and processing, enabling real-time monitoring. Next, the intelligent identification layer utilizes advanced data mining and intelligent analysis technologies to significantly enhance anomaly identification and early warning capabilities, enabling timely detection of potential risks. Finally, the comprehensive evaluation layer combines historical data with real-time status to conduct a comprehensive evaluation and optimization of the containment, not only improving the operational reliability of the containment but also providing a scientific basis for life extension assessments and aging management. The entire system, through layer-by-layer progressive data processing and analysis, not only ensures the safe operation of the nuclear power plant but also optimizes monitoring efficiency and cost, demonstrating the significant application value of intelligent monitoring technology in the nuclear power field.

[0149] In an exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram is shown in Figure 10. The computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores intelligent evaluation data for the safety monitoring of nuclear power plant containment structures. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements an intelligent evaluation method for the safety monitoring of nuclear power plant containment structures.

[0150] Those skilled in the art will understand that the structure shown in Figure 10 is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or may combine certain components, or may have different component arrangements.

[0151] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0152] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0153] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0154] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0155] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0156] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0157] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0158] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for intelligent evaluation of the safety monitoring of a nuclear power plant containment, characterized in that, The intelligent evaluation method for monitoring the safety of the nuclear power plant containment includes: Construct a sensing front-end layer to acquire containment monitoring data through the sensing front-end layer; The containment monitoring data is filtered, analyzed, and noise-reduced through the information acquisition layer; The intelligent recognition layer is used to perform in-depth mining and intelligent analysis on the containment monitoring data in the data acquisition instrument to obtain analysis results; The comprehensive evaluation layer performs a full intelligent analysis and evaluation of the containment structure based on the analysis results.

2. The intelligent evaluation method for nuclear power plant containment safety monitoring according to claim 1, characterized in that, The step of constructing a sensing front-end layer to obtain containment monitoring data includes: Front-end sensors are installed to collect data on the containment vessel's strain, temperature, displacement, prestress, and appearance.

3. The intelligent evaluation method for nuclear power plant containment safety monitoring according to claim 1, characterized in that, The step of filtering, analyzing, and reducing noise in the containment monitoring data through the information acquisition layer includes: The data collected by the sensing front-end layer is transmitted to the data acquisition instrument using data and communication interfaces, and the data is initially screened, preprocessed and efficiently stored using distributed database technology.

4. The intelligent evaluation method for nuclear power plant containment safety monitoring according to claim 1, characterized in that, After the step of using the intelligent recognition layer to perform in-depth mining and intelligent analysis of the containment monitoring data in the data acquisition instrument to obtain the analysis results, the method further includes: Construct a multi-level monitoring network to achieve real-time and accurate monitoring and trend prediction of the containment status, as well as intelligent identification and immediate response to abnormal data.

5. The intelligent evaluation method for nuclear power plant containment safety monitoring according to claim 4, characterized in that, The steps of using the intelligent recognition layer to perform in-depth mining and intelligent analysis of the containment monitoring data in the data acquisition instrument to obtain analysis results include: Based on the containment monitoring data, the actual elastic modulus and Poisson's ratio of the containment concrete are calculated, and an accurate data prediction model is constructed. The analysis results are obtained through in-depth mining and intelligent analysis using the data prediction model.

6. The intelligent evaluation method for nuclear power plant containment safety monitoring according to claim 1, characterized in that, The steps of performing comprehensive intelligent analysis and evaluation of the containment structure based on the analysis results through a comprehensive evaluation layer include: The analysis results are used to perform intelligent damage diagnosis and refined analysis using a state assessment model based on deep learning algorithms, and the layout of monitoring points and data acquisition strategies are optimized based on the amount of monitoring data and assessment requirements.

7. The intelligent evaluation method for nuclear power plant containment safety monitoring according to claim 6, characterized in that, The steps of using a state assessment model based on deep learning algorithms to perform intelligent damage diagnosis and refined analysis on the analysis results include: A state assessment model based on a CNN neural network algorithm is used to identify the characteristics of the containment vessel under different damage states and locate the damage location.

8. A smart analysis system for monitoring the safety of a nuclear power plant containment vessel, characterized in that, The intelligent analysis system for monitoring the safety of the nuclear power plant containment includes: The perception front-end layer is used to acquire containment monitoring data; The information acquisition layer is used to aggregate the containment monitoring data to the data acquisition instrument. The intelligent recognition layer is used for in-depth mining and intelligent analysis of the aggregated data; The comprehensive assessment layer is used for comprehensive monitoring, assessment, and optimization of the containment structure.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the intelligent evaluation method for nuclear power plant containment safety monitoring according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the intelligent evaluation method for nuclear power plant containment safety monitoring as described in any one of claims 1-7.

Images