Intelligent risk identification method and system based on nuclear power plant work order maintenance process

Abstract

The invention discloses an intelligent risk identification method and system based on a nuclear power plant work order maintenance process, and relates to the technical field of nuclear power plant work management, and the method comprises the steps: collecting maintenance related data in real time through a first collection module; preprocessing the maintenance related data, and performing first feature extraction; establishing a risk identification model by using a first intelligent algorithm, performing risk assessment on the real-time data by using the trained risk identification model, and generating a first risk assessment result; and carrying out maintenance feedback, and regularly carrying out updating maintenance on the risk identification model according to the first risk assessment result. According to the method, data acquired by the risk intelligent identification method is more comprehensive, and continuous and effective maintenance data can be provided for an intelligent algorithm; according to the method, a result feedback and evaluation mechanism is established, and an intelligent algorithm is further optimized through a result. The intelligent algorithm is further optimized, and the calculation result is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of nuclear power plant work management, and in particular to a method and system for intelligently identifying risks in a maintenance process based on a work order of a nuclear power plant. Background Art

[0002] The risk assessment of nuclear power plants in the process of work order maintenance faces many challenges, including the difficulty of data collection and integration, the establishment and optimization of risk assessment models, the complexity of risk identification and analysis, the application of risk assessment results, compliance with regulations and standards, the establishment of a maintenance effectiveness evaluation system, and personnel training and management issues. Solving these problems requires nuclear power plants to invest a lot of resources and technology, and also requires continuous innovation and adaptation to the changing external environment. Only through continuous efforts and improvements can we ensure that the risk assessment of work order maintenance in nuclear power plants can not only ensure safety, but also improve efficiency, and ultimately achieve long-term stable operation of nuclear power plants.

[0003] Specifically, the work order maintenance process of nuclear power plants requires not only efficient identification and assessment of risks, but also the maximum realization of risk warning and dynamic feedback in the process, so as to timely adjust the maintenance strategy to ensure equipment safety and stable operation of the plant. The above-mentioned deficiencies in the existing technology are particularly evident in the actual application of nuclear power plants. For example, the inconsistency of data collection often leads to delays in risk identification, which makes it impossible to respond to risks in a timely manner. At the same time, traditional risk assessment models are difficult to provide comprehensive and accurate assessments when facing complex and sophisticated equipment systems such as nuclear power plants. Some risks may be ignored during maintenance, which leads to the failure to discover or repair equipment hazards in time. In addition, it is difficult for risk assessment results to be effectively linked with management and maintenance teams, affecting the scientific nature of maintenance decisions. The high safety requirements of the nuclear power plant industry make risk management not only limited to the identification of potential problems, but also require a continuous and forward-looking risk warning mechanism. Existing technical means cannot fully meet this demand. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: the existing technology for risk assessment methods of nuclear power plants has the problem of how to realize intelligent, automated and real-time risk identification and assessment during the work order maintenance process, which has become a technical problem that nuclear power plants need to solve urgently.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for intelligently identifying risks in the maintenance process of a nuclear power plant work order, comprising:

[0007] Collect maintenance-related data in real time through a first collection module;

[0008] Preprocessing maintenance-related data and performing first feature extraction;

[0009] Using the first intelligent algorithm to establish a risk identification model, using the trained risk identification model to perform risk assessment on real-time data, and generating a first risk assessment result;

[0010] Provide feedback during maintenance and regularly update and maintain the risk identification model based on the results of the first risk assessment.

[0011] As a preferred solution of the method for intelligent risk identification in the maintenance process of a nuclear power plant work order according to the present invention, the first collection module collects maintenance-related data in real time, including calling the first collection module from the monitoring system of the nuclear power plant to collect maintenance-related data in real time.

[0012] As a preferred solution of the method for intelligent risk identification in the maintenance process of nuclear power plant work orders described in the present invention, the preprocessing of maintenance-related data includes cleaning, normalizing and encoding the collected data to eliminate noise and inconsistent formats.

[0013] As a preferred solution of the method for intelligent risk identification in the maintenance process of a nuclear power plant work order according to the present invention, the first feature extraction includes extracting the first feature related to the maintenance risk and selecting the first feature subset that is most conducive to risk identification.

[0014] As a preferred scheme of the method for intelligent risk identification in the nuclear power plant work order maintenance process described in the present invention, the use of the first intelligent algorithm to establish a risk identification model includes using the extracted features and the selected feature subset, using the first intelligent algorithm to establish a risk identification model, and training and verifying the model.

[0015] As a preferred solution of the method for intelligent risk identification in the nuclear power plant work order maintenance process described in the present invention, the generating of the first risk assessment result includes performing a risk assessment on the data collected in real time, and generating a corresponding risk level and early warning signal according to the assessment result.

[0016] As a preferred solution of the method for intelligent risk identification in the nuclear power plant work order maintenance process described in the present invention, the regular updating and maintenance of the first model based on the first risk assessment result includes updating and maintaining the risk identification model based on feedback after maintenance and regular performance evaluation results to ensure that the recognition accuracy of the first intelligent algorithm is continuously optimized over time.

[0017] Another object of the present invention is to provide an intelligent risk identification system based on the maintenance process of nuclear power plant work orders, which can use an intelligent algorithm to establish and train a risk identification model through one of the schemes, and perform risk assessment on real-time data, thereby solving the problem of current nuclear energy assessment technology in both ensuring safety and improving efficiency.

[0018] As a preferred solution of the risk intelligent identification system based on the maintenance process of nuclear power plant work orders described in the present invention, it includes: a data collection module, a data preprocessing module, a model building module, and a risk assessment module;

[0019] The data collection module is responsible for collecting data related to the maintenance of nuclear power plant work orders in real time, including a real-time data collection function called from the nuclear power plant monitoring system;

[0020] The data preprocessing module processes the collected maintenance data, cleans, normalizes, and encodes the data, removes noise and unifies the data format, extracts features related to maintenance risks, and selects the most suitable feature subset for risk identification;

[0021] The model building module uses intelligent algorithms to build and train risk identification models, conduct risk assessment on real-time data, and generate risk assessment results;

[0022] The risk assessment module publishes risk levels and early warning signals, and updates and maintains the risk identification model based on feedback and regular performance evaluation results.

[0023] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a method for intelligently identifying risks in a maintenance process based on a work order of a nuclear power plant.

[0024] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for intelligently identifying risks in a maintenance process based on a work order of a nuclear power plant.

[0025] Beneficial effects of the present invention: The data collected by the intelligent risk identification method in the maintenance process of nuclear power plant work orders provided by the present invention is more comprehensive, and can provide continuous and effective maintenance data for the intelligent algorithm; the method establishes a result feedback and evaluation mechanism, and further optimizes the intelligent algorithm through the results. The intelligent algorithm is further optimized, and the calculation results are more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0027] Figure 1 An overall flow chart of a method for intelligently identifying risks in a nuclear power plant work order maintenance process provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0029] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides a method for intelligently identifying risks in a maintenance process based on a work order of a nuclear power plant, comprising:

[0030] S1: Collect maintenance-related data in real time through the first collection module.

[0031] Furthermore, the first collecting module collecting the maintenance related data in real time includes calling the first collecting module from the monitoring system of the nuclear power plant to collect the maintenance related data in real time.

[0032] In an embodiment of the present application, the first collection module may be a module for collecting information on monitoring equipment of a monitoring system of a nuclear power plant, a sensor network of the equipment, a historical maintenance record database, and work order information input by maintenance personnel.

[0033] Furthermore, data is collected from the monitoring equipment of the monitoring system, and the steps include equipment parameter collection, data preprocessing, and data transmission; data collection is performed using a sensor network, which involves sensor layout, data collection, and preliminary data processing;.

[0034] It should be noted that the analysis of historical maintenance records to guide current maintenance work includes record screening, fault analysis, and maintenance strategy formulation; the integration of work order maintenance information and operation guidelines includes the main steps of work order information extraction, operation guide writing, and maintenance process planning.

[0035] S2: Preprocess the maintenance-related data and perform the first feature extraction.

[0036] In the embodiment of the present application, preprocessing the maintenance-related data may be to clean, normalize and encode the collected data to eliminate noise and inconsistency in format, specifically including A1 to A5:

[0037] A1. Cleaning data with format and content errors: unify the formats of data from different sources, such as time, date, and value, ensure that the content meets the field requirements, and remove unnecessary characters.

[0038] A2, logical error cleaning: deduplicate data, replace or delete unreasonable values, and ensure logical consistency of data.

[0039] A3, Remove unnecessary data: Selectively delete data fields that are irrelevant or unimportant to the analysis to reduce the complexity of model building and improve efficiency.

[0040] A4, correlation verification: When merging multiple data sources, verify the correlation between the data to ensure that the feature attributes are correct.

[0041] A5, Outlier processing: Process outliers in the data by deleting or filling them, and use mean, median or mode statistical methods to fill them.

[0042] The normalization is to perform a linear transformation on the original data, map the data to the interval [0,1], and use the Min-Max normalization method, which is expressed as follows:

[0043]

[0044] Converting the data to a normal distribution with a mean of 0 and a standard deviation of 1 is suitable for situations where there are outliers in the data. This is called Z-score standardization, and the formula is:

[0045]

[0046] Among them, mean is the mean and σ is the standard deviation. This method has little impact on outliers and helps improve the robustness of the algorithm.

[0047] In an optional embodiment, the data can also be preprocessed by: normalization or standardization, which normalizes or standardizes the data to make the scales of different features relatively balanced and avoid the influence of dimensional differences between features on clustering results. Feature selection or dimensionality reduction, if the data dimension is too high, dimensionality reduction methods (such as principal component analysis PCA) can be used to simplify the data structure and reduce the computational complexity.

[0048] The first feature extraction includes extracting first features related to maintenance risks and selecting a first feature subset that is most conducive to risk identification.

[0049] In an embodiment of the present application, the first feature extraction method may be to extract key features related to maintenance risks by using data exploratory analysis, key feature extraction, and feature selection.

[0050] Step 1, the data exploratory analysis includes: Statistical analysis: basic statistical analysis of the data set, such as calculating the mean, median, standard deviation, extreme value, etc., to understand the distribution and central tendency of the data. Correlation analysis: linear relationships between features are analyzed through correlation coefficient matrix to identify variables that may be related to maintenance risks. Outlier detection: outliers in the data are detected using box plots and scatter plots, which have a negative impact on the accuracy of the model.

[0051] Step 2, the key feature extraction is specifically as follows: filtering features: removing features irrelevant to maintenance risks based on business knowledge and statistical results, reducing dimensions and computational complexity; extracting statistical features: extracting important statistical features from time series data, which can reflect the stability and reliability of equipment operation; extracting text and image features: if the data contains text or image information, using a bag-of-words model, TF-IDF method or deep learning algorithm to extract feature vectors;

[0052] Step 3, the feature selection is specifically as follows: Model-based feature selection: Use linear models, decision trees or Lasso regression methods to evaluate the importance of each feature and select the most useful feature subset for maintenance risk prediction. Recursive feature elimination: A method of selecting the best features by repeatedly building models. Expert knowledge-based selection: Combine the knowledge of domain experts to manually select features to ensure that the selected features have clear professional meaning and explanation.

[0053] In an optional embodiment, the first feature extraction method may also be feature extraction based on clustering: B1, executing a clustering algorithm to divide the data into a number of clusters. The data points in each cluster have certain similarities, and there are significant differences between clusters.

[0054] B2, extract clustering features. After clustering is completed, new features can be extracted based on the clustering results.

[0055] B3, cluster labels as new features: Assign a cluster label to each data point and input it into the subsequent model as a new classification feature. This feature helps to distinguish different distribution patterns of data.

[0056] B4, statistical information of intra-cluster features, calculates the statistical features (such as mean, standard deviation, maximum value, minimum value, etc.) within each cluster and uses them as new features. This method can provide information on the central tendency and variability of data in the clustering dimension.

[0057] B5, calculates the distance between the data point and the center of its cluster and uses the distance as a new feature. This helps reflect the degree of outlier and stability of the data point.

[0058] It should be noted that this method can provide additional feature information to the model, making the feature representation of the data richer and more multidimensional, thereby improving the model's learning ability and prediction effect.

[0059] Example 2, reference Figure 1 , which is an embodiment of the present invention, provides a method for intelligently identifying risks in a maintenance process based on a work order of a nuclear power plant, comprising:

[0060] S3: Use the first intelligent algorithm to establish a risk identification model, use the trained risk identification model to perform risk assessment on real-time data, and generate a first risk assessment result.

[0061] Furthermore, the use of the first intelligent algorithm to establish a risk identification model includes using the extracted features and the selected feature subsets, adopting the first intelligent algorithm to establish a risk identification model, and training and verifying the model.

[0062] In an embodiment of the present application, the use of the first intelligent algorithm to establish a risk identification model may be to use model selection, data preparation, and model training to establish the risk identification model, and to train and verify the model.

[0063] Furthermore, the data preparation is specifically as follows:

[0064] C1, data partitioning: the data set is divided into training set, validation set and test set, with a ratio of 70% training set, 15% validation set and 15% test set;

[0065] C2, feature scaling: normalize or standardize the features to make them at the same scale;

[0066] C3, Data Encoding: Perform one-hot encoding or label encoding on categorical features to convert text or categorical data into numerical data.

[0067] C4 uses the trained risk identification model to conduct risk assessment on the data collected in real time, and generates corresponding risk levels and early warning signals based on the assessment results.

[0068] In an optional embodiment, the risk identification model established using the first intelligent algorithm may also be a data-enhanced prediction model, specifically including D1 to D5:

[0069] D1, for cases where there are few data samples or the data is unbalanced, oversampling (such as SMOTE), undersampling or data synthesis methods are used to generate more samples or balance the category distribution.

[0070] D2, before building the model, apply more complex feature engineering methods, such as building interactive features, generating new derived features, time series feature extraction, or using clustering labels, to further optimize the expression of the data.

[0071] D3, select important features through algorithms, such as information gain, recursive feature elimination (RFE) or Lasso regression to select the feature set that is most helpful for risk identification.

[0072] D4: During the model training process, use methods such as Grid Search or Random Search to optimize and adjust the key parameters of the model.

[0073] D5, based on the integration technology of multiple different models, such as Bagging, Boosting, Stacking, etc., combines the prediction results of multiple models to improve the prediction ability and robustness of the overall model.

[0074] It should be noted that in the data preparation stage, abnormal data and noise data are cleaned and processed to ensure the quality of input data, thereby improving the risk identification effect of the model. If the data volume is large and has complex patterns, deep neural networks (such as convolutional neural networks, recurrent neural networks, etc.) can be introduced for modeling to better capture the complex features and relationships in the data.

[0075] S4: Provide feedback for maintenance and regularly update and maintain the risk identification model based on the first risk assessment results.

[0076] Furthermore, generating the first risk assessment result includes performing a risk assessment on the data collected in real time, and generating a corresponding risk level and warning signal according to the assessment result.

[0077] The regularly updating and maintaining the first model according to the first risk assessment result includes updating and maintaining the risk identification model according to feedback after maintenance and regular performance evaluation results to ensure that the recognition accuracy of the first intelligent algorithm is continuously optimized over time.

[0078] It should be noted that the risk assessment results and early warning signals are presented to the operators in a visual form, and the feedback information of the maintenance personnel is recorded. The specific process steps for further optimization and adjustment of the algorithm include designing a visual interface, real-time data display, multi-dimensional data analysis, cross-cloud alarm and notification functions, and user interaction and feedback.

[0079] Example 3: The following is an embodiment of the present invention, which provides a method for intelligently identifying risks in the maintenance process of a nuclear power plant work order. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0080] First, in order to verify the effectiveness and advantages of the intelligent risk identification method based on the maintenance process of nuclear power plant work orders, an experimental platform was established to simulate the actual processing process of maintenance work orders in nuclear power plants. In the experiment, the first collection module was used to collect maintenance-related data in real time from the monitoring system of the nuclear power plant, including equipment operating status, fault history, environmental parameters, maintenance time, maintenance personnel operation records, etc. Data collection is carried out continuously to ensure the timeliness and diversity of the data.

[0081] After data collection is completed, the data is preprocessed. The specific operations are as follows: First, all data are cleaned to remove missing data, unreasonable values, and noise interference data to ensure data quality. Then, the numerical data is normalized to be within the range of [0,1] to eliminate the scale difference between the data. For categorical data, unique hot encoding is used to convert it into numerical form. After preprocessing, the data enters the first feature extraction stage.

[0082] In the feature extraction process, correlation analysis and data exploratory analysis are used to extract features that are highly correlated with maintenance risks, such as equipment operating time, failure frequency, ambient temperature, and maintenance time. Furthermore, feature selection methods are used to determine the feature subset that is most conducive to risk identification, so as to reduce the complexity of the model and improve the recognition efficiency of the model.

[0083] In order to ensure the long-term effectiveness of the model, a regular feedback and update mechanism is set. According to the feedback and evaluation results after the maintenance operation, the model is regularly retrained and adjusted to ensure that the model can continuously adapt to the new maintenance data environment. All data and model performance are recorded during the experiment for subsequent analysis and verification. Some experimental results are shown in Table 1.

[0084] Table 1 Experimental data record table

[0085]

[0086]

[0087] From the above experimental data table, it can be seen that the use of intelligent risk identification methods based on the maintenance process of nuclear power plant work orders can effectively predict and evaluate equipment maintenance risks. Model sensitivity to failure frequency: As can be seen from the table, equipment C has the highest failure frequency, and the model assesses its risk level as 5, and issues a red warning signal. In contrast, equipment B and equipment G have lower failure frequencies, and the risk level assessed by the model is also lower, only 3, and the warning signal is yellow. This shows that the model is highly sensitive and accurate to the key feature of failure frequency.

[0088] The experimental data also shows that the risk levels of devices A, D, and E with higher ambient temperatures are generally higher. This shows that the model can identify potential risks in combination with environmental parameters. In contrast, device B with lower ambient temperature has lower risk, which further verifies the model's advantage in comprehensive multi-feature evaluation.

[0089] Example 4: The following is an embodiment of the present invention, which provides a risk intelligent identification system based on the maintenance process of a nuclear power plant work order, including a data collection module, a data preprocessing module, a model building module, and a risk assessment module.

[0090] The data collection module is responsible for collecting data related to the work order maintenance of the nuclear power plant in real time, including a real-time data collection function called from the nuclear power plant monitoring system.

[0091] The data preprocessing module processes the collected maintenance data, cleans, normalizes, and encodes the data, removes noise and unifies the data format, extracts features related to maintenance risks, and selects the most suitable feature subset for risk identification.

[0092] The model building module uses intelligent algorithms to build and train risk identification models, conduct risk assessment on real-time data, and generate risk assessment results.

[0093] The risk assessment module publishes risk levels and early warning signals, and updates and maintains the risk identification model based on feedback and regular performance evaluation results.

[0094] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0095] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0096] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0097] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0098] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for intelligently identifying risks in the maintenance process of nuclear power plant work orders, characterized in that: include: Collect maintenance-related data in real time through a first collection module; Preprocessing maintenance-related data and performing first feature extraction; Using the first intelligent algorithm to establish a risk identification model, using the trained risk identification model to perform risk assessment on real-time data, and generating a first risk assessment result; Provide feedback during maintenance and regularly update and maintain the risk identification model based on the results of the first risk assessment.

2. The method for intelligently identifying risks in the maintenance process of a nuclear power plant work order according to claim 1, characterized in that: The first collecting module collecting the maintenance related data in real time includes calling the first collecting module from the monitoring system of the nuclear power plant to collect the maintenance related data in real time.

3. The method for intelligently identifying risks in the maintenance process of a nuclear power plant work order according to claim 2, characterized in that: The preprocessing of maintenance-related data includes cleaning, normalizing and encoding the collected data to eliminate noise and inconsistency in format.

4. The method for intelligently identifying risks in the maintenance process of a nuclear power plant work order according to claim 3, characterized in that: The first feature extraction comprises: The first features related to maintenance risk are extracted, and a subset of the first features that is most conducive to risk identification is selected.

5. The method for intelligently identifying risks in the maintenance process of a nuclear power plant work order according to claim 4, characterized in that: The use of the first intelligent algorithm to establish a risk identification model includes using the extracted features and the selected feature subset, adopting the first intelligent algorithm to establish a risk identification model, and training and verifying the model.

6. The method for intelligently identifying risks in the maintenance process of a nuclear power plant work order according to claim 5, characterized in that: The generating of the first risk assessment result includes performing a risk assessment on the data collected in real time, and generating a corresponding risk level and a warning signal according to the assessment result.

7. The method for intelligently identifying risks in the maintenance process of a nuclear power plant work order according to claim 6, characterized in that: The regularly updating and maintaining the first model according to the first risk assessment result includes updating and maintaining the risk identification model according to feedback after maintenance and regular performance evaluation results to ensure that the recognition accuracy of the first intelligent algorithm is continuously optimized over time.

8. A system using the method for intelligent risk identification in the maintenance process of a nuclear power plant work order according to any one of claims 1 to 7, characterized in that: It includes data collection module, data preprocessing module, model building module and risk assessment module; The data collection module is responsible for collecting data related to the maintenance of nuclear power plant work orders in real time, including a real-time data collection function called from the nuclear power plant monitoring system; The data preprocessing module processes the collected maintenance data, cleans, normalizes, and encodes the data, removes noise and unifies the data format, extracts features related to maintenance risks, and selects the most suitable feature subset for risk identification; The model building module uses intelligent algorithms to build and train risk identification models, conduct risk assessment on real-time data, and generate risk assessment results; The risk assessment module publishes risk levels and early warning signals, and updates and maintains the risk identification model based on feedback and regular performance evaluation results.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for intelligently identifying risks in the maintenance process of a nuclear power plant work order are implemented as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for intelligently identifying risks in a nuclear power plant work order maintenance process according to any one of claims 1 to 7 are implemented.

Images