Nuclear equipment fault monitoring method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of nuclear power plant equipment safety, in particular to a nuclear equipment fault monitoring method and device, electronic equipment and a storage medium. According to the nuclear equipment fault monitoring method provided by the embodiment of the invention, the historical monitoring matrix and the real-time monitoring data of the target nuclear equipment need to be acquired firstly; performing increment processing on the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix; constructing an integrated multivariate state estimation model based on the dynamic memory matrix; inputting the real-time monitoring data into the integrated multivariate state estimation model to calculate a state estimation vector corresponding to the integrated multivariate state estimation model; performing core equipment state evaluation based on the state estimation vector and the real-time monitoring data to obtain core equipment state information; and determining that the state information of the core equipment is in a fault state in response to the condition that the state information of the core equipment meets a preset fault judgment condition. Therefore, the monitoring accuracy of the nuclear equipment can be improved.

Description

Technical Field

[0001] The present application relates to the field of nuclear power plant equipment safety technology, and in particular to a nuclear equipment fault monitoring method and device, electronic equipment, and storage medium. Background Art

[0002] Nuclear equipment failure prediction and health management are important aspects of safe, efficient and reliable operation of nuclear power plants. However, during the long-term service of nuclear equipment, due to factors such as operating errors of nuclear power plant operators, natural environmental disasters, and weak digitalization, the performance of nuclear equipment will accelerate under improper use conditions, and the probability of failure will continue to increase. In serious cases, it may cause the nuclear power plant to be shut down for maintenance unexpectedly or even cause major safety accidents such as nuclear leakage.

[0003] At present, nuclear power plants generally adopt regular preventive maintenance measures for nuclear equipment based on experience and statistical data. However, in fact, the occurrence of most nuclear equipment failures is not strongly correlated with the use time. Frequent preventive maintenance of nuclear equipment under unnecessary circumstances has caused excessive maintenance to a certain extent, resulting in increased maintenance costs and reduced unit availability. Therefore, how to improve the monitoring accuracy of nuclear equipment is still a difficult problem that needs to be solved in the industry. Summary of the invention

[0004] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a nuclear equipment fault monitoring method and device, electronic equipment, and storage medium, which can improve the monitoring accuracy of nuclear equipment.

[0005] A nuclear equipment fault monitoring method according to an embodiment of the first aspect of the present application includes:

[0006] Obtain historical monitoring matrix and real-time monitoring data of target nuclear equipment;

[0007] Incrementally process the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix;

[0008] Constructing an integrated multivariate state estimation model based on the dynamic memory matrix;

[0009] Inputting the real-time monitoring data into the integrated multivariate state estimation model to calculate a state estimation vector corresponding to the integrated multivariate state estimation model;

[0010] Based on the state estimation vector and the real-time monitoring data, nuclear equipment state evaluation is performed to obtain nuclear equipment state information;

[0011] In response to the core device status information satisfying a preset fault judgment condition, it is determined that the core device status information is in a fault state.

[0012] According to some embodiments of the present application, the historical monitoring matrix includes a plurality of historical monitoring vectors, and the incremental processing of the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix includes:

[0013] Based on the real-time monitoring data, the closeness calculation of the historical monitoring matrix is ​​performed to obtain the parameter closeness corresponding to each of the historical monitoring vectors;

[0014] Reordering the historical monitoring vectors in the historical monitoring matrix based on the parameter closeness to obtain a target monitoring matrix;

[0015] Sampling is performed on the target monitoring matrix to obtain the dynamic memory matrix.

[0016] According to some embodiments of the present application, the closeness calculation of the historical monitoring matrix based on the real-time monitoring data to obtain the parameter closeness corresponding to each of the historical monitoring vectors includes:

[0017] The historical monitoring matrix and the real-time monitoring data are respectively standardized to obtain a standardized historical matrix corresponding to the historical monitoring matrix and a standardized real-time vector corresponding to the real-time monitoring data;

[0018] Constructing a correlation monitoring matrix according to the real-time monitoring data and each of the historical monitoring vectors;

[0019] The parameter closeness corresponding to each of the historical monitoring vectors is calculated based on the historical monitoring matrix, the real-time monitoring data and the correlation matrix.

[0020] According to some embodiments of the present application, constructing a correlation monitoring matrix according to the real-time monitoring data and each of the historical monitoring vectors includes:

[0021] Calculate the historical monitoring mean corresponding to each of the historical monitoring vectors;

[0022] Determining each correlation matrix element based on the ratio of the real-time monitoring data to each of the historical monitoring averages;

[0023] The relevance matrix elements are integrated into a diagonal matrix to construct the relevance monitoring matrix.

[0024] According to some embodiments of the present application, constructing an integrated multivariate state estimation model based on the dynamic memory matrix includes:

[0025] Partitioning the dynamic memory matrix to obtain a first number of cluster center parameters and a sub-dynamic memory matrix corresponding to each of the cluster center parameters;

[0026] Determine the base model estimation parameter corresponding to each cluster center parameter according to the sub-dynamic memory matrix;

[0027] The first number of base model estimation parameters and the cluster center parameters corresponding to each of the base model estimation parameters are integrated to obtain the integrated multivariate state estimation model.

[0028] According to some embodiments of the present application, the dynamic memory matrix includes a plurality of historical monitoring vectors obtained by sampling, each of the historical monitoring vectors is configured with a corresponding parameter closeness, and the partitioning process is performed on the dynamic memory matrix to obtain a first number of cluster center parameters and a sub-dynamic memory matrix corresponding to each of the cluster center parameters, including:

[0029] Based on the parameter closeness of each of the historical monitoring vector configurations, configuring a reference value for the dynamic memory matrix;

[0030] The reference degree value is used to perform neighbor propagation clustering on the dynamic memory matrix to partition the dynamic memory matrix, thereby obtaining a first number of cluster center parameters and the sub-dynamic memory matrices corresponding to each cluster center parameter.

[0031] According to some embodiments of the present application, inputting the real-time monitoring data into the integrated multivariate state estimation model to calculate a state estimation vector corresponding to the integrated multivariate state estimation model includes:

[0032] Calculate the initial weight corresponding to each cluster center parameter according to the real-time monitoring data and each cluster center parameter;

[0033] Performing weighting processing on the initial weights corresponding to the cluster center parameters to update the initial weights and obtain the base model contribution corresponding to the base model estimation parameters;

[0034] The state estimation vector is calculated based on the estimated parameters of each base model and the contribution of the base model corresponding to each base model estimated parameter.

[0035] According to some embodiments of the present application, the state estimation vector is calculated according to each of the base model estimation parameters and the base model contribution corresponding to each of the base model estimation parameters, including:

[0036] For each of the base model estimation parameters, the product of the base model estimation parameter and the corresponding base model contribution is determined as a contribution component element;

[0037] Summing a first number of the contribution component elements to obtain a first state estimation element;

[0038] Summing the contribution of the first number of base models to obtain a second state estimation element;

[0039] The ratio of the first state estimation element to the second state estimation element is determined as the state estimation vector.

[0040] According to some embodiments of the present application, the nuclear equipment state assessment based on the state estimation vector and the real-time monitoring data to obtain the nuclear equipment state information includes:

[0041] Performing residual calculation according to the state estimation vector and the real-time monitoring data to obtain a residual sequence; wherein the residual sequence includes residual values ​​corresponding to each observation time;

[0042] Performing covariance calculation on each residual value in the residual sequence to obtain a first covariance matrix at a target observation moment and a second covariance matrix at a moment immediately before the target observation moment;

[0043] A health index is calculated based on the first covariance matrix and the second covariance matrix to obtain the nuclear equipment status information.

[0044] According to some embodiments of the present application, before determining that the nuclear device status information is in a fault state in response to the nuclear device status information satisfying a preset fault judgment condition, the method further includes:

[0045] Acquire a warning threshold coefficient and past device status information corresponding to a plurality of observation times before the target observation time;

[0046] Perform threshold calculation based on the warning threshold coefficient and each of the past device status information to obtain a fault determination threshold;

[0047] In response to the nuclear device status information satisfying a preset fault judgment condition, determining that the nuclear device status information is in a fault state includes:

[0048] In response to the core device status information exceeding the fault judgment threshold, it is determined that the core device status information satisfies the fault judgment condition, and the core device status information is determined to be in the fault state.

[0049] According to some embodiments of the present application, the historical monitoring matrix includes a plurality of historical monitoring vectors, and after the threshold value is calculated according to the warning threshold coefficient and each of the past device status information to obtain the fault determination threshold value, it further includes:

[0050] In response to the nuclear equipment status information not exceeding the fault determination threshold, incorporating the real-time monitoring data into the historical monitoring matrix;

[0051] After incorporating the real-time monitoring data into the historical monitoring matrix, performing a closeness calculation on the historical monitoring matrix based on the real-time monitoring data to obtain a parameter closeness corresponding to each of the historical monitoring vectors;

[0052] The historical monitoring vectors whose parameter closeness is less than a preset closeness threshold are deleted from the historical monitoring matrix to update the historical monitoring matrix.

[0053] According to some embodiments of the present application, the target nuclear equipment is configured with sensor measurement points, and the step of obtaining the historical monitoring matrix and real-time monitoring data of the target nuclear equipment includes:

[0054] Acquire monitoring data sets at multiple observation times; wherein each monitoring data set includes monitoring sensor data collected by each sensor measuring point at one observation time;

[0055] Integrate the monitoring sensor data collected at each observation time into an operation data vector corresponding to the observation time;

[0056] Determine the operation data vector corresponding to the target observation time as real-time monitoring data;

[0057] The operation data vectors corresponding to the multiple observation times before the target observation time are integrated into the historical monitoring matrix.

[0058] According to the second aspect of the present application, a nuclear equipment fault monitoring device includes:

[0059] A data acquisition module is used to obtain the historical monitoring matrix and real-time monitoring data of the target nuclear equipment;

[0060] An incremental processing module, used for performing incremental processing on the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix;

[0061] An estimation model building module, used for building an integrated multivariate state estimation model based on the dynamic memory matrix;

[0062] A vector calculation module, used for inputting the real-time monitoring data into the integrated multivariate state estimation model to calculate a state estimation vector corresponding to the integrated multivariate state estimation model;

[0063] A state assessment module, used to conduct a nuclear equipment state assessment based on the state estimation vector and the real-time monitoring data to obtain nuclear equipment state information;

[0064] The fault judgment module is used to determine that the nuclear equipment status information is in a fault state in response to the nuclear equipment status information satisfying a preset fault judgment condition.

[0065] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements a nuclear equipment fault monitoring method as described in any one of the embodiments of the first aspect of the present application.

[0066] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement a nuclear equipment fault monitoring method as described in any one of the embodiments of the first aspect of the present application.

[0067] The nuclear equipment fault monitoring method and device, electronic device, and storage medium according to the embodiments of the present application have at least the following beneficial effects:

[0068] According to the nuclear equipment fault monitoring method of the embodiment of the present application, it is necessary to first obtain the historical monitoring matrix and real-time monitoring data of the target nuclear equipment; perform incremental processing on the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix; construct an integrated multivariate state estimation model based on the dynamic memory matrix; input the real-time monitoring data into the integrated multivariate state estimation model to calculate the state estimation vector corresponding to the integrated multivariate state estimation model; perform nuclear equipment state assessment based on the state estimation vector and the real-time monitoring data to obtain nuclear equipment state information; in response to the nuclear equipment state information satisfying the preset fault judgment condition, determine that the nuclear equipment state information is in a fault state. In this way, the monitoring accuracy of nuclear equipment can be improved.

[0069] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0071] Figure 1 A schematic diagram of a process flow of a nuclear equipment fault monitoring method provided in an embodiment of the present application;

[0072] Figure 2 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0073] Figure 3 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0074] Figure 4 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0075] Figure 5 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0076] Figure 6 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0077] Figure 7 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0078] Figure 8 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0079] Fig. 9 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0080] Fig.10 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0081] Fig.11 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0082] Fig.12 Another schematic diagram of a process flow of a method for monitoring nuclear equipment failures provided in an embodiment of the present application;

[0083] Fig.13 is a schematic diagram of the structure of a nuclear equipment fault monitoring device provided in an embodiment of the present application;

[0084] Fig.14 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0085] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0086] In the description of this application, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed", etc. are understood to exclude the number itself, and "above", "below", "within", etc. are understood to include the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0087] In the description of the present application, it should be understood that descriptions involving orientation, such as up, down, left, right, front, back, etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0088] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0089] In the description of this application, it should be noted that, unless otherwise clearly defined, the terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in this application in combination with the specific content of the technical solution. In addition, the identification of specific steps below does not represent a limitation on the order of steps and execution logic. The execution order and execution logic between each step should be understood and inferred with reference to the contents described in the embodiment.

[0090] Nuclear equipment failure prediction and health management are important aspects of safe, efficient and reliable operation of nuclear power plants. However, during the long-term service of nuclear equipment, due to factors such as operating errors of nuclear power plant operators, natural environmental disasters, and weak digitalization, the performance of nuclear equipment will accelerate under improper use conditions, and the probability of failure will continue to increase. In serious cases, it may cause the nuclear power plant to be shut down for maintenance unexpectedly or even cause major safety accidents such as nuclear leakage.

[0091] At present, nuclear power plants generally adopt regular preventive maintenance measures for nuclear equipment based on experience and statistical data. However, in fact, the occurrence of most nuclear equipment failures is not strongly correlated with the use time. Frequent preventive maintenance of nuclear equipment under unnecessary circumstances has caused excessive maintenance to a certain extent, resulting in increased maintenance costs and reduced unit availability.

[0092] Therefore, adopting a state-based predictive operation strategy to improve the comprehensiveness, timeliness and accuracy of operating status monitoring of key equipment in nuclear power plants and issuing early warnings before equipment failures occur is of great significance for ensuring the safety and reliability of nuclear power plant operations and reducing the occurrence of nuclear accidents.

[0093] With the rapid development of advanced sensor technology and industrial Internet of Things, the real-time operation signals of nuclear power plant equipment can be perceived, managed and analyzed through sensors or data acquisition terminals, laying the foundation for nuclear equipment performance monitoring and early warning analysis.

[0094] Fault identification methods based on signal processing technology are classic methods for equipment performance monitoring and early warning analysis, including time domain, frequency domain, time-frequency domain, modal decomposition, and various filtering methods (Kalman filtering, blind filtering), etc. This type of method extracts the fault characteristics of monitoring data through deep information mining of equipment operation signals, and establishes a mapping relationship between fault characteristics and fault modes.

[0095] In recent years, equipment performance monitoring and early warning analysis methods based on multivariate state estimation technology (including principal component analysis, independent component analysis, linear discriminant analysis, and partial least squares analysis) have been widely studied. These methods project the high-dimensional data composed of equipment operation signals into a low-dimensional manifold space, and then mine the correlation information between equipment operation signals.

[0096] However, due to the influence of the radiation environment of nuclear power plants and the high safety requirements of operation, the above methods face some problems in implementation and application:

[0097] First, the high computational complexity makes it difficult to efficiently process large-scale monitoring data, resulting in low timeliness of equipment performance monitoring and failure to meet the on-site operation requirements of nuclear power plants.

[0098] Secondly, it has poor generalization and cannot be updated adaptively as the working mechanism of the equipment changes. Therefore, when the external working conditions fluctuate, the accuracy of the model warning is greatly reduced.

[0099] The present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a nuclear equipment fault monitoring method and device, electronic equipment, and storage medium, which can improve the monitoring accuracy of nuclear equipment.

[0100] The following is a further explanation based on the attached drawings:

[0101] Reference Figure 1 The nuclear equipment fault monitoring method according to the embodiment of the present application may include, but is not limited to:

[0102] Step S101, obtaining the historical monitoring matrix and real-time monitoring data of the target nuclear equipment;

[0103] Step S102, performing incremental processing on the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix;

[0104] Step S103, constructing an integrated multivariate state estimation model based on the dynamic memory matrix;

[0105] Step S104, inputting the real-time monitoring data into the integrated multivariate state estimation model to calculate a state estimation vector corresponding to the integrated multivariate state estimation model;

[0106] Step S105, evaluating the nuclear equipment status based on the state estimation vector and the real-time monitoring data to obtain the nuclear equipment status information;

[0107] Step S106, in response to the nuclear device status information satisfying a preset fault judgment condition, determining that the nuclear device status information is in a fault state.

[0108] The nuclear equipment fault monitoring method shown in step S101 to step S106 of the embodiment of the present application needs to first obtain the historical monitoring matrix and real-time monitoring data of the target nuclear equipment; perform incremental processing on the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix; construct an integrated multivariate state estimation model based on the dynamic memory matrix; input the real-time monitoring data into the integrated multivariate state estimation model to calculate the state estimation vector corresponding to the integrated multivariate state estimation model; perform nuclear equipment state evaluation based on the state estimation vector and real-time monitoring data to obtain nuclear equipment state information; in response to the nuclear equipment state information meeting the preset fault judgment condition, determine that the nuclear equipment state information is in a fault state. In this way, the embodiment of the present application can realize real-time and accurate monitoring of the nuclear equipment state, effectively solve the technical problems of nuclear power plants in equipment monitoring and fault prediction, and provide a strong guarantee for the safe, efficient and reliable operation of nuclear power plants.

[0109] In step S101 of some embodiments, a comprehensive data foundation is provided for the embodiments of the present application by first acquiring the historical monitoring matrix and real-time monitoring data of the target nuclear equipment. The historical monitoring matrix contains a large amount of historical data of the equipment under normal operation, while the real-time monitoring data reflects the current operation status of the equipment. The combination of the two enables the embodiments of the present application to fully understand the operation history and current status of the equipment.

[0110] Reference Figure 2 According to some embodiments of the present application, the target nuclear equipment is configured with sensor measuring points, and step S101 obtains the historical monitoring matrix and real-time monitoring data of the target nuclear equipment, which may include, but is not limited to:

[0111] Step S201, obtaining a plurality of monitoring data sets at observation times; wherein each monitoring data set includes monitoring sensor data collected by each sensor measuring point at an observation time;

[0112] Step S202, integrating each monitoring sensor data collected at each observation time into an operation data vector corresponding to the observation time;

[0113] Step S203, determining the operation data vector corresponding to the target observation time as real-time monitoring data;

[0114] Step S204, integrating the operation data vectors corresponding to multiple observation times before the target observation time into a historical monitoring matrix.

[0115] In some embodiments of the present application, the target nuclear equipment is configured with sensor measurement points, which are responsible for collecting the operating status information of the equipment in real time. In step S101, the embodiment of the present application obtains the historical monitoring matrix and real-time monitoring data of the target nuclear equipment to provide a data basis for subsequent analysis and fault prediction. This process may include several sub-steps.

[0116] In step S201 of some embodiments, the embodiments of the present application collect monitoring data sets at multiple observation times. The monitoring data set at each observation time contains the monitoring sensor data collected by all sensor measuring points at that time. These data sets provide the embodiments of the present application with detailed operating status information of the equipment at different time points, which is the basis for constructing a historical monitoring matrix and determining real-time monitoring data.

[0117] In step S202 of some embodiments, the embodiments of the present application integrate the monitoring sensor data collected at each observation time to form an operation data vector corresponding to the observation time. This operation data vector is a mathematical representation of the operation state of the device at a specific time. It contains the data of all relevant sensor measurement points and can be used for subsequent state estimation and fault diagnosis.

[0118] In step S203 of some embodiments, the embodiment of the present application determines the operation data vector corresponding to the target observation time as real-time monitoring data. This real-time monitoring data is the current focus of the embodiment of the present application. It reflects the latest operation status of the device and is crucial for timely discovering changes in device performance.

[0119] In step S204 of some embodiments, the embodiments of the present application integrate the operation data vectors corresponding to multiple observation times before the target observation time into a historical monitoring matrix. This matrix records the changes in the operation status of the equipment over a period of time and is an important data source for trend analysis and state estimation. Through the historical monitoring matrix, the embodiments of the present application can identify the long-term trends and patterns of equipment operation and provide a basis for predicting the future operation status of the equipment.

[0120] The embodiment of the present application shown by step S201 to step S204 can comprehensively collect and integrate the monitoring data of the equipment, providing solid data support for accurate status assessment and fault prediction. This method not only improves the accuracy of monitoring, but also enhances the ability to predict the future status of the equipment by analyzing historical data, thereby providing a strong guarantee for the safe operation of the nuclear power plant.

[0121] In step S102 of some embodiments, a dynamic memory matrix is ​​obtained by incrementally processing the historical monitoring matrix. This step allows the embodiments of the present application to dynamically update and adjust the historical data to reflect the latest operating status of the device, thereby solving the problem of outdated data in traditional monitoring methods. The construction of the dynamic memory matrix enables the embodiments of the present application to adapt to changes in the operating status of the device and improves the timeliness and accuracy of monitoring.

[0122] In some more specific embodiments, the historical monitoring matrix X his It can be expressed as:

[0123]

[0124] Among them, the historical monitoring matrix X his The element x in ij Indicates that at t i The measurement data of the jth measuring point monitored at all times, x i Indicates that at t i The historical monitoring vector is composed of the measurement data of all the measuring points monitored at all times. n represents the number of monitoring vectors, and m represents the number of measuring points, where: i = 1, ..., n; j = 1, ..., m.

[0125] Reference Figure 3 According to some embodiments of the present application, the historical monitoring matrix includes multiple historical monitoring vectors. Step S102 performs incremental processing on the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix, which may include, but is not limited to:

[0126] Step S301, performing closeness calculation on the historical monitoring matrix based on the real-time monitoring data to obtain the parameter closeness corresponding to each historical monitoring vector;

[0127] Step S302, reordering the historical monitoring vectors in the historical monitoring matrix based on the parameter closeness to obtain a target monitoring matrix;

[0128] Step S303: sampling the target monitoring matrix to obtain a dynamic memory matrix.

[0129] In some embodiments of the present application, the construction and processing of the historical monitoring matrix is ​​the core part of the nuclear equipment fault monitoring method. The historical monitoring matrix is ​​composed of multiple historical monitoring vectors, each of which represents the operating status of the equipment at different observation times. In step S102, the embodiment of the present application performs incremental processing on the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix. This process includes several key operations.

[0130] In step S301 of some embodiments, the embodiments of the present application perform a closeness calculation on the historical monitoring matrix based on the real-time monitoring data to obtain the parameter closeness corresponding to each historical monitoring vector. Parameter closeness is an indicator that measures the similarity between real-time monitoring data and historical monitoring data. By calculating the parameter closeness corresponding to each historical monitoring vector, the embodiments of the present application can evaluate the closeness between real-time data and historical data. The purpose of this step is to identify the historical data that is most similar to the current device status and provide a reference for subsequent monitoring and analysis.

[0131] In step S302 of some embodiments, the embodiments of the present application reorder the historical monitoring vectors in the historical monitoring matrix based on the calculated parameter proximity. The purpose of this operation is to prioritize the historical data closest to the real-time monitoring data, thereby forming a target monitoring matrix. It should be understood that the construction of the target monitoring matrix enables the embodiments of the present application to pay more attention to the data that is most relevant to the current equipment status when performing fault monitoring on the target nuclear equipment, thereby improving the pertinence and effectiveness of the monitoring.

[0132] In step S303 of some embodiments, the embodiments of the present application sample the target monitoring matrix to obtain a dynamic memory matrix. It should be noted that sampling is a data reduction technology that selects a portion of the most representative historical monitoring vectors from the target monitoring matrix to construct a dynamic memory matrix. This dynamic memory matrix not only contains the data most relevant to the current device status, but also reduces the size of the data through sampling, thereby improving the efficiency of data processing.

[0133] The embodiment of the present application provided through steps S301 to S303 can effectively process and update historical monitoring data, construct a dynamic memory matrix that reflects the current state of the target nuclear equipment and has historical continuity, improves the accuracy of monitoring, and provides data support for accurate state assessment and fault prediction.

[0134] Reference Figure 4 According to some embodiments of the present application, step S301 performs closeness calculation on the historical monitoring matrix based on the real-time monitoring data to obtain the parameter closeness corresponding to each historical monitoring vector, which may include, but is not limited to:

[0135] Step S401, standardizing the historical monitoring matrix and the real-time monitoring data respectively to obtain a standardized historical matrix corresponding to the historical monitoring matrix and a standardized real-time vector corresponding to the real-time monitoring data;

[0136] Step S402, constructing a correlation monitoring matrix based on the real-time monitoring data and each historical monitoring vector;

[0137] Step S403, calculating the parameter closeness corresponding to each historical monitoring vector according to the historical monitoring matrix, the real-time monitoring data and the correlation matrix.

[0138] In some embodiments of the present application, the closeness calculation of the historical monitoring matrix is ​​to evaluate the similarity between the real-time monitoring data and the historical data, so as to provide a basis for the condition monitoring and fault prediction of the nuclear equipment. This process starts with step S301, which includes several key sub-steps.

[0139] In step S401 of some embodiments, the embodiments of the present application perform standardization processing on the historical monitoring matrix and the real-time monitoring data. Standardization is to eliminate the dimensional differences between different sensor measurement points and ensure that all data are compared on the same scale. Through this processing, a standardized historical monitoring matrix and a standardized real-time monitoring vector are obtained. The purpose of this step is to improve the comparability of the data and make the closeness calculation more accurate.

[0140] In step S402 of some embodiments, the embodiments of the present application construct a correlation monitoring matrix based on the real-time monitoring data and each historical monitoring vector. This matrix reflects the degree of correlation between the real-time monitoring data and the historical monitoring data, and is the basis for calculating the closeness. The construction of the correlation monitoring matrix may involve a variety of mathematical methods, such as distance measurement, similarity measurement, etc., the purpose of which is to quantify the similarity between the real-time data and the historical data.

[0141] Reference Figure 5 According to some embodiments of the present application, step S402 constructs a correlation monitoring matrix based on the real-time monitoring data and each historical monitoring vector, which may include, but is not limited to:

[0142] Step S501, calculating the historical monitoring mean corresponding to each historical monitoring vector;

[0143] Step S502, determining each correlation matrix element based on the ratio of the real-time monitoring data to each historical monitoring mean;

[0144] Step S503: Integrate the elements of each correlation matrix into a diagonal matrix to construct a correlation monitoring matrix.

[0145] In some embodiments of the present application, constructing a correlation monitoring matrix is ​​one of the key steps in implementing a nuclear equipment fault monitoring method. This process is performed in step S402, and is intended to quantify the correlation between real-time monitoring data and historical monitoring data in order to more accurately assess the current state of the equipment. The process of constructing a correlation monitoring matrix may include the following steps:

[0146] In step S501 of some embodiments, the embodiments of the present application calculate the historical monitoring mean corresponding to each historical monitoring vector. This step involves statistical analysis of the data in the historical monitoring matrix to determine the central trend of each historical monitoring vector. The calculation of the historical monitoring mean provides a benchmark for subsequent correlation evaluation, so that the embodiments of the present application can measure the degree of deviation between real-time data and historical data.

[0147] In step S502 of some embodiments, the embodiments of the present application determine each relevance matrix element based on the ratio of the real-time monitoring data to each historical monitoring mean. This step evaluates the similarity between the real-time data and the historical data by comparing the ratio of the real-time monitoring data to the historical mean. The calculation result of the ratio will be used as an element of the relevance matrix, reflecting the strength of association between the real-time monitoring data and each historical monitoring vector.

[0148] In step S503 of some embodiments, the embodiments of the present application integrate the elements of each correlation matrix into a diagonal matrix to construct a correlation monitoring matrix. A diagonal matrix is ​​a special matrix in which all elements on the non-diagonal line are zero and only the elements on the diagonal line have values. In the present application, the diagonal elements of the diagonal matrix represent the correlation between the real-time monitoring data and the historical monitoring vector. In this way, the correlation monitoring matrix can concisely and effectively express the correlation between the real-time data and the historical data.

[0149] The embodiment of the present application provided by steps S501 to S503 can construct a correlation monitoring matrix, which quantifies the correlation between real-time monitoring data and historical monitoring data. This quantitative correlation evaluation provides important information for the embodiment of the present application, which helps to more accurately evaluate the current state of the equipment and predict potential failure risks, thereby improving the safety and reliability of nuclear power plant equipment operation.

[0150] In step S403 of some embodiments, the embodiments of the present application calculate the parameter closeness corresponding to each historical monitoring vector based on the historical monitoring matrix, the real-time monitoring data and the correlation monitoring matrix. Parameter closeness is a key indicator that reflects the similarity between the real-time monitoring data and each historical monitoring vector. By calculating the closeness, the embodiments of the present application can identify the historical data that is most similar to the current device state, which is crucial for understanding the current operating state of the device and predicting its future behavior.

[0151] The embodiment of the present application provided through steps S401 to S403 can accurately evaluate the similarity between real-time monitoring data and historical data, providing important data support for subsequent state estimation and fault prediction. This method not only improves the accuracy of monitoring, but also enables the embodiment of the present application to adapt to changes in the operating state of the equipment through dynamic updates, providing solid data support for accurate state evaluation and fault prediction. The application of this method is of great significance for improving the safety and reliability of nuclear power plant equipment operation.

[0152] In some more specific embodiments of the present application, the historical monitoring matrix X his and real-time monitoring data x obs Perform standardization processing respectively to obtain the standardized historical matrix corresponding to the historical monitoring matrix Standardized real-time vector corresponding to real-time monitoring data It can be expressed as:

[0153]

[0154] Among them, l n+1 is a column vector whose elements are 1.

[0155] Further, the normalized real-time observation vector is calculated With standardized historical monitoring matrix The parameter closeness of the monitoring vector in i :

[0156]

[0157] Where Λ represents a diagonal matrix, which is calculated by normalizing the real-time observation vector With standardized historical monitoring matrix The larger the ratio, the higher the monitoring importance of the corresponding measuring point. The expression is as follows:

[0158]

[0159] Based on the parameter closeness r i Standardized Historical Monitoring Matrix The historical monitoring vectors in ascending order can be obtained by reordering the standardized historical monitoring matrix

[0160] In step S103 of some embodiments, an integrated multivariate state estimation model is constructed based on a dynamic memory matrix. The integrated multivariate state estimation model is an analysis model that integrates multiple base models and can estimate and predict the device state from multiple angles and levels. This integration method improves the accuracy and robustness of state estimation, so that the embodiments of the present application can more accurately capture the changing trend of device performance.

[0161] Reference Figure 6 According to some embodiments of the present application, step S103 constructs an integrated multivariate state estimation model based on a dynamic memory matrix, which may include, but is not limited to:

[0162] Step S601, partitioning the dynamic memory matrix to obtain a first number of cluster center parameters and a sub-dynamic memory matrix corresponding to each cluster center parameter;

[0163] Step S602, determining the base model estimation parameter corresponding to each cluster center parameter according to the sub-dynamic memory matrix;

[0164] Step S603: Integrate the first number of base model estimation parameters and the cluster center parameters corresponding to each base model estimation parameter to obtain an integrated multivariate state estimation model.

[0165] In some embodiments of the present application, constructing an integrated multivariate state estimation model (EMSET) is one of the core steps of the nuclear equipment fault monitoring method, which aims to improve the estimation accuracy of the equipment state by integrating the prediction results of multiple base models. This process is performed in step S103 and may include the following sub-steps.

[0166] In step S601 of some embodiments, the embodiments of the present application perform partition processing on the dynamic memory matrix to obtain a certain number of cluster center parameters and sub-dynamic memory matrices corresponding to these cluster center parameters. This step may include cluster analysis techniques, such as the improved neighbor propagation (AP) clustering method, which can identify natural groupings or patterns in the data. The cluster center parameters represent the center or typical representative of each subset (or cluster), and the sub-dynamic memory matrix contains data points related to each cluster center parameter. The purpose of this partitioning process is to decompose complex data sets into smaller, more manageable parts, each of which is represented by a cluster center parameter, thereby simplifying subsequent analysis and processing.

[0167] Reference Figure 7According to some embodiments of the present application, the dynamic memory matrix includes a plurality of historical monitoring vectors obtained by sampling, each historical monitoring vector is configured with a corresponding parameter closeness, and step S601 performs partition processing on the dynamic memory matrix to obtain a first number of cluster center parameters and a sub-dynamic memory matrix corresponding to each cluster center parameter, which may include, but is not limited to:

[0168] Step S701, configuring a reference value for a dynamic memory matrix based on the closeness of parameters configured by each historical monitoring vector;

[0169] Step S702, performing neighbor propagation clustering on the dynamic memory matrix by using the reference degree value to partition the dynamic memory matrix to obtain a first number of cluster center parameters and sub-dynamic memory matrices corresponding to each cluster center parameter.

[0170] In some embodiments of the present application, the construction and processing of the dynamic memory matrix is ​​to improve the accuracy and efficiency of nuclear equipment fault monitoring. The dynamic memory matrix includes historical monitoring vectors obtained by sampling, each of which is configured with a corresponding parameter closeness, which reflects the similarity between real-time monitoring data and historical data. In step S601, the embodiment of the present application performs partition processing on the dynamic memory matrix to obtain a certain number of cluster center parameters and sub-dynamic memory matrices corresponding to these cluster center parameters.

[0171] In step S701 of some embodiments, the embodiments of the present application configure reference values ​​for the dynamic memory matrix based on the closeness of the parameters configured for each historical monitoring vector. The reference value is a key parameter for controlling the formation of clusters during the clustering process, which can help determine which data points should be classified into the same category. Based on the closeness of the parameters configured for each historical monitoring vector, the reference value is configured for the dynamic memory matrix to ensure that the embodiments of the present application can pay more attention to historical data similar to the real-time monitoring data during the clustering process, thereby improving the accuracy and relevance of clustering.

[0172] In step S702 of some embodiments, the embodiments of the present application perform neighbor propagation clustering on the dynamic memory matrix through reference values. Neighbor propagation clustering is a clustering algorithm based on graph theory, which forms clusters by iteratively transferring similarity information between data points. In the present application, neighbor propagation clustering uses the configured reference value to guide the clustering process, so that historical monitoring vectors similar to real-time monitoring data are more likely to be classified into the same cluster. The result of this step is to obtain a certain number of cluster center parameters, each cluster center parameter represents the center or typical representative of a class of data, and sub-dynamic memory matrices corresponding to these cluster center parameters, which contain historical monitoring data similar to each cluster center parameter.

[0173] The embodiment of the present application shown by step S701 to step S702 can effectively divide the dynamic memory matrix into multiple related subsets, each subset having a certain similarity with the real-time monitoring data. This partitioning process not only simplifies the complexity of the data, but also makes the subsequent state estimation and fault prediction more accurate and efficient. The application of this method is of great significance for improving the safety and reliability of nuclear power plant equipment operation, because it can ensure that the embodiment of the present application pays more attention to the data most relevant to the current equipment status, thereby providing more accurate fault warning and status assessment.

[0174] In step S602 of some embodiments, the embodiments of the present application determine the base model estimation parameters corresponding to each cluster center parameter according to each sub-dynamic memory matrix. This step involves selecting or constructing a most suitable base model for each subset, where the base model can accurately capture and describe the characteristics and dynamics of the data in the subset. The base model estimation parameters are key parameters corresponding to the base model, which are used to characterize the corresponding output parameters after the base model captures and describes the characteristics and dynamics of the data in the subset.

[0175] In step S603 of some embodiments, the embodiments of the present application integrate all the estimated parameters of the base models and their corresponding cluster center parameters to obtain the final integrated multivariate state estimation model. This integrated multivariate state estimation model combines the characteristics of all base models and can estimate and predict the equipment status from multiple angles and levels. In this way, the integrated multivariate state estimation model can provide more comprehensive and accurate equipment status information than a single model, thereby providing strong support for fault monitoring and health management of nuclear equipment.

[0176] The embodiment of the present application shown by steps S601 to S603 can construct an integrated multivariate state estimation model. This integrated multivariate state estimation model improves the estimation accuracy of the nuclear equipment state by integrating the prediction results of multiple base models, providing a reliable basis for achieving accurate state assessment and fault prediction.

[0177] In some more specific embodiments of the present application, a fixed sampling interval δ can be selected according to a preset value n to reorder the standardized historical monitoring matrix Take samples.

[0178] Select the sampled historical monitoring vector to construct the dynamic memory matrix D dyn , in order to reduce the dynamic memory matrix D under the premise of ensuring the diversity of historical monitoring vector distribution dyn scale.

[0179] According to the parameter proximity, the negative value of the parameter proximity is assigned to the reference value of the Affinity Propagation (AP) clustering to control the number of cluster centers and complete the improvement of the Affinity Propagation clustering algorithm.

[0180] Furthermore, the improved neighbor propagation clustering algorithm is used to transform the dynamic memory matrix D dyn Perform partition processing to obtain l cluster centers and corresponding clusters, and construct a sub-dynamic memory matrix for each cluster:

[0181] AP(D dyn )=[c1,L,c k ,L,c l ,sD dyn1 ,K,sD dynk ,K,sD dynl ]

[0182] Among them, AP(·) is the improved neighbor propagation clustering algorithm, c k is the kth cluster center, sD dynk is the kth sub-dynamic memory matrix, where l is the number of clusters, i.e., the first number, k=1,…,l.

[0183] Furthermore, the base model is trained according to the sub-dynamic memory matrix, which is expressed as:

[0184]

[0185] in, It is the output result of the kth base model MSETk, that is, the base model estimation parameters corresponding to the kth base model.

[0186] Furthermore, the output result of the base model MSETk is and the cluster center c k Arranged in column direction, they constitute the integrated multivariate state estimation model EMSET.

[0187] In step S104 of some embodiments, the real-time monitoring data is input into the integrated multivariate state estimation model to calculate the state estimation vector, which is the best estimate of the current state of the equipment and provides an important basis for subsequent state evaluation. By comparing the state estimation vector and the real-time monitoring data, the method can accurately evaluate the state of the nuclear equipment in step S105 and obtain the state information of the equipment.

[0188] Reference Figure 8 According to some embodiments of the present application, step S104 inputs the real-time monitoring data into the integrated multivariate state estimation model to calculate the state estimation vector corresponding to the integrated multivariate state estimation model, which may include, but is not limited to:

[0189] Step S801, calculating the initial weight corresponding to each cluster center parameter according to the real-time monitoring data and each cluster center parameter;

[0190] Step S802, performing weighting processing on the initial weights corresponding to the parameters of each cluster center to update the initial weights and obtain the contribution of each base model estimation parameter to the base model;

[0191] Step S803, calculating a state estimation vector according to each base model estimation parameter and the base model contribution corresponding to each base model estimation parameter.

[0192] In some embodiments of the present application, the real-time monitoring data is input into the integrated multivariate state estimation model in order to calculate the corresponding state estimation vector, which is a key step in realizing nuclear equipment fault monitoring. The state estimation vector can provide an accurate estimate of the current operating state of the equipment, providing an important basis for subsequent state assessment and fault prediction. In step S104, this calculation process includes several key operations:

[0193] In step S801 of some embodiments, the embodiments of the present application calculate the initial weights corresponding to the parameters of each cluster center based on the real-time monitoring data and the parameters of each cluster center. It should be noted that the initial weights reflect the similarity between the real-time monitoring data and the parameters of each cluster center, and are the starting point for determining the importance of each base model in the integrated multivariate state estimation model. By calculating the initial weights, the embodiments of the present application can identify the base model that is most relevant to the real-time monitoring data, providing a basis for subsequent weight adjustment and state estimation.

[0194] In step S802 of some embodiments, the embodiments of the present application perform variable weight processing on the initial weights corresponding to the parameters of each cluster center to update the initial weights and obtain the base model contribution corresponding to each base model estimation parameter. Variable weight processing is a method of adjusting weights, which takes into account the performance of the base model on different data subsets and the relationship between the base models. The base model contribution corresponding to each base model estimation parameter is obtained by variable weight processing, so that the contribution of each base model to the final state estimation can be more accurately evaluated through the base model contribution, thereby improving the accuracy and reliability of the state estimation.

[0195] In step S803 of some embodiments, the embodiment of the present application calculates a state estimation vector based on each base model estimation parameter and the base model contribution corresponding to each base model estimation parameter. This step involves weighted averaging each base model estimation parameter according to its corresponding base model contribution to form a final state estimation vector. The state estimation vector integrates the information of all base models, provides a comprehensive estimate of the current operating state of the equipment, and provides an important basis for the state assessment and fault prediction of nuclear equipment.

[0196] Reference Fig. 9 According to some embodiments of the present application, step S803 calculates the state estimation vector according to each base model estimation parameter and the base model contribution corresponding to each base model estimation parameter, which may include, but is not limited to:

[0197] Step S901, for each base model estimation parameter, the product of the base model estimation parameter and the corresponding base model contribution is determined as a contribution component element;

[0198] Step S902, summing up the first number of contribution component elements to obtain a first state estimation element;

[0199] Step S903, summing up the contribution of the first number of base models to obtain a second state estimation element;

[0200] Step S904: determine the ratio of the first state estimation element to the second state estimation element as a state estimation vector.

[0201] In some embodiments of the present application, the process of calculating the state estimation vector is a key component of the integrated multivariate state estimation model, which involves combining the estimation results of each base model with its contribution to produce a comprehensive device state estimation. In step S803, this process includes several key operations:

[0202] In step S901 of some embodiments, the embodiments of the present application multiply each base model estimation parameter by the corresponding base model contribution to determine it as a contribution component. The purpose of this step is to quantify the contribution of each base model to the overall state estimation, and the result of the product reflects the weight of the base model estimation in the final state vector.

[0203] In step S902 of some embodiments, the present application embodiment sums up the contributing elements of all base models to obtain a first state estimation element. This summation operation actually integrates the information of all base models to form a comprehensive estimation value, which represents one aspect of the device state.

[0204] In step S903 of some embodiments, the present application embodiment sums up the contribution of all base models to obtain a second state estimation element. This summation operation provides an indicator to measure the total contribution of all base models, which helps to consider the relative importance of base models in the final state estimation.

[0205] In step S904 of some embodiments, the present application embodiment determines the ratio of the first state estimation element to the second state estimation element as a state estimation vector. This ratio operation is a normalization process, so that the state estimation vector can more accurately reflect the actual state of the device.

[0206] Through steps S901 to S904 shown in the embodiment of the present application, a comprehensive state estimation vector can be calculated, which not only takes into account the estimation results of each base model, but also takes into account their contribution to the final estimation. This method improves the accuracy and reliability of state estimation and provides important information for state assessment and fault prediction of target nuclear equipment. This method of calculating the state estimation vector enables the embodiment of the present application to more accurately capture the performance changes of the equipment, thereby providing strong support for the safe operation of nuclear power plants.

[0207] Through steps S801 to S803 shown in the embodiment of the present application, the real-time monitoring data can be effectively input into the integrated multivariate state estimation model to calculate a state estimation vector that accurately reflects the current state of the device. This method not only improves the accuracy of state estimation, but also makes the state estimation more comprehensive and reliable by considering the contribution between the base models.

[0208] According to some more specific embodiments of the present application, based on standardized real-time monitoring data and the parameters of each cluster center c1,L,c k ,L,c l , calculate the initial weight w corresponding to each cluster center parameter k , expressed as:

[0209]

[0210] The initial weight w corresponding to each cluster center parameter k Perform variable weight processing to update the initial weights and obtain the contribution w of each base model estimation parameter to the base model k ′, expressed as:

[0211]

[0212] Estimate parameters based on each basic model The base model contribution w′ corresponding to the estimated parameters of each base model k , calculate the state estimation vector x est , expressed as:

[0213]

[0214] in, To contribute to the composition elements, is the first state estimation element, is the second state estimation element.

[0215] In step S105 of some embodiments, when the nuclear equipment status information meets the preset fault judgment conditions, the embodiments of the present application can determine that the equipment is in a fault state and issue a warning in time. This state-based monitoring and warning mechanism can not only reduce unnecessary preventive maintenance and maintenance costs, but also improve the availability of the unit. More importantly, it can timely discover potential fault risks and avoid major safety accidents such as accidental shutdowns and nuclear leaks.

[0216] Reference Fig.10 According to some embodiments of the present application, step S105 performs nuclear equipment state assessment based on the state estimation vector and the real-time monitoring data to obtain nuclear equipment state information, which may include, but is not limited to:

[0217] Step S1001, performing residual calculation based on the state estimation vector and the real-time monitoring data to obtain a residual sequence; wherein the residual sequence includes residual values ​​corresponding to each observation time;

[0218] Step S1002, performing covariance calculation on each residual value in the residual sequence to obtain a first covariance matrix at the target observation time and a second covariance matrix at the previous time immediately before the target observation time;

[0219] Step S1003, calculating the health index based on the first covariance matrix and the second covariance matrix to obtain nuclear equipment status information.

[0220] In some embodiments of the present application, nuclear equipment state assessment is a key step, which determines the current state of the equipment based on the state estimation vector and real-time monitoring data. This process is performed in step S105 and includes several key operations, which are intended to assess the health status of the equipment by analyzing the deviations and changes in the data.

[0221] In step S1001 of some embodiments, the embodiments of the present application perform residual calculation based on the state estimation vector and the real-time monitoring data to obtain a residual sequence. The residual refers to the difference between the real-time monitoring data and the state estimation vector, and the residual sequence includes the residual values ​​corresponding to each observation time. The purpose of this step is to identify the deviation between the actual operating state of the equipment and the state predicted by the model. The residual sequence can reveal fluctuations and potential anomalies in equipment performance.

[0222] In step S1002 of some embodiments, the embodiments of the present application perform covariance calculation for each residual value in the residual sequence to obtain a first covariance matrix at the target observation time and a second covariance matrix at the previous moment adjacent to the target observation time. The covariance matrix is ​​a statistical tool used to describe the correlation between residual values, which can reflect the pattern and trend of changes in device status. By calculating the covariance matrix at different times, the embodiments of the present application can evaluate the stability and consistency of the device status.

[0223] In step S1003 of some embodiments, the embodiments of the present application calculate the health index based on the first covariance matrix and the second covariance matrix to obtain the nuclear equipment status information. The nuclear equipment status information is a comprehensive indicator that combines the information in the covariance matrix to quantify the health status of the equipment. By comparing the covariance matrices at different times, the nuclear equipment status information can reveal the long-term change trend of the equipment status and provide support for the maintenance and fault prevention of the target nuclear equipment.

[0224] Through steps S1001 to S1003 provided in the embodiment of the present application, an in-depth state assessment can be performed based on the state estimation vector and real-time monitoring data. This method can not only identify the current state of the equipment, but also predict the future performance of the equipment by analyzing the residuals and covariances, thereby providing strong support for the safe operation of the nuclear power plant. The application of this state assessment method enables the embodiment of the present application to more accurately capture the performance changes of the equipment and promptly discover potential failure risks.

[0225] According to some more specific embodiments of the present application, according to the state estimation vector x est and standardized real-time monitoring data The residual is calculated as:

[0226]

[0227] Among them, e obs is the observation time t obs The residual value of .

[0228] Further, according to the residual value e obs , construct the residual sequence E obs , expressed as:

[0229] E obs =[e obs-N+1 ,K,e obs-s+1 ,...,e obs ]

[0230] Among them, E obs is the observation time t obs The residual sequence, e obs-s+1 is the observation time t obs-s+1 The residual value of , N is the length of the residual sequence, s = 1,…,N.

[0231] Furthermore, the observation time t obs Determine the target observation time, for the residual sequence E obs The covariance of each residual value in is calculated to obtain the target observation time t obs The first covariance matrix C obs .

[0232] It should be noted that near the observation time t obs The covariance matrix of the continuous residual values ​​is calculated as the first covariance matrix C obs , expressed as;

[0233]

[0234] Similarly, we get the target observation time t obs The previous moment t obs-1 The second covariance matrix C obs-1 .

[0235] Furthermore, based on the first covariance matrix C obs and the second covariance matrix C obs-1 Calculate the health index and obtain the nuclear equipment status information h obs , expressed as:

[0236]

[0237] Among them, h obs is the target observation time t obs The health index of the core device, i.e. the status information of the core device. ||.||1 is the L1 norm.

[0238] Reference Fig.11 According to some embodiments of the present application, in step S106, in response to the nuclear equipment status information satisfying the preset fault judgment condition, before determining that the nuclear equipment status information is in a fault state, it may also include, but is not limited to:

[0239] Step S1101, obtaining the warning threshold coefficient and the past device status information corresponding to multiple observation times before the target observation time;

[0240] Step S1102, performing threshold calculation based on the warning threshold coefficient and each past device status information to obtain a fault determination threshold.

[0241] In some embodiments of the present application, before determining whether the nuclear equipment status information is in a fault state, the embodiments of the present application perform a series of preprocessing steps to ensure the accuracy and reliability of the fault determination. These steps include obtaining the early warning threshold coefficient and the past equipment status information corresponding to multiple observation times before the target observation time, and performing threshold calculation based on this information to obtain the fault determination threshold.

[0242] In step S1101 of some embodiments, the early warning threshold coefficient and past equipment status information are obtained. It should be noted that the early warning threshold coefficient is used to adjust the sensitivity of fault determination, which can be set according to the specific operating conditions of the equipment and historical fault data. Past equipment status information provides the operating status of the equipment before the target observation time, which is crucial for understanding the state changes of the target nuclear equipment and identifying potential failure modes.

[0243] In step S1102 of some embodiments, the embodiments of the present application perform threshold calculation based on the warning threshold coefficient and each past device status information to obtain a fault judgment threshold. This calculation process may involve statistical analysis and machine learning techniques to determine a reasonable threshold that can distinguish between the normal operating state and potential fault state of the equipment. The setting of the fault judgment threshold needs to take into account the operating characteristics and safety requirements of the target nuclear equipment to ensure that a warning can be issued in time before the target nuclear equipment fails.

[0244] The embodiment of the present application can compare the nuclear equipment status information with the fault judgment threshold in the subsequent steps to determine whether the equipment is in a fault state. If the nuclear equipment status information meets the preset fault judgment condition, that is, exceeds the fault judgment threshold, the embodiment of the present application will determine that the nuclear equipment status information is in a fault state and take corresponding measures. This method not only improves the accuracy of fault judgment, but also enhances the predictive ability and preventive maintenance ability of the embodiment of the present application by considering the historical operation data of the equipment.

[0245] Through step S1101 to step S1102 shown in the embodiment of the present application, by obtaining the warning threshold coefficient and past equipment status information, and calculating the fault judgment threshold accordingly, the nuclear equipment fault monitoring method of the embodiment of the present application can provide timely warning before the equipment fails, thereby providing strong support for the safe operation of the nuclear power plant. This data-driven fault judgment method enables the embodiment of the present application to more accurately capture the performance changes of the equipment, timely discover potential fault risks, and avoid major safety accidents such as accidental shutdowns and nuclear leaks.

[0246] According to some more specific embodiments, the warning threshold coefficient and the past device status information corresponding to multiple observation times before the target observation time are obtained, and the threshold is calculated according to the warning threshold coefficient and each past device status information to obtain the fault judgment threshold, which can be expressed as:

[0247]

[0248] Among them, λ obs is the observation time t obs a is the fault judgment threshold. a is the early warning threshold coefficient, which can be determined based on the on-site equipment operation experience.

[0249] Reference Fig.12 According to some embodiments of the present application, the historical monitoring matrix includes multiple historical monitoring vectors. After the threshold value is calculated according to the warning threshold coefficient and each past device status information in step S1102 to obtain the fault judgment threshold value, it may also include, but is not limited to:

[0250] Step S1201, in response to the nuclear equipment status information not exceeding the fault determination threshold, incorporating the real-time monitoring data into the historical monitoring matrix;

[0251] Step S1202, after incorporating the real-time monitoring data into the historical monitoring matrix, closeness calculation is performed on the historical monitoring matrix based on the real-time monitoring data to obtain the parameter closeness corresponding to each historical monitoring vector;

[0252] Step S1203: delete the historical monitoring vectors whose parameter closeness is less than a preset closeness threshold from the historical monitoring matrix to update the historical monitoring matrix.

[0253] In some embodiments of the present application, the maintenance and update of the historical monitoring matrix is ​​an important part of the nuclear equipment fault monitoring method. After completing the threshold calculation and obtaining the fault judgment threshold, the embodiment of the present application will decide how to process the real-time monitoring data and the historical monitoring matrix according to whether the nuclear equipment status information exceeds the fault judgment threshold.

[0254] In step S1201 of some embodiments, if the nuclear equipment status information does not exceed the fault determination threshold, the embodiments of the present application will incorporate the real-time monitoring data into the historical monitoring matrix. The purpose of this step is to continuously enrich the historical data set in order to better capture the long-term trends and patterns of equipment operation. The incorporation of real-time monitoring data enables the historical monitoring matrix to reflect the latest operating status of the equipment, providing the latest information for subsequent state estimation and fault prediction.

[0255] In step S1202 of some embodiments, after incorporating the real-time monitoring data into the historical monitoring matrix, the embodiments of the present application will perform a closeness calculation on the historical monitoring matrix based on the real-time monitoring data to obtain the parameter closeness corresponding to each historical monitoring vector. Parameter closeness is an indicator of the similarity between the historical monitoring vector and the real-time monitoring data, which helps to identify which historical data is most relevant to the current device status.

[0256] In step S1203 of some embodiments, the embodiments of the present application will delete the historical monitoring vectors whose parameter closeness is less than the preset closeness threshold from the historical monitoring matrix to update the historical monitoring matrix. The purpose of this step is to ensure that the data in the historical monitoring matrix are highly relevant to the current equipment status, thereby avoiding the interference of noise and irrelevant data in the historical data on the monitoring results. By deleting irrelevant data, the historical monitoring matrix can maintain its accuracy and effectiveness, providing more reliable data support for the status assessment and fault prediction of the target nuclear equipment.

[0257] Through steps S1201 to S1203 provided in the embodiment of the present application, not only can the historical monitoring matrix be updated in time to ensure that it reflects the latest operating status of the equipment, but also the quality of the historical monitoring matrix can be improved through closeness calculation and data screening, thereby providing more accurate and reliable monitoring results for the safe operation of the nuclear power plant. The application of this method enables the embodiment of the present application to more effectively identify and predict equipment failures and avoid major safety accidents such as accidental shutdowns and nuclear leaks.

[0258] In step S106 of some embodiments, when the nuclear equipment status information meets the preset fault judgment conditions, the embodiments of the present application can determine that the equipment is in a fault state and issue a warning in time. This state-based monitoring and warning mechanism can not only reduce unnecessary preventive maintenance and reduce maintenance costs, but also improve the availability of the unit. More importantly, it can timely discover potential fault risks and avoid major safety accidents such as accidental shutdowns and nuclear leaks.

[0259] According to some embodiments provided by the present application, step S106, in response to the nuclear device status information satisfying a preset fault judgment condition, determines that the nuclear device status information is in a fault state, which may include, but is not limited to:

[0260] In response to the nuclear device status information exceeding the fault determination threshold, it is determined that the nuclear device status information meets the fault determination condition, and the nuclear device status information is determined to be in a fault state.

[0261] It should be noted that the embodiment of the present application needs to compare different nuclear equipment status information with a preset fault judgment threshold to determine whether the target nuclear equipment meets the fault judgment condition. When the nuclear equipment status information exceeds the fault judgment threshold, the embodiment of the present application will determine that the nuclear equipment status information meets the fault judgment condition and determine the nuclear equipment status information as a fault state. It should be understood that the key to this process lies in the setting of the fault judgment threshold, which is calculated based on historical monitoring data, real-time monitoring data and possible warning threshold coefficients. The fault judgment threshold is a critical value used to distinguish between the normal operating state and the fault state of the equipment. When the actual status information of the nuclear equipment exceeds this threshold, it means that the equipment may be experiencing performance degradation or abnormal conditions, which may indicate potential failures or performance degradation. By comparing the nuclear equipment status information with the fault judgment threshold and determining that the equipment is in a fault state when the threshold is exceeded, the nuclear equipment fault monitoring method of the embodiment of the present application can provide an effective fault detection and response mechanism for nuclear power plants. This method can not only improve the accuracy of fault detection, but also enhance the prevention and response capabilities of nuclear power plants to equipment failures, thereby ensuring the safe and reliable operation of nuclear power plants.

[0262] In some more specific embodiments, after the nuclear equipment status information is determined to be in a fault state, the embodiments of the present application can take corresponding measures, such as sounding an alarm, notifying maintenance personnel, or automatically triggering a preventive maintenance procedure. Such a response mechanism is crucial to ensuring the safe operation of a nuclear power plant, because it can promptly detect and handle problems that may affect equipment safety, thereby avoiding the occurrence and expansion of accidents.

[0263] In some more specific embodiments, the embodiments of the present application can further analyze the nature and cause of the fault in order to take more accurate maintenance or adjustment measures. For example, historical monitoring data is reviewed, operating conditions before and after the fault are checked, and possible fault modes are identified. In this way, nuclear power plants can not only respond to current faults, but also learn from them, improve future monitoring and maintenance strategies, and improve the overall equipment health management level.

[0264] Reference Fig.13 According to the second aspect of the present application, the nuclear equipment fault monitoring device may include, but is not limited to:

[0265] The data acquisition module 1301 is used to acquire the historical monitoring matrix and real-time monitoring data of the target nuclear equipment;

[0266] The incremental processing module 1302 is used to perform incremental processing on the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix;

[0267] An estimation model building module 1303, used to build an integrated multivariate state estimation model based on a dynamic memory matrix;

[0268] A vector calculation module 1304 is used to input the real-time monitoring data into the integrated multivariate state estimation model to calculate the state estimation vector corresponding to the integrated multivariate state estimation model;

[0269] A state evaluation module 1305 is used to evaluate the state of nuclear equipment based on the state estimation vector and real-time monitoring data to obtain nuclear equipment state information;

[0270] The fault judgment module 1306 is used to determine that the nuclear equipment status information is in a fault state in response to the nuclear equipment status information satisfying a preset fault judgment condition.

[0271] It can be seen that the contents of the above-mentioned nuclear equipment fault monitoring method embodiment are all applicable to the embodiment of the present nuclear equipment fault monitoring device. The functions specifically implemented by the present nuclear equipment fault monitoring device embodiment are the same as those in the above-mentioned nuclear equipment fault monitoring method embodiment, and the beneficial effects achieved are also the same as those achieved by the above-mentioned nuclear equipment fault monitoring method embodiment.

[0272] Reference Fig.14 , Fig.14 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0273] The processor 1401 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0274] The memory 1402 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1402 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1402, and the processor 1401 calls and executes the nuclear equipment fault monitoring method of the embodiment of this application;

[0275] Input / output interface 1403, used to implement information input and output;

[0276] The communication interface 1404 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0277] A bus 1405 that transmits information between the various components of the device (e.g., the processor 1401, the memory 1402, the input / output interface 1403, and the communication interface 1404);

[0278] The processor 1401 , the memory 1402 , the input / output interface 1403 and the communication interface 1404 are connected to each other in communication within the device via a bus 1405 .

[0279] The embodiment of the present application further provides a computer program product, which includes a computer program. A processor of a computer device reads and executes the computer program, so that the computer device executes and implements the above-mentioned nuclear equipment fault monitoring method.

[0280] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "comprises" and "comprising" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0281] It should be understood that in the present disclosure, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, which may include, but is not limited to, any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0282] It should be understood that in the description of the embodiments of the present application, the meaning of multiple (or multiple items) is more than two, greater than, less than, exceed, etc. are understood to not include the number, and above, below, within, etc. are understood to include the number.

[0283] In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0284] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0285] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0286] If the integrated unit 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 disclosure is essentially or the part that contributes to the prior art or all or 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 and may include, but is not limited to, a number of instructions for a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0287] It should also be understood that the various implementations provided in the embodiments of the present application can be combined arbitrarily to achieve different technical effects.

[0288] The above is a specific description of the implementation methods of the present disclosure, but the present disclosure is not limited to the above implementation methods. Technical personnel familiar with the art can also make various equivalent modifications or substitutions without violating the spirit of the present disclosure. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present disclosure.

Claims

1. A method for monitoring nuclear equipment failure, characterized in that: include: Obtain historical monitoring matrix and real-time monitoring data of target nuclear equipment; Incrementally process the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix; Constructing an integrated multivariate state estimation model based on the dynamic memory matrix; Inputting the real-time monitoring data into the integrated multivariate state estimation model to calculate a state estimation vector corresponding to the integrated multivariate state estimation model; Based on the state estimation vector and the real-time monitoring data, nuclear equipment state evaluation is performed to obtain nuclear equipment state information; In response to the core device status information satisfying a preset fault judgment condition, it is determined that the core device status information is in a fault state.

2. The method according to claim 1, characterized in that The historical monitoring matrix includes a plurality of historical monitoring vectors, and the historical monitoring matrix is ​​incrementally processed according to the real-time monitoring data to obtain a dynamic memory matrix, including: Based on the real-time monitoring data, the closeness calculation of the historical monitoring matrix is ​​performed to obtain the parameter closeness corresponding to each of the historical monitoring vectors; Reordering the historical monitoring vectors in the historical monitoring matrix based on the parameter closeness to obtain a target monitoring matrix; Sampling is performed on the target monitoring matrix to obtain the dynamic memory matrix.

3. The method according to claim 2, characterized in that The closeness calculation of the historical monitoring matrix based on the real-time monitoring data to obtain the parameter closeness corresponding to each historical monitoring vector includes: The historical monitoring matrix and the real-time monitoring data are respectively standardized to obtain a standardized historical matrix corresponding to the historical monitoring matrix and a standardized real-time vector corresponding to the real-time monitoring data; Constructing a correlation monitoring matrix according to the real-time monitoring data and each of the historical monitoring vectors; The parameter closeness corresponding to each of the historical monitoring vectors is calculated based on the historical monitoring matrix, the real-time monitoring data and the correlation matrix.

4. The method according to claim 3, characterized in that The step of constructing a correlation monitoring matrix according to the real-time monitoring data and each of the historical monitoring vectors includes: Calculate the historical monitoring mean corresponding to each of the historical monitoring vectors; Determining each correlation matrix element based on the ratio of the real-time monitoring data to each of the historical monitoring averages; The relevance matrix elements are integrated into a diagonal matrix to construct the relevance monitoring matrix.

5. The method according to claim 1, characterized in that The method of constructing an integrated multivariate state estimation model based on the dynamic memory matrix includes: Partitioning the dynamic memory matrix to obtain a first number of cluster center parameters and a sub-dynamic memory matrix corresponding to each of the cluster center parameters; Determine the base model estimation parameter corresponding to each cluster center parameter according to the sub-dynamic memory matrix; The first number of base model estimation parameters and the cluster center parameters corresponding to each of the base model estimation parameters are integrated to obtain the integrated multivariate state estimation model.

6. The method according to claim 5, characterized in that The dynamic memory matrix includes a plurality of historical monitoring vectors obtained by sampling, each of the historical monitoring vectors is configured with a corresponding parameter closeness, and the partitioning process is performed on the dynamic memory matrix to obtain a first number of cluster center parameters and a sub-dynamic memory matrix corresponding to each of the cluster center parameters, including: Based on the parameter closeness of each of the historical monitoring vector configurations, configuring a reference value for the dynamic memory matrix; The reference degree value is used to perform neighbor propagation clustering on the dynamic memory matrix to partition the dynamic memory matrix, thereby obtaining a first number of cluster center parameters and the sub-dynamic memory matrices corresponding to each cluster center parameter.

7. The method according to claim 5, characterized in that The step of inputting the real-time monitoring data into the integrated multivariate state estimation model to calculate a state estimation vector corresponding to the integrated multivariate state estimation model includes: Calculate the initial weight corresponding to each cluster center parameter according to the real-time monitoring data and each cluster center parameter; Performing weighting processing on the initial weights corresponding to the cluster center parameters to update the initial weights and obtain the base model contribution corresponding to the base model estimation parameters; The state estimation vector is calculated based on the estimated parameters of each base model and the contribution of the base model corresponding to each base model estimated parameter.

8. The method according to claim 7, characterized in that The step of calculating the state estimation vector according to each of the base model estimation parameters and the base model contribution corresponding to each of the base model estimation parameters comprises: For each of the base model estimation parameters, the product of the base model estimation parameter and the corresponding base model contribution is determined as a contribution component element; Summing a first number of the contribution component elements to obtain a first state estimation element; Summing the contribution of the first number of base models to obtain a second state estimation element; The ratio of the first state estimation element to the second state estimation element is determined as the state estimation vector.

9. The method according to claim 1, characterized in that: The nuclear equipment state assessment based on the state estimation vector and the real-time monitoring data to obtain nuclear equipment state information includes: Performing residual calculation according to the state estimation vector and the real-time monitoring data to obtain a residual sequence; wherein the residual sequence includes residual values ​​corresponding to each observation time; Performing covariance calculation on each residual value in the residual sequence to obtain a first covariance matrix at a target observation moment and a second covariance matrix at a moment immediately before the target observation moment; A health index is calculated based on the first covariance matrix and the second covariance matrix to obtain the nuclear equipment status information.

10. The method according to claim 9, characterized in that Before the step of determining that the nuclear device status information is in a fault state in response to the nuclear device status information satisfying a preset fault judgment condition, the step further includes: Acquire a warning threshold coefficient and past device status information corresponding to a plurality of observation times before the target observation time; Perform threshold calculation based on the warning threshold coefficient and each of the past device status information to obtain a fault determination threshold; In response to the nuclear device status information satisfying a preset fault judgment condition, determining that the nuclear device status information is in a fault state includes: In response to the core device status information exceeding the fault judgment threshold, it is determined that the core device status information satisfies the fault judgment condition, and the core device status information is determined to be in the fault state.

11. The method according to claim 10, characterized in that The historical monitoring matrix includes a plurality of historical monitoring vectors, and after the threshold value is calculated according to the early warning threshold coefficient and each of the past device status information to obtain the fault determination threshold value, it also includes: In response to the nuclear equipment status information not exceeding the fault determination threshold, incorporating the real-time monitoring data into the historical monitoring matrix; After incorporating the real-time monitoring data into the historical monitoring matrix, performing a closeness calculation on the historical monitoring matrix based on the real-time monitoring data to obtain a parameter closeness corresponding to each of the historical monitoring vectors; The historical monitoring vectors whose parameter closeness is less than a preset closeness threshold are deleted from the historical monitoring matrix to update the historical monitoring matrix.

12. The method according to claim 1, characterized in that The target nuclear equipment is configured with sensor measuring points, and the step of obtaining the historical monitoring matrix and real-time monitoring data of the target nuclear equipment includes: Acquire monitoring data sets at multiple observation times; wherein each monitoring data set includes monitoring sensor data collected by each sensor measuring point at one observation time; Integrate the monitoring sensor data collected at each observation time into an operation data vector corresponding to the observation time; Determine the operation data vector corresponding to the target observation time as real-time monitoring data; The operation data vectors corresponding to the multiple observation times before the target observation time are integrated into the historical monitoring matrix.

13. A nuclear equipment fault monitoring device, characterized in that: include: A data acquisition module is used to obtain the historical monitoring matrix and real-time monitoring data of the target nuclear equipment; An incremental processing module, used for performing incremental processing on the historical monitoring matrix according to the real-time monitoring data to obtain a dynamic memory matrix; An estimation model building module, used for building an integrated multivariate state estimation model based on the dynamic memory matrix; A vector calculation module, used for inputting the real-time monitoring data into the integrated multivariate state estimation model to calculate a state estimation vector corresponding to the integrated multivariate state estimation model; A state assessment module, used to conduct a nuclear equipment state assessment based on the state estimation vector and the real-time monitoring data to obtain nuclear equipment state information; The fault judgment module is used to determine that the nuclear equipment status information is in a fault state in response to the nuclear equipment status information satisfying a preset fault judgment condition.

14. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and the processor implements the nuclear equipment fault monitoring method as described in any one of claims 1 to 12 when executing the computer program.

15. A computer-readable storage medium, characterized in that: The storage medium stores a program, and the program is executed by a processor to implement the nuclear equipment fault monitoring method as described in any one of claims 1 to 12.

Images