Knowledge graph-based nuclear power plant fault diagnosis and display method and system, electronic device, and storage medium

Abstract

A knowledge graph-based nuclear power plant fault diagnosis and display method and system, an electronic device, and a storage medium. The method comprises the following steps: monitoring, in real time, a key device to be monitored, and acquiring real-time information of said key device (S101); on the basis of the real-time information, determining whether there is an anomaly early-warning for said key device (S102); if there is an anomaly early-warning, identifying a fault data feature of said key device (S103); acquiring a related knowledge graph model in a knowledge base and calling the knowledge graph model (S104); performing complex knowledge decomposition and reasoning on the basis of the related knowledge graph model and the fault data feature to obtain a fault diagnosis result (S105); and displaying the fault diagnosis result (S106).

Description

Knowledge Graph-Based Fault Diagnosis and Display Methods, Systems, Electronic Equipment, and Storage Media for Nuclear Power Plants Technical Field

[0001] This invention relates to the field of nuclear power detection, and more specifically, to a method, system, electronic device, and storage medium for nuclear power plant fault diagnosis and display based on knowledge graphs. Background Technology

[0002] Failures in critical and sensitive nuclear power plant equipment can have a significant impact on the safety of the power plant system. Quickly and accurately locating the cause of a failure in its early stages is crucial for preventing its escalation and improving the safety of power plant operations. While the vast amounts of historical data, maintenance experience, and repair reports stored during power plant operation contain invaluable information, the lack of effective technical means prevents them from providing guidance for the maintenance of critical equipment from a data and knowledge perspective. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method, system, electronic device and storage medium for nuclear power plant fault diagnosis and display based on knowledge graph, which addresses the problems existing in the prior art.

[0004] The technical solution adopted by this invention to solve its technical problem is: to construct a method for fault diagnosis and display in nuclear power plants based on knowledge graphs, including the following steps:

[0005] Real-time monitoring of the key equipment to be monitored, and acquisition of real-time information of the key equipment to be monitored;

[0006] Based on the real-time information, determine whether the key equipment to be monitored has an abnormal warning;

[0007] If no abnormal warning occurs, continue monitoring of the key equipment to be monitored;

[0008] If an abnormal warning is issued, the fault data characteristics of the key equipment to be monitored will be identified.

[0009] Retrieve and invoke relevant knowledge graph models from the knowledge base;

[0010] Based on the relevant knowledge graph model and the fault data characteristics, complex knowledge decomposition and reasoning are performed to obtain fault diagnosis results;

[0011] The fault diagnosis results are then displayed.

[0012] In the knowledge graph-based nuclear power plant fault diagnosis and display method described in this invention, the method further includes:

[0013] A knowledge graph model is constructed based on nuclear power plant information; the nuclear power plant information includes: equipment information, fault information, and expert knowledge.

[0014] In the knowledge graph-based nuclear power plant fault diagnosis and display method of the present invention, the construction of a knowledge graph model based on nuclear power plant information includes:

[0015] Obtain the device information, the fault information, and the expert knowledge;

[0016] The key knowledge elements of the equipment fault are obtained by extracting the equipment information, the fault information, and the expert knowledge.

[0017] The knowledge graph model is constructed based on the key knowledge elements;

[0018] The knowledge graph model is stored in a knowledge base.

[0019] In the knowledge graph-based nuclear power plant fault diagnosis and display method of the present invention, the step of constructing the knowledge graph model based on the key knowledge elements includes:

[0020] Based on the aforementioned key knowledge elements, different types of key knowledge elements are obtained;

[0021] Identify the different types of key knowledge elements to obtain key knowledge elements with the same attributes;

[0022] The key knowledge elements with the same attributes are merged and complex knowledge is decomposed to obtain entity relationship triples.

[0023] Obtain the nodes and directed edges of the entity relationship triples;

[0024] The network topology is constructed based on the nodes and directed edges of the entity relationship triples.

[0025] Obtain the relational attributes of the network topology;

[0026] Construct a conditional probability table;

[0027] The knowledge graph model is constructed based on the network topology, the relational attributes, and the conditional probability table.

[0028] In the knowledge graph-based nuclear power plant fault diagnosis and display method of the present invention, the identification of fault data features of the key equipment to be monitored includes:

[0029] Obtain abnormal data of the key equipment to be monitored;

[0030] Retrieve instance information of key devices from the knowledge base;

[0031] Based on the abnormal data and the instance information of key equipment in the knowledge base, the fault data characteristics of the key equipment to be monitored are obtained by matching and identifying them.

[0032] In the knowledge graph-based nuclear power plant fault diagnosis and display method of the present invention, the step of performing complex knowledge decomposition and reasoning based on the relevant knowledge graph model and the fault data features to obtain fault diagnosis results includes:

[0033] Based on the characteristics of the fault data, the fault types of the key equipment to be monitored are obtained;

[0034] Based on the fault types of the key equipment to be monitored, determine the relevant entity relationship triples for each fault type;

[0035] Based on the entity relationship triples, network topology structures are formed for each fault type;

[0036] Determine the causal relationships between node variables in the network topology under each fault type;

[0037] Determine the strength of causal relationships between nodes in the network topology under each fault type;

[0038] The fault diagnosis result is obtained by analyzing and reasoning based on the causal relationship, the strength of the causal relationship, and the relevant knowledge graph model.

[0039] In the knowledge graph-based nuclear power plant fault diagnosis and display method of the present invention, determining the causal relationship between node variables in the network topology under each fault type includes:

[0040] Based on expert knowledge, the variable nodes and directed edges in the network topology under various types are added, removed, or adjusted to determine the causal relationships between node variables in the network topology under various fault types.

[0041] This invention also provides a knowledge graph-based nuclear power plant fault diagnosis and display system, comprising:

[0042] The monitoring unit is used to perform real-time monitoring of the key equipment to be monitored and to acquire real-time information of the key equipment to be monitored.

[0043] An anomaly detection unit is used to determine whether an anomaly warning has occurred for the key equipment to be monitored based on the real-time information; if no anomaly warning has occurred, the monitoring of the key equipment to be monitored continues.

[0044] The feature data identification unit is used to identify the fault data features of the key equipment to be monitored if an abnormal warning occurs.

[0045] The invocation unit is used to retrieve and invoke relevant knowledge graph models from the knowledge base.

[0046] The fault diagnosis unit is used to perform complex knowledge decomposition and reasoning based on the relevant knowledge graph model and the fault data features to obtain fault diagnosis results.

[0047] The diagnostic results display unit is used to display the fault diagnosis results.

[0048] The present invention also provides a storage medium storing a computer program adapted for loading by a processor to execute the steps of the knowledge graph-based nuclear power plant fault diagnosis and display method as described above.

[0049] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps of the knowledge graph-based nuclear power plant fault diagnosis and display method as described above by calling the computer program stored in the memory.

[0050] The knowledge graph-based nuclear power plant fault diagnosis and display method, system, electronic equipment, and storage medium of this invention have the following beneficial effects: The method includes the following steps: real-time monitoring of key equipment to be monitored and acquisition of real-time information of the key equipment; determination of whether an abnormal warning has occurred based on the real-time information; identification of fault data characteristics of the key equipment if an abnormal warning occurs; acquisition and invocation of relevant knowledge graph models from the knowledge base; complex knowledge decomposition and reasoning based on the relevant knowledge graph models and fault data characteristics to obtain fault diagnosis results; and display of the fault diagnosis results. This invention forms a complete and reliable fault diagnosis reasoning model for key equipment; based on this, and addressing the interpretability requirements of fault diagnosis, it utilizes knowledge graph methods to achieve visualized display of fault reasoning results, thereby satisfying the model's universality and practicality and providing fault handling suggestions for maintenance personnel. Attached Figure Description

[0051] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0052] Figure 1 is a flowchart illustrating an embodiment of the knowledge graph-based nuclear power plant fault diagnosis and display method provided by the present invention.

[0053] Figure 2 is a flowchart illustrating Embodiment 2 of the nuclear power plant fault diagnosis and display method based on knowledge graph provided by the present invention;

[0054] Figure 3 is a logic block diagram of the nuclear power plant fault diagnosis and display system based on knowledge graph provided by the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Rapidly and accurately locating the cause of a fault in its early stages is an effective means to prevent the fault from escalating and improve the safety of power plant operation. Given the vast amount of historical data and information stored during power plant operation, this invention extracts existing fault expert knowledge on critical and sensitive nuclear power equipment to form usable fault diagnosis knowledge. This knowledge is then structured and processed using knowledge graph technology to achieve knowledge mapping and structured modeling, providing a foundation for efficient knowledge utilization. Specifically, this invention provides a knowledge graph-based method for fault diagnosis and visualization in nuclear power plants, utilizing qualitative factor reasoning. This method, based on knowledge graphs, constructs, updates, and displays fault reasoning models, improving the adaptability of the fault reasoning model and the interpretability of diagnostic results.

[0057] Specifically, as shown in Figure 1, in this embodiment, the nuclear power plant fault diagnosis and display method based on knowledge graph includes the following steps:

[0058] Step S101: Perform real-time monitoring of the key equipment to be monitored and obtain real-time information of the key equipment to be monitored.

[0059] Furthermore, prior to performing step S101, the knowledge graph-based nuclear power plant fault diagnosis and visualization method also includes the following steps:

[0060] A knowledge graph model is constructed based on nuclear power plant information. This information includes equipment information, fault information, and expert knowledge.

[0061] The process of building a knowledge graph model based on nuclear power plant information includes: acquiring equipment information, fault information, and expert knowledge; extracting key knowledge elements of equipment faults from the equipment information, fault information, and expert knowledge; building a knowledge graph model based on the key knowledge elements; and storing the knowledge graph model in a knowledge base.

[0062] In this embodiment of the invention, constructing a knowledge graph model based on key knowledge elements includes: classifying key knowledge elements to obtain different types of key knowledge elements; identifying key knowledge elements of different types to obtain key knowledge elements with the same attributes; merging key knowledge elements with the same attributes and performing complex knowledge decomposition to obtain entity-relation triples; obtaining the nodes and directed edges of the entity-relation triples; constructing a network topology based on the nodes and directed edges of the entity-relation triples; obtaining the relation attributes of the network topology; constructing a conditional probability table; and constructing a knowledge graph model based on the network topology, relation attributes, and conditional probability table.

[0063] Step S102: Determine whether there is an abnormal warning for the key equipment to be monitored based on real-time information; if no abnormal warning is detected, continue monitoring the key equipment to be monitored.

[0064] Step S103: If an abnormal warning occurs, identify the fault data characteristics of the key equipment to be monitored.

[0065] In some embodiments, identifying fault data features of the key equipment to be monitored includes: acquiring abnormal data of the key equipment to be monitored; acquiring instance information of the key equipment in the knowledge base; and matching and identifying the abnormal data and the instance information of the key equipment in the knowledge base to obtain fault data features of the key equipment to be monitored.

[0066] Step S104: Obtain and invoke the relevant knowledge graph model from the knowledge base.

[0067] Step S105: Perform complex knowledge decomposition and reasoning based on relevant knowledge graph models and fault data characteristics to obtain fault diagnosis results.

[0068] In some embodiments, complex knowledge decomposition and reasoning based on relevant knowledge graph models and fault data characteristics are performed to obtain fault diagnosis results, including: classifying fault data characteristics to obtain fault types of key equipment to be monitored; determining relevant entity relationship triples in each fault type according to the fault types of key equipment to be monitored; forming network topology structures under each fault type based on entity relationship triples; determining causal relationships between node variables in the network topology structures under each fault type; determining the causal association strength between nodes in the network topology structures under each fault type; and performing analysis and reasoning based on causal relationships, causal relationship strength, and relevant knowledge graph models to obtain fault diagnosis results.

[0069] In this embodiment of the invention, determining the causal relationship between node variables in the network topology under each fault type includes: adding, removing, or adjusting the variable nodes and directed edges of the network topology under each type based on expert knowledge, thereby determining the causal relationship between node variables in the network topology under each fault type.

[0070] Step S106: Display the fault diagnosis results.

[0071] Specifically, as shown in Figure 2, a knowledge graph model is first constructed, and then fault diagnosis is performed based on the constructed knowledge graph model.

[0072] Specifically, as shown in Figure 2, the construction of a knowledge graph model includes the following steps:

[0073] The first step is to obtain information such as equipment information, fault information, and expert knowledge.

[0074] The second step involves extracting the acquired equipment information, fault information, and expert knowledge, classifying the extracted knowledge elements, and storing them in a structured format in a knowledge base. The key knowledge elements related to equipment faults are primarily derived from the nuclear power plant's iCCM basic database, FMEA data tables, and historical fault safety data extracted from the iCCM electronic history, including fault modes and their related entity and relationship information.

[0075] The extracted knowledge elements can be categorized into three types: text-based, semi-structured, and structured. Text-based elements include: historical data examples from iCCM electronic resumes; semi-structured elements include: FMEA data tables, extracted fault modes and their related entity and relationship information; and structured elements include: FTA fault trees.

[0076] The third step is to combine expert knowledge and sample data to construct a knowledge graph model that matches the knowledge base.

[0077] The specific steps are as follows:

[0078] (1) Obtain the nodes and directed edges of the entity relation triple.

[0079] (2) Construct the network topology based on the nodes and directed edges of the entity relationship triples. The network topology can be constructed from expert knowledge or learned from historical data.

[0080] (3) Determine the relational attributes of the network topology.

[0081] (4) Form a conditional probability table; where the network topology can be constructed by expert knowledge or learned based on historical data.

[0082] (5) Integrate the network topology, relational attributes and conditional probability tables to obtain a knowledge graph model.

[0083] The sample data includes real-time operational data, historical data, and fault sample data, storing a large amount of operational data during the operation of the nuclear power plant. Specifically, entity relation triples can be obtained by merging knowledge with the same attributes and decomposing complex knowledge. The same attribute refers to the same entity described by different expert knowledge constructed by personnel at different times and in different environments, such as the same equipment, structure, measurement point, or fault. Merging refers to integrating the sub-network structures within all expert knowledge bases into a unified knowledge structure, facilitating knowledge storage, modification, and retrieval, while avoiding the generation of contradictory knowledge and fault reasoning. The merging process involves traversing the entity nodes of the substructures. If the unified structure contains the same entity, the valid structure in the substructure is directly attached to the unified structure, and contradiction checks are performed on the structure. If the unified structure does not contain the same entity, the relevant structure is directly stored in the unified graph without contradiction checks. This unified structure is a Bayesian network topology of "node-edge-node".

[0084] It should be noted that the knowledge graph model in the knowledge base can be expanded and updated. Specifically, by converting between the knowledge base ontology and the network topology of the Bayesian network, the knowledge graph model can be rapidly updated when the knowledge in the knowledge base is expanded and updated. Similarly, when the network topology is reasonably added or deleted based on expert knowledge, the knowledge structure and its relationships in the knowledge base will also be reasonably adjusted. In addition, considering that ordinary Bayesian networks are difficult to express the strong coupling characteristics between parameters and cannot consider the impact of changes in the causal relationship strength between state variables such as fault causes and fault effects over time on fault diagnosis results, the idea of ​​polymorphic Bayesian networks is introduced based on the fixed network structure. As knowledge is continuously accumulated and updated, the state of network nodes can be refined from binary to polymorphic, dividing the fault development process over time into early, middle, and late stages, and further subdividing the state of network nodes. This allows the causal relationship between variables to be adjusted according to the changes in fault development over time, making the model more consistent with the actual situation in engineering applications, and the effect of the inference model will gradually improve, forming a virtuous cycle.

[0085] The specific steps for fault diagnosis are as follows:

[0086] The first step is to monitor the key equipment in real time and obtain real-time information.

[0087] The second step is to determine whether any abnormal warnings have been issued for critical equipment based on real-time information; if no abnormal warnings have been issued, monitoring continues.

[0088] The third step involves the inference engine model identifying fault data characteristics based on abnormal data from key equipment if an anomaly warning is detected. These fault data characteristics include, but are not limited to, time-series data, vibration and acoustic signature features.

[0089] The fourth step involves retrieving and calling relevant knowledge graph models from the matching knowledge base to decompose and reason about complex knowledge. The fused overall structure constitutes the complex knowledge. This complex knowledge decomposition is a necessary step in the fault diagnosis process. During fault diagnosis, the inference engine, based on actual operational data, repeatedly traverses the overall structure to remove entities and entity relationships unrelated to the alarm nodes, thereby reducing the overall knowledge and obtaining a diagnostic reasoning model based on the fault patterns related to the abnormal data (where fault patterns matching the knowledge structure of the alarm nodes are the relevant fault patterns). This model is then used for fault diagnosis reasoning, accelerating the fault diagnosis process, reducing computational load, improving reasoning efficiency, quickly identifying the cause of the fault, and providing maintenance recommendations. Since operational experience is associated with fault patterns, corresponding operational suggestions are automatically provided after the fault is diagnosed.

[0090] In particular, the qualitative factor reasoning method is used to infer the cause of the fault during fault diagnosis. The specific method is as follows:

[0091] Based on instance information of key equipment stored in the knowledge base, such as CRF (Circulating Water System) circulating pumps, RCV (Chemical and Volumetric Control System) charging pumps, and main generators, this information is divided into four parts: structure, function, attributes, and operating procedures. For example, taking the CRF circulating water pump as an example, the CRF... The product is divided into three levels: pump body module, motor module, gear reducer module, and auxiliary modules. The pump body module includes pressure-bearing component module, hydraulic component module, and pump shaft module. The motor module includes electromagnetic module and motor shaft module. The gear reducer module mainly includes planetary gear module and coupling module. The auxiliary modules mainly include shaft seal module and cooling / lubrication module. Each module stores function, attribute, and operating procedure information. By dividing the original entity structure according to fault types, and based on the related entity relationship triples in each fault type, a preliminary Bayesian network topology can be formed for each fault type. Then, by combining expert knowledge, the variable nodes and directed edges in each initial topology are manually added, removed, or adjusted to further determine the Bayesian network structure for each fault type and clarify the causal relationships between node variables.

[0092] After determining the Bayesian network topology under various failure modes, the quantitative characteristics of inter-node dependencies are further identified. Specifically, based on the feature attributes of each variable node instance already existing in the knowledge base, these are transformed into quantitative features between nodes. Then, expert knowledge is used to adjust the conditional probabilities between each node in the diagnostic inference model, ultimately determining the strength of factor associations between nodes in the Bayesian network. The threshold, prior probability, and conditional probability of nodes in the diagnostic inference model can be manually edited, enabling quantitative assignment of variable values.

[0093] In addition, the inference model provided by this invention also supports offline input of inference criteria, enabling functions such as fault data analysis and auxiliary source tracing.

[0094] It should be noted that the knowledge graph model referred to in this invention is the knowledge structure stored in the knowledge base, and the diagnostic reasoning model is a complete fault diagnosis algorithm. The two have a calling relationship, that is, the diagnostic reasoning model calls the knowledge graph model.

[0095] This invention first requires extracting existing expert knowledge on faults in critical and sensitive nuclear power equipment. It extracts fault-related knowledge of key equipment from highly fragmented data, analyzes various known fault modes, and summarizes the correlation between fault modes, fault effects, fault phenomena, and fault causes. The knowledge is then structured, and key pilot equipment is used as the research object to realize knowledge mapping and structured modeling, providing a foundation for the efficient use of knowledge.

[0096] Due to the complexity of equipment structures, the diverse failure modes, causes, and types, and the uncertainty in the correspondence between failure symptoms and causes, general diagnostic methods struggle to express the correlation between failure symptoms and causes. This invention, combined with practical needs, researches expert knowledge base construction and fault qualitative causal reasoning techniques. By combining expert knowledge with historical data to construct a reasoning model that matches the knowledge base, it achieves the transformation of knowledge into a network topology structure for the reasoning model, ultimately forming a complete and reliable fault diagnosis reasoning model for key equipment. Based on this, to address the interpretability requirements of fault diagnosis, it utilizes knowledge graph methods to visualize the fault reasoning results, satisfying the model's universality and practicality, and providing fault handling suggestions for maintenance personnel.

[0097] Figure 3 illustrates the nuclear power plant fault diagnosis and display system based on knowledge graphs provided by this invention. Specifically, as shown in Figure 3, the nuclear power plant fault diagnosis and display system based on knowledge graphs includes:

[0098] The monitoring unit 301 is used to monitor the key equipment to be monitored in real time and obtain real-time information of the key equipment to be monitored.

[0099] The anomaly detection unit 302 is used to determine whether an anomaly warning has occurred on the key equipment to be monitored based on real-time information; if no anomaly warning has occurred, the monitoring of the key equipment to be monitored will continue.

[0100] The feature data identification unit 303 is used to identify the fault data features of the key equipment to be monitored if an abnormal warning occurs.

[0101] Unit 304 is used to retrieve and invoke relevant knowledge graph models from the knowledge base.

[0102] The fault diagnosis unit 305 is used to perform complex knowledge decomposition and reasoning based on relevant knowledge graph models and fault data characteristics to obtain fault diagnosis results.

[0103] The diagnostic results display unit 306 is used to display the fault diagnosis results.

[0104] Specifically, the specific coordination and operation process between the various units in the knowledge graph-based nuclear power plant fault diagnosis and display system can be referred to the above-mentioned knowledge graph-based nuclear power plant fault diagnosis and display method, and will not be repeated here.

[0105] Furthermore, an electronic device of the present invention includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the knowledge graph-based nuclear power plant fault diagnosis and display method as described above. Specifically, according to embodiments of the present invention, the processes described above with reference to the flowchart can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, when the computer program is downloaded, installed, and executed by an electronic device, it performs the functions defined in the methods of the embodiments of the present invention. The electronic device in the present invention can be a terminal such as a laptop, desktop computer, tablet computer, or smartphone, or it can be a server.

[0106] Furthermore, one type of storage medium of the present invention stores a computer program thereon, which, when executed by a processor, implements the knowledge graph-based nuclear power plant fault diagnosis and display method described above. Specifically, it should be noted that the storage medium described above in the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. This propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0107] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0108] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0109] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0110] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0111] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They do not limit the scope of protection of the present invention. All equivalent changes and modifications made within the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. A knowledge graph-based nuclear power plant fault diagnosis and display method, characterized in that, The method comprises the following steps: real-time monitoring of a key device to be monitored and obtaining real-time information of the key device to be monitored; determining whether an abnormal early warning occurs in the key device to be monitored based on the real-time information; if no abnormal early warning occurs, continuing to monitor the key device to be monitored; if an abnormal early warning occurs, identifying a fault data feature of the key device to be monitored; obtaining a related knowledge graph model in a knowledge base and calling the related knowledge graph model; based on the related knowledge graph model and the fault data feature, carrying out complex knowledge decomposition and reasoning to obtain a fault diagnosis result; displaying the fault diagnosis result. 2.The knowledge graph based nuclear power plant fault diagnosis and display method according to claim 1, characterized in that, The method further comprises: constructing a knowledge graph model based on nuclear power plant information; the nuclear power plant information comprises device information, fault information and expert knowledge. 3.The knowledge graph based nuclear power plant fault diagnosis and display method according to claim 2, characterized in that, The method of constructing the knowledge graph model based on the nuclear power plant information comprises: obtaining the device information, the fault information and the expert knowledge; extracting the device information, the fault information and the expert knowledge to obtain key knowledge elements of device faults; constructing the knowledge graph model based on the key knowledge elements; storing the knowledge graph model in a knowledge base. 4.The knowledge graph-based nuclear power plant fault diagnosis and display method according to claim 3, characterized in that, The method of constructing the knowledge graph model based on the key knowledge elements comprises: classifying the key knowledge elements based on the key knowledge elements to obtain key knowledge elements of different categories; identifying the key knowledge elements of different categories to obtain key knowledge elements with the same attribute; merging and decomposing the key knowledge elements with the same attribute to obtain entity relationship triples; obtaining nodes and directed edges of the entity relationship triples; constructing a network topology structure based on the nodes and directed edges of the entity relationship triples; obtaining relationship attributes of the network topology structure; constructing a conditional probability table; constructing the knowledge graph model according to the network topology structure, the relationship attributes and the conditional probability table. 5.The knowledge graph based nuclear power plant fault diagnosis and display method according to claim 1, characterized in that, The method of identifying the fault data feature of the key device to be monitored comprises: obtaining abnormal data of the key device to be monitored; obtaining instance information of the key device in the knowledge base; matching and identifying based on the abnormal data and the instance information of the key device in the knowledge base to obtain the fault data feature of the key device to be monitored. 6.The knowledge graph based nuclear power plant fault diagnosis and display method according to claim 1, characterized in that, The method of carrying out complex knowledge decomposition and reasoning based on the related knowledge graph model and the fault data feature to obtain a fault diagnosis result comprises: dividing based on the fault data feature to obtain a fault type of the key device to be monitored; determining related entity relationship triples in each fault type according to the fault type of the key device to be monitored; forming a network topology structure under each fault type based on the entity relationship triples; determining a causal relationship between node variables in the network topology structure under each fault type; determining a causal association strength between nodes in the network topology structure under each fault type; based on the causal relationship, the causal relationship strength and the related knowledge graph model, analyzing and reasoning to obtain the fault diagnosis result. 7.The knowledge graph based nuclear power plant fault diagnosis and display method according to claim 6, characterized in that, The method of determining a causal relationship between node variables in the network topology structure under each fault type comprises: Based on expert knowledge, the variable nodes and the relationship directed edges in the network topology structure under each type are added, removed or adjusted to determine the causal relationship between the node variables in the network topology structure under each fault type. 8.A knowledge graph-based nuclear power plant fault diagnosis and display system, characterized in that, The method comprises the steps of: a monitoring unit for monitoring a key device in real time and obtaining real-time information of the key device; an abnormality judgment unit for judging whether the key device has an abnormal warning based on the real-time information; if there is no abnormal warning, the key device is continuously monitored; a feature data identification unit for identifying fault data features of the key device if there is an abnormal warning; a calling unit for obtaining a related knowledge graph model in a knowledge base and calling the model; a fault diagnosis unit for complex knowledge decomposition and reasoning based on the related knowledge graph model and the fault data features to obtain a fault diagnosis result; a diagnosis result display unit for displaying the fault diagnosis result.

9. A storage medium, characterized by The storage medium stores a computer program, and the computer program is suitable for being loaded by the processor to execute the steps of the nuclear power plant fault diagnosis and display method based on the knowledge graph according to any one of claims 1 to 7.

10. An electronic device, comprising: The device comprises a memory and a processor, and the memory stores a computer program, and the processor executes the steps of the nuclear power plant fault diagnosis and display method based on the knowledge graph according to any one of claims 1 to 7 by calling the computer program stored in the memory.

Images