CN115601013A

CN115601013A - Nuclear reactor failure determination method, device, apparatus, storage medium, and product

Info

Publication number: CN115601013A
Application number: CN202211309243.4A
Authority: CN
Inventors: 刘敏; 李文淮; 丁鹏; 胡硕文; 夏文卿; 于枫婉; 陈澍; 段承杰; 崔大伟; 林继铭
Original assignee: China General Nuclear Power Corp; China Nuclear Power Technology Research Institute Co Ltd; CGN Power Co Ltd
Current assignee: China General Nuclear Power Corp; China Nuclear Power Technology Research Institute Co Ltd; CGN Power Co Ltd
Priority date: 2022-10-25
Filing date: 2022-10-25
Publication date: 2023-01-13
Also published as: WO2024087404A1

Abstract

The present application relates to a nuclear reactor fault determination method, apparatus, device, storage medium, and product. The method comprises the following steps: firstly, acquiring running state data when a target event in a nuclear reactor fails; analyzing the running state data through a fault reasoning model to obtain the fault probability of a plurality of candidate events logically associated with the target event; and determining the fault judgment result of the nuclear reactor according to the fault probability of each candidate event. The fault inference model is constructed based on the operation state data of the historical fault events of the nuclear reactor and the conditional fault probability among the historical fault events. By adopting the method, the fault judgment result of the nuclear reactor can be accurately obtained.

Description

Nuclear reactor failure determination method, device, apparatus, storage medium, and product

Technical Field

The present application relates to the field of nuclear power systems, and in particular, to a nuclear reactor fault determining method, apparatus, device, storage medium, and product.

Background

With the development of nuclear power technology, parts, components, equipment and system software and hardware of a nuclear reactor system are more and more complex, and the corresponding hidden trouble of failure is increased. If a nuclear power plant system or equipment fails, critical components may be burned and a large amount of radioactivity is released, which poses a great threat to the life safety of people. Therefore, performing fault analysis on a nuclear reactor is very important for safe operation and maintenance of the whole nuclear power plant.

In the traditional method, a fault tree analysis method is used for carrying out fault analysis on a nuclear reactor, all related events in the nuclear reactor are represented through a logic gate, and then fault early warning or fault diagnosis is carried out on the nuclear reactor according to a system design structure, so that targeted maintenance is carried out.

However, the above method cannot accurately obtain the failure determination result of the nuclear reactor, and thus cannot effectively maintain the nuclear reactor system.

Disclosure of Invention

In view of the foregoing, it is desirable to provide a nuclear reactor failure determination method, apparatus, device, storage medium, and product that can accurately analyze a nuclear reactor failure.

In a first aspect, the present application provides a nuclear reactor fault determination method. The method comprises the following steps:

acquiring operating state data of a nuclear reactor when a target event fails;

analyzing the running state data through a fault reasoning model to obtain the fault probability of a plurality of candidate events logically associated with the target event; the fault reasoning model is constructed based on the operation state data of the historical fault events of the nuclear reactor and the conditional fault probability among the historical fault events;

and determining the fault judgment result of the nuclear reactor according to the fault probability of each candidate event.

In one embodiment, if the target event is a top-level event, the candidate events logically associated with the target event are bottom-level events of different types;

if the target event is a bottom-level event, a plurality of candidate events logically associated with the target event are a plurality of top-level events of different types;

the top layer event represents the event of the nuclear reactor body failure, and the bottom layer event represents the event of the replaceable component failure in the nuclear reactor.

In one embodiment, the target event is a top-level event, and the determining the fault determination result of the nuclear reactor according to the fault probability of each candidate event includes:

according to the fault probability of each bottom layer event, acquiring a target bottom layer event of which the fault probability meets a preset first fault probability condition;

determining a failure determination for the nuclear reactor that the target bottom event is an event that causes a failure of the top event.

In one embodiment, the method further comprises:

and outputting a maintenance instruction of the target bottom layer event, wherein the maintenance instruction is used for instructing a user to replace a part corresponding to the target bottom layer event to maintain the fault of the nuclear reactor.

In one embodiment, the target event is a bottom-level event, and determining a fault determination result of the nuclear reactor according to the fault probability of each candidate event includes:

acquiring a target top-level event with the fault probability meeting a preset second fault probability condition according to the fault probability of each top-level event;

determining that a failure determination of the nuclear reactor is that a failure of a bottom-level event will cause a failure of a targeted top-level event.

In one embodiment, the method further comprises:

constructing a reliability quantization function of the nuclear reactor according to the fault probability of the target top-level event;

determining the residual life of the nuclear reactor after the target moment according to the reliability quantization function and the nuclear reactor operating state data at the target moment; the target time is any time after the current time.

In one embodiment, the fault inference model is constructed by:

generating node information of each node and conditional fault probability of each associated node in a fault inference model through operation state data of historical fault events of the nuclear reactor; each node corresponds to at least one event, each node comprises nodes of three levels, namely a top level node, a middle node and a bottom level node, and the nodes of different levels correspond to different types of events;

and constructing a fault inference model according to the node information of each node and the conditional fault probability of each associated node in the fault inference model.

In one embodiment, generating the conditional failure probability for each associated node from operating state data for historical failure events of the nuclear reactor comprises:

acquiring the running time of historical fault events of all nodes in a nuclear reactor and the times of different types of faults of all nodes;

acquiring the failure times of each associated node in the nuclear reactor according to the times of different types of failures of each node;

and acquiring the conditional fault probability of each associated node according to the fault times of each associated node in the nuclear reactor.

In one embodiment, acquiring the number of times that different types of faults occur to each node includes:

acquiring event parameter intervals when different types of faults occur to each node;

and counting the frequency of different types of faults of each node according to the event parameter interval and the operating state data corresponding to each node when the nuclear reactor has a historical fault event.

In a second aspect, the present application further provides a nuclear reactor fault determination device. The device includes:

the data acquisition module is used for acquiring running state data when a target event in the nuclear reactor fails;

the probability acquisition module is used for analyzing the running state data through a fault reasoning model and acquiring the fault probability of a plurality of candidate events logically related to the target event; the fault reasoning model is constructed based on the operation state data of the historical fault events of the nuclear reactor and the conditional fault probability among the historical fault events;

and the result acquisition module is used for determining the fault judgment result of the nuclear reactor according to the fault probability of each candidate event.

In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the steps of the method in any of the embodiments of the first aspect when the computer program is executed by the processor.

In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method in any of the embodiments of the first aspect described above.

In a fifth aspect, the present application further provides a computer program product. The computer program product comprising a computer program that, when executed by a processor, performs the steps of the method in any of the embodiments of the first aspect described above.

According to the method, the device, the equipment, the storage medium and the product for judging the nuclear reactor faults, firstly, the running state data of a target event in the nuclear reactor when the target event fails is obtained; analyzing the running state data through a fault reasoning model to obtain the fault probability of a plurality of candidate events logically associated with the target event; and determining the fault judgment result of the nuclear reactor according to the fault probability of each candidate event. The fault inference model is constructed based on the operation state data of the historical fault events of the nuclear reactor and the conditional fault probability among the historical fault events. The method determines the fault judgment result of the nuclear reactor through a fault inference model, and the fault inference model is built based on the operation state data and the conditional fault probability of each event of the fault nuclear reactor. In the process of reasoning the nuclear reactor fault events, a plurality of operating states of each event of the nuclear reactor are considered, and the relationship among the plurality of states of each fault event is represented in a probability mode, so that the obtained fault judgment result has higher discrimination, the fault judgment result of the nuclear reactor can be accurately obtained, and the effective maintenance of a nuclear reactor system is facilitated.

Drawings

FIG. 1 is a diagram of an exemplary implementation of a method for determining a nuclear reactor fault;

FIG. 2 is a schematic flow diagram of a nuclear reactor fault determination method according to an embodiment;

FIG. 3 is a schematic flow chart illustrating the steps of a nuclear reactor fault diagnosis decision in one embodiment;

FIG. 4 is a schematic flow chart illustrating the steps of a nuclear reactor failure prediction decision in one embodiment;

FIG. 5 is a schematic flow chart illustrating the remaining life determining step of the nuclear reactor according to an embodiment;

FIG. 6 is a flowchart illustrating the steps of fault inference model construction in one embodiment;

FIG. 7 is a schematic diagram of a network of event relationships of a nuclear reactor in accordance with an embodiment;

FIG. 8 is a flowchart illustrating the conditional failure probability obtaining step in one embodiment;

FIG. 9 is a flowchart illustrating a failure number obtaining step according to an embodiment;

FIG. 10 is a block diagram showing a configuration of a nuclear reactor failure determining apparatus according to an embodiment;

FIG. 11 is a diagram illustrating an internal structure of a computer device in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.

The nuclear reactor fault determination method provided by the embodiment of the application can be applied to the application environment shown in fig. 1. The acquisition device 102 is configured to acquire operation state data of a nuclear reactor in a fault, and is in communication connection with the server 104 to transmit the operation state data to the server 104, and the terminal 106 is in communication with the server 104 through a network. The data storage system may store data that the server 104 needs to process. The data storage system may be integrated on the server 104, or may be located on the cloud or other network server. The data storage system stores the operating state data of the historical fault events of the nuclear reactor and the operating state data of the target events when faults occur, the server 104 constructs a fault inference model according to the operating state data of the historical fault events of the nuclear reactor, and then the fault probabilities of a plurality of candidate events logically associated with the target events are obtained by combining the operating state data of the target events when faults occur. The terminal 106 displays the failure probability of each candidate event. The terminal 106 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things devices, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.

A failure generally refers to a system or equipment being abnormal during operation and failing to meet expected performance requirements, and is characterized by a parameter of operational performance exceeding a specified limit, which may cause the equipment to be partially or completely disabled. During operation of a nuclear reactor, a fault sometimes occurs. Therefore, two important requirements always exist when the nuclear power plant unit operates: (1) Abnormal fault signs can be detected in the early stage of the accident process, namely fault early warning; (2) When a failure occurs, it is possible to identify the root cause of the occurrence of the abnormality (failure diagnosis).

How to effectively analyze the faults of nuclear reactors is an important research topic of personnel in the technical field of nuclear power.

Fault Tree Analysis (FTA) is a fundamental tool for performing reliability and safety Analysis on complex systems of nuclear power plants, and a Fault state which is most undesirable to occur in the system is used as a top-level target of Fault Analysis, then all factors causing the final Fault are searched according to a system design structure, and then all direct factors causing the next level of the Fault are found until some factors which do not need to be deeply researched fundamentally. The fault tree starts from top fault, reasoning and analyzing the fault forming reason according to certain logic key from the whole to the component one-layer and one-layer refinement, and finally determining the initial basic reason, the influence degree and the occurrence probability of the fault. Therefore, the fault tree analysis is beneficial to clear the fault mode of the nuclear power generating unit or the system, and is also beneficial to finding out weak links when predicting and diagnosing the faults of the system or the equipment, so that improvement is facilitated in design, or spare parts are added in maintenance, effective maintenance measures are taken, and the faults are prevented.

Specifically, the fault tree includes top-level events, intermediate events, and basic events, and an and gate, an or gate, a roll-in/roll-out, and the like. Starting from the top level events, the base events are connected by logical relationships for each intermediate event. Obviously, the process of searching for a basic event starting from a top event belongs to a fault diagnosis problem after a fault is found, and the process of solving the top event starting from the basic event belongs to the problem of predicting the possibility of occurrence of a top event. However, the fault tree analysis method is only suitable for a system with a clear fault logical relationship and a determined fault mechanism, and the reasons are summarized as follows:

(1) The status of an event in the fault tree is only two statuses (normal or fault), but during actual operation of the nuclear power plant, more statuses may exist (for example, for a crack of a device or a rotating component, the crack is propagated from a small crack to a large crack, and the failure degree of the crack is difficult to define by a binary division method), and the fault tree is difficult to describe a multi-status event.

(2) Logic gate descriptions (and, or, etc.) in the fault tree are all deterministic logic relations, that is, there is a definite causal relation between the previous-level event and the next-level event, but for a complex system of a nuclear power plant, there may be many possibilities that cause the fault to occur, and the occurrence is progressive in a form of a certain probability distribution, and at this time, the fault tree is difficult to characterize the fault diagnosis problem of uncertainty existing between the previous-level and the next-level.

In order to accurately obtain a determination result of a nuclear reactor fault, the present application provides a nuclear reactor fault determination method, where the nuclear reactor fault determination method in this embodiment is described by taking a terminal 106 applied to fig. 1 as an example, and as shown in fig. 2, the determination method includes the following steps:

s202, acquiring running state data when a target event in the nuclear reactor fails.

Among them, nuclear reactors are also called atomic energy reactors or reactors, which are devices capable of maintaining a controlled self-sustaining chain type nuclear fission reaction to realize nuclear energy utilization. The nuclear reactor can generate a self-sustaining chain type nuclear fission process in the nuclear reactor without adding a neutron source by reasonably arranging nuclear fuel.

Specifically, when a fault occurs in a known nuclear reactor, the operation state data is acquired, a target event is determined, a series of operation parameters, such as the fault type, the fault level and the fault time of the target event, are determined according to the fault occurring in the target event, and the series of operation parameters are used as the operation state data when the fault occurs in the target event.

S204, analyzing the running state data through a fault reasoning model to obtain fault probabilities of a plurality of candidate events logically associated with the target event; a fault inference model is constructed based on operating state data of historical fault events of a nuclear reactor and conditional fault probabilities between the historical fault events.

It should be understood that the failure probability in the embodiments of the present application is the probability of a failure characterizing an event of a nuclear reactor. For example, if event a is included in the nuclear reactor and failure a occurs in event a, P (a = a) represents the failure probability of failure a occurring in event a.

The conditional failure probability in the embodiments of the present application refers to the probability of failure of at least two associated events of a nuclear reactor. For example, if a nuclear reactor includes event a and event B, and event a and event B are associated, event a has a fault a, and event B has a fault B, then P (a = a | B = B) represents the conditional fault probability of event a having a fault a when event B has a fault B.

The fault reasoning model is a model for representing the relation of events of the nuclear reactor, and specifically comprises nodes and directional arrows connecting the nodes. The nodes in the fault inference model refer to events of the nuclear reactor, the position relation of the nodes refers to the correlation events of the nuclear reactor, and the directional arrows connecting the nodes refer to the logical relation of the correlation events of the nuclear reactor.

It should be noted that the directional arrows in the failure inference model are described as conditional failure probabilities, and the conditional failure probabilities are established based on historical failure events of the nuclear reactor, that is, the failure inference model links events in the nuclear reactor by probability values. Correspondingly, the output of the fault inference model is also characterized in the form of probability values. Alternatively, the historical failure event may be a real event based on the actual situation of the nuclear reactor, or a simulated event of failure of the nuclear reactor under laboratory conditions.

After acquiring the fault reasoning model, the fault probability needs to be acquired by combining the operation state data. Specifically, a plurality of candidate events associated with the target event are obtained according to the target event, and in the fault inference model, nodes corresponding to the target event and the candidate events associated with the target event and the conditional fault probabilities of the nodes are obtained, and then the fault probabilities of the candidate events associated with the target event are obtained by combining the conditional fault probabilities in the fault inference model.

And S206, determining the fault judgment result of the nuclear reactor according to the fault probability of each candidate event.

When a nuclear reactor fails, it is contemplated that the failure-causing element may comprise one or more failure elements, and that the failure may also cause one or more events to fail. That is, the failure probability of each candidate event may be a plurality of probabilities corresponding to a plurality of failure factors, or may be a plurality of probabilities corresponding to a plurality of failure predictions. Therefore, it is necessary to determine the failure determination result of the nuclear reactor based on the failure probability.

In the embodiment of the application, a fault probability range can be set according to a certain probability threshold value to determine a fault judgment result, and a specific fault probability value can also be directly determined to determine the fault judgment result. And the meaning of the corresponding fault probability expression is different when the candidate events are different.

Optionally, when the failure probability of the candidate event is a probability corresponding to multiple failure factors, the failure probability at this time indicates a degree of influence of the candidate event on the failure event. The larger the fault probability is, the larger the influence of the candidate event corresponding to the fault probability on the fault event is; the smaller the failure probability is, the smaller the influence of the candidate event corresponding to the failure probability on the failure event is.

Optionally, when the failure probability of the candidate event is a probability corresponding to the multiple failure predictions, the failure probability represents a probability that the candidate event occurs to the failure event. The higher the failure probability is, the higher the probability of occurrence of the candidate event corresponding to the failure probability is, and the lower the failure probability is, the lower the probability of occurrence of the candidate event corresponding to the failure probability is.

In the embodiment of the application, firstly, the operation state data of a nuclear reactor when a target event fails is obtained; analyzing the running state data through a fault reasoning model to obtain the fault probability of a plurality of candidate events logically associated with the target event; and determining the fault judgment result of the nuclear reactor according to the fault probability of each candidate event. The fault inference model is constructed based on the operation state data of the historical fault events of the nuclear reactor and the conditional fault probability among the historical fault events. The method determines the fault judgment result of the nuclear reactor through a fault inference model, and the fault inference model is built based on the operation state data and the conditional fault probability of each event of the fault nuclear reactor. In the process of reasoning the nuclear reactor fault events, a plurality of operating states of each event of the nuclear reactor are considered, and the relationship among the plurality of states of each fault event is represented in a probability mode, so that the obtained fault judgment result has higher discrimination, the fault judgment result of the nuclear reactor can be accurately obtained, and the effective maintenance of a nuclear reactor system is facilitated.

When constructing information of each node in the fault inference model, node determination is generally required to be performed in combination with various events of the nuclear reactor to obtain an effective fault inference model. Based on this, a specific judgment method of a nuclear reactor event will be described below by way of an example.

In one embodiment, if the target event is a top-level event, the candidate events logically associated with the target event are bottom-level events of different types; if the target event is a bottom-level event, the candidate events logically associated with the target event are top-level events of different types. The top layer event represents the event of the nuclear reactor body failure, and the bottom layer event represents the event of the replaceable component failure in the nuclear reactor.

The target event refers to a composite event generated by the failure and the combined action of a plurality of factors in the nuclear reactor, such as the events of core damage, the temperature of fuel, the release of radioactive factors or the failure of a safety system, and the like; the bottom layer events refer to basic events which are not required to be deeply studied in the nuclear reactor, such as the aging of devices, the damage of components due to sudden faults and the like.

When the target event is a top-level event, it indicates that the target event is a composite event generated by the failure and the joint action of multiple bottom-level events, and at this time, the multiple events are taken as multiple candidate events logically associated with the target event, and the candidate events are different types. For example, the failure of the top-level event T is caused by the failure a of the bottom-level event a and the failure B of the bottom-level event B, the target event is the top-level event T, and the candidate events are the bottom-level event a with the failure a and the bottom-level event B with the failure B.

When the target event is a bottom-level event, the target event is a basic event which does not need to be deeply studied in the nuclear reactor, and the target event can cause different types of top-level events to occur. For example, if the bottom-layer event a fails, it may cause the top-layer event T1 to fail T1, and also cause the top-layer event T2 to fail T2, then the target event is the bottom-layer event, and the candidate events are the top-layer event T1 that has failed T1 and the top-layer event T2 that has failed T2.

In the embodiment of the application, the target event and the candidate event are logically associated to construct the fault reasoning model, so that fault diagnosis from a top event to a bottom event can be realized, fault prediction from the bottom event to the top event can also be realized, and the comprehensiveness of the fault reasoning model is improved.

In the process of fault diagnosis, the fault diagnosis method can be directly obtained from operation data of a fault site, and can also be used for analyzing according to an internal mechanism generated by the fault so as to accurately judge the reason of the fault of the top-level event. Based on this, a specific determination procedure of the failure diagnosis will be described below by an embodiment.

In one embodiment, as shown in fig. 3, the target event is a top-level event, and the process of determining the failure determination result of the nuclear reactor according to the failure probability of each candidate event includes the following steps:

s302, according to the fault probability of each bottom layer event, a target bottom layer event with the fault probability meeting a preset first fault probability condition is obtained.

The first failure probability condition may be that a failure probability range is set according to a certain probability threshold to determine a target bottom layer event, or a specific failure probability value is directly determined to determine the target bottom layer event. For example, if the failure probability of the bottom layer event a is 50%, the failure probability of the bottom layer event B is 45%, and the failure probability of the bottom layer event C is 5%, and if the bottom layer event with the failure probability greater than 40% is set as the target bottom layer event, the target bottom layer event obtained at this time is the bottom layer event a and the bottom layer event B. If the bottom layer event with the failure probability of 50% is set as the target bottom layer event, the obtained target bottom layer event is the bottom layer event A.

S304, determining that the target bottom-layer event is the event causing the top-layer event to have a fault as the fault determination result of the nuclear reactor.

A fault of a nuclear reactor may determine one or more target underlying events. When the target bottom layer event is one event, the target bottom layer event is an event causing the top layer event to be in failure, and when the target bottom layer event is multiple, the target bottom layer events are combined to be used as a reason for causing the top layer event to be in failure.

In the embodiment of the application, the nuclear reactor fault is diagnosed according to the first fault probability condition, so that the influence degree of each bottom layer event on the top layer event fault can be further distinguished while each bottom layer event is obtained, and the nuclear reactor can be maintained in a targeted mode conveniently.

After fault diagnosis, the underlying events corresponding to the fault factors are generally repaired in conjunction with fault signals to quickly restore the normal state of the nuclear reactor. Based on this, the following describes a specific implementation process of the fault repair by an embodiment.

In one embodiment, the nuclear reactor fault determining method provided in the embodiments of the present application further includes: and outputting a maintenance instruction of the target bottom layer event, wherein the maintenance instruction is used for instructing a user to replace a component corresponding to the target bottom layer event to maintain the fault of the nuclear reactor.

The maintenance instruction of the target bottom layer event comprises information such as the position and the fault probability of the target bottom layer event, and the maintenance instruction can be displayed in a buzzer, an alarm device or a display screen. When the instruction of maintenance instruction is sent out, the maintenance personnel can quickly position the part corresponding to the target bottom layer event and execute the action of replacing the part.

In the embodiment of the application, the user is reminded to carry out fault maintenance through maintenance instruction, and the personnel do not need to patrol and examine to monitor each bottom layer event in real time under the line, and the target bottom layer event is searched for maintenance, so that the efficiency of nuclear reactor fault diagnosis is improved.

Corresponding to the process of fault diagnosis, the process of fault prediction can be directly obtained from the operation data of a fault site, and analysis can also be carried out according to an internal mechanism generated by the fault so as to judge the probability of the fault occurrence of the top-level event. Based on this, a specific determination procedure of the failure prediction will be described below by an embodiment.

In one embodiment, as shown in fig. 4, the target event is a bottom-layer event, and the determining the fault determination result of the nuclear reactor according to the fault probability of each candidate event includes the following steps:

s402, according to the fault probability of each top-level event, acquiring a target top-level event with the fault probability meeting a preset second fault probability condition.

And the second failure probability condition is to determine a target top-level event according to the maximum value of the failure probability of each top-level event. For example, the failure probability of the top-level event a is 45%, the failure probability of the top-level event B is 55%, and the target bottom-level event acquired at this time is the top-level event B. If the failure probability of both the top-level event a and the top-level event B is 50%, the target top-level events acquired at this time are the top-level event a and the top-level event B.

S404, determining that the failure judgment result of the nuclear reactor is that the failure of the bottom layer event can cause the failure of the target top layer event.

A fault of a nuclear reactor may determine one or more target top-level events. When the target top-level event is one event, the failure of the bottom-level event is indicated to cause the failure of one top-level event, and when the target top-level events are a plurality of events, the failure of the bottom-level event is indicated to cause the failure of a plurality of top-level events.

In the embodiment of the application, the nuclear reactor fault is predicted according to the second fault probability condition, and the most probable fault event can be obtained by comparing the fault probabilities of a plurality of top-layer events under the condition that the bottom-layer event has a fault, so that the nuclear reactor can be maintained conveniently.

After the fault prediction is carried out, the fault operation data of the fault inference model can be obtained by recording the operation state of the nuclear reactor, and a series of performance evaluations of the nuclear reactor can be completed by calculating the fault data. Based on this, a specific procedure for evaluating a failure of a nuclear reactor will be described below by way of an example.

In one embodiment, as shown in fig. 5, the nuclear reactor fault determining method provided in the embodiment of the present application further includes the following steps:

s502, according to the fault probability of the target top-level event, a reliability quantization function of the nuclear reactor is constructed.

Wherein the reliability quantification function is a function for calculating the remaining life of the nuclear reactor.

Under the condition that the target top-level event has the risk of failure and the nuclear reactor can still run safely, a reliability quantization function of the nuclear reactor can be constructed according to the failure probability of the target top-level event:

t (T) = R (a) × P (T = a) + K (T) formula 1

T (T) in equation 1 is a reliability quantification function, R (a) and P (T = a) represent the probability and reliability values for the top event T at fault a, and K (T) is the rated life time of the nuclear reactor at the top event T at fault a. Since the nuclear reactor is in a continuous operation state, the failure of the top level event also changes, and correspondingly, K (t) also changes along with the failure change of the top level event.

S504, determining the residual life of the nuclear reactor after the target time according to the reliability quantization function and the nuclear reactor operating state data at the target time; the target time is any time after the current time.

Given or assuming, in combination with the measurements at all times in the future, dynamically estimating all the reliability quantization functions at the future times, the remaining life of the nuclear reactor is estimated as:

t in equation 2 _k Indicating any time after the current time. Obviously, the remaining lifetime is obtained based on the integral operation of the reliability quantization function on the event, with t _k Will gradually decrease the remaining life. Furthermore, as the nuclear reactor continues to operate, the probability of failure and the rated life time change, and T (T) changes, affecting the ultimate remaining life of the nuclear reactor.

In the embodiment of the application, the reliability quantization function is constructed according to the fault probability, the residual life of the nuclear reactor is calculated through the integral of the reliability quantization function on time, and the residual life can be obtained through a simple calculation mode, so that frequent predictive maintenance is realized, and support is provided for the residual life of the system.

In diagnosing a nuclear reactor fault, the fault is typically diagnosed by constructing a fault tree or model to effectively characterize the nuclear reactor. Based on this, a specific construction process of the fault inference model is described below by an embodiment.

In one embodiment, as shown in fig. 6, the construction process of the fault inference model includes the following steps:

s602, generating node information of each node and conditional fault probability of each associated node in a fault inference model through operation state data of historical fault events of the nuclear reactor; each node corresponds to at least one event, each node comprises three levels of nodes including a top level node, a middle node and a bottom level node, and the nodes of different levels correspond to different types of events.

The node information of each node represents the position relation of each node, and is specifically determined based on the logic relation between the events of the nuclear reactor faults, and the conditional fault probability of the associated node is determined based on the fault probabilities of two or more nodes.

In the fault inference model, a top level node corresponds to a top level event of the nuclear reactor; the intermediate node corresponds to an intermediate event of the nuclear reactor; the bottom node corresponds to a bottom event of the nuclear reactor. Wherein an intermediate node refers to a node that is capable of associating a top level node with a bottom level node. For example, in the fault inference model, if there is a top-level event, a network structure diagram of a relationship between events of the nuclear reactor is shown in fig. 7, where X1, X2, X3, and X4 represent bottom-level nodes corresponding to the bottom-level event, M1, M2, and M3 represent intermediate nodes corresponding to the intermediate event, and T represents the top-level node corresponding to the top-level event. In addition, the directional arrows in fig. 7 represent the logical relationship between the nodes, and it can be seen that the middle node M2 is a node generated by the cooperation of the bottom node X2 and the bottom node X3, the middle node M3 is a node generated by the sole action of the bottom node X4, the middle node M1 is a node generated by the cooperation of the middle node M2 and the middle node M3, and the top node T is a node generated by the cooperation of the bottom node X1 and the middle node M1. It should be noted that, in the entire network structure diagram, the logical relationship represented by the directional arrow includes location information and probability information. The position information refers to the directivity of the associated event and is represented by the direction of an arrow; the probability information refers to the influence degree of the associated nodes, and specifically, the conditional fault probability is assigned to the directed arrows connected with the associated nodes.

S604, a fault inference model is constructed according to the node information of each node and the conditional fault probability of each associated node in the fault inference model.

After the node information of each node and the conditional fault probability of each associated node are obtained, a specific mode for constructing a fault inference model is as follows: firstly, a framework of a fault inference model is built according to node information of each node, the nodes are connected in a directional arrow mode, each associated node corresponds to a directional arrow, conditional fault probability is given to each associated node, and the fault inference model is built.

According to the method and the device, the fault reasoning model is built through the node information and the conditional fault probability to represent the relation of events of the nuclear reactor, the building idea of the model is clear, and the model has high interpretability.

When obtaining the conditional fault probability, a calculation is generally performed by counting the operational data of the associated events to obtain a more accurate fault inference probability model. Based on this, a specific acquisition process of the conditional failure probability is described below by an embodiment.

In one embodiment, as shown in FIG. 8, a process for generating a conditional failure probability for each associated node from operating state data for historical failure events of a nuclear reactor includes the steps of:

s802, acquiring the running time of historical fault events of all nodes in the nuclear reactor and the times of different types of faults of all nodes.

Since the conditional probabilities in the fault inference model are statistically derived based on discrete data, and the operating condition data of the nuclear reactor is continuous, it is necessary to obtain parameters related to historical fault events during the operating time of the historical fault events. Specifically, the type of the fault and the measurement time are determined, then the nodes are measured for a limited number of times within the measurement time, and each measurement is performed on the operation state data of each node at the same time, so that whether each node has the fault or not and the type of the fault can be recorded for each measurement, and then the number of times of different types of faults of each node can be counted by combining the type of the fault of each node on the basis of the limited number of measurements.

S804, acquiring the failure times of each associated node in the nuclear reactor according to the times of different types of failures of each node.

The associated node is a node having a logical relationship in the fault inference model, that is, the associated node includes at least two nodes, and correspondingly, the fault frequency of the associated node includes at least two fault frequencies.

Optionally, the associated node includes a node a and a node B, and when performing the limited-time measurement, the number of times that the node a fails a needs to be obtained, and when the node a fails a, the number of times that the node B fails and the type of the failure are obtained. If the node a includes multiple types of faults, the number of times and the type of the fault of the node B when the node a has different faults need to be obtained in sequence according to the different types of faults and the number of times of the node a having different faults.

Optionally, the associated node includes a node a, a node B, and a node C, and when performing the limited measurement, the number of times that the node a fails a needs to be obtained, and under the measurement condition that the node a fails a, the number of times that the node B fails and the type of the failure, and the number of times that the node B fails and the type of the failure are obtained. If the node A comprises multiple types of faults, the number of times of faults of the node B and the number of times of faults of the node C and the fault type of the node B under the condition that different faults occur to the node A need to be obtained in sequence according to the faults and the times of faults of different types of the node A.

And S806, acquiring the conditional fault probability of each associated node according to the fault frequency of each associated node in the nuclear reactor.

After the failure times of each associated node are obtained, the failure probability of each node is obtained first, and then the conditional failure probability is obtained by counting the failure times of each associated node.

The failure probability of each node can be expressed as:

P(X ₁ ＝X ₁₁ ) = (R/K)/T equation 3

T in equation 3 represents a given operating time, K represents the number of consecutive measurements, and R represents node X ₁ Occurrence of failure X ₁₁ Number of failures of type.

The conditional failure probability of two associated nodes can be expressed as:

P(M＝M ₁₁ |X ₁ ＝X ₁₁ ) = L/R formula 4

R in equation 4 represents node X ₁ Occurrence of failure X ₁₁ Number of failures of type, L denotes node X ₁ Occurrence of failure X ₁₁ When the type of the node M has a fault M ₁₁ Number of failures of type.

The conditional failure probability of a plurality of associated nodes may be expressed as:

P(M＝M ₁₁ |X ₁ ＝X ₁₁ ,X ₂ ＝X ₂₁ ) = Q/P equation 5

P in equation 5 is represented at node X ₁ Occurrence of failure X ₁₁ Type of failure, node X ₂ Occurrence of failure X ₂₁ Number of failures of type, Q, at node X ₁ Occurrence of failure X ₁₁ Type of failure, node X ₂ Occurrence of failure X ₂₁ Type of failure, node M fails ₁₁ The number of types of failures.

In the embodiment of the application, the conditional fault probability is obtained by counting the fault times of the associated events in the running time of the fault events, and the continuous measurement signals are converted into the discrete fault signals, so that the conditional fault probability can be conveniently updated subsequently while the validity of the conditional probability is ensured.

When obtaining the failure probability of the nuclear reactor, generally, the failure times of each node need to be obtained for statistics, so as to obtain an effective failure inference model. Based on this, a specific process for acquiring the number of different types of failures of each node is described below by an embodiment.

In one embodiment, as shown in fig. 9, the process of acquiring the number of times that different types of faults occur to each node includes the following steps:

and S902, acquiring event parameter intervals when different types of faults occur to each node.

During the operation of the reactor, a plurality of different types of nodes are included, and correspondingly, appropriate monitoring equipment is arranged aiming at each node so as to collect relevant operation state information, such as physical parameters of temperature, pressure, voltage, current and the like. For physical parameters such as temperature and pressure, event parameter intervals of different types of faults can be determined respectively.

Exemplarily, if the parameter interval of the event a is extended into five types, the event parameter interval can be divided into a normal interval, an alarm value interval, a low fault interval, a medium fault interval and a high fault interval by the physical parameter of the event a, and correspondingly, the event of the normal interval is a ₁ Type, event of alarm value interval is A ₂ Type, event of low fault interval is A ₃ Type, event in middle fault interval is A ₄ The event of type and high fault interval is A ₅ Type (b).

It should be noted that the probability of the failure of the bottom node follows a negative exponential distribution. The failure rate refers to the probability that a product which has not failed at a certain moment is failed in a unit time after the moment. Generally denoted as λ, which is also a function of time t, and hence also denoted as λ (t), is referred to as the failure rate function. For the lifetime of the aging element, the cumulative distribution of the probability of failure is:

F(x,λ)＝1-e ^-λx equation 6

Therefore, the probability of the normal state to be changed to the alarm state, the small fault, the medium fault or the large fault (λ value in the above formula) is different, and λ is set to be ₁ ,λ ₂ ,λ ₃ ,λ ₄ The corresponding cumulative failure probabilities are respectively F ₁ 、F ₂ 、F ₃ 、F ₄ Then the probability of the event normal state is (1-F) ₁ -F ₂ -F ₃ -F ₄ ) And the parameters are determined according to the difference of the components and the service environment for comprehensive evaluation.

And S904, counting the times of different types of faults of each node according to the event parameter interval and the operating state data corresponding to each node when the historical fault event occurs to the nuclear reactor.

It should be understood that each event parameter interval corresponds to a type of fault, and the parameters of the operating state data of the nodes correspond to the parameters of the parameter interval.

Specifically, limited state data is acquired through running state data corresponding to each node, the limited state data sequentially corresponds to event parameter intervals, and the times of different event parameter intervals corresponding to the data acquired by each node, namely the times of different types of faults of each node, are acquired.

In the embodiment of the application, different types of faults of each node are defined according to the event parameter interval, and the types of the nodes can be defined according to the difference of the nodes, so that the constructed fault inference model can maximally represent the relation of each event of the nuclear reactor, and the accuracy of the fault determination result of the nuclear reactor is enhanced.

In one embodiment, a nuclear reactor fault determination method is provided, the embodiment including:

(1) And acquiring operating state data when a target event in the nuclear reactor fails.

(2) Generating node information of each node in a fault inference model through the operation state data of the historical fault events of the nuclear reactor; each node corresponds to at least one event, each node comprises three levels of nodes including a top level node, a middle node and a bottom level node, and the nodes of different levels correspond to different types of events.

(3) The method comprises the steps of obtaining the running time of historical fault events of all nodes in the nuclear reactor and the times of different types of faults of all nodes.

(4) And acquiring event parameter intervals when different types of faults occur to each node.

(5) And counting the frequency of different types of faults of each node according to the event parameter interval and the running state data corresponding to each node when the nuclear reactor has a historical fault event.

(6) And acquiring the conditional fault probability of each associated node according to the fault frequency of each associated node in the nuclear reactor and the running time of the historical fault event of each node.

(7) And constructing a fault inference model according to the node information of each node and the conditional fault probability of each associated node in the fault inference model.

(8) And if the target event is a top-level event, the candidate events logically associated with the target event are bottom-level events of different types.

(9) And acquiring a target bottom layer event with the fault probability meeting a preset first fault probability condition according to the fault probability of each bottom layer event.

(10) And determining that the fault determination result of the nuclear reactor is that the target bottom layer event is the event causing the top layer event to have the fault.

(11) And outputting a maintenance instruction of the target bottom layer event, wherein the maintenance instruction is used for instructing a user to replace a component corresponding to the target bottom layer event to maintain the fault of the nuclear reactor.

(12) And if the target event is a bottom-level event, the candidate events logically associated with the target event are top-level events of different types.

(13) And acquiring a target top-level event with the fault probability meeting a preset second fault probability condition according to the fault probability of each top-level event.

(14) And determining that the failure determination result of the nuclear reactor is that the failure of the bottom layer event will cause the failure of the target top layer event.

(15) And constructing a reliability quantization function of the nuclear reactor according to the fault probability of the target top-level event.

(16) Determining the residual life of the nuclear reactor after the target moment according to the reliability quantization function and the nuclear reactor operating state data at the target moment; the target time is any time after the current time.

The nuclear reactor fault determination method provided by the embodiment of the application comprises the steps of firstly, acquiring running state data when a target event in a nuclear reactor fails; analyzing the running state data through a fault reasoning model to obtain the fault probability of a plurality of candidate events logically associated with the target event; and determining the fault judgment result of the nuclear reactor according to the fault probability of each candidate event. The fault inference model is constructed based on the operation state data of the historical fault events of the nuclear reactor and the conditional fault probability among the historical fault events. The method determines the fault judgment result of the nuclear reactor through a fault inference model, and the fault inference model is built based on the operation state data and the conditional fault probability of each event of the fault nuclear reactor. In the process of reasoning the nuclear reactor fault events, a plurality of operating states of each event of the nuclear reactor are considered, and the relationship among the plurality of states of each fault event is represented in a probability mode, so that the obtained fault judgment result has higher discrimination, the fault judgment result of the nuclear reactor can be accurately obtained, and the effective maintenance of a nuclear reactor system is facilitated.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the above embodiments may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.

Based on the same inventive concept, the embodiment of the present application further provides a nuclear reactor fault determination device for implementing the nuclear reactor fault determination method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so specific limitations in one or more embodiments of the nuclear reactor fault determination device provided below may refer to the limitations in the above nuclear reactor fault determination method, and details are not described here.

In one embodiment, as shown in fig. 10, there is provided a nuclear reactor fault determining apparatus 100 including: a data acquisition module 1002, a probability acquisition module 1004, and a result determination module 1006, wherein:

a data acquisition module 1002 is configured to acquire operating state data of a nuclear reactor at a failure of a target event.

A probability obtaining module 1004, configured to analyze the operation state data through a fault inference model, and obtain fault probabilities of multiple candidate events logically associated with the target event; a fault inference model is constructed based on operating state data of historical fault events of a nuclear reactor and conditional fault probabilities between the historical fault events.

An outcome determination module 1006 for determining a failure determination outcome for the nuclear reactor based on the failure probability of each candidate event.

In one embodiment, if the target event is a top-level event, the candidate events logically associated with the target event are bottom-level events of different types; if the target event is a bottom-level event, a plurality of candidate events logically associated with the target event are a plurality of top-level events of different types; the top layer event represents the event of the nuclear reactor body failure, and the bottom layer event represents the event of the replaceable component failure in the nuclear reactor.

In one embodiment, the result determination module 1006 includes:

the first acquisition unit is used for acquiring a target bottom layer event of which the fault probability meets a preset first fault probability condition according to the fault probability of each bottom layer event;

and the first determination unit is used for determining that the target bottom-layer event is the event causing the top-layer event to have a fault as the fault determination result of the nuclear reactor.

In one embodiment, the nuclear reactor fault determining apparatus 100 further includes:

and the maintenance instruction module is used for outputting a maintenance instruction of the target bottom layer event, and the maintenance instruction is used for instructing a user to replace a component corresponding to the target bottom layer event to maintain the fault of the nuclear reactor.

In one embodiment, the result determining module 1006 further comprises:

the second acquisition unit is used for acquiring a target top-level event of which the fault probability meets a preset second fault probability condition according to the fault probability of each top-level event;

and the second determining unit is used for determining that the failure determination result of the nuclear reactor is that the failure of the bottom layer event can cause the failure of the target top layer event.

In one embodiment, the nuclear reactor failure determination apparatus 100 further includes:

the function construction module is used for constructing a reliability quantization function of the nuclear reactor according to the fault probability of the target top-level event;

the life determining module is used for determining the residual life of the nuclear reactor after the target moment according to the reliability quantization function and the nuclear reactor operating state data at the target moment; the target time is any time after the current time.

the model generation module is used for generating node information of each node and conditional fault probability of each associated node in the fault inference model according to the running state data of the historical fault events of the nuclear reactor; each node corresponds to at least one event, each node comprises nodes of three levels, namely a top node, a middle node and a bottom node, and the nodes of different levels correspond to different types of events;

and the model construction module is used for constructing the fault inference model according to the node information of each node in the fault inference model and the conditional fault probability of each associated node.

In one embodiment, a model generation unit includes:

the third acquisition unit is used for acquiring the running time of historical fault events of all nodes in the nuclear reactor and the times of different types of faults of all nodes;

the fourth acquisition unit is used for acquiring the failure times of each associated node in the nuclear reactor according to the times of different types of failures of each node;

and the fifth acquisition unit is used for acquiring the conditional fault probability of each associated node according to the fault frequency of each associated node in the nuclear reactor.

In an embodiment, the fourth obtaining unit is further configured to obtain event parameter intervals when different types of faults occur to each node; and counting the frequency of different types of faults of each node according to the event parameter interval and the operating state data corresponding to each node when the nuclear reactor has a historical fault event.

The respective modules in the nuclear reactor failure determination apparatus may be wholly or partially implemented by software, hardware, or a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 11. The computer device includes a processor, a memory, an Input/Output interface (I/O for short), and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and the computer program to run on the non-volatile storage medium. The database of the computer device is used to store operating state data for historical time to failure of the nuclear reactor. The input/output interface of the computer device is used for exchanging information between the processor and an external device. The communication interface of the computer device is used for connecting and communicating with an external terminal through a network. The computer program is executed by a processor to implement a nuclear reactor fault determination method.

Those skilled in the art will appreciate that the architecture shown in fig. 11 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a computer device is provided, comprising a memory and a processor, the memory having a computer program stored therein, the processor implementing the following steps when executing the computer program:

acquiring operating state data of a nuclear reactor when a target event fails;

analyzing the running state data through a fault reasoning model to obtain fault probabilities of a plurality of candidate events logically related to the target event; the fault reasoning model is constructed based on the operation state data of the historical fault events of the nuclear reactor and the conditional fault probability among the historical fault events;

In one embodiment, the processor when executing the computer program further performs the steps of:

determining that the failure determination result of the nuclear reactor is that the target bottom event is an event causing a failure of the top event.

and outputting a maintenance instruction of the target bottom layer event, wherein the maintenance instruction is used for instructing a user to replace a component corresponding to the target bottom layer event to maintain the fault of the nuclear reactor.

In one embodiment, the processor, when executing the computer program, further performs the steps of:

generating node information of each node and conditional fault probability of each associated node in a fault inference model through operating state data of historical fault events of the nuclear reactor; each node corresponds to at least one event, each node comprises nodes of three levels, namely a top node, a middle node and a bottom node, and the nodes of different levels correspond to different types of events;

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

acquiring operating state data of a nuclear reactor when a target event fails;

generating node information of each node and conditional fault probability of each associated node in a fault inference model through operating state data of historical fault events of the nuclear reactor; each node corresponds to at least one event, each node comprises nodes of three levels, namely a top level node, a middle node and a bottom level node, and the nodes of different levels correspond to different types of events;

In one embodiment, a computer program product is provided, comprising a computer program which when executed by a processor performs the steps of:

acquiring operating state data of a nuclear reactor when a target event fails;

It should be noted that, the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) referred to in the present application are information and data authorized by the user or sufficiently authorized by each party, and the collection, use and processing of the related data need to comply with the relevant laws and regulations and standards of the relevant country and region.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, the computer program can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), magnetic Random Access Memory (MRAM), ferroelectric Random Access Memory (FRAM), phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases involved in the embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above examples only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, and these are all within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A nuclear reactor failure determination method, the method comprising:

acquiring operating state data of a nuclear reactor when a target event fails;

analyzing the running state data through a fault reasoning model to obtain fault probabilities of a plurality of candidate events logically related to the target event; the fault inference model is constructed based on operating state data of historical fault events of the nuclear reactor and conditional fault probabilities between each of the historical fault events;

determining a fault determination for the nuclear reactor based on the fault probability for each of the candidate events.

2. The method of claim 1, wherein if the target event is a top-level event, the candidate events logically associated with the target event are bottom-level events of different types;

if the target event is a bottom-level event, the candidate events logically associated with the target event are top-level events of different types;

wherein the top level event represents an event of a failure of the nuclear reactor body and the bottom level event represents an event of a failure of a replaceable component in the nuclear reactor.

3. The method of claim 2, wherein the target event is a top-level event, and wherein determining the fault determination for the nuclear reactor based on the fault probability for each of the candidate events comprises:

determining that the target bottom layer event is the event causing the top layer event to fail as a result of the failure determination of the nuclear reactor.

4. The method of claim 3, further comprising:

5. The method of claim 2, wherein the target event is an underlying event, and wherein determining the fault determination for the nuclear reactor based on the probability of failure for each of the candidate events comprises:

according to the fault probability of each top-level event, acquiring a target top-level event of which the fault probability meets a preset second fault probability condition;

determining that the failure determination of the nuclear reactor is that failure of the bottom layer event will cause failure of the target top layer event.

6. The method of claim 5, further comprising:

determining the remaining life of the nuclear reactor after the target time according to the reliability quantization function and the nuclear reactor operating state data of the target time; the target time is any time after the current time.

7. The method according to any one of claims 1-6, wherein the fault inference model is constructed by:

generating node information of each node and conditional fault probability of each associated node in the fault inference model through operating state data of historical fault events of the nuclear reactor; each node corresponds to at least one event, each node comprises nodes of three levels, namely a top level node, a middle node and a bottom level node, and the nodes of different levels correspond to different types of events;

and constructing the fault inference model according to the node information of each node in the fault inference model and the conditional fault probability of the associated node.

8. The method of claim 7, wherein generating a conditional failure probability for each of the associated nodes from operating state data for historical failure events of the nuclear reactor comprises:

acquiring the running time of historical fault events of all nodes in the nuclear reactor and the times of different types of faults of all the nodes;

9. The method of claim 8, wherein obtaining the number of times each of the nodes has a different type of failure comprises:

acquiring event parameter intervals when different types of faults occur to the nodes;

10. A nuclear reactor fault analysis apparatus, comprising:

the probability acquisition module is used for analyzing the running state data through a fault reasoning model and acquiring the fault probability of a plurality of candidate events logically related to the target event; the fault inference model is constructed based on operating state data of historical fault events of the nuclear reactor and conditional fault probabilities between each of the historical fault events;

and the result determination module is used for determining the fault judgment result of the nuclear reactor according to the fault probability of each candidate event.

11. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor realizes the steps of the method of any one of claims 1 to 9 when executing the computer program.

12. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 9.

13. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 9.