Research reactor fault detection method, device, system and storage medium

Abstract

The present application discloses a research reactor fault detection method, device, system and storage medium, which relates to the field of nuclear power technology, in order to at least solve the problems of fuzzy and inaccurate fault detection results, large troubleshooting workload and low fault handling efficiency. The method is applied to a research reactor system, including event nodes, state nodes and preset change trends of state nodes associated with event nodes caused by fault events corresponding to event nodes; including: in the case of detecting that a target state node is in a fault state, determining a first event node associated with the target state node; determining a first state change amount of a state variable of a first state node associated with the first event node; from the first event node, selecting a second event node whose change trend of the first state change amount associated with the event is the same as the target preset change trend corresponding to the first state node, so as to determine the fault event of the second event node as the target fault event.

Description

Technical Field

[0001] The present application relates to the field of nuclear power technology, and in particular to a research reactor fault detection method, device, system and storage medium. Background Art

[0002] Research reactors have numerous equipment and complex structures, and their safety requirements are extremely high. The monitoring and diagnosis of the operating status of research reactors are directly related to the safety and reliability of the reactor. If the research reactor failure cannot be detected and eliminated in time during operation, it may further deteriorate and may even cause more serious consequences. The current research reactor system directly determines whether a failure has occurred based on whether each state variable exceeds its threshold, and performs fault warning. However, an abnormal state variable in the research reactor system may be a different fault event. Therefore, the fault warning results obtained based on the current fault detection method are low in accuracy and very vague, requiring fault handling personnel to check multiple fault events one by one, which has a large workload and low fault handling efficiency. Summary of the invention

[0003] The present invention provides a research reactor fault detection method, device, system and storage medium to at least solve the problems of fuzzy and inaccurate fault detection results, large troubleshooting workload and low fault handling efficiency in related technologies. The technical solution of the present invention is as follows:

[0004] According to a first aspect of an embodiment of the present invention, a research reactor fault detection method is provided, which is applied to a research reactor system, and includes event nodes, state nodes, and preset change trends of state nodes associated with the event nodes caused by fault events corresponding to the event nodes; each event node is associated with at least one state node; the state nodes represent state variables generated under the operating state of the research reactor, and the event nodes represent fault events of the research reactor system; the method includes: when it is detected that the target state node is in a fault state, determining a first event node associated with the target state node; determining a first state change amount of the state variable of the first state node associated with the first event node; from the first event node, selecting a second event node whose change trend of the first state change amount associated with the event is the same as the target preset change trend corresponding to the first state node, so as to determine the fault event of the second event node as the target fault event.

[0005] As an implementation method, before determining the first event node associated with the target state node when detecting that the target state node is in a fault state, the method also includes: determining that the target state node is in a fault state when the current state variable of the target state node is greater than or equal to the first variable threshold corresponding to the target state node.

[0006] As an implementation method, the method also includes: when the current state variable is less than the first variable threshold, obtaining multiple state variable values ​​of the target state node within a preset time; performing linear fitting on the multiple state variable values ​​to obtain the current change trend of the current state variable; if it is determined that the current change trend is consistent with the preset change condition corresponding to the target state node, then determining that the target state node is in a fault state.

[0007] As an implementation method, the method also includes: when the current change trend is an upward trend and the rising speed is greater than or equal to the preset rising speed, determining that the target state node is in a fault state; or; when the current change trend is an upward trend and the rising speed is less than the preset rising speed, determining that the target state node is in a non-fault state; or; when the current change trend is a downward trend, determining that the target state node is in a non-fault state.

[0008] As an implementation method, the research reactor system includes a knowledge graph, which includes a tree branch composed of each event node and the corresponding state node. The tree branch includes a directed symbol representing a preset change trend of a state node associated with the event node due to a fault event corresponding to the event node, wherein the directed symbol includes a positive symbol representing an increase in the corresponding state variable caused by the fault event or a negative symbol representing a decrease in the corresponding state variable caused by the fault event.

[0009] As an implementation method, determining the first event node associated with the target state node includes: determining the first target tree branch connected to the target state node from the knowledge graph; and determining the event node connected to the first target tree branch as the first event node.

[0010] As an implementation method, from the first event node, a second event node is selected, in which the change trend of the first state change quantity associated with the event is the same as the target preset change trend corresponding to the first state node, including: determining the second target tree branch connected to the first event node from the knowledge graph; determining the state node connected to the second target tree branch as the first state node; based on the target preset change trend represented by the directed symbol of the first state node, determining whether the first state change quantity of the first state node is the same as the target preset change trend; determining the first event node in the first event node in which the first state nodes connected by the event node are the same as the target preset change trend as the second event node, wherein the fault event includes one or more of the following: the above-mentioned fault event includes one or more of the following: booster pump failure, main pump failure, main coolant heat transfer tube rupture, water loss event, main coolant leakage, normal secondary water flow loss, uncontrolled extraction of a single control rod during power operation, fuel assembly damage, and plant-wide power outage accident, and a fault event is associated with at least one state variable.

[0011] According to a second aspect of an embodiment of the present invention, a research reactor fault detection device is provided, which is applied to a research reactor system, and includes event nodes, state nodes, and preset change trends of state nodes associated with the event nodes caused by fault events corresponding to the event nodes; each event node is associated with at least one state node; the state node represents the state variables generated when the research reactor is in an operating state, and the event node represents the fault event of a fault in the research reactor system; the device includes: a first determination unit, which is used to determine a first event node associated with a target state node when detecting that the target state node is in a fault state; a second determination unit, which is used to determine a first state change amount of the state variable of the first state node associated with the first event node; and a fault diagnosis unit, which is used to select a second event node from the first event node, whose change trends of the first state change amounts associated with the event are the same as the target preset change trends corresponding to the first state node, so as to determine the fault event of the second event node as the target fault event.

[0012] According to a third aspect of an embodiment of the present invention, a research reactor system is provided, which includes event nodes, state nodes, and preset change trends of state nodes associated with the event nodes caused by fault events corresponding to the event nodes; each event node is associated with at least one state node; the state nodes represent state variables generated under the operating state of the research reactor, and the event nodes represent fault events of the research reactor system. The system is configured to execute the research reactor fault detection method as in the first aspect and any possible implementation method thereof.

[0013] According to a fourth aspect of an embodiment of the present invention, a computer device is provided, comprising: a processor and a memory for storing processor executable instructions; wherein the processor is configured to execute the executable instructions to implement a research reactor fault detection method as in the first aspect and any possible implementation thereof.

[0014] According to a fifth aspect of an embodiment of the present invention, a research reactor system is provided to implement the research reactor fault detection method as described in the first aspect and any possible implementation thereof.

[0015] According to a sixth aspect of an embodiment of the present invention, there is provided a computer-readable storage medium on which instructions are stored. When the instructions in the computer-readable storage medium are executed by a processor of a computer device, the computer device is enabled to execute the research reactor fault detection method as described in the first aspect and any possible implementation thereof.

[0016] According to a seventh aspect of an embodiment of the present application, a computer program product is provided. The computer program product includes computer instructions. When the computer instructions are executed on a computer device, the computer device executes the research reactor fault detection method of the first aspect and any possible implementation thereof.

[0017] The technical solution provided by the embodiments of the present invention brings at least the following beneficial effects: when detecting an abnormality of a target state variable, the present application first determines the first event node of a related candidate fault event that can cause the abnormality of the target state variable, and then further determines the first state change amount of the candidate state variable (i.e., the first state variable) affected by the candidate fault event associated with the first event node, and judges whether the change trend of the first state change amount of all candidate state variables associated with each candidate fault event is consistent with the target preset change trend that the candidate state variable should produce when affected by the corresponding candidate fault event. If they are consistent, it means that the candidate fault event is the target fault event that causes the abnormality of the target state variable. If they are inconsistent, it means that the candidate fault event does not cause the abnormality of the target state variable. Through the above-mentioned fault detection method, the inverse relationship between the state variable and the fault event is used to reversely infer the candidate fault events that can be caused, and then the forward relationship between the fault event and the state variable is used to forward infer the target fault event among the candidate fault events, so as to analyze and determine based on the change trends of multiple state variables and the influence relationship between each candidate fault event, so as to accurately determine the target fault event that causes the variable abnormality, and accurately determine the fault source based on the target fault event, so that the fault handling personnel only need to directly handle the fault corresponding to the target fault event, which simplifies the fault troubleshooting steps, reduces the troubleshooting workload, and greatly improves the fault handling efficiency.

[0018] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application, and do not constitute an improper limitation on the present application.

[0020] Figure 1 is a flow chart showing a method for detecting a research reactor fault according to an exemplary embodiment;

[0021] Figure 2 is a schematic diagram of a knowledge graph constructed by fault events and state variables according to an exemplary embodiment;

[0022] Figure 3 is a schematic diagram of a fault event reasoning process according to an exemplary embodiment;

[0023] Figure 4 is a schematic diagram showing a specific process of research reactor modeling and fault diagnosis according to an exemplary embodiment;

[0024] Figure 5 is a block diagram of a research reactor fault detection device according to an exemplary embodiment;

[0025] Figure 6 is a schematic diagram of a control device according to an exemplary embodiment. DETAILED DESCRIPTION

[0026] In order to enable ordinary persons in the art to better understand the technical solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0028] Before introducing in detail the research reactor fault detection method provided in the embodiment of the present application, a brief introduction to the application scenarios involved in the embodiment of the present application is first given.

[0029] Research reactors have a wide variety of equipment and complex structures, and their safety requirements are extremely high. The monitoring and diagnosis of the operating status of research reactors are directly related to the safety and reliability of the reactor. If the fault cannot be discovered and eliminated in time during operation, it may further deteriorate and may even cause more serious consequences. Although a certain research reactor currently has complete accident and event handling procedures and alarm handling procedures, there are many alarm signals, and each alarm signal may be caused by several or even multiple reasons. Once a fault occurs, it is difficult for the operator to determine the source of the alarm in a short time and take correct measures in the face of numerous alarm signals. In addition, alarm components are not necessarily all fault sources. If the alarm components are directly operated as fault sources, it may not only fail to eliminate the fault, but also expand the impact of the fault, resulting in unpredictable serious consequences. Therefore, it is of great significance to conduct fault diagnosis research on the research reactor to help operators provide operation support.

[0030] In response to the above problems, the present application provides a research reactor fault detection method. When a target state variable is detected to be abnormal, the first event node of a related candidate fault event that can cause the target state variable to be abnormal is first determined, and then the first state change amount of the candidate state variable (i.e., the first state variable) affected by the candidate fault event associated with the first event node is further determined, and it is judged whether the change trend of the first state change amount of all candidate state variables associated with each candidate fault event is consistent with the target preset change trend that the candidate state variable should produce when affected by the corresponding candidate fault event. If they are consistent, it means that the candidate fault event is the target fault event that causes the target state variable to be abnormal. If they are inconsistent, it means that the candidate fault event does not cause the target state variable to be abnormal. Through the above-mentioned fault detection method, the inverse relationship between the state variable and the fault event is used to reversely infer the candidate fault events that can be caused, and then the forward relationship between the fault event and the state variable is used to forward infer the target fault event among the candidate fault events. The target fault event that causes the variable abnormality can be accurately determined, and the fault source can be accurately determined based on the target fault event, so that the fault handling personnel only need to directly handle the fault corresponding to the target fault event, which simplifies the fault troubleshooting steps, reduces the troubleshooting workload, and greatly improves the fault handling efficiency.

[0031] The research reactor fault detection method provided in the embodiment of the present application can be applied to a research reactor system or a control device (such as a server or a controller) having a research reactor fault detection function.

[0032] For ease of understanding, the research reactor fault detection method provided in the present application is specifically introduced below with reference to the accompanying drawings.

[0033] Figure 1 is a flow chart of a research reactor fault detection method according to an exemplary embodiment. Figure 1 As shown, the research reactor fault detection method includes the following steps.

[0034] S11, when detecting that the target state node is in a fault state, determining a first event node associated with the target state node.

[0035] When an abnormality of the target state variable is detected, a first event node of a related candidate fault event that can cause the abnormality of the target state variable is determined.

[0036] S12, determining a first state change amount of a state variable of a first state node associated with the first event node.

[0037] Determine the first state change of the candidate state variable (ie, the first state variable) affected by the candidate fault event associated with the first event node. The first state variable includes the state variable directly associated with the first event node (eg,Figure 2 medium pressurized flow rate) and indirectly related state variables (such as, Figure 2 medium degasser liquid level and make-up water cycle).

[0038] S13. From the first event nodes, select second event nodes where the change trends of the first state change amounts associated with the events are all the same as the target preset change trends corresponding to the first state nodes, so as to determine the fault event of the second event node as the target fault event.

[0039] Judge whether the change trends of the first state change amounts of all candidate state variables associated with each candidate fault event are consistent with the target preset change trends that should be generated when the corresponding candidate state variables are affected by the candidate fault event. If they are consistent, it indicates that the candidate fault event is the target fault event that causes the abnormality of the target state variable. If they are not consistent, it indicates that the candidate fault event is not the one that causes the abnormality of the target state variable.

[0040] Through the above implementation manners, the target fault event that causes the variable abnormality can be accurately determined, so as to accurately determine the fault source based on the target fault event, enabling the fault handling personnel to directly perform fault handling on the fault corresponding to the target fault event only, streamlining the fault troubleshooting steps, reducing the troubleshooting workload, and greatly improving the fault handling efficiency.

[0041] As a fault determination method, when the current state variable of the target state node is greater than or equal to the first variable threshold corresponding to the target state node, it is determined that the target state node is in a fault state.

[0042] The first variable threshold is used to indicate the critical value from non-fault to fault.

[0043] As another fault determination method, when the current state variable is less than the first variable threshold, obtain multiple state variable values of the target state node within a preset time; perform linear fitting on the multiple state variable values to obtain the current change trend of the current state variable; if it is determined that the current change trend is consistent with the preset change condition corresponding to the target state node, it is determined that the target state node is in a fault state.

[0044] Correspondingly, if it is determined that the current change trend is inconsistent with the preset change condition corresponding to the target state node, it is determined that the target state node is in a non-fault state.

[0045] Furthermore, in order to improve the fault efficiency, a second variable threshold is set, and the second variable threshold is smaller than the first variable threshold. When the current state variable is greater than or equal to the second variable threshold, it means that the current state variable is likely to increase to the first variable threshold within the preset time. Then, linear fitting is performed on multiple state variable values ​​adjacent to the current state variable time to obtain the current change trend of the current state variable; and it is judged whether the current change trend is consistent with the preset change condition corresponding to the target state node. If they are consistent, it is determined that the target state node is in a faulty state.

[0046] The second variable threshold is used to indicate the monitored state variable value that will reach the first variable threshold within a preset time range according to the maximum increase rate of the monitored state variable increase rate or the state variable increase rate that is greater than the preset variable increase rate.

[0047] As a trend judgment method, the following method is used to judge whether the change trend of the current state variable within the preset time is consistent with the preset change condition: when the current change trend is an upward trend and the rising speed is greater than or equal to the preset rising speed, it is determined that the target state node is in a fault state; or; when the current change trend is an upward trend and the rising speed is less than the preset rising speed, it is determined that the target state node is in a non-fault state; or; when the current change trend is a downward trend, it is determined that the target state node is in a non-fault state.

[0048] As an implementation mode, the research reactor system includes a knowledge graph, the knowledge graph includes a tree branch consisting of each event node and a corresponding state node, the tree branch includes a directed symbol representing a preset change trend of a state node associated with the event node due to a fault event corresponding to the event node, wherein the directed symbol includes a positive symbol representing an increase in a corresponding state variable caused by the fault event or a negative symbol representing a decrease in the corresponding state variable caused by the fault event.

[0049] As an implementation method, determining the first event node associated with the target state node specifically includes: determining the first target tree branch connected to the target state node from the knowledge graph; and determining the event node connected to the first target tree branch as the first event node.

[0050] As another implementation, from the first event node, a second event node is selected whose change trend of the first state change amount associated with the event is the same as the target preset change trend corresponding to the first state node, specifically including: determining the second target tree branch connected to the first event node from the knowledge graph; determining the state node connected to the second target tree branch as the first state node; determining whether the first state change amount of the first state node is the same as the target preset change trend based on the target preset change trend represented by the directed symbol of the first state node; and determining the first event node in the first event node whose first state nodes connected to the event node are the same as the target preset change trend as the second event node.

[0051] In some embodiments, Figure 2 In the knowledge graph shown, fault events and state variables are composed of solid and dotted lines with directed arrows. The solid line indicates that the fault event will cause the state variable associated with the solid line to increase, or the increase of the state variable associated with the solid line will cause the fault event associated with the solid line; the dotted line indicates that the fault event will cause the state variable associated with the dotted line to decrease, or the decrease of the state variable associated with the dotted line will cause the fault event associated with the dotted line. If the arrow points from the fault event to the state variable, it means that the fault event causes the state variable to change; if the arrow points from the state variable to the fault event, it means that the change of the state variable causes the fault event.

[0052] The above failure events include one or more of the following: booster pump failure, main pump failure, main cooler heat transfer tube rupture, loss of coolant event (i.e., LOCA), main coolant leakage, loss of normal secondary water flow, uncontrolled withdrawal of a single control rod during power operation, fuel assembly damage, and plant power outage (i.e., SBO).

[0053] The booster pump failure is directly associated with the following state variables: booster pump drive end vibration, booster pump non-drive end bearing temperature, booster pump motor three-phase coil temperature, booster pump current, booster pump drive end bearing temperature and booster flow rate; and is indirectly associated with the following state variables: degasser liquid level and water replenishment cycle.

[0054] The main pump fault is associated with the following state variables: main pump motor stator winding temperature, main pump bearing temperature, main pump thrust end casing vibration, main pump thrust end shaft displacement, main system flow, stack inlet pressure and degassing flow and container pressure, stack outlet temperature, and thermal power.

[0055] The rupture of the main cooler heat transfer tube is associated with the following state variables: secondary water dosage, pressurization flow, deaerator liquid level, water replenishment cycle, stack inlet pressure, container pressure, stack outlet pressure and container level.

[0056] The water loss event is associated with the following state variables: closed building pressure, pressure vessel liquid level, pit level in the small chamber under the pile, pit level in the main process room, liquid level in the small liquid lift, γ dose in the main process room, pile inlet pressure, degassing flow, container make-up vessel pressure, container make-up vessel level, pressurization flow, degassing vessel level, and water replenishment cycle.

[0057] The main coolant leakage is associated with the following state variables: main process room pit level, pile chamber pit level, large liquid lift level, main process room gamma dose, pressurized flow, degasser level, and water replenishment cycle.

[0058] The loss of normal secondary water flow is associated with the following state variables: the main heat secondary water inlet and outlet pressure difference, the total secondary water flow, the heat exchanger water inlet main pipe pressure, the stack outlet temperature, and the thermal power.

[0059] The uncontrolled extraction of a single control rod in power operation is associated with the following state variables: nitrogen-16 (N16) activity, reactor outlet temperature, thermal power, and automatic rod swing.

[0060] Fuel assembly damage is associated with the following state variables: total γ activity concentration after failure, fuel assembly outlet temperature, stack outlet temperature, thermal power, chimney gas effluent, automatic rod swing, delayed neutron number, and main process room γ dose.

[0061] The plant blackout accident (ie, SBO) is associated with the following state variables: main pump stop, stack inlet pressure, degassing flow, degasser liquid level, water replenishment cycle, booster pump stop, UHA / UHB power loss, UEA / UEB power loss, and diesel engine failure to start automatically.

[0062] In some embodiments, in order to improve the accuracy and resolution of fault diagnosis, the SDG model is used to perform fault diagnosis on fault events caused by abnormal state variables in the knowledge graph. Specifically, a hybrid reasoning method combining reverse reasoning and forward reasoning can be used. The so-called reverse reasoning is to reversely explore all compatible paths from the current alarm node to all possible cause nodes in the known SDG transient samples, and select candidate fault sources. The process of reverse reasoning is consistent with fault diagnosis, so it is also called a fault diagnosis mode; forward reasoning is to start from the selected candidate fault source and forwardly explore all compatible paths in combination with the observed nodes. In the forward reasoning process, if it is found that it is inconsistent with the known facts, the candidate fault source is a false solution and should be discarded. Similarly, if it is consistent with the observed facts, the candidate fault source is considered to be credible and can be used as one of the possible fault sources. The process of forward reasoning is consistent with safety evaluation, so it is also called a safety evaluation mode.

[0063] For example, Figure 3As shown in the figure, the specific fault event reasoning process assumes that the instantaneous state variables are: A(+), B(-), C(0). According to reverse reasoning, two compatible paths can be obtained: B→A→R1, B→A→R2, and R1 and R2 are candidate fault sources. Then the forward reasoning verification is performed: if R1 is the fault source, then node C should be too large or too small, but the state of node C in the instantaneous sample obtained is "0", that is, it is within the normal value range, which is inconsistent with the observation situation, so the candidate fault source R1 is a false solution and should be discarded. Similarly, the forward reasoning verification of R2 is consistent with the actual observation value when R2 is the fault source, indicating that R2 is credible as the fault source.

[0064] As an implementation method, the specific steps of fault diagnosis are as follows.

[0065] 1) According to the research reactor accident event handling procedures and relevant process system knowledge, the SDG model under the research reactor failure condition is established.

[0066] 2) The system parameters of the research reactor are monitored online. First, the confirmation threshold is used for judgment. If the confirmation threshold reaches the alarm, the state of the parameter is obtained. If the parameter does not reach the confirmation threshold, the sensitive threshold and QTA method are used for state monitoring to obtain trend fragments.

[0067] 3) The parameters that exceed the sensitive threshold but do not exceed the confirmation threshold are placed in a sliding window. When the length of the sliding window data reaches 5 seconds, data fitting is used to obtain the parameter trend.

[0068] 4) According to the judged parameter status, SDG reverse reasoning is performed to perform fault diagnosis.

[0069] 5) Based on the originating fault obtained in the previous step, forward reasoning is used to verify and eliminate false solutions.

[0070] 6) When a fault is diagnosed, the fault result and related treatment measures are given.

[0071] The research reactor is a pressurized water pool-shell reactor, with light water as coolant and moderator, and beryllium as reflector. The reactor adopts digital, integrated, intelligent instrumentation and control, power distribution, radiation protection and other systems. When the reactor is in operation, the reactor coolant system continuously removes the heat generated by the core and transfers it to the secondary water system through the main heat exchanger, and then the secondary water system transfers the heat to the final heat sink - river water. The systems included in the reactor mainly include the main coolant system, emergency residual heat removal system, emergency core cooling, secondary water system, purification system, air compression system, high-pressure water system, reactor water pool cooling and purification system, special drainage system, drain exhaust system, production water system ventilation system, power distribution system, etc.

[0072] The research reactor can have 7 operating conditions, but most of the time it is operated at rated power. This application takes the operation at rated power as an example to establish a knowledge graph under the SDG model to indicate the model of a certain stage of the system, usually referring to the initial response stage, the intermediate response stage and the final response stage. For the research reactor system, its initial response stage model should be established because there are a large number of protection systems in the reactor. Even after a system failure or even a serious accident, some parameters of the system can be restored to the normal range after a period of time under the action of the protection system. This will cause some compatible paths to be disconnected, which is very unfavorable to the diagnosis process. In addition, fault diagnosis requires the source of the fault to be determined in the shortest time so that effective measures can be taken quickly to control the expansion of the fault and prevent more serious consequences.

[0073] The specific process of research reactor modeling and fault diagnosis is as follows Figure 4 As shown. Based on the modeling steps and the accident event handling procedures of the research reactor, the main faults of the research reactor are modeled as units, where the main faults include primary water leakage, main cooler heat transfer tube rupture, LOCA, main pump failure, booster pump failure, loss of normal secondary water flow, fuel assembly damage, overpower, uncontrolled extraction of a single control rod during power operation, SBO, etc. Finally, the entire fault model of the research reactor is integrated based on the unit model, and the constructed model is as follows Figure 2 The above can be used to design the subsequent research reactor fault diagnosis system based on the overall SDG model.

[0074] In order to achieve the above functions, the research reactor fault detection device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0075] The present application also provides a method Figure 5 The research reactor fault detection device shown. The device is applied to the research reactor system, which includes event nodes, state nodes and preset change trends of state nodes associated with the event nodes caused by fault events corresponding to the event nodes; each event node is associated with at least one state node; the state node represents the state variables generated in the operating state of the research reactor, and the event node represents the fault event of the research reactor system; the device includes: a first determination unit 501, a second determination unit 502 and a fault diagnosis unit 503.

[0076] The first determination unit 501 is configured to determine a first event node associated with the target status node when it is detected that the target status node is in a fault state.

[0077] The second determination unit 502 is configured to determine a first state change amount of a state variable of a first status node associated with the first event node.

[0078] The fault diagnosis unit 503 is configured to select, from the first event nodes, a second event node whose change trends of the first state change amounts associated with the events are all the same as a target preset change trend corresponding to the first status node, so as to determine the fault event of the second event node as the target fault event.

[0079] As an implementation manner, before determining the first event node associated with the target status node when it is detected that the target status node is in a fault state, the first determination unit 501 is further configured to: determine that the target status node is in a fault state when the current state variable of the target status node is greater than or equal to a first variable threshold corresponding to the target status node.

[0080] As an implementation manner, the first determination unit 501 is further configured to: when the current state variable is less than the first variable threshold, obtain multiple state variable values of the target status node within a preset time; perform linear fitting on the multiple state variable values to obtain the current change trend of the current state variable; if it is determined that the current change trend is consistent with a preset change condition corresponding to the target status node, determine that the target status node is in a fault state.

[0081] As an implementation manner, the first determination unit 501 is configured to: determine that the target status node is in a fault state when the current change trend is an upward trend and the upward speed is greater than or equal to a preset upward speed; or; determine that the target status node is not in a fault state when the current change trend is an upward trend and the upward speed is less than the preset upward speed; or; determine that the target status node is not in a fault state when the current change trend is a downward trend.

[0082] As an implementation manner, the research reactor system includes a knowledge graph, and the knowledge graph includes a tree-like branch formed by each event node and a corresponding status node. The tree-like branch includes a directed symbol representing a preset change trend of the status node associated with the event node caused by the fault event corresponding to the event node. Among them, the directed symbol includes a positive symbol representing that the fault event causes an increase in the corresponding state variable or a negative symbol representing that the fault event causes a decrease in the corresponding state variable.

[0083] As an implementation manner, the first determination unit 501 is specifically configured to: determine a first target tree-like branch connected to the target status node from the knowledge graph; determine the event node connected to the first target tree-like branch as the first event node.

[0084] As an implementation method, the second determination unit 502 is specifically used to: determine from the knowledge graph a second target tree branch connected to the first event node; determine the state node connected to the second target tree branch as the first state node; determine whether the first state change amount of the first state node is the same as the target preset change trend based on the target preset change trend represented by the directed symbol of the first state node; determine the first event node in the first event node, in which the first state nodes connected by the event node are all the same as the target preset change trend, as the second event node, wherein the fault event includes one or more of the following: the above-mentioned fault event includes one or more of the following: booster pump failure, main pump failure, main coolant heat transfer tube rupture, water loss event, main coolant leakage, loss of normal secondary water flow, uncontrolled extraction of a single control rod during power operation, fuel assembly damage, and plant-wide power outage, and a fault event is associated with at least one state variable.

[0085] Regarding the device in the above embodiment, the specific manner in which each unit module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0086] Figure 6 is a schematic diagram of a control device provided by this application. Figure 6 The control device 80 may include at least one processor 801 and a memory 803 for storing processor executable instructions. The processor 801 is configured to execute instructions in the memory 803 to implement the research reactor fault detection method in the following embodiment.

[0087] In addition, the control device 80 may further include a communication bus 802 , at least one communication interface 804 , an input device 806 , and an output device 805 .

[0088] The processor 801 may be a central processing unit (CPU), a microprocessing unit, an ASIC, or one or more integrated circuits for controlling the execution of the program of the present application.

[0089] The communication bus 802 may include a pathway for transmitting information between the above-mentioned components.

[0090] The communication interface 804 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0091] The input device 806 is used to receive input signals and the output device 805 is used to output signals.

[0092] The memory 803 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may exist independently and be connected to the processing unit through a bus. The memory may also be integrated with the processing unit.

[0093] The memory 803 is used to store instructions for executing the solution of the present application, and the execution is controlled by the processor 801. The processor 801 is used to execute the instructions stored in the memory 803, thereby realizing the functions of the method of the present application.

[0094] In a specific implementation, as an embodiment, the processor 801 may include one or more CPUs, such as Figure 6 CPU0 and CPU1 in.

[0095] In a specific implementation, as an embodiment, the control device 80 may include multiple processors, such as Figure 6 801 and processor 807 in the embodiment of the present invention. Each of these processors may be a single-CPU processor or a multi-CPU processor. The processor herein may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0096] The control device is Figure 6 The system shown includes: a processor 801 and a memory 803 for storing executable instructions of the processor 801; wherein the processor 801 is configured to execute the executable instructions to implement a research reactor fault detection method as described in any possible implementation method described above. The same technical effect can be achieved, and to avoid repetition, it will not be described here.

[0097] The embodiment of the present application also provides a research reactor system, which is configured to execute any possible implementation method of the research reactor fault detection method described above, and can achieve the same technical effect, so it will not be described here to avoid repetition.

[0098] The embodiment of the present application also provides a computer-readable storage medium, when the instructions in the computer-readable storage medium are executed by the processor of the control device or control equipment, the control device or control equipment can execute the research reactor fault detection method of any possible implementation method as described above. And the same technical effect can be achieved, so it will not be repeated here to avoid repetition.

[0099] The embodiment of the present application also provides a computer program product, including a computer program or instruction, which is executed by a processor as a research reactor fault detection method in any possible implementation manner described above. The same technical effect can be achieved, and to avoid repetition, it will not be described here.

[0100] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0101] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for detecting a research reactor fault, characterized in that: Applied to a research reactor system, including event nodes, state nodes, and preset change trends of state nodes associated with the event nodes due to fault events corresponding to the event nodes; each event node is associated with at least one state node; The state node represents the state variable generated in the operating state of the research reactor, and the event node represents the fault event of the research reactor system. The method includes: In case of detecting that the target state node is in a fault state, determining a first event node associated with the target state node; Determine a first state change amount of a state variable of a first state node associated with the first event node; From the first event node, select a second event node whose change trend of the first state change quantity associated with the event is the same as the target preset change trend corresponding to the first state node, so as to determine the fault event of the second event node as the target fault event; The research reactor system includes a knowledge graph, the knowledge graph includes a tree branch composed of each event node and a corresponding state node, the tree branch includes a directed symbol representing a preset change trend of a state node associated with the event node due to a fault event corresponding to the event node, wherein the directed symbol includes a positive symbol representing an increase in a corresponding state variable caused by the fault event or a negative symbol representing a decrease in the corresponding state variable caused by the fault event; The determining of the first event node associated with the target state node comprises: determining a first target tree branch connected to the target state node from the knowledge graph; determining the event node connected to the first target tree branch as the first event node; The selecting, from the first event node, a second event node whose change trend of the first state change quantity associated with the event is the same as the target preset change trend corresponding to the first state node, includes: determining from the knowledge graph a second target tree branch connected to the first event node; determining the state node connected to the second target tree branch as the first state node; determining whether the first state change quantity of the first state node is the same as the target preset change trend based on the target preset change trend represented by the directed symbol of the first state node; determining the first event node in the first event node where the first state nodes connected by the event node are the same as the target preset change trend as the second event node, wherein the fault event includes one or more of the following: the above-mentioned fault event includes one or more of the following: booster pump failure, main pump failure, main cooler heat transfer tube rupture, water loss event, main coolant leakage, normal secondary water flow loss, uncontrolled extraction of a single control rod during power operation, fuel assembly damage, and plant-wide power outage accident, and one fault event is associated with at least one state variable.

2. The method for detecting a research reactor fault according to claim 1, characterized in that: Before determining the first event node associated with the target state node in the case where the target state node is detected to be in a fault state, the method further includes: When the current state variable of the target node is greater than or equal to a first variable threshold corresponding to the target node, it is determined that the target state node is in a fault state.

3. The research reactor fault detection method according to claim 2, characterized in that: The method further comprises: When the current state variable is less than the first variable threshold, obtaining multiple state variable values ​​of the target node within a preset time; Performing linear fitting on the multiple state variable values ​​to obtain the current change trend of the current state variable; If it is determined that the current change trend is consistent with the preset change condition corresponding to the target state node, it is determined that the target state node is in a fault state.

4. The research reactor fault detection method according to claim 3, characterized in that: The method further comprises: When the current change trend is an upward trend and the rising speed is greater than or equal to a preset rising speed, determining that the target state node is in a fault state; Or; when the current change trend is an upward trend and the rising speed is less than a preset rising speed, determining that the target state node is in a non-fault state; Or; when the current change trend is a downward trend, it is determined that the target state node is in a non-fault state.

5. A research reactor fault detection device, characterized in that: The research reactor fault detection device is applied to a research reactor system, including an event node, a state node, and a preset change trend of a state node associated with the event node caused by a fault event corresponding to the event node; Each event node is associated with at least one state node; The state node represents the state variable generated in the operating state of the research reactor, and the event node represents the fault event of the research reactor system failure; The research reactor fault detection device comprises: a first determination unit, configured to determine a first event node associated with the target state node when detecting that the target state node is in a fault state; A second determining unit, configured to determine a first state change amount of a state variable of a first state node associated with the first event node; A fault diagnosis unit, configured to select, from the first event node, a second event node whose change trend of the first state change quantity associated with the event is the same as the target preset change trend corresponding to the first state node, so as to determine the fault event of the second event node as the target fault event; The research reactor system includes a knowledge graph, the knowledge graph includes a tree branch composed of each event node and a corresponding state node, the tree branch includes a directed symbol representing a preset change trend of a state node associated with the event node due to a fault event corresponding to the event node, wherein the directed symbol includes a positive symbol representing an increase in a corresponding state variable caused by the fault event or a negative symbol representing a decrease in the corresponding state variable caused by the fault event; The first determination unit is specifically used to: determine a first target tree branch connected to the target state node from the knowledge graph; determine an event node connected to the first target tree branch as the first event node; The fault diagnosis unit is specifically used to: determine from the knowledge graph a second target tree branch connected to the first event node; determine the state node connected to the second target tree branch as the first state node; determine whether the first state change amount of the first state node is the same as the target preset change trend based on the target preset change trend represented by the directed symbol of the first state node; determine the first event node in the first event node where all the first state nodes connected by the event node are the same as the target preset change trend as the second event node, wherein the fault event includes one or more of the following: the above-mentioned fault event includes one or more of the following: booster pump failure, main pump failure, main cooler heat transfer tube rupture, water loss event, main coolant leakage, loss of normal secondary water flow, uncontrolled extraction of a single control rod during power operation, fuel assembly damage, and plant-wide power outage, and one of the fault events is associated with at least one state variable.

6. A research reactor system, characterized in that: It includes event nodes, state nodes and preset change trends of state nodes associated with event nodes caused by fault events corresponding to the event nodes; each event node is associated with at least one state node; The state node represents the state variables generated when the research reactor is in operation, and the event node represents the fault event of the research reactor system; and is configured to execute the research reactor fault detection method as described in any one of claims 1 to 4.

7. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor, the processor is enabled to perform the research reactor fault detection method according to any one of claims 1 to 4.

Images