Method and device for detecting reliability of residual heat removal system of nuclear reactor and residual heat removal system of nuclear reactor

Abstract

The invention relates to a reliability detection method and device for a residual heat removal system of a nuclear reactor, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring first system boundary state data of a system boundary of a nuclear reactor residual heat removal system; calling a preset thermal hydraulic model according to the first system boundary state data, and predicting second system boundary state data under a plurality of time steps after the nuclear reactor waste heat removal system changes along with time; and performing reliability detection on the nuclear reactor waste heat removal system according to the second system boundary state data under the plurality of time steps. By adopting the method, the detection real-time performance and the detection efficiency of the residual heat removal system of the nuclear reactor can be improved.

Description

Technical Field

[0001] This application relates to the field of reactor reliability testing technology, and in particular to a method, apparatus and system for testing the reliability of a nuclear reactor waste heat removal system. Background Technology

[0002] Currently, for the secondary passive nuclear reactor residual heat removal system, the usual approach is to first conduct large-scale Monte Carlo sampling of the secondary passive nuclear reactor residual heat removal system to obtain system status data, and then conduct reliability analysis and testing on the secondary passive nuclear reactor residual heat removal system based on the system status data.

[0003] However, since large-scale Monte Carlo sampling requires a long sampling process, the reliability analysis and testing process has poor real-time performance and takes a long time, resulting in poor real-time performance and low efficiency. Summary of the Invention

[0004] Therefore, it is necessary to provide a reliability detection method, device, and system for nuclear reactor waste heat removal systems that can improve the real-time detection performance and efficiency of the detection of nuclear reactor waste heat removal systems, in order to address the aforementioned technical problems.

[0005] In a first aspect, this application provides a method for reliability testing of a nuclear reactor waste heat removal system, including:

[0006] Acquire the first system boundary state data of the nuclear reactor waste heat removal system;

[0007] Based on the first system boundary state data, a preset thermal-hydraulic model is invoked to predict the second system boundary state data at multiple time steps after the nuclear reactor waste heat removal system changes over time.

[0008] The reliability of the nuclear reactor waste heat removal system is tested based on the second system boundary state data at the multiple time steps.

[0009] In one embodiment, the training process of the preset thermal-hydraulic model includes:

[0010] Acquire training data, wherein the training data includes system boundary state sample data, which is generated by a thermal-hydraulic analysis program;

[0011] Based on the system boundary state sample data, the system boundary state prediction data of the nuclear reactor waste heat removal system at future time steps after the change over time is generated using the preset thermal-hydraulic model.

[0012] Based on the system boundary state sample data and the system boundary state prediction data, the preset thermal-hydraulic model is iteratively optimized.

[0013] In one embodiment, the step of iteratively optimizing the preset thermo-hydraulic model based on the system boundary state sample data and the system boundary state prediction data includes:

[0014] Based on the system boundary state sample data and the system boundary state prediction data, data fitting loss, control equation residual loss of the thermal-hydraulic analysis program, and system condition loss are constructed.

[0015] The preset thermal-hydraulic model is iteratively optimized based on the data fitting loss, the residual loss of the control equation, and the system condition loss.

[0016] In one embodiment, the iterative optimization of the preset thermal-hydraulic model based on the data fitting loss, the control equation residual loss, and the system condition loss includes:

[0017] The target model loss is obtained by weighted and fused the data fitting loss, the control equation residual loss, and the system condition loss.

[0018] If the target model loss converges, then the preset thermal-hydraulic model is determined to have completed iterative optimization.

[0019] If the target model loss does not converge, the preset thermal-hydraulic model is updated based on the target model loss, and the process returns to the step of obtaining training data until the calculated target model loss converges.

[0020] In one embodiment, the residual loss of the governing equations includes a first residual loss of the continuity equations for the gas and liquid phases in the nuclear reactor waste heat removal system, a second residual loss of the momentum conservation equation, and a third residual loss of the energy conservation equation.

[0021] In one embodiment, the system condition loss includes an initial condition loss; constructing the system condition loss includes:

[0022] Based on the system boundary state sample data, construct the first initial system state features corresponding to the nuclear reactor waste heat removal system;

[0023] Based on the system boundary state prediction data, construct the second initial system state characteristics corresponding to the nuclear reactor waste heat removal system;

[0024] An initial conditional loss is generated based on the difference between the first initial system state characteristics and the second initial system state characteristics.

[0025] In one embodiment, the system condition loss includes boundary condition loss; constructing the system condition loss includes:

[0026] Determine the system boundary condition type for the nuclear reactor waste heat removal system;

[0027] According to the preset mapping rules corresponding to the system boundary condition type, the system boundary state prediction data is converted into system boundary condition features;

[0028] Based on the system boundary condition characteristics and the preset boundary condition characteristic threshold, a boundary condition loss is generated.

[0029] In one embodiment, the reliability of the nuclear reactor waste heat removal system is tested based on the second system boundary state data at the plurality of time steps, including:

[0030] Based on the second system boundary state data at each time step, the preset system limit state prediction module is invoked to generate multiple system state values ​​of the nuclear reactor waste heat removal system.

[0031] Based on the multiple system state values ​​and the preset failure criterion threshold, the system failure probability of the nuclear reactor waste heat removal system is generated.

[0032] The reliability of the nuclear reactor waste heat removal system is determined based on the system failure probability.

[0033] Secondly, this application also provides a reliability testing device for a nuclear reactor waste heat removal system, comprising:

[0034] The acquisition module is used to acquire the first system boundary state data of the system boundary of the nuclear reactor waste heat removal system;

[0035] The sampling module is used to call a preset thermal-hydraulic model based on the first system boundary state data to predict the second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after the changes over time.

[0036] The detection module is used to perform reliability detection on the nuclear reactor waste heat removal system based on the second system boundary state data at the multiple time steps.

[0037] Thirdly, this application also provides a nuclear reactor waste heat removal system, the nuclear reactor waste heat removal system including a controller, the controller being used for:

[0038] Acquire first system boundary state data of the nuclear reactor waste heat removal system; based on the first system boundary state data, call a preset thermal-hydraulic model to predict second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after changes over time; perform reliability testing on the nuclear reactor waste heat removal system based on the second system boundary state data at the multiple time steps.

[0039] The aforementioned reliability testing method, apparatus, and system for nuclear reactor waste heat removal systems can first acquire first system boundary state data of the system boundary of the nuclear reactor waste heat removal system. Then, based on the first system boundary state data, a preset thermal-hydraulic model can be invoked to predict second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after changes over time. This realizes the generation of large-scale system boundary state data of the nuclear reactor waste heat removal system using an artificial intelligence model, replacing the large-scale Monte Carlo sampling process. Compared with the large-scale Monte Carlo sampling process, the acquisition of second system boundary state data at multiple time steps in this application is faster and more efficient. Therefore, reliability testing of the nuclear reactor waste heat removal system based on the second system boundary state data at multiple time steps can improve the real-time performance and efficiency of the nuclear reactor waste heat removal system testing. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart illustrating a reliability testing method for a nuclear reactor waste heat removal system in one embodiment of this application.

[0042] Figure 2 This is a flowchart illustrating the training process of a preset thermal-hydraulic model in one embodiment of this application.

[0043] Figure 3 This is a structural block diagram of a reliability testing device for a nuclear reactor waste heat removal system in one embodiment of this application;

[0044] Figure 4 This is an internal structural diagram of a computer device in one embodiment of this application. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0046] In one exemplary embodiment, such as Figure 1 As shown, a reliability testing method for a nuclear reactor waste heat removal system is provided, including steps 202 to 206. Wherein:

[0047] Step 202: Obtain the first system boundary state data of the system boundary of the nuclear reactor waste heat removal system.

[0048] In this embodiment, the nuclear reactor waste heat removal system can be the secondary side passive nuclear reactor waste heat removal system. The system boundary of the nuclear reactor waste heat removal system can be the system inlet position, system outlet position, and channel wall position, etc. The first system boundary state data can be the physical quantity parameters of the system boundary, such as flow rate, mass, pressure, temperature, and phase fraction, etc.

[0049] As an example, the first system boundary state data can be a feature composed of multiple physical quantity parameters at multiple system boundary points, and this feature can exist in the form of a vector or a matrix.

[0050] Step 204: Based on the first system boundary state data, call the preset thermal-hydraulic model to predict the second system boundary state data at multiple time steps after the nuclear reactor waste heat removal system changes over time.

[0051] The preset thermal-hydraulic model can be a neural network model, which can be a time series model, such as a PINN network model or an LSTM network model. In this embodiment, the first system boundary state data can be used as input data, and the preset thermal-hydraulic model can be used to quickly predict and generate the second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after the current time step, thus realizing the rapid acquisition of large-scale system boundary state data.

[0052] Step 206: Based on the second system boundary state data at multiple time steps, perform reliability testing on the nuclear reactor residual heat removal system.

[0053] As an example, step 206 includes: calculating the failure probability of the nuclear reactor residual heat removal system based on the second system boundary state data at multiple time steps; and determining whether the nuclear reactor residual heat removal system is reliable based on the success probability.

[0054] As an example, reliability testing of a nuclear reactor residual heat removal system is performed based on second system boundary state data at multiple time steps, including:

[0055] Based on the second system boundary state data at each time step, the preset system limit state prediction module is invoked to generate multiple system state values ​​for the nuclear reactor waste heat removal system; based on the multiple system state values ​​and the preset failure criterion threshold, the system failure probability of the nuclear reactor waste heat removal system is generated; based on the system failure probability, the reliability of the nuclear reactor waste heat removal system is determined.

[0056] In this embodiment, the second system boundary state data at each time point can be used as input to call the preset system limit state prediction module to generate multiple system state values ​​for the nuclear reactor residual heat removal system. Based on the difference between the multiple system state values ​​at each time point and the preset failure criterion threshold, the number of first state values ​​that are greater than the preset failure criterion threshold is counted. The ratio between the number of first state values ​​and the total number of multiple system state values ​​is calculated to obtain the system failure probability. If the system failure probability is greater than the preset probability threshold, the nuclear reactor residual heat removal system is determined to be unreliable; if the system failure probability is not greater than the preset probability threshold, the nuclear reactor residual heat removal system is determined to be reliable.

[0057] As an example, the aforementioned preset system limit state prediction module can be the system limit state function corresponding to the nuclear reactor residual heat removal system; the formula for calculating the system failure probability is as follows:

[0058]

[0059] in, This represents the system failure probability. Let be the system limit state function. For the first Second system boundary state data, For indicator functions, indicator functions satisfy Time characterizes the functional failure of the nuclear reactor residual heat removal system. This represents the total number of boundary state data for multiple second systems.

[0060] In the aforementioned reliability testing method for a nuclear reactor waste heat removal system, the first system boundary state data of the system boundary of the nuclear reactor waste heat removal system can be obtained first. Then, based on the first system boundary state data, a preset thermal-hydraulic model can be invoked to predict the second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after changes over time. This realizes the generation of large-scale system boundary state data of the nuclear reactor waste heat removal system using an artificial intelligence model, replacing the large-scale Monte Carlo sampling process. Compared with the large-scale Monte Carlo sampling process, this embodiment obtains the second system boundary state data at multiple time steps faster and more efficiently. Therefore, by performing reliability testing on the nuclear reactor waste heat removal system based on the second system boundary state data at multiple time steps, the real-time performance and efficiency of the nuclear reactor waste heat removal system testing can be improved.

[0061] In one exemplary embodiment, such as Figure 2 As shown, the training process of the preset thermal hydraulic model includes:

[0062] Step 302: Obtain training data, which includes system boundary state sample data, generated by a thermal-hydraulic analysis program.

[0063] Among them, the thermal-hydraulic analysis program can be a simulation program, which can be used to generate system boundary state sample data.

[0064] Step 304: Based on the system boundary state sample data, use a preset thermal-hydraulic model to generate system boundary state prediction data for the nuclear reactor waste heat removal system at future time steps after the changes over time.

[0065] In this embodiment, the system boundary state sample data at the current time step can be used as input, and a preset thermal-hydraulic model can be used to predict the changes in the nuclear reactor waste heat removal system over time, so as to obtain the system boundary state prediction data for future time steps.

[0066] Step 306: Iteratively optimize the preset thermal-hydraulic model based on the system boundary state sample data and the system boundary state prediction data.

[0067] As an example, step 306 includes: obtaining sample label data corresponding to the system boundary state sample data, wherein the sample label data can be the real boundary data of the nuclear reactor residual heat removal system at a future time step corresponding to the system boundary state sample data; and iteratively optimizing the preset thermal-hydraulic model based on the difference between the sample label data and the system boundary state prediction data.

[0068] As an example, based on system boundary state sample data and system boundary state prediction data, a preset thermo-hydraulic model is iteratively optimized, including:

[0069] Based on the system boundary state sample data and system boundary state prediction data, we construct the data fitting loss, the control equation residual loss of the thermal-hydraulic analysis program, and the system condition loss; based on the data fitting loss, the control equation residual loss, and the system condition loss, we iteratively optimize the preset thermal-hydraulic model.

[0070] In this embodiment, sample label data corresponding to system boundary state sample data can be obtained. Based on the sample label data, system boundary state sample data, and system boundary state prediction data, data fitting loss, control equation residual loss of the thermal-hydraulic analysis program, and system condition loss are constructed. The data fitting loss, control equation residual loss, and system condition loss are weighted and fused to obtain the target model loss. Based on the target model loss, the preset thermal-hydraulic model is iteratively optimized.

[0071] In some embodiments, a preset thermal-hydraulic model is iteratively optimized based on data fitting loss, control equation residual loss, and system condition loss, including:

[0072] The data fitting loss, the control equation residual loss, and the system condition loss are weighted and fused to obtain the target model loss. If the target model loss converges, the preset thermal-hydraulic model is considered to have completed iterative optimization. If the target model loss does not converge, the preset thermal-hydraulic model is updated based on the target model loss, and the process returns to the step of obtaining training data until the calculated target model loss converges.

[0073] Among them, the data fitting loss is used to characterize the data fitting accuracy of the preset thermal-hydraulic model, the control equation residual loss is used to ensure that the preset thermal-hydraulic model conforms to the control equations of the thermal-hydraulic analysis procedure, that is, conforms to the basic physical constraints, and the system condition loss is used to ensure that the preset thermal-hydraulic model conforms to the basic system conditions. In this way, the preset thermal-hydraulic model can be iteratively optimized based on the data fitting loss, control equation residual loss and system condition loss, which can ensure the accuracy of the preset thermal-hydraulic model.

[0074] Specifically, a data fitting loss is constructed based on the sample label data corresponding to the system boundary state sample data and the system boundary state prediction data; a residual loss of the control equations of the thermal-hydraulic analysis program is constructed based on the system boundary state prediction data; a system condition loss is constructed based on the system boundary state sample data and the system boundary state prediction data; the data fitting loss, the residual loss of the control equations of the thermal-hydraulic analysis program, and the system condition loss are weighted and fused to obtain the target model loss; if the target model loss converges, the preset thermal-hydraulic model is determined to be iteratively optimized; if the target model loss does not converge, the preset thermal-hydraulic model is updated by backpropagation based on the gradient information calculated by the target model loss, and the steps of obtaining training data are executed in reverse order until the calculated target model loss converges.

[0075] As an example, system conditional loss includes boundary conditional loss and initial conditional loss. The formula for calculating the target model loss is as follows:

[0076]

[0077] in, For the target model loss, For data fitting loss, for The corresponding weighted weights, The residual loss of the governing equations in the thermal-hydraulic analysis program, for The corresponding weighted weights, For boundary condition loss, for The corresponding weighted weights, Loss due to initial conditions for The corresponding weighted weights.

[0078] In some embodiments, the residual loss of the governing equations includes a first residual loss of the continuity equations for the gas and liquid phases in the nuclear reactor residual heat removal system, a second residual loss of the momentum conservation equation, and a third residual loss of the energy conservation equation.

[0079] The continuity equations for the gas and liquid phases are as follows:

[0080] Gas phase:

[0081]

[0082] Liquid phase:

[0083]

[0084] The momentum conservation equations for the gas and liquid phases are as follows:

[0085] Gas phase:

[0086]

[0087] Liquid phase:

[0088]

[0089] The energy conservation equations for the gas and liquid phases are as follows:

[0090] Gas phase:

[0091]

[0092] Liquid phase:

[0093]

[0094] Taking the continuity equation of the gas phase as an example, the corresponding residual loss is:

[0095]

[0096] Subscript and They represent the gas phase and the liquid phase, respectively. It is the volume fraction. This refers to the mass transfer rate, expressed in kg / m³·s. The velocity is expressed in m / s. FWG and FWF represent the drag coefficients of the gas and liquid phases, respectively, with units of s⁻¹. FIG and FIF represent the drag coefficients of the gas and liquid phase interfaces, respectively, with units of s⁻¹. This is a volume force, with units of N / m³. It is a volumetric heat source, with units of W / m³. For the system boundary state prediction data, the continuity equations, momentum conservation equations, and energy conservation equations for the gas and liquid phases mentioned above are all known equations and will not be elaborated here.

[0097] In some embodiments, the formula for calculating the data fitting loss is as follows:

[0098]

[0099] in, This refers to the number of data acquisition points at the system boundary points. For sample label data, This is the system boundary state prediction data output from the preset thermal-hydraulic model.

[0100] In some embodiments, the system condition loss includes the initial condition loss; constructing the system condition loss includes:

[0101] Based on system boundary state sample data at multiple time steps, a first initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; based on system boundary state prediction data at future time steps, a second initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; and based on the difference between the first initial system state feature and the second initial system state feature, an initial conditional loss is generated.

[0102] The initial condition is that the nuclear reactor waste heat removal system is in its initial operating state. The system state can be characterized by the distribution of physical parameters of the nuclear reactor waste heat removal system, such as the distribution of physical parameters such as temperature, pressure, flow rate and phase fraction at each data acquisition point in the nuclear reactor waste heat removal system. Both the first initial system state feature and the second initial system state feature can be feature vectors.

[0103] Specifically, based on system boundary state sample data at multiple time steps, a first initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed. This first initial system state feature characterizes the state of the nuclear reactor waste heat removal system at the initial operating moment of the current time step and can be composed of multiple physical quantity characteristic values ​​of the gas and liquid phases of the nuclear reactor waste heat removal system at the current time step. These physical quantity characteristics can include temperature, pressure, flow rate, and phase fraction, etc. Based on system boundary state prediction data at future time steps, a second initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed. This second initial system state feature characterizes the state of the nuclear reactor waste heat removal system at the initial operating moment of the future time step and can be composed of multiple physical quantity characteristic values ​​of the gas and liquid phases of the nuclear reactor waste heat removal system at the current time step. These physical quantity characteristics can include temperature, pressure, flow rate, and phase fraction, etc. Initial condition loss is generated based on the difference between the first and second initial system state features.

[0104] As an example, in this embodiment, the root mean square error between the first initial system state characteristics and the second initial system state characteristics corresponding to multiple data acquisition points in the nuclear reactor waste heat removal system can be calculated as the initial condition loss.

[0105] In some embodiments, the system condition loss includes boundary condition loss; constructing the system condition loss includes:

[0106] Determine the system boundary condition type of the nuclear reactor waste heat removal system; convert the system boundary state prediction data at future time steps into system boundary condition features according to the preset mapping rules corresponding to the system boundary condition type; generate boundary condition loss based on the system boundary condition features and preset boundary condition feature thresholds.

[0107] In this embodiment, different preset mapping rules are set for different boundary condition types. These preset mapping rules are used to generate boundary condition features that characterize the boundary conditions of the nuclear reactor waste heat removal system.

[0108] Specifically, based on the system boundary condition type of the nuclear reactor waste heat removal system, a corresponding preset mapping rule is determined; based on the preset mapping rule, the system boundary state prediction data at future time steps is converted into system boundary condition features; and based on the mean square error between all system boundary condition features and the preset boundary condition feature threshold, the boundary condition loss is obtained.

[0109] As an example, the formula for calculating boundary condition loss is as follows:

[0110]

[0111] in, For boundary condition loss, The number of data collection points. For the first System boundary state prediction data from each data collection point. To preset the boundary condition feature threshold, This is the preset mapping rule.

[0112] In this embodiment, the preset thermal-hydraulic model can be iteratively optimized by setting data fitting loss, control equation residual loss of the thermal-hydraulic analysis program, and system condition loss. This ensures the accuracy of the data fitting of the preset thermal-hydraulic model and that the preset thermal-hydraulic model can meet the basic system conditions and physical constraints, thus guaranteeing the accuracy of the preset thermal-hydraulic model.

[0113] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0114] Based on the same inventive concept, this application also provides a reliability testing device for a nuclear reactor waste heat removal system for implementing the reliability testing method for the nuclear reactor waste heat removal system described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the reliability testing device for a nuclear reactor waste heat removal system provided below can be found in the limitations of the reliability testing method for a nuclear reactor waste heat removal system described above, and will not be repeated here.

[0115] In one exemplary embodiment, such as Figure 3 As shown, a reliability testing device for a nuclear reactor waste heat removal system is provided, comprising: an acquisition module 402, a sampling module 404, and a detection module 406, wherein:

[0116] The acquisition module 402 is used to acquire the first system boundary state data of the system boundary of the nuclear reactor waste heat removal system.

[0117] The sampling module 404 is used to predict the second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after the nuclear reactor waste heat removal system changes over time, based on the first system boundary state data and by calling a preset thermal-hydraulic model.

[0118] The detection module 406 is used to perform reliability detection on the nuclear reactor waste heat removal system based on the second system boundary state data at the multiple time steps.

[0119] In one exemplary embodiment, the apparatus further includes:

[0120] A training module is used to acquire training data, wherein the training data includes system boundary state sample data, which is generated by a thermal-hydraulic analysis program; based on the system boundary state sample data, using the preset thermal-hydraulic model, system boundary state prediction data of the nuclear reactor waste heat removal system at future time steps after time changes are generated; based on the system boundary state sample data and the system boundary state prediction data, the preset thermal-hydraulic model is iteratively optimized.

[0121] In one exemplary embodiment, the training module is further configured to:

[0122] Based on the system boundary state sample data and the system boundary state prediction data, construct the data fitting loss, the control equation residual loss of the thermal-hydraulic analysis program, and the system condition loss; based on the data fitting loss, the control equation residual loss, and the system condition loss, iteratively optimize the preset thermal-hydraulic model.

[0123] In one exemplary embodiment, the training module is further configured to:

[0124] The data fitting loss, the control equation residual loss, and the system condition loss are weighted and fused to obtain the target model loss. If the target model loss converges, the preset thermal-hydraulic model is determined to be iteratively optimized. If the target model loss does not converge, the preset thermal-hydraulic model is updated according to the target model loss, and the process returns to the step of obtaining training data until the calculated target model loss converges.

[0125] In an exemplary embodiment, the residual loss of the governing equations includes a first residual loss of the continuity equations for the gas and liquid phases in the nuclear reactor residual heat removal system, a second residual loss of the momentum conservation equation, and a third residual loss of the energy conservation equation.

[0126] In an exemplary embodiment, the system conditional loss includes an initial conditional loss; the training module is further configured to:

[0127] Based on the system boundary state sample data, a first initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; based on the system boundary state prediction data, a second initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; based on the difference between the first initial system state feature and the second initial system state feature, an initial condition loss is generated.

[0128] In one exemplary embodiment, the system conditional loss includes boundary conditional loss; the training module is further configured to:

[0129] The system boundary condition type of the nuclear reactor waste heat removal system is determined; the system boundary state prediction data is converted into system boundary condition features according to the preset mapping rule corresponding to the system boundary condition type; and the boundary condition loss is generated according to the system boundary condition features and the preset boundary condition feature threshold.

[0130] In one exemplary embodiment, the detection module is further configured to:

[0131] Based on the second system boundary state data at each time step, a preset system limit state prediction module is invoked to generate multiple system state values ​​for the nuclear reactor waste heat removal system; based on the multiple system state values ​​and a preset failure criterion threshold, the system failure probability of the nuclear reactor waste heat removal system is generated; based on the system failure probability, the reliability of the nuclear reactor waste heat removal system is determined.

[0132] Each module in the reliability testing device for the aforementioned nuclear reactor waste heat removal system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0133] In one exemplary embodiment, a nuclear reactor waste heat removal system is provided, which includes a controller for performing the contents of the above-described embodiment of the reliability detection method for the nuclear reactor waste heat removal system, which will not be repeated here.

[0134] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a reliability testing method for a nuclear reactor waste heat removal system.

[0135] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0136] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0137] Acquire first system boundary state data of the nuclear reactor waste heat removal system; based on the first system boundary state data, call a preset thermal-hydraulic model to predict second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after changes over time; perform reliability testing on the nuclear reactor waste heat removal system based on the second system boundary state data at the multiple time steps.

[0138] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0139] Acquire training data, wherein the training data includes system boundary state sample data, which is generated by a thermal-hydraulic analysis program; based on the system boundary state sample data, use the preset thermal-hydraulic model to generate system boundary state prediction data for the nuclear reactor waste heat removal system at future time steps after changes over time; iteratively optimize the preset thermal-hydraulic model based on the system boundary state sample data and the system boundary state prediction data.

[0140] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0141] Based on the system boundary state sample data and the system boundary state prediction data, construct the data fitting loss, the control equation residual loss of the thermal-hydraulic analysis program, and the system condition loss; based on the data fitting loss, the control equation residual loss, and the system condition loss, iteratively optimize the preset thermal-hydraulic model.

[0142] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0143] The data fitting loss, the control equation residual loss, and the system condition loss are weighted and fused to obtain the target model loss. If the target model loss converges, the preset thermal-hydraulic model is determined to be iteratively optimized. If the target model loss does not converge, the preset thermal-hydraulic model is updated according to the target model loss, and the process returns to the step of obtaining training data until the calculated target model loss converges.

[0144] In one embodiment, the residual loss of the governing equations includes a first residual loss of the continuity equations for the gas and liquid phases in the nuclear reactor residual heat removal system, a second residual loss of the momentum conservation equation, and a third residual loss of the energy conservation equation.

[0145] In one embodiment, the system condition loss includes initial condition loss; the processor, when executing the computer program, also performs the following steps:

[0146] Based on the system boundary state sample data, a first initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; based on the system boundary state prediction data, a second initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; based on the difference between the first initial system state feature and the second initial system state feature, an initial condition loss is generated.

[0147] In one embodiment, the system condition loss includes boundary condition loss; the processor, when executing the computer program, further implements the following steps:

[0148] The system boundary condition type of the nuclear reactor waste heat removal system is determined; the system boundary state prediction data is converted into system boundary condition features according to the preset mapping rule corresponding to the system boundary condition type; and the boundary condition loss is generated according to the system boundary condition features and the preset boundary condition feature threshold.

[0149] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0150] Based on the second system boundary state data at each time step, a preset system limit state prediction module is invoked to generate multiple system state values ​​for the nuclear reactor waste heat removal system; based on the multiple system state values ​​and a preset failure criterion threshold, the system failure probability of the nuclear reactor waste heat removal system is generated; based on the system failure probability, the reliability of the nuclear reactor waste heat removal system is determined.

[0151] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0152] Acquire first system boundary state data of the nuclear reactor waste heat removal system; based on the first system boundary state data, call a preset thermal-hydraulic model to predict second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after changes over time; perform reliability testing on the nuclear reactor waste heat removal system based on the second system boundary state data at the multiple time steps.

[0153] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0154] Acquire training data, wherein the training data includes system boundary state sample data, which is generated by a thermal-hydraulic analysis program; based on the system boundary state sample data, use the preset thermal-hydraulic model to generate system boundary state prediction data for the nuclear reactor waste heat removal system at future time steps after changes over time; iteratively optimize the preset thermal-hydraulic model based on the system boundary state sample data and the system boundary state prediction data.

[0155] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0156] Based on the system boundary state sample data and the system boundary state prediction data, construct the data fitting loss, the control equation residual loss of the thermal-hydraulic analysis program, and the system condition loss; based on the data fitting loss, the control equation residual loss, and the system condition loss, iteratively optimize the preset thermal-hydraulic model.

[0157] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0158] The data fitting loss, the control equation residual loss, and the system condition loss are weighted and fused to obtain the target model loss. If the target model loss converges, the preset thermal-hydraulic model is determined to be iteratively optimized. If the target model loss does not converge, the preset thermal-hydraulic model is updated according to the target model loss, and the process returns to the step of obtaining training data until the calculated target model loss converges.

[0159] In one embodiment, the residual loss of the governing equations includes a first residual loss of the continuity equations for the gas and liquid phases in the nuclear reactor residual heat removal system, a second residual loss of the momentum conservation equation, and a third residual loss of the energy conservation equation.

[0160] In one embodiment, the system condition loss includes initial condition loss; the computer program, when executed by the processor, further performs the following steps:

[0161] Based on the system boundary state sample data, a first initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; based on the system boundary state prediction data, a second initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; based on the difference between the first initial system state feature and the second initial system state feature, an initial condition loss is generated.

[0162] In one embodiment, the system condition loss includes boundary condition loss; the computer program, when executed by a processor, further implements the following steps:

[0163] The system boundary condition type of the nuclear reactor waste heat removal system is determined; the system boundary state prediction data is converted into system boundary condition features according to the preset mapping rule corresponding to the system boundary condition type; and the boundary condition loss is generated according to the system boundary condition features and the preset boundary condition feature threshold.

[0164] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0165] Based on the second system boundary state data at each time step, a preset system limit state prediction module is invoked to generate multiple system state values ​​for the nuclear reactor waste heat removal system; based on the multiple system state values ​​and a preset failure criterion threshold, the system failure probability of the nuclear reactor waste heat removal system is generated; based on the system failure probability, the reliability of the nuclear reactor waste heat removal system is determined.

[0166] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0167] Acquire first system boundary state data of the nuclear reactor waste heat removal system; based on the first system boundary state data, call a preset thermal-hydraulic model to predict second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after changes over time; perform reliability testing on the nuclear reactor waste heat removal system based on the second system boundary state data at the multiple time steps.

[0168] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0169] Acquire training data, wherein the training data includes system boundary state sample data, which is generated by a thermal-hydraulic analysis program; based on the system boundary state sample data, use the preset thermal-hydraulic model to generate system boundary state prediction data for the nuclear reactor waste heat removal system at future time steps after changes over time; iteratively optimize the preset thermal-hydraulic model based on the system boundary state sample data and the system boundary state prediction data.

[0170] In one embodiment, based on the system boundary state sample data and the system boundary state prediction data, a data fitting loss, a control equation residual loss of the thermal-hydraulic analysis program, and a system condition loss are constructed; based on the data fitting loss, the control equation residual loss, and the system condition loss, a preset thermal-hydraulic model is iteratively optimized.

[0171] In one embodiment, the data fitting loss, the control equation residual loss, and the system condition loss are weighted and fused to obtain the target model loss; if the target model loss converges, it is determined that the preset thermal-hydraulic model has been iteratively optimized; if the target model loss does not converge, the preset thermal-hydraulic model is updated according to the target model loss, and the step of obtaining training data is returned to until the calculated target model loss converges.

[0172] In one embodiment, the residual loss of the governing equations includes a first residual loss of the continuity equations for the gas and liquid phases in the nuclear reactor residual heat removal system, a second residual loss of the momentum conservation equation, and a third residual loss of the energy conservation equation.

[0173] In one embodiment, the system condition loss includes initial condition loss; the computer program, when executed by the processor, further performs the following steps:

[0174] Based on the system boundary state sample data, a first initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; based on the system boundary state prediction data, a second initial system state feature corresponding to the nuclear reactor waste heat removal system is constructed; based on the difference between the first initial system state feature and the second initial system state feature, an initial condition loss is generated.

[0175] In one embodiment, the system condition loss includes boundary condition loss; the computer program, when executed by a processor, further implements the following steps:

[0176] The system boundary condition type of the nuclear reactor waste heat removal system is determined; the system boundary state prediction data is converted into system boundary condition features according to the preset mapping rule corresponding to the system boundary condition type; and the boundary condition loss is generated according to the system boundary condition features and the preset boundary condition feature threshold.

[0177] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0178] Based on the second system boundary state data at each time step, a preset system limit state prediction module is invoked to generate multiple system state values ​​for the nuclear reactor waste heat removal system; based on the multiple system state values ​​and a preset failure criterion threshold, the system failure probability of the nuclear reactor waste heat removal system is generated; based on the system failure probability, the reliability of the nuclear reactor waste heat removal system is determined.

[0179] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. 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). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0180] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0181] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for reliability testing of a nuclear reactor waste heat removal system, characterized in that, The method includes: Acquire the first system boundary state data of the nuclear reactor waste heat removal system; Based on the first system boundary state data, a preset thermal-hydraulic model is invoked to predict the second system boundary state data at multiple time steps after the nuclear reactor waste heat removal system changes over time. The reliability of the nuclear reactor waste heat removal system is tested based on the second system boundary state data at the multiple time steps.

2. The method according to claim 1, characterized in that, The training process of the preset thermal-hydraulic model includes: Acquire training data, wherein the training data includes system boundary state sample data, which is generated by a thermal-hydraulic analysis program; Based on the system boundary state sample data, the system boundary state prediction data of the nuclear reactor waste heat removal system at future time steps after the change over time is generated using the preset thermal-hydraulic model. Based on the system boundary state sample data and the system boundary state prediction data, the preset thermal-hydraulic model is iteratively optimized.

3. The method according to claim 2, characterized in that, The step of iteratively optimizing the preset thermal-hydraulic model based on the system boundary state sample data and the system boundary state prediction data includes: Based on the system boundary state sample data and the system boundary state prediction data, data fitting loss, control equation residual loss of the thermal-hydraulic analysis program, and system condition loss are constructed. The preset thermal-hydraulic model is iteratively optimized based on the data fitting loss, the residual loss of the control equation, and the system condition loss.

4. The method according to claim 3, characterized in that, The step of iteratively optimizing the preset thermal-hydraulic model based on the data fitting loss, the residual loss of the control equation, and the system condition loss includes: The target model loss is obtained by weighted and fused the data fitting loss, the control equation residual loss, and the system condition loss. If the target model loss converges, then the preset thermal-hydraulic model is determined to have completed iterative optimization. If the target model loss does not converge, the preset thermal-hydraulic model is updated based on the target model loss, and the process returns to the step of obtaining training data until the calculated target model loss converges.

5. The method according to claim 3, characterized in that, The residual loss of the governing equations includes the first residual loss of the continuity equation for the gas and liquid phases in the nuclear reactor waste heat removal system, the second residual loss of the momentum conservation equation, and the third residual loss of the energy conservation equation.

6. The method according to claim 3, characterized in that, The system condition loss includes the initial condition loss; constructing the system condition loss includes: Based on the system boundary state sample data, construct the first initial system state features corresponding to the nuclear reactor waste heat removal system; Based on the system boundary state prediction data, construct the second initial system state characteristics corresponding to the nuclear reactor waste heat removal system; An initial conditional loss is generated based on the difference between the first initial system state characteristics and the second initial system state characteristics.

7. The method according to claim 3, characterized in that, The system condition loss includes boundary condition loss; constructing the system condition loss includes: Determine the system boundary condition type for the nuclear reactor waste heat removal system; According to the preset mapping rules corresponding to the system boundary condition type, the system boundary state prediction data is converted into system boundary condition features; Based on the system boundary condition characteristics and the preset boundary condition characteristic threshold, a boundary condition loss is generated.

8. The method according to claim 1, characterized in that, Based on the second system boundary state data at the multiple time steps, the reliability of the nuclear reactor residual heat removal system is tested, including: Based on the second system boundary state data at each time step, the preset system limit state prediction module is invoked to generate multiple system state values ​​of the nuclear reactor waste heat removal system. Based on the multiple system state values ​​and the preset failure criterion threshold, the system failure probability of the nuclear reactor waste heat removal system is generated. The reliability of the nuclear reactor waste heat removal system is determined based on the system failure probability.

9. A reliability testing device for a nuclear reactor waste heat removal system, characterized in that, The device includes: The acquisition module is used to acquire the first system boundary state data of the system boundary of the nuclear reactor waste heat removal system; The sampling module is used to call a preset thermal-hydraulic model based on the first system boundary state data to predict the second system boundary state data of the nuclear reactor waste heat removal system at multiple time steps after the changes over time. The detection module is used to perform reliability detection on the nuclear reactor waste heat removal system based on the second system boundary state data at the multiple time steps.

10. A nuclear reactor waste heat removal system, characterized in that, The nuclear reactor waste heat removal system includes a controller for performing the steps of the method as described in any one of claims 1-8.

Images