Nuclear steam pressure control system fault detection method, device and equipment, control system and medium

Abstract

The embodiment of the invention provides a nuclear steam pressure control system fault detection method, device and equipment, a control system and a medium, and relates to the technical field of nuclear reactors. The method comprises the steps that a first thermal parameter set of the nuclear steam pressure control system under the current operation working condition is obtained, and parameters in the first thermal parameter set are related to the current operation working condition of the nuclear steam pressure control system; the first thermal parameter set is input into a pre-constructed target fault detection model, a detection result is obtained, the detection result is used for indicating that the actuator is in a normal state or a fault state, and under the condition that the detection result indicates that the actuator is in the fault state, the detection result comprises a first fault type and a first fault degree of the actuator; the target fault detection model is obtained through training according to a plurality of second thermal parameter groups of the nuclear steam pressure control system under a plurality of operation conditions. According to the embodiment of the invention, the efficiency of detecting the fault of the actuator in the nuclear steam pressure control system can be improved.

Description

Technical Field

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

[0002] As the main control system of a nuclear reactor, the nuclear steam pressure control system can control the steam discharge and feed water flow to ensure that the heat of the nuclear reactor core is safely and stably transferred to the secondary circuit. The actuator of the nuclear steam pressure control system is the operating drive element of the nuclear steam pressure system. Its working state is often in harsh environments such as high temperature, high pressure, and high radiation. Therefore, the actuator is prone to drift, stuck, and other faults. When an actuator fails, fault diagnosis usually relies on the experience of engineers and operators. However, relying on manual search among many parameters and then obtaining the fault diagnosis results of the actuator based on the searched parameters takes a lot of time and has the problem of low detection efficiency. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to propose a nuclear steam pressure control system fault detection method, device, equipment, control system and medium, aiming to improve the efficiency of detecting actuator faults in a nuclear steam pressure control system.

[0004] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present application provides a nuclear steam pressure control system fault detection method, which is applied to a nuclear steam pressure control system, wherein the nuclear steam pressure control system includes an actuator, and the actuator is used to drive the nuclear steam pressure control system to operate. The method includes:

[0005] Acquire a first thermal parameter group of the nuclear steam pressure control system under a current operating condition, wherein parameters in the first thermal parameter group are related to the current operating condition of the nuclear steam pressure control system;

[0006] The first thermal parameter group is input into a pre-built target fault detection model to obtain a detection result, wherein the detection result is used to indicate that the actuator is in a normal state or a faulty state. When the detection result indicates that the actuator is in the faulty state, the detection result includes a first fault type and a first fault degree of the actuator. The target fault detection model is trained based on multiple second thermal parameter groups of the nuclear steam pressure control system under multiple operating conditions.

[0007] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a nuclear steam pressure control system fault detection device, wherein the nuclear steam pressure control system includes an actuator, and the actuator is used to drive the nuclear steam pressure control system to operate, and the device includes:

[0008] A data acquisition module, used for acquiring a first thermal parameter group of the nuclear steam pressure control system under a current operating condition, wherein the parameters in the first thermal parameter group are related to the current operating condition of the nuclear steam pressure control system;

[0009] A detection module is used to input the first thermal parameter group into a pre-built target fault detection model to obtain a detection result, wherein the detection result is used to indicate that the actuator is in a normal state or a faulty state. When the detection result indicates that the actuator is in the faulty state, the detection result includes a first fault type and a first fault degree of the actuator. The target fault detection model is trained based on multiple second thermal parameter groups of the nuclear steam pressure control system under multiple operating conditions.

[0010] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the nuclear steam pressure control system fault detection method described in the first aspect.

[0011] To achieve the above-mentioned purpose, a fourth aspect of the embodiments of the present application provides a nuclear steam pressure control system, comprising:

[0012] an actuator, the actuator being used to drive the nuclear steam pressure control system to operate; and

[0013] The electronic device described in the third aspect.

[0014] To achieve the above-mentioned purpose, the fifth aspect of an embodiment of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the nuclear steam pressure control system fault detection method described in the first aspect above.

[0015] The nuclear steam pressure control system fault detection method, device, equipment, control system and medium proposed in the present application obtain the first thermal parameter group of the nuclear steam pressure control system under the current operating condition, and input the first thermal parameter group into the pre-built target fault detection model to obtain the detection result, and the detection result can indicate that the actuator is in a normal state or a fault state. When the detection result indicates that the actuator is in a fault state, the detection result includes the first fault type and the first fault degree of the actuator. In the above manner, the actuator can be quickly detected through the target fault detection model, thereby improving the efficiency of detecting the fault of the actuator in the nuclear steam pressure control system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of a nuclear steam pressure control system fault detection method provided by an embodiment of the present application;

[0017] Figure 2 is another flow chart of the nuclear steam pressure control system fault detection method provided by an embodiment of the present application;

[0018] Figure 3 It is a structural schematic diagram of a nuclear steam pressure control system fault detection device provided in an embodiment of the present application;

[0019] Figure 4 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0021] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0023] To solve the problems of the prior art, the embodiments of the present application provide a nuclear steam pressure control system fault detection method, device, equipment, control system and medium, aiming to improve the efficiency of detecting faults of actuators in a nuclear steam pressure control system.

[0024] The nuclear steam pressure control system fault detection method, device, equipment, control system and medium provided in the embodiments of the present application are specifically illustrated through the following embodiments. First, the nuclear steam pressure control system fault detection method in the embodiments of the present application is described.

[0025] The nuclear steam pressure control system fault detection method provided in the embodiment of the present application relates to the field of nuclear reactor technology. The nuclear steam pressure control system fault detection method provided in the embodiment of the present application can be applied to a terminal, can also be applied to a server side, and can also be software running in a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or a server cluster or a distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the nuclear steam pressure control system fault detection method, etc., but is not limited to the above forms.

[0026] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0027] Figure 1 The present invention is a flow chart of a method for detecting a fault in a nuclear steam pressure control system provided in an embodiment of the present invention. The method for detecting a fault in a nuclear steam pressure control system provided in an embodiment of the present invention can be applied to a nuclear steam pressure control system, and specifically to electronic equipment in a nuclear steam pressure control system. The nuclear steam pressure control system also includes an actuator, and the actuator is used to drive the operation of the nuclear steam pressure control system. Figure 1 The method may include but is not limited to steps 101 to 102.

[0028] Step 101, obtaining a first thermal parameter group of the nuclear steam pressure control system under a current operating condition, wherein parameters in the first thermal parameter group are related to the current operating condition of the nuclear steam pressure control system;

[0029] Step 102: Input the first thermal parameter group into a pre-built target fault detection model to obtain a detection result, wherein the detection result is used to indicate that the actuator is in a normal state or a faulty state. When the detection result indicates that the actuator is in the faulty state, the detection result includes a first fault type and a first fault degree of the actuator. The target fault detection model is trained based on multiple second thermal parameter groups of the nuclear steam pressure control system under multiple operating conditions.

[0030] The electronic device can pre-build a target fault detection model. The target fault detection model is obtained by training multiple second thermal parameter groups under multiple operating conditions of the nuclear steam pressure control system, and can be used to detect the actuator. After the construction of the target fault detection model is completed, the state of the actuator can be detected by the target fault detection model. Specifically, through multiple sensors set in the nuclear steam pressure control system, the electronic device can obtain the first thermal parameter group of the actuator of the nuclear steam pressure control system under the current operating condition, and input the first thermal parameter group into the target fault detection model to obtain the detection result output by the target fault detection model. Among them, the parameters in the first thermal parameter group are related to the current operating condition of the nuclear steam pressure control system, and may include parameters such as reactor power setting value, actual reactor output power, coolant inlet temperature, coolant outlet temperature, steam generator feed water flow, steam flow, actuator opening and pressure drop, bypass valve opening and steam pressure. The detection result can indicate that the actuator is in a normal state or a fault state. The representation form of the detection result is not limited, and it can be an identifier corresponding to the normal state or an identifier corresponding to the fault state, or other forms that can distinguish different states of the actuator.

[0031] If the detection result indicates that the actuator is in a fault state, at this time, the detection result also includes the first fault type and the first fault degree of the actuator. Among them, the first fault type is used to indicate the type of fault that occurs in the actuator, for example, it can be a constant gain fault, a constant deviation fault or a stuck fault. The first fault degree is used to indicate the degree of deviation of the opening of the actuator in the first thermal parameter group compared to the theoretical opening of the actuator under the current operating conditions. Exemplarily, if the first fault type is a constant gain fault, the first fault degree can be any value from 0.4 to 1.6. Alternatively, if the first fault type is a constant deviation fault, the first fault degree can be any value from -20% to 20%. Alternatively, if the first fault type is a stuck fault, the first fault degree is a fixed value. It should be noted that the first fault degree is not limited to the range given in the above example. Through the above steps, the rapid detection of the actuator can be completed through the target fault detection model.

[0032] In the steps 101 to 102 shown in the embodiment of the present application, the first thermal parameter group of the nuclear steam pressure control system under the current operating condition is obtained, and the first thermal parameter group is input into the pre-built target fault detection model to obtain the detection result, which can indicate that the actuator is in a normal state or a fault state. When the detection result indicates that the actuator is in a fault state, the detection result includes the first fault type and the first fault degree of the actuator. In the above manner, the actuator can be quickly detected through the target fault detection model, thereby improving the efficiency of detecting the fault of the actuator in the nuclear steam pressure control system.

[0033] In some embodiments, inputting the first thermal parameter group into a pre-built target fault detection model includes:

[0034] The first thermal parameter group is input into the target fault detection model according to a preset data format, wherein the preset data format is determined based on a preset time step and a characteristic dimension of the first thermal parameter group.

[0035] The electronic device may pre-configure a preset data format, and input the first thermal parameter group into the target fault detection model according to the preset data format, so that the input first thermal parameter group meets the data format requirements of the target fault detection model. Exemplarily, the preset data format may be: 1×T×N, where T is the preset time step; N is the characteristic dimension of the first thermal parameter group, that is, the number of parameters included in the first thermal parameter group. In this way, the first thermal parameter group may be input into the target fault detection model, so that the target fault detection model can perform fault detection on the actuator according to the input first thermal parameter group.

[0036] In some embodiments, the target fault detection model is obtained according to the following process:

[0037] Constructing a sample data set, wherein the sample data set includes a training sample set and a test sample set;

[0038] According to the training sample set, the initial fault detection model is trained until the loss function of the initial fault detection model converges to obtain an intermediate fault detection model;

[0039] Testing the intermediate fault detection model according to the test sample set to obtain a detection accuracy rate of the intermediate fault detection model;

[0040] If the detection accuracy is greater than a preset accuracy, the intermediate fault detection model is determined as the target fault detection model.

[0041] The electronic device can pre-build a target fault detection model. Specifically, the electronic device can pre-configure a variety of operating conditions and obtain the second thermal parameter group under different operating conditions, the second thermal parameter group includes but is not limited to the reactor power setting value, the actual output power of the reactor, the coolant inlet temperature, the coolant outlet temperature, the steam generator feed water flow, the steam flow, the steam pressure, the actuator opening and pressure drop, the bypass valve opening and other parameters, and further based on the data generation program of the simulation platform, generate sample data according to multiple second thermal parameter groups and configured fault labels, so as to construct a sample data set including a training sample set and a test sample set according to the sample data. The sample data includes input data and output data, wherein the input data is the second thermal parameter group or the third thermal parameter group, and the output data is the normal label corresponding to the second thermal parameter group, or the fault label corresponding to the third thermal parameter group.

[0042] In this way, based on the training sample set, the electronic device can train the initial fault detection model. The initial fault detection model can adopt a fixed-weight time series classification neural network model, a multi-layer long short-term memory network or a convolutional neural network. Taking the fixed-weight time series classification neural network model as an example, the initial fault detection model can include multiple fault detection sub-models, each fault detection sub-model includes a time series classification neural network layer, a global average pooling layer, a fully connected layer and a normalized index layer, and multiple fault detection sub-models share the output layer. There is a problem of high deviation in the accuracy of a single fault detection sub-model. During the training process, the accuracy of the training sample set can easily reach 100%. By integrating multiple fault detection sub-models, the possibility of overfitting can be reduced.

[0043] For each fault detection sub-model, the time series classification neural network layer is used to extract features from the input data to obtain a feature map. The time series classification neural network layer may include multiple residual blocks, each of which is composed of multiple time series classification neural network submodules. The input data is input to the time series classification neural network submodule in the time series classification neural network layer in the form of a T×N-dimensional time series, and the bottleneck layer in the time series classification neural network submodule performs feature extraction. The bottleneck layer can reduce the dimension of the input data to reduce the complexity of the model and the possibility of overfitting. After the bottleneck layer, the time series classification neural network submodule can extract features through filters, so that matrices with larger sizes can be reduced in dimension, and the input data can be aggregated at different sizes to facilitate the extraction of features from different time steps. The input of each residual block is transmitted to the input of the next residual block through a shortcut linear connection, thereby alleviating the gradient vanishing problem by allowing the direct flow of gradients. The global average pooling layer reduces each feature map to a single value (i.e., the average value) by performing an average pooling operation on the feature map of the entire time length. The global average pooling layer operates on the entire feature map, so it can retain the spatial position information of the feature map while retaining the global features. The fully connected layer consists of a layer of neurons, and an activation function is used to introduce nonlinear factors to improve the learning ability of the initial fault detection model. The fully connected layer is mainly used to map the features extracted from the time series classification neural network layer to the label space of the fault data to facilitate subsequent classification by the classifier. The normalized exponential layer performs mathematical calculations through the normalized exponential function to normalize a numerical vector to a probability distribution vector. The initial fault detection model is composed of a collection of multiple fault detection sub-models with the same structure. Multiple fault detection sub-models share the output layer. The output layer determines the output data based on the average value of the probability distribution vectors obtained by the normalized exponential layers of multiple fault detection sub-models.

[0044] During the training process, according to the input data in the training sample set, forward reasoning can be performed through the initial fault detection model to obtain the output data of the initial fault detection model, and then the loss value of the initial fault detection model is determined according to the output data of the initial fault detection model and the output data and loss function included in the training sample set. Subsequently, back propagation is performed to update the model parameters of the initial fault detection model. After the model parameters are updated, if the loss function of the initial fault detection model converges (that is, the loss value fluctuates within a preset range or the loss value is less than a preset threshold), then an intermediate fault detection model can be obtained. If the loss function of the initial fault detection model does not converge, the step of forward reasoning based on the input data in the training sample set is re-executed.

[0045] Furthermore, after the intermediate fault detection model is obtained, the intermediate fault detection model can be tested according to the test sample set, and the detection accuracy of the intermediate fault detection model satisfies formula (1):

[0046]

[0047] Among them, P Accuracy It represents the detection accuracy, TP is the number of correctly identified normal sample data, TN is the number of correctly identified fault sample data, FP is the number of fault sample data predicted incorrectly, and FN is the number of normal sample data predicted incorrectly.

[0048] When the detection accuracy is greater than the preset accuracy, the electronic device can determine the intermediate fault detection model as the target fault detection model. The preset accuracy can be adjusted according to the actual detection accuracy requirements. In this way, the construction of the target fault detection model can be completed, which is convenient for subsequent fault detection of the actuator through the target fault detection model.

[0049] In some embodiments, constructing a sample data set includes:

[0050] Acquire a plurality of second thermal parameter groups of the nuclear steam pressure control system under a plurality of operating conditions within a preset historical time period, wherein the plurality of operating conditions include steady-state conditions and transient conditions, and the parameters in each of the second thermal parameter groups are related to each of the operating conditions of the nuclear steam pressure control system;

[0051] According to the second thermal parameter group under each of the operating conditions and the preconfigured plurality of second fault types and plurality of second fault degrees, a fault simulation is performed on the nuclear steam pressure control system under each of the operating conditions to obtain a plurality of third thermal parameter groups under each of the operating conditions and a fault label corresponding to each of the third thermal parameter groups, wherein the fault label is used to indicate the second fault type and second fault degree corresponding to each of the third thermal parameter groups;

[0052] constructing a plurality of fault sample data according to a plurality of third thermal parameter groups under a plurality of the operating conditions and the fault labels respectively corresponding to the plurality of third thermal parameter groups under a plurality of the operating conditions;

[0053] configuring a normal label for each second thermal parameter group under the operating condition, and obtaining a plurality of normal sample data according to a plurality of second thermal parameter groups under the plurality of operating conditions and the normal label corresponding to each second thermal parameter group;

[0054] Obtaining a plurality of sample data according to the plurality of fault sample data and the plurality of normal sample data;

[0055] Dividing the plurality of sample data according to a preset ratio to obtain the training sample set and the test sample set;

[0056] The sample data set is obtained according to the training sample set and the test sample set.

[0057] The electronic device can construct a sample data set. Specifically, first, a preset historical time period and multiple operating conditions can be configured. The preset historical time period can be determined according to the actual data volume requirement. Multiple operating conditions may include steady-state conditions and transient conditions, and transient conditions may include step-change loads and linear variable loads. Exemplarily, steady-state conditions may include three conditions: 100% full power (Full Power, FP), 60% FP, and 30% FP. Step-change loads may include three conditions: 100% FP-90% FP, 60% FP-50% FP, and 30% FP-20% FP. Linear variable loads may include 100% FP-90% FP, 60% FP-50% FP, and 30% FP-20% FP. The division of operating conditions is not limited to the aforementioned examples and can be adjusted according to actual conditions.

[0058] In this way, the electronic device can obtain multiple second thermal parameter groups of the core steam pressure control system under multiple operating conditions in a preset historical time period. Each second thermal parameter group corresponds to an operating condition, and the parameters in each second thermal parameter group are related to an operating condition corresponding to each second thermal parameter group. Each second thermal parameter group includes but is not limited to parameters such as reactor power setting value, actual reactor output power, coolant inlet temperature, coolant outlet temperature, steam generator feed water flow, steam flow, steam pressure, actuator opening and pressure drop, bypass valve opening, etc. Then, after pre-configuring the simulation platform with data such as the fault introduction time, simulation time, and fault object, the electronic device can perform fault simulation on the nuclear steam pressure control system under each operating condition through the simulation platform according to the second thermal parameter group under each operating condition, the pre-configured multiple second fault types (such as constant gain fault, constant deviation fault, and stuck fault), and the pre-configured multiple second fault degrees, obtain multiple third thermal parameter groups under each operating condition and the fault labels corresponding to each third thermal parameter group, and obtain multiple fault sample data according to the multiple third thermal parameter groups under multiple operating conditions and the fault labels corresponding to the multiple third thermal parameter groups under multiple operating conditions. Among them, each second fault type has corresponding multiple second fault degrees, the fault label is used to indicate the second fault type and the second fault degree corresponding to each third thermal parameter group, and each fault sample data includes a third thermal parameter group under an operating condition, and a fault label corresponding to a third thermal parameter group.

[0059] At the same time, the electronic device can configure a normal label for the second thermal parameter group under each operating condition, and construct multiple normal sample data according to multiple second thermal parameter groups under multiple operating conditions and the normal labels corresponding to each second thermal parameter group, each normal sample data includes a second thermal parameter group under an operating condition, and a normal label corresponding to a second thermal parameter group. In this way, multiple sample data (including multiple fault sample data and multiple normal sample data) can be obtained. According to a preset ratio (for example, 8:2), after dividing the multiple sample data, a training sample set and a test sample set can be obtained, and then a sample data set containing the training sample set and the test sample set can be obtained.

[0060] Through the above steps, a sample data set can be constructed, so that a target fault detection model can be constructed based on the sample data set in the future.

[0061] In some embodiments, before training the initial fault detection model according to the training sample set, the method further includes:

[0062] In response to a first input to the initial fault detection model, the following steps are performed:

[0063] Setting the activation function of the initial fault detection model as a target activation function;

[0064] Setting the loss function of the initial fault detection model as a target loss function;

[0065] Setting the optimizer of the initial fault detection model to a target optimization algorithm;

[0066] Setting the learning rate of the initial fault detection model to a target value;

[0067] The first input includes an identifier of the target activation function, an identifier of the target loss function, an identifier of the target optimization algorithm, and the target value.

[0068] Before training the initial fault detection model according to the training sample set, the electronic device can also receive a first input to the initial fault detection model, and in response to the first input, adjust the training strategy of the initial fault detection model to ensure that the model can capture key features through effective training and has a certain generalization ability. Specifically, the first input includes the identifier of the target activation function (e.g., ReLU), the identifier of the target loss function (e.g., cross entropy loss function), the identifier of the target optimization algorithm (e.g., Adaptive Moment Estimation, Adam), and the target value (e.g., 0.001). Then, according to the identifier of the target activation function, the identifier of the target loss function, the identifier of the target optimization algorithm, and the target value contained in the first input, the electronic device can set the activation function of the initial fault detection model to the target activation function, set the loss function of the initial fault detection model to the target loss function, set the optimizer of the initial fault detection model to the target optimization algorithm, and set the learning rate of the initial fault detection model to the target value. In this way, the configuration of the initial fault detection model can be completed.

[0069] In some embodiments, after inputting the first thermal parameter group into a pre-built target fault detection model to obtain a detection result, the method further includes:

[0070] Display the detection page;

[0071] In response to a second input to the detection page, the detection result is displayed.

[0072] After the first thermal parameter group is input into the target fault detection model and the detection result is obtained, the detection result can be displayed for easy viewing of the detection result. Specifically, the electronic device can display a detection page. The electronic device can receive a second input to the detection page, and the second input is used to indicate the display of detailed information in the detection result. When the second input is received, the detection result can be displayed through the detection page. In addition, if the detection result indicates that the actuator is in a faulty state, the first fault type and the first fault degree in the detection result can be displayed. In addition, the time when the actuator fails can also be displayed. In this way, the operator can view the detection results and then maintain the nuclear steam pressure control system according to the detection results.

[0073] Figure 2 is another flow chart of the nuclear steam pressure control system fault detection method provided in the embodiment of the present application, please refer to Figure 2 The following is an example of a nuclear steam pressure control system fault detection method provided in an embodiment of the present application. The nuclear steam pressure control system fault detection method comprises the following steps:

[0074] Step 201, obtaining a plurality of second thermal parameter groups of a nuclear steam pressure control system under a plurality of operating conditions.

[0075] First, multiple operating conditions are set, including steady-state conditions and transient conditions. The steady-state conditions include three types of operating conditions: 100% FP, 60% FP, and 30% FP. The transient conditions include 100% FP-90% FP step change load, 60% FP-50% FP step change load, 30% FP-20% FP step change load, 100% FP-90% FP linear change load, 60% FP-50% FP linear change load, and 30% FP-20% FP linear change load, for a total of 9 types of operating conditions. The range of the fault degree is set. For constant deviation faults, the range of the fault degree is -20% to 20%. For constant gain faults, the range of the fault degree is 0.4 to 1.6. For stuck faults, the fault degree is a fixed value.

[0076] Through the simulation platform, the third thermal parameter group corresponding to the above 9 operating conditions can be generated according to the configured second fault degree, so as to carry out the subsequent offline training of the initial fault detection model. The simulation platform can store the generated data, and the data format is an M×T×N cell array set, where M is the number of simulations, N is the characteristic dimension of the second thermal parameter group, and T is the preset time step. Since the sampling frequency is 10 Hz, the relationship between T and the simulation time t(s) is: T=t×10+1. The format of the fault label is a C×1 cell array, where C is the number of configured second fault degrees, and each cell is a T×1 string array.

[0077] Step 202, constructing a sample data set according to a plurality of second thermal parameter groups;

[0078] In one embodiment, by inserting faults at 10 time points, a total of 3110 sets of data (i.e., sample data, including fault sample data and normal sample data) can be generated in the actuator fault state and normal state. The multiple sample data are divided into a training sample set and a test sample set in a ratio of 8:2 to complete the construction of the sample data set.

[0079] Step 203, training the initial fault detection model according to the sample data set to obtain a target fault detection model with fixed weights;

[0080] Batch training is performed during model training. The batch size can be set. For example, the batch size can be set to 32, that is, the input data dimension in one batch is 32×3001 (preset time step)×12 (feature dimension of thermal parameter group). To avoid the gradient vanishing problem, the ReLU activation function is used as the activation function; fault detection is a multi-classification task, so the cross entropy loss function is used as the loss function, and the Adam algorithm is selected as the optimizer of this solution, and the learning rate is set to 0.001. In addition, the data format during offline training is M×T×N, which can be used to forward the initial fault detection model multiple times to obtain output data. The format of the fault label for offline training is C×1, which can be used to calculate the difference between the output data of each output and the output data in the sample data set to obtain the loss value.

[0081] The initial fault detection model is constructed based on the time series classification neural network. The architecture of the initial fault detection model has five layers, including a time series classification neural network layer, a global average pooling layer, a fully connected layer, a normalized exponential layer, and an output layer. Due to the high deviation problem of the accuracy of a single time series classification neural network, it is very easy to reach 100%, but the integration of multiple time series classification neural networks can overcome the overfitting problem. Therefore, the initial fault detection model is composed of a collection of four time series classification neural networks (i.e., fault detection sub-models) with the same structure, and the output data corresponding to a single input data will be determined according to the average value of the probability distribution vector obtained by the normalized exponential layer of the four single time series classification neural networks. Among them, the four time series classification neural networks all include a time series classification neural network layer, a global average pooling layer, a fully connected layer, a normalized exponential layer, and the four time series classification neural networks share an output layer.

[0082] Among them, the time series classification neural network layer is composed of two different residual blocks, and each residual block is composed of three time series classification neural network submodules. The input data is a T×N-dimensional time series. After the input data is input into the initial fault detection model, the time series classification neural network layer can be used to extract features and obtain feature maps. First, feature extraction is performed through the bottleneck layer in the time series classification neural network submodule. The bottleneck layer uses n filters with a length of 1 and a step size of 1 to convert a T×N-dimensional signal into a T×n-dimensional signal, where n is much smaller than N, thereby achieving the purpose of dimensionality reduction. The operation of the bottleneck layer can reduce the complexity of the model and alleviate the overfitting problem. In this way, different convolutional layers can be combined in parallel, and the result matrices processed by different convolutional layers are spliced ​​together in the depth dimension to form a deeper matrix. After the bottleneck layer, the time series classification neural network submodule uses 4 filters with different scale lengths (1, 10, 20, 40) for feature extraction. The advantage of this method is that it can first reduce the dimensionality of matrices with larger sizes while aggregating input data at different sizes, which is convenient for extracting features of input data from different time steps. The input of each residual block is transmitted to the input of the next residual block through a shortcut linear connection, thereby alleviating the vanishing gradient problem by allowing the direct flow of gradients.

[0083] The global average pooling layer reduces each feature map to a single value (i.e., the average value) by performing an average pooling operation on the feature maps of the entire time length. Global average pooling operates on the entire feature map, so it is able to retain the spatial location information of the feature map while retaining the global features.

[0084] The fully connected layer consists of a layer of neurons, and an activation function is used to introduce nonlinear factors to improve the learning ability of the neural network model. The fully connected layer is mainly used to map the features extracted from the time series classification neural network layer to the label space of the fault data to facilitate subsequent classification by the classifier.

[0085] The normalized exponential layer performs mathematical calculations using a normalized exponential function to normalize a numerical vector into a probability distribution vector.

[0086] The output layer determines the label with the highest probability according to the probability distribution vector given by the normalized exponential layer and obtains the detection result.

[0087] After completing the training of the initial fault detection model, the intermediate fault detection model can be obtained. The performance of the intermediate fault detection model can be evaluated using the test sample set, and the evaluation indicator is the detection accuracy. The detection accuracy is the ratio of the number of correctly detected sample data in all fault sample data and normal sample data to the total number of sample data. The higher the detection accuracy, the better the detection performance of the intermediate fault detection model. The detection accuracy satisfies formula (1). In this way, a target fault detection model with fixed weights can be obtained.

[0088] Step 204, obtaining a first thermal parameter group of the nuclear steam pressure control system under the current operating condition;

[0089] After completing the construction of the target fault detection model, the electronic equipment can obtain the first thermal parameter group of the nuclear steam pressure control system under the current operating conditions. The first thermal parameter group includes the reactor power setting value, the actual reactor output power, the coolant inlet temperature, the coolant outlet temperature, the steam generator feed water flow, the steam flow, the actuator opening and pressure drop, the bypass valve opening and the steam pressure.

[0090] Step 205: input the first thermal parameter group into the target fault detection model to obtain the detection result under the current operating condition.

[0091] The input data format of the target fault detection model is 1×T×N. The detection result corresponding to the first thermal parameter group, that is, the detection result of the actuator under the current operating condition, can be obtained through a fixed-weight time series classification neural network (i.e., the target fault detection model).

[0092] The nuclear steam pressure control system fault detection method provided in the embodiment of the present application can reduce the impact of actuator faults on the nuclear steam pressure control system. Through continuous state monitoring, detection knowledge can be adaptively learned from the collected data. Moreover, without adding new sensing equipment, the existing thermal parameter group can be used to improve the accuracy and efficiency of detection, so that the nuclear steam pressure control system can adapt to complex and changeable working conditions.

[0093] Figure 3 This is a schematic diagram of the structure of the nuclear steam pressure control system fault detection device provided in the embodiment of the present application. Figure 3 The embodiment of the present application also provides a nuclear steam pressure control system fault detection device 300, the nuclear steam pressure control system includes an actuator, the actuator is used to drive the nuclear steam pressure control system to operate, the nuclear steam pressure control system fault detection device 300 can implement the above-mentioned nuclear steam pressure control system fault detection method, the device 300 includes:

[0094] A data acquisition module 301 is used to acquire a first thermal parameter group of the nuclear steam pressure control system under the current operating condition, wherein the parameters in the first thermal parameter group are related to the current operating condition of the nuclear steam pressure control system;

[0095] The detection module 302 is used to input the first thermal parameter group into a pre-built target fault detection model to obtain a detection result, wherein the detection result is used to indicate that the actuator is in a normal state or a faulty state. When the detection result indicates that the actuator is in the faulty state, the detection result includes a first fault type and a first fault degree of the actuator. The target fault detection model is trained based on multiple second thermal parameter groups of the nuclear steam pressure control system under multiple operating conditions.

[0096] The specific implementation of the nuclear steam pressure control system fault detection device 300 is basically the same as the specific implementation of the nuclear steam pressure control system fault detection method described above, and will not be repeated here.

[0097] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned nuclear steam pressure control system fault detection method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0098] Figure 4 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. Figure 4 . Electronic equipment includes:

[0099] The processor 401 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0100] The memory 402 may be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 402 may store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 402, and the processor 401 calls and executes the nuclear steam pressure control system fault detection method of the embodiment of this application;

[0101] Input / output interface 403, used to implement information input and output;

[0102] Communication interface 404, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0103] Bus 405 , which transmits information between various components of the device (e.g., processor 401 , memory 402 , input / output interface 403 , and communication interface 404 );

[0104] The processor 401 , the memory 402 , the input / output interface 403 and the communication interface 404 are connected to each other in communication within the device via the bus 405 .

[0105] The embodiment of the present application also provides a nuclear steam pressure control system, including:

[0106] an actuator, the actuator being used to drive the nuclear steam pressure control system to operate; and

[0107] Electronic equipment.

[0108] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned nuclear steam pressure control system fault detection method is implemented.

[0109] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0110] The nuclear steam pressure control system fault detection method, device, equipment, control system and medium proposed in the present application obtain the first thermal parameter group of the nuclear steam pressure control system under the current operating condition, and input the first thermal parameter group into the pre-built target fault detection model to obtain the detection result, and the detection result can indicate that the actuator is in a normal state or a fault state. When the detection result indicates that the actuator is in a fault state, the detection result includes the first fault type and the first fault degree of the actuator. In the above manner, the actuator can be quickly detected through the target fault detection model, thereby improving the efficiency of detecting the fault of the actuator in the nuclear steam pressure control system.

[0111] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0112] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0113] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0114] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0115] The terms "first", "second", "third", "fourth", etc. (if any) in the specification 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. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0116] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0117] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0118] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0119] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0120] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0121] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A nuclear steam pressure control system fault detection method, applied to a nuclear steam pressure control system, wherein the nuclear steam pressure control system comprises an actuator, and the actuator is used to drive the nuclear steam pressure control system to operate, characterized in that: The method comprises: Acquire a first thermal parameter group of the nuclear steam pressure control system under a current operating condition, wherein parameters in the first thermal parameter group are related to the current operating condition of the nuclear steam pressure control system; The first thermal parameter group is input into a pre-built target fault detection model to obtain a detection result, wherein the detection result is used to indicate that the actuator is in a normal state or a faulty state. When the detection result indicates that the actuator is in the faulty state, the detection result includes a first fault type and a first fault degree of the actuator. The target fault detection model is trained based on multiple second thermal parameter groups of the nuclear steam pressure control system under multiple operating conditions.

2. The method according to claim 1, characterized in that Inputting the first thermal parameter group into the target fault detection model includes: The first thermal parameter group is input into the target fault detection model according to a preset data format, wherein the preset data format is determined based on a preset time step and a characteristic dimension of the first thermal parameter group.

3. The method according to claim 1, characterized in that: The target fault detection model is obtained according to the following process: Constructing a sample data set, wherein the sample data set includes a training sample set and a test sample set; According to the training sample set, the initial fault detection model is trained until the loss function of the initial fault detection model converges to obtain an intermediate fault detection model; Testing the intermediate fault detection model according to the test sample set to obtain a detection accuracy rate of the intermediate fault detection model; If the detection accuracy is greater than a preset accuracy, the intermediate fault detection model is determined as the target fault detection model.

4. The method according to claim 3, characterized in that The constructing of the sample data set includes: Acquire a plurality of second thermal parameter groups of the nuclear steam pressure control system under a plurality of operating conditions within a preset historical time period, wherein the plurality of operating conditions include steady-state conditions and transient conditions, and the parameters in each of the second thermal parameter groups are related to each of the operating conditions of the nuclear steam pressure control system; According to the second thermal parameter group under each of the operating conditions and the preconfigured plurality of second fault types and plurality of second fault degrees, a fault simulation is performed on the nuclear steam pressure control system under each of the operating conditions to obtain a plurality of third thermal parameter groups under each of the operating conditions and a fault label corresponding to each of the third thermal parameter groups, wherein the fault label is used to indicate the second fault type and second fault degree corresponding to each of the third thermal parameter groups; constructing a plurality of fault sample data according to a plurality of third thermal parameter groups under a plurality of the operating conditions and the fault labels respectively corresponding to the plurality of third thermal parameter groups under a plurality of the operating conditions; configuring a normal label for each second thermal parameter group under the operating condition, and obtaining a plurality of normal sample data according to a plurality of second thermal parameter groups under the plurality of operating conditions and the normal label corresponding to each second thermal parameter group; Obtaining a plurality of sample data according to the plurality of fault sample data and the plurality of normal sample data; Dividing the plurality of sample data according to a preset ratio to obtain the training sample set and the test sample set; The sample data set is obtained according to the training sample set and the test sample set.

5. The method according to claim 3, characterized in that: Before training the initial fault detection model according to the training sample set, the method further includes: In response to a first input to the initial fault detection model, the following steps are performed: Setting the activation function of the initial fault detection model as a target activation function; Setting the loss function of the initial fault detection model as a target loss function; Setting the optimizer of the initial fault detection model to a target optimization algorithm; Setting the learning rate of the initial fault detection model to a target value; The first input includes an identifier of the target activation function, an identifier of the target loss function, an identifier of the target optimization algorithm, and the target value.

6. The method according to claim 1, characterized in that After inputting the first thermal parameter group into a pre-built target fault detection model to obtain a detection result, the method further includes: Display the detection page; In response to a second input to the detection page, the detection result is displayed.

7. A nuclear steam pressure control system fault detection device, characterized in that: The nuclear steam pressure control system comprises an actuator, and the actuator is used to drive the nuclear steam pressure control system to operate. The device comprises: A data acquisition module, used for acquiring a first thermal parameter group of the nuclear steam pressure control system under a current operating condition, wherein the parameters in the first thermal parameter group are related to the current operating condition of the nuclear steam pressure control system; A detection module is used to input the first thermal parameter group into a pre-built target fault detection model to obtain a detection result, wherein the detection result is used to indicate that the actuator is in a normal state or a faulty state. When the detection result indicates that the actuator is in the faulty state, the detection result includes a first fault type and a first fault degree of the actuator. The target fault detection model is trained based on multiple second thermal parameter groups of the nuclear steam pressure control system under multiple operating conditions.

8. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the nuclear steam pressure control system fault detection method according to any one of claims 1 to 6 when executing the computer program.

9. A nuclear steam pressure control system, characterized in that: include: an actuator, the actuator being used to drive the nuclear steam pressure control system to operate; and The electronic device according to claim 8.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the nuclear steam pressure control system fault detection method according to any one of claims 1 to 6 is implemented.

Images