Method, device, computer equipment and storage medium for evaluating control loop of auxiliary steam distribution system of nuclear power plant

Abstract

The present invention specifically relates to a control loop evaluation method, device, computer equipment and storage medium for an auxiliary steam distribution system of a nuclear power plant, and belongs to the field of process monitoring technology. The present invention uses historical data of each valve control loop in the auxiliary steam distribution system to train a neural network model to obtain a trained neural network model; uses the trained neural network model to classify the real-time data of each valve control loop in the auxiliary steam distribution system to determine the working valve control loop data and the standby valve control loop data; and evaluates the working valve control loop data to obtain the control performance index of the working valve control loop. The present invention automatically determines the one-in-use and one-standby status of valves in the auxiliary steam distribution system of a nuclear power plant, and realizes accurate evaluation of the working valve control loop.

Description

Technical Field

[0001] The present invention relates to the technical field of process monitoring, and in particular to a control loop evaluation method, device, computer equipment and storage medium for an auxiliary steam distribution system of a nuclear power plant. Background Art

[0002] As a complex industrial system, the normal operation of a nuclear power plant requires the coordinated work of many control loops, and the auxiliary steam distribution system (abbreviated as "SVA system") is a very critical control loop. As a public steam system for nuclear power units, the auxiliary steam distribution system provides auxiliary steam to related auxiliary equipment and systems to meet the different operating conditions of nuclear power units.

[0003] Auxiliary steam is produced by auxiliary boilers during the shutdown and startup of nuclear power units, and by steam converters during the operation of nuclear power units. Two parallel pressure reducing regulating valves SVA303VV and SVA306VV are installed on the low-pressure steam main pipe for auxiliary steam supplied from the auxiliary boiler, and two parallel pressure reducing regulating valves SVA583VV and SVA586VV are installed on the main pipe for auxiliary steam supplied from the independent steam system.

[0004] The auxiliary steam distribution system distributes auxiliary steam to each user. Auxiliary steam systems include but are not limited to shaft seal systems and deaerator systems. Since there are many auxiliary steam systems, the main task of the auxiliary steam distribution system is to maintain the balance between auxiliary steam supply and demand. In order to maintain the stability of auxiliary steam pressure, the operating variables of the auxiliary steam distribution system include pressure reducing regulating valve SVA303VV, pressure reducing regulating valve SVA306VV, pressure reducing regulating valve SVA583VV and pressure reducing regulating valve SVA586VV.

[0005] In the auxiliary steam distribution system, each valve establishes a control loop. The control loops of the pressure reducing control valve SVA303VV and the pressure reducing control valve SVA306VV are: the controller outputs a signal to control the pressure reducing control valve according to the difference between the set value and the process variable of the low-pressure steam main pipe pressure after the pressure reducing control valve, so that the process variable of the low-pressure steam main pipe pressure after the pressure reducing control valve is as close to the set value as possible; the process variable of the low-pressure steam main pipe pressure after the pressure reducing control valve is the measured value of the low-pressure steam main pipe pressure after the pressure reducing control valve. The control loops of the pressure reducing control valve SVA583VV and the pressure reducing control valve SVA586VV are: the controller outputs a signal to control the pressure reducing control valve according to the difference between the set value and the process variable of the main steam main pipe pressure after the pressure reducing control valve, so that the process variable of the main steam main pipe pressure after the pressure reducing control valve is as close to the set value as possible; the process variable of the main steam main pipe pressure after the pressure reducing control valve is the average value of the three measured values ​​of the main steam main pipe pressure after the pressure reducing control valve.

[0006] In order to analyze the working condition of each valve, it is necessary to evaluate each valve control loop. However, in each valve control loop of the auxiliary steam distribution system, there is a situation where one is in use and one is in standby. The pressure reducing control valve SVA303VV and the pressure reducing control valve SVA306VV are one in use and one in standby, and the pressure reducing control valve SVA583VV and the pressure reducing control valve SVA586VV are one in use and one in standby. For the evaluation of the valve control loop of the auxiliary steam distribution system, it is necessary to determine whether it is a standby valve control loop or a working valve control loop. Only the working valve control loop needs to be evaluated.

[0007] In terms of control loop evaluation, traditional control loop evaluation methods often start from the error between the measured signal and the set value, analyze the variance and oscillation of the error signal, and thus evaluate the control effect of the control loop; or evaluate the viscosity and other properties of the valve through the relationship between the controller output and the control deviation. In the face of a valve control loop with one in use and one in standby, the use of traditional control loop evaluation methods often requires manual input of a flag signal to determine whether the valve control loop is in working condition. However, due to the limitations of nuclear power objects, it is often impossible to obtain all the required signal values. In the auxiliary steam distribution system, the flag signal of whether the two valves are working is not available, which makes it difficult to evaluate the valve control loop. Summary of the invention

[0008] The purpose of the present invention is to provide a control loop evaluation method for an auxiliary steam distribution system of a nuclear power plant based on a neural network, which automatically determines whether the valve control loop in the auxiliary steam distribution system of a nuclear power plant is a main valve control loop or a working valve control loop, and automatically evaluates the working valve control loop.

[0009] In order to achieve the above object, the present invention provides a control loop evaluation method for an auxiliary steam distribution system of a nuclear power plant, comprising the following steps:

[0010] Step 1, using the historical data of each valve control loop in the auxiliary steam distribution system to train the neural network model, to obtain a trained neural network model; the trained neural network model is used to distinguish the working valve control loop data and the standby valve control loop data; the historical data of each valve control loop in the auxiliary steam distribution system includes the set value, the process variable and the controller output signal and the state, and the state of the valve control loop is working or standby;

[0011] Step 2, obtaining real-time data of each valve control loop in the auxiliary steam distribution system; the real-time data of each valve control loop in the auxiliary steam distribution system includes set values, process variables and controller output signals;

[0012] Step 3: Use the trained neural network model to classify the real-time data of each valve control loop in the auxiliary steam distribution system, and determine the working valve control loop data and the standby valve control loop data;

[0013] Step 4: Evaluate the working valve control loop data to obtain the control performance index of the working valve control loop.

[0014] As one of the achievable methods, in step 1, the neural network model is a BP neural network model; the BP neural network model is trained using historical data of each valve control loop in the auxiliary steam distribution system, including the following steps:

[0015] Establish a BP neural network model including input layer, hidden layer and output layer;

[0016] The BP neural network model is trained by using the historical data of each valve control loop in the auxiliary steam distribution system, and the back propagation weights of the BP neural network model are updated based on the gradient descent method.

[0017] As one of the feasible ways, the back propagation weight update formula is:

[0018]

[0019]

[0020] in, is the weight connecting the ith neuron to the jth neuron, is the weight The amount of adjustment; is the bias of the jth neuron; is bias The adjustment amount; L is the loss function; η is the learning rate.

[0021] As one of the possible implementations, the loss function is the mean square error.

[0022] As one of the feasible methods, in step 4, the control performance index of the working valve control loop is:

[0023]

[0024] in, It is the control performance index of the working valve control loop; It is the oscillation index of the working valve control loop; It is an evaluation index of the working valve control loop; It is the nonlinear judgment correlation coefficient index of the working valve control loop; , and Average constant.

[0025] As one possible approach, the oscillation indicator of the operating valve control loop for:

[0026]

[0027] in, It is the distance between the first maximum value and the first two minimum values ​​in the control deviation autocorrelation diagram of the working valve control loop; It is the distance between the first minimum value and the first maximum value in the control deviation autocorrelation diagram of the working valve control loop.

[0028] As one of the achievable methods, the control deviation autocorrelation diagram of the working valve control loop is obtained according to the control deviation autocorrelation function of the working valve control loop; the control deviation autocorrelation function of the working valve control loop is:

[0029]

[0030] in, is the control deviation autocorrelation function of the working valve control loop; It is the time series of the process variable of the working valve control loop; It is the mean value of the process variable of the working valve control loop at the current moment; Is the process variable hysteresis of the working valve control loop time units later; N is the length of the sequence.

[0031] As one of the possible ways, the evaluation index of the working valve control loop for:

[0032]

[0033] in, It is the control deviation of the working valve control loop; is the reference control deviation of the working valve control loop, and is the set value.

[0034] As one possible approach, the control deviation of the operating valve control loop for:

[0035]

[0036] Where N is the length of the sequence; is the process variable of the working valve control loop at time t; It is the set value of the working valve control loop at time t.

[0037] As one of the feasible ways, the nonlinear judgment correlation coefficient index of the working valve control loop is for:

[0038]

[0039] in, = ;

[0040] The control deviation signal of the working valve control loop is at the frequency The complex spectrum component indicates that the deviation signal of the working valve control loop is at the frequency Amplitude and phase information at ;

[0041] The frequency of the deviation signal of the working valve control loop The complex spectrum component indicates that the deviation signal of the working valve control loop is at the frequency Amplitude and phase information at ;

[0042] The frequency of the deviation signal of the working valve control loop The complex spectrum component indicates that the deviation signal of the working valve control loop is at the frequency Amplitude and phase information at ;

[0043] The frequency of the deviation signal of the working valve control loop The complex conjugate of the spectral component at ;

[0044] It is expectation; is the normalized bispectral energy, with values ​​between 0 and 1.

[0045] As one of the feasible methods, step 4 also includes: comparing the control performance index of the working valve control loop with the threshold value, if it exceeds the threshold value, issuing a fault warning to instruct the operator to repair it; otherwise, scoring and displaying the normal operation of the working valve control loop according to the set normal operation scoring rules of the working valve control loop.

[0046] In order to achieve the above object, the present invention also provides a control loop evaluation device for an auxiliary steam distribution system of a nuclear power plant, comprising:

[0047] A model training module is used to train a neural network model using historical data of each valve control loop in the auxiliary steam distribution system, obtain the trained neural network model and send it to the server; the trained neural network model is used to distinguish between working valve control loop data and standby valve control loop data; the historical data of each valve control loop in the auxiliary steam distribution system includes set values, process variables, controller output signals and states, and the state of the valve control loop is working or standby;

[0048] A sensor is used to obtain real-time data of each valve control loop in the auxiliary steam distribution system and send it to the server; the real-time data of each valve control loop in the auxiliary steam distribution system includes set values, process variables and controller output signals;

[0049] The server is used to use the trained neural network model to classify the real-time data of each valve control loop in the auxiliary steam distribution system, determine the working valve control loop data and the standby valve control loop data; evaluate the working valve control loop data, and obtain the control performance index of the working valve control loop.

[0050] As one of the feasible ways, the server is also used to compare the control performance index of the working valve control loop with the threshold value. If the threshold value is exceeded, an early warning is issued to instruct the operator to repair it; otherwise, the normal operation of the working valve control loop is scored and displayed according to the set normal operation scoring rules of the working valve control loop.

[0051] In order to achieve the above-mentioned purpose, the present invention also provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the above-mentioned nuclear power plant auxiliary steam distribution system control loop evaluation method are implemented.

[0052] In order to achieve the above objectives, the present invention also provides a computer-readable storage medium, on which computer-readable instructions are stored. When the computer-readable instructions are executed, the steps of the above-mentioned nuclear power plant auxiliary steam distribution system control loop evaluation method are implemented.

[0053] Beneficial technical effects of the present invention:

[0054] The neural network-based nuclear power plant auxiliary steam distribution system control loop evaluation method, device, computer equipment and storage medium of the present invention automatically determine the one-in-use and one-in-standby status of valves in the nuclear power plant auxiliary steam distribution system, and realize accurate evaluation of the working valve control loop. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1A flowchart of an embodiment of a control loop evaluation method for an auxiliary steam distribution system of a nuclear power plant according to the present invention;

[0056] Figure 2 A schematic diagram of an embodiment of a low-pressure steam main valve control loop for an auxiliary steam distribution system of a nuclear power plant;

[0057] Figure 3 A schematic diagram of an embodiment of a main steam main pipe valve control loop of an auxiliary steam distribution system of a nuclear power plant;

[0058] Figure 4 It is a structural schematic diagram of an embodiment of a control loop evaluation device for an auxiliary steam distribution system of a nuclear power plant according to the present invention;

[0059] Figure 5 The figure is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used in the specification of the application are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned figure descriptions are intended to cover non-exclusive inclusions.

[0061] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0062] The technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings and specific embodiments.

[0063] See also Figure 1 This embodiment provides a control loop evaluation method for an auxiliary steam distribution system of a nuclear power plant, comprising the following steps:

[0064] Step 1, using the historical data of each valve control loop in the auxiliary steam distribution system to train the neural network model, to obtain a trained neural network model; the trained neural network model is used to distinguish the working valve control loop data and the standby valve control loop data; the historical data of each valve control loop in the auxiliary steam distribution system includes the set value, the process variable and the controller output signal and the state, and the state of the valve control loop is working or standby;

[0065] Step 2, obtaining real-time data of each valve control loop in the auxiliary steam distribution system; the real-time data of each valve control loop in the auxiliary steam distribution system includes set values, process variables and controller output signals;

[0066] Step 3: Use the trained neural network model to classify the real-time data of each valve control loop in the auxiliary steam distribution system, and determine the working valve control loop data and the standby valve control loop data;

[0067] Step 4: Evaluate the working valve control loop data to obtain the control performance index of the working valve control loop.

[0068] See also Figure 4 In this embodiment, as one of the achievable methods, in step 1, the neural network model is a BP neural network model; the BP neural network model is trained using the historical data of each valve control loop in the auxiliary steam distribution system, including the following steps:

[0069] Establish a BP neural network model including input layer, hidden layer and output layer;

[0070] The BP neural network model is trained by using the historical data of each valve control loop in the auxiliary steam distribution system, and the back propagation weights of the BP neural network model are updated based on the gradient descent method.

[0071] Take the historical data of a valve control loop as an example to illustrate how to train a neural network model:

[0072] The input of the neural network model is the historical data of the valve control loop; the historical data of the valve control loop is , where SP(t) is the set value time series of the valve control loop, is the time series of the process variable of the valve control loop, is the output signal time series of the valve control loop, S(t) is the working state value time series of the valve control loop; the working state value of the valve control loop includes 0 and 1, 0 indicates that the valve control loop is a standby valve control loop, and 1 indicates that the valve control loop is a working valve control loop;

[0073] In the process of training the neural network model through the historical data of the valve control loop, the neural network model is based on the valve control loop set value time series SP (t), process variable time series And the output signal time series , obtain the working state probability distribution of the valve control loop; the working state probability distribution of the valve control loop is the working state probability of the valve control loop at each moment, and the working state probability of the valve control loop includes the probability that the valve control loop is a working valve control loop and the probability that the valve control loop is a standby valve control loop;

[0074] Then the neural network model calculates the difference between the working state probability distribution of the valve control loop and the actual working state distribution through the loss function, and uses the difference as the basis for back propagation weight update based on the gradient descent method, so that the working state probability distribution of the valve control loop output by it is as close as possible to the actual working state distribution, and obtains the trained neural network model.

[0075] Take the real-time data of a valve control loop as an example to illustrate how to use the trained neural network model to determine whether the real-time data of the valve control loop is the working valve control loop data or the standby valve control loop data:

[0076] The real-time data of the valve control loop is , is the set value of the valve control loop at time t, is the process variable of the valve control loop at time t, is the output signal of the valve control loop at time t;

[0077] The real-time data of the valve control loop is input into the trained neural network model; the trained neural network model obtains the working state probability of the valve control loop according to the real-time data of the valve control loop; the working state probability of the valve control loop includes the probability that the valve control loop is a working valve control loop and the probability that the valve control loop is a standby valve control loop;

[0078] When the probability of a working valve control loop is greater than or equal to the probability of a standby valve control loop, the real-time data of the valve control loop is determined to be the working valve control loop data; when the probability of a working valve control loop is less than the probability of a standby valve control loop, the real-time data of the valve control loop is determined to be the standby valve control loop data; when the probability of a working valve control loop is equal to the probability of a standby valve control loop, the real-time data of the valve control loop is determined to be invalid data and will not be included in the subsequent control performance index evaluation.

[0079] In this embodiment, as one of the achievable methods, the back propagation weight update formula is:

[0080]

[0081]

[0082] in, is the weight connecting the ith neuron to the jth neuron, is the weight The adjustment amount, is the loss function L with respect to weights The gradient of the weights is updated to optimize the mapping between input and output; is the bias of the jth neuron; is bias The amount of adjustment; is the loss function L with respect to the bias The gradient of the bias adjustment can more flexibly control the activation of neurons; L is the loss function, which measures the difference between the predicted value and the true value of the BP neural network model; η is the learning rate, which determines the step size of each update.

[0083] Backpropagation calculates the gradient layer by layer through the chain rule and uses the gradient descent method to update the weights and biases to minimize the loss function.

[0084] In this embodiment, as one of the achievable methods, the loss function is a mean square error.

[0085] In this embodiment, as one of the achievable methods, in step 4, the control performance index of the working valve control loop is:

[0086]

[0087] in, It is the control performance index of the working valve control loop; It is the oscillation index of the working valve control loop; It is an evaluation index of the working valve control loop; It is the nonlinear judgment correlation coefficient index of the working valve control loop; , and Average constant.

[0088] See also Figure 5 In this embodiment, as one of the possible implementations, the oscillation index of the working valve control loop is for:

[0089]

[0090] in, It is the distance between the first maximum value and the first two minimum values ​​in the control deviation autocorrelation diagram of the working valve control loop; It is the distance between the first minimum value and the first maximum value in the control deviation autocorrelation diagram of the working valve control loop.

[0091] See also Figure 5 In this embodiment, as one of the achievable methods, the control deviation autocorrelation diagram of the working valve control loop is obtained according to the control deviation autocorrelation function of the working valve control loop; the control deviation autocorrelation function of the working valve control loop is:

[0092]

[0093] in, is the control deviation autocorrelation function of the working valve control loop; It is the time series of the process variable of the working valve control loop; It is the mean value of the process variable of the working valve control loop at the current moment; Is the process variable hysteresis of the working valve control loop time units later; N is the length of the sequence

[0094] In this embodiment, as one of the achievable methods, the evaluation index of the working valve control loop is for:

[0095]

[0096] in, It is the control deviation of the working valve control loop; is the reference control deviation of the working valve control loop, and is the set value.

[0097] In this embodiment, as one of the achievable methods, the control deviation of the working valve control loop is for:

[0098]

[0099] Where N is the length of the sequence; is the process variable of the working valve control loop at time t; It is the set value of the working valve control loop at time t.

[0100] In this embodiment, as one of the achievable methods, the nonlinear judgment correlation coefficient index of the working valve control loop is for:

[0101]

[0102] in, = ;

[0103] The control deviation signal of the working valve control loop is at the frequency The complex spectrum component indicates that the deviation signal of the working valve control loop is at the frequency Amplitude and phase information at ;

[0104] The frequency of the deviation signal of the working valve control loop The complex spectrum component indicates that the deviation signal of the working valve control loop is at the frequency Amplitude and phase information at ;

[0105] The frequency of the deviation signal of the working valve control loop The complex spectrum component indicates that the deviation signal of the working valve control loop is at the frequency Amplitude and phase information at ;

[0106] The frequency of the deviation signal of the working valve control loop The complex conjugate of the spectral component at ;

[0107] It is the expected value. By collecting the deviation signal samples of the working valve control loop for many times, Fourier transform is performed on each deviation signal sample of the working valve control loop to obtain each frequency signal. All frequency signals are averaged to obtain the expected value. It is used to smooth out the influence of random noise, so as to extract the stable statistical characteristics of the deviation signal of the working valve control loop.

[0108] is the normalized bispectral energy, with values ​​between 0 and 1.

[0109] For a signal at a frequency The complex spectral components of , its expected value is:

[0110]

[0111] Among them, if is the sampling frequency of the signal, then In the interval [0, fs], the interval is divided into N equally spaced frequency points, and then through a double loop, let and Take the values ​​of these frequency points to calculate .

[0112] refer to Figure 4As an implementation of the above method, the present invention provides an embodiment of a control loop evaluation device for an auxiliary steam distribution system of a nuclear power plant. The device embodiment corresponds to the embodiment of the control loop evaluation method for an auxiliary steam distribution system of a nuclear power plant, and the device can be specifically applied to various electronic devices. The control loop evaluation device for an auxiliary steam distribution system of a nuclear power plant described in this embodiment includes:

[0113] A model training module is used to train a neural network model using historical data of each valve control loop in the auxiliary steam distribution system, obtain the trained neural network model and send it to the server; the trained neural network model is used to distinguish between working valve control loop data and standby valve control loop data; the historical data of each valve control loop in the auxiliary steam distribution system includes set values, process variables, controller output signals and states, and the state of the valve control loop is working or standby;

[0114] A sensor is used to obtain real-time data of each valve control loop in the auxiliary steam distribution system and send it to the server; the real-time data of each valve control loop in the auxiliary steam distribution system includes set values, process variables and controller output signals;

[0115] The server is used to use the trained neural network model to classify the real-time data of each valve control loop in the auxiliary steam distribution system, determine the working valve control loop data and the standby valve control loop data; evaluate the working valve control loop data, and obtain the control performance index of the working valve control loop.

[0116] In this embodiment, as one of the feasible methods, the server is also used to compare the control performance index of the working valve control loop with the threshold value. If the threshold value is exceeded, an early warning is issued to instruct the operator to repair it; otherwise, the normal operation of the working valve control loop is scored and displayed according to the set normal operation scoring rules of the working valve control loop.

[0117] See also Figure 5 As an implementation of the above method, the present invention provides an embodiment of a computer device, which corresponds to the embodiment of the above-mentioned nuclear power plant auxiliary steam distribution system control loop evaluation method.

[0118] The computer device described in this embodiment includes a memory, a processor and a network interface that are interconnected and communicated through a system bus. It should be noted that the figure only shows a computer device with a memory, a processor and a network interface, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to a microprocessor, an application-specific integrated circuit, a programmable gate array, a digital processor, an embedded device, etc.

[0119] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0120] The memory includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory, random access memory, static random access memory, read-only memory, electrically erasable programmable read-only memory, programmable read-only memory, magnetic memory, disk, optical disk, etc. In some embodiments, the memory may be an internal storage unit of the computer device, such as a hard disk or memory of the computer device. In other embodiments, the memory may also be an external storage device of the computer device, such as a plug-in hard disk, smart memory card, secure digital card, flash memory card, etc. equipped on the computer device. Of course, the memory may also include both the internal storage unit of the computer device and its external storage device. In this embodiment, the memory is generally used to store the operating system and various application software installed on the computer device, such as the computer-readable instructions of the control loop evaluation method of the auxiliary steam distribution system of the nuclear power plant mentioned above. In addition, the memory may also be used to temporarily store various types of data that have been output or are to be output.

[0121] In some embodiments, the processor may be a central processing unit, a controller, a microcontroller, a microprocessor, or other data processing chip. The processor is generally used to control the overall operation of the computer device. In this embodiment, the processor is used to run the computer-readable instructions stored in the memory or process data, such as running the computer-readable instructions of the control loop evaluation method of the auxiliary steam distribution system of the nuclear power plant.

[0122] The network interface may include a wireless network interface or a wired network interface, which is generally used to establish a communication connection between the computer device and other electronic devices.

[0123] As an implementation of the above method, the present invention provides an embodiment of a computer-readable storage medium, and the embodiment of the computer-readable storage medium corresponds to the embodiment of the control loop evaluation method of the auxiliary steam distribution system of a nuclear power plant.

[0124] The computer-readable storage medium described in this embodiment stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned nuclear power plant auxiliary steam distribution system control loop evaluation method.

[0125] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A control loop evaluation method for an auxiliary steam distribution system of a nuclear power plant, characterized in that: The following steps are involved: Step 1, using the historical data of each valve control loop in the auxiliary steam distribution system to train the neural network model, to obtain a trained neural network model; the trained neural network model is used to distinguish the working valve control loop data and the standby valve control loop data; the historical data of each valve control loop in the auxiliary steam distribution system includes the set value, the process variable and the controller output signal and the state, and the state of the valve control loop is working or standby; Step 2, obtaining real-time data of each valve control loop in the auxiliary steam distribution system; the real-time data of each valve control loop in the auxiliary steam distribution system includes set values, process variables and controller output signals; Step 3: Use the trained neural network model to classify the real-time data of each valve control loop in the auxiliary steam distribution system, and determine the working valve control loop data and the standby valve control loop data; Step 4: Evaluate the working valve control loop data to obtain the control performance index of the working valve control loop.

2. The control loop evaluation method for the auxiliary steam distribution system of a nuclear power plant according to claim 1, characterized in that: In step 1, the neural network model is a BP neural network model; the BP neural network model is trained using historical data of each valve control loop in the auxiliary steam distribution system, including the following steps: Establish a BP neural network model including input layer, hidden layer and output layer; The BP neural network model is trained by using the historical data of each valve control loop in the auxiliary steam distribution system, and the back propagation weights of the BP neural network model are updated based on the gradient descent method.

3. The control loop evaluation method for the auxiliary steam distribution system of a nuclear power plant according to claim 2, characterized in that: The back propagation weight update formula is: in, is the weight connecting the ith neuron to the jth neuron, is the weight The amount of adjustment; is the bias of the jth neuron; is bias The adjustment amount; L is the loss function; η is the learning rate.

4. The control loop evaluation method for the auxiliary steam distribution system of a nuclear power plant according to claim 1, characterized in that: In step 4, the control performance index of the working valve control loop is: in, It is the control performance index of the working valve control loop; It is the oscillation index of the working valve control loop; It is an evaluation index of the working valve control loop; It is the nonlinear judgment correlation coefficient index of the working valve control loop; , and Average constant.

5. The control loop evaluation method for the auxiliary steam distribution system of a nuclear power plant according to claim 4, characterized in that: Oscillation indicator of the operating valve control loop for: in, It is the distance between the first maximum value and the first two minimum values ​​in the control deviation autocorrelation diagram of the working valve control loop; It is the distance between the first minimum value and the first maximum value in the control deviation autocorrelation diagram of the working valve control loop.

6. The control loop evaluation method for the auxiliary steam distribution system of a nuclear power plant according to claim 5, characterized in that: The control deviation autocorrelation diagram of the working valve control loop is obtained according to the control deviation autocorrelation function of the working valve control loop; the control deviation autocorrelation function of the working valve control loop is: in, is the control deviation autocorrelation function of the working valve control loop; It is the time series of the process variable of the working valve control loop; It is the mean value of the process variable of the working valve control loop at the current moment; Is the process variable hysteresis of the working valve control loop time units later; N is the length of the sequence.

7. The control loop evaluation method for the auxiliary steam distribution system of a nuclear power plant according to claim 4, characterized in that: Evaluation indicators of working valve control loop for: in, It is the control deviation of the working valve control loop; is the reference control deviation of the working valve control loop, and is the set value; Control deviation of the operating valve control circuit for: Where N is the length of the sequence; is the process variable of the working valve control loop at time t; It is the set value of the working valve control loop at time t.

8. The control loop evaluation method for the auxiliary steam distribution system of a nuclear power plant according to claim 4, characterized in that: Nonlinearity judgment correlation coefficient index of working valve control loop for: in, = ; The control deviation signal of the working valve control loop is at the frequency The complex spectrum component indicates that the deviation signal of the working valve control loop is at the frequency Amplitude and phase information at ; The frequency of the deviation signal of the working valve control loop The complex spectrum component indicates that the deviation signal of the working valve control loop is at the frequency Amplitude and phase information at ; The frequency of the deviation signal of the working valve control loop The complex spectrum component indicates that the deviation signal of the working valve control loop is at the frequency Amplitude and phase information at ; The frequency of the deviation signal of the working valve control loop The complex conjugate of the spectral component at ; It is expectation; is the normalized bispectral energy, with values ​​between 0 and 1.

9. The control loop evaluation method for the auxiliary steam distribution system of a nuclear power plant according to claim 1, characterized in that: Step 4 also includes: comparing the control performance index of the working valve control loop with the threshold value. If the threshold value is exceeded, a fault warning is issued to prompt the operator to repair it; otherwise, the normal operation of the working valve control loop is scored and displayed according to the set normal operation scoring rules of the working valve control loop.

10. A control loop evaluation device for an auxiliary steam distribution system of a nuclear power plant, characterized in that: include: A model training module is used to train a neural network model using historical data of each valve control loop in the auxiliary steam distribution system, obtain the trained neural network model and send it to the server; the trained neural network model is used to distinguish between working valve control loop data and standby valve control loop data; the historical data of each valve control loop in the auxiliary steam distribution system includes set values, process variables, controller output signals and states, and the state of the valve control loop is working or standby; A sensor is used to obtain real-time data of each valve control loop in the auxiliary steam distribution system and send it to the server; the real-time data of each valve control loop in the auxiliary steam distribution system includes set values, process variables and controller output signals; The server is used to use the trained neural network model to classify the real-time data of each valve control loop in the auxiliary steam distribution system, determine the working valve control loop data and the standby valve control loop data; evaluate the working valve control loop data, and obtain the control performance index of the working valve control loop.

11. A computer device comprising a memory and a processor, wherein the memory stores computer-readable instructions, characterized in that: When the processor executes the computer-readable instructions, the steps of the nuclear power plant auxiliary steam distribution system control loop evaluation method as described in any one of claims 1-9 are implemented.

12. A computer-readable storage medium having computer-readable instructions stored thereon, characterized in that: When the computer-readable instructions are executed, the steps of the control loop evaluation method for the auxiliary steam distribution system of a nuclear power plant are implemented as described in any one of claims 1-9.

Images