Lead-bismuth fast reactor primary circuit lead-bismuth alloy flow prediction and control method and system

Abstract

The application discloses a kind of lead bismuth fast reactor primary circuit lead bismuth alloy flow prediction and control method and system, comprising: obtaining the original lead bismuth alloy flow signal of lead-based fast reactor primary circuit;Using the HOFEM module of NL-FCNN to the lead bismuth alloy flow signal is carried out multi-scale decomposition, obtains each scale lead bismuth alloy flow component;The each scale lead bismuth alloy flow component is input into ANSA dynamic optimization network structure and is carried out feature extraction, obtains multi-scale feature vector;The multi-scale feature vector is input into NL-FCNN model, obtains the optimized lead bismuth alloy flow prediction value;According to the optimized lead bismuth alloy flow prediction value control lead-based fast reactor primary circuit lead bismuth alloy flow, the method and system can solve the technical problems that reactor variable load operation period primary circuit lead bismuth alloy flow fluctuation is big, unit load regulation performance is poor.

Description

Technical Field

[0001] This invention belongs to the field of nuclear power engineering technology and relates to a method and system for predicting and controlling the flow rate of lead-bismuth alloy in the primary loop of a lead-bismuth fast reactor. Background Technology

[0002] Lead-based fast reactors are fast neutron reactors that use molten lead or lead-bismuth alloy as a coolant. Due to their high safety, economic efficiency, and sustainability, they have become one of the mainstream fourth-generation nuclear energy candidate reactor types internationally. In lead-based fast reactors, liquid lead-bismuth alloy circulates in the reactor's primary loop as a coolant, and its flow control is crucial for the reactor's safe and stable operation. Specifically, the functions of liquid lead-bismuth alloy flow control include: 1) Maintaining stable reactor power: By adjusting the flow rate of helium-liquid lead-bismuth alloy, the reactor's power output can be controlled; 2) Ensuring safety: Liquid lead-bismuth alloy flow control helps prevent overheating or overcooling of the reactor. Appropriate flow rate ensures that the core temperature remains within a safe range, avoiding safety accidents caused by excessively high or low temperatures; 3) Optimizing performance: By precisely controlling the flow rate of liquid lead-bismuth alloy, the reactor's thermal efficiency and economic efficiency can be improved. Reasonable flow rate regulation can reduce energy loss and improve the overall system efficiency; 4) Controlling oxygen concentration: Liquid lead-bismuth alloy is corrosive to reactor structural materials, while a suitable oxygen concentration can form a dense oxide film on the pipe wall for corrosion protection. Since the solubility of solid lead oxide in lead-bismuth alloy varies with the alloy's temperature and flow rate, the dissolution and precipitation of lead oxide can be controlled by regulating the flow rate of the liquid lead-bismuth alloy, thereby adjusting the oxygen concentration in the alloy. For example, changes in flow rate will alter the dissolution rate of lead oxide, thus affecting the oxygen concentration and maintaining it within a reasonable range that provides corrosion protection without causing excessive oxidation.

[0003] Existing lead-based fast reactors employ both reactor-to-reactor and reactor-to-reactor operation modes, both of which suffer from poor reactor operational flexibility and poor transient response to rapid and significant load changes. The primary reason for this is the poor prediction accuracy and unstable control of the primary loop lead-bismuth alloy flow prediction and control technology. Regarding parameter prediction, existing PSO (Particle Swarm Optimization) algorithms are prone to getting trapped in local optima and their fast convergence speed leads to low convergence accuracy. RNN (Recurrent Neural Networks) are prone to gradient vanishing and exploding problems, making it difficult for the model to learn long-range dependencies in sequences. Traditional fully connected neural networks (FCNNs) often rely on simple activation functions (such as ReLU and Sigmoid) and fixed network structures when dealing with complex nonlinear problems, limiting their performance when handling high-dimensional, nonlinear data. Furthermore, existing methods lack dynamic adaptability in parameter optimization and feature extraction, making it difficult to effectively capture high-order nonlinear relationships in the data. Simultaneously, in terms of actuator control, traditional DAC (Digital-to-Analog Converter) struggles to eliminate nonlinear errors caused by component mismatch, resulting in low conversion accuracy. It is necessary to develop more accurate methods for predicting and controlling the primary loop lead-bismuth alloy flow rate, which is crucial for the safe and stable operation of lead-bismuth fast reactor nuclear power plants. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and system for predicting and controlling the primary loop lead-bismuth alloy flow rate of a lead-bismuth fast reactor. This method and system can solve the technical problems of large fluctuations in the primary loop lead-bismuth alloy flow rate and poor unit load regulation performance during reactor variable load operation.

[0005] To achieve the above objectives, this invention discloses a method for predicting and controlling the primary loop lead-bismuth alloy flow rate of a lead-bismuth fast reactor, comprising:

[0006] Obtain the original lead-bismuth alloy flow signal in the primary loop of the lead-based fast reactor;

[0007] The HOFEM module of NL-FCNN was used to perform multi-scale decomposition on the lead-bismuth alloy flow signal to obtain the lead-bismuth alloy flow components at each scale.

[0008] The lead-bismuth alloy flow components at each scale are input into the ANSA dynamic optimization network structure for feature extraction to obtain multi-scale feature vectors.

[0009] The multi-scale feature vectors are input into the NL-FCNN model to obtain the optimized lead-bismuth alloy flow prediction value;

[0010] The lead-bismuth alloy flow rate in the primary loop of the lead-based fast reactor is controlled based on the optimized lead-bismuth alloy flow rate prediction value.

[0011] The further improvement of the lead-bismuth alloy flow prediction and control method for the primary loop of the lead-bismuth fast reactor described in this invention lies in:

[0012] Furthermore, prior to obtaining the raw lead-bismuth alloy flow signal of the primary loop of the lead-based fast reactor, the following steps are also included:

[0013] The original lead-bismuth alloy flow signal is preprocessed to eliminate differences in dimensions and amplitude.

[0014] Furthermore, the NL-FCNN model is optimized based on RTLM and MEFA.

[0015] Furthermore, the loss function used in optimizing the NL-FCNN model based on RTLM and MEFA is:

[0016]

[0017] Among them, y i f(x) is the true value. i ;θ t ) represents the predicted value of the lead-bismuth alloy flow model.

[0018] Furthermore, the process of controlling the lead-bismuth alloy flow rate in the primary loop of the lead-based fast reactor based on the optimized lead-bismuth alloy flow rate prediction value is as follows:

[0019] The optimal lead-bismuth alloy flow prediction value is converted into an electrical signal based on the DEM algorithm. The electrical signal is then input into the drive mechanism of the lead-bismuth main pump to drive the lead-bismuth main pump to operate at a variable frequency, thereby controlling the lead-bismuth alloy flow in the primary loop of the lead-based fast reactor.

[0020] Furthermore, the electrical signal is represented as:

[0021]

[0022] Among them, v out(n) f is the simulated voltage of the lead-bismuth alloy flow rate output at the nth sampling time. i，n z is the conversion factor. l，n The input digital signal value of the lead-bismuth alloy flow rate at the nth sampling time.

[0023] This invention discloses a lead-bismuth alloy flow prediction and control system for the primary loop of a lead-bismuth fast reactor, comprising:

[0024] The acquisition module is used to acquire the original lead-bismuth alloy flow signal of the primary loop of the lead-based fast reactor;

[0025] The decomposition module is used to perform multi-scale decomposition of the lead-bismuth alloy flow signal using the HOFEM module of NL-FCNN to obtain the lead-bismuth alloy flow components at each scale.

[0026] The extraction module is used to input the lead-bismuth alloy flow components at each scale into the ANSA dynamic optimization network structure for feature extraction, and obtain multi-scale feature vectors.

[0027] The prediction module is used to input the multi-scale feature vector into the NL-FCNN model to obtain the optimal lead-bismuth alloy flow prediction value;

[0028] The control module is used to control the lead-bismuth alloy flow rate in the primary loop of the lead-based fast reactor based on the optimized lead-bismuth alloy flow rate prediction value.

[0029] The further improvement of the lead-bismuth alloy flow prediction and control system for the primary loop of the lead-bismuth fast reactor described in this invention lies in:

[0030] Furthermore, the process of controlling the lead-bismuth alloy flow rate in the primary loop of the lead-based fast reactor based on the optimized lead-bismuth alloy flow rate prediction value is as follows:

[0031] The optimal lead-bismuth alloy flow prediction value is converted into an electrical signal based on the DEM algorithm. The electrical signal is then input into the drive mechanism of the lead-bismuth main pump to drive the lead-bismuth main pump to operate at a variable frequency, thereby controlling the lead-bismuth alloy flow in the primary loop of the lead-based fast reactor.

[0032] This invention discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the lead-bismuth alloy flow prediction and control method for the primary loop of the lead-bismuth fast reactor.

[0033] This invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for predicting and controlling the primary loop lead-bismuth alloy flow rate of the lead-bismuth fast reactor.

[0034] The present invention has the following beneficial effects:

[0035] The lead-bismuth alloy flow prediction and control method and system for the primary loop of the lead-bismuth fast reactor described in this invention employs an innovative algorithm based on a nonlinear fully connected neural network (NL-FCNN) and proposes a dynamic nonlinear activation function (DNAF). This function can adaptively adjust the shape and parameters of the activation function according to the distribution characteristics of the input data. An adaptive network structure adjustment strategy based on gradient feedback (ANSA) dynamically optimizes the number of network layers and neurons. A high-order feature extraction module (HOFEM) is introduced to capture high-order nonlinear relationships in the data through multi-layer nonlinear transformation. A parameter optimization strategy based on a hybrid optimization algorithm (HOA) is used, combined with gradient descent and global optimization algorithms for parameter updates. A real-time dynamic learning mechanism (RTLM) is developed, which can dynamically adjust the model parameters and structure according to changes in the input data. By introducing the dynamic nonlinear activation mechanism, adaptive network structure adjustment, and efficient parameter optimization strategy, the modeling ability of the lead-bismuth alloy flow model for complex nonlinear data is significantly improved, greatly enhancing the flow prediction accuracy of the lead-bismuth alloy in the primary loop of the lead-bismuth fast reactor. Simultaneously, it addresses the problems of traditional DAC (digital-to-analog converter) in eliminating nonlinear errors caused by component mismatch and low conversion accuracy in actuator control. This invention is based on DEM (Dynamic Element Matching) digital-to-analog conversion technology, which can reduce nonlinear errors caused by component mismatch and effectively improve the control accuracy of lead-bismuth main pump. Attached Figure Description

[0036] The accompanying drawings, which form part of this specification, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0037] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] In the description of this invention, it should be understood that the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0040] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0041] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Additionally, the character " / " in this invention generally indicates that the preceding and following objects have an "or" relationship.

[0042] It should be understood that although terms such as first, second, third, etc., may be used in the embodiments of the present invention to describe the preset range, these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from one another. For example, without departing from the scope of the embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0043] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0045] The accompanying drawings illustrate various structural schematic diagrams according to embodiments disclosed in this invention. These drawings are not to scale, and some details have been enlarged for clarity, and some details may have been omitted. The shapes of the various regions and layers shown in the drawings, as well as their relative sizes and positional relationships, are merely exemplary and may deviate from reality due to manufacturing tolerances or technical limitations. Furthermore, those skilled in the art can design regions / layers with different shapes, sizes, and relative positions as needed.

[0046] Example 1

[0047] refer to Figure 1 The method for predicting and controlling the lead-bismuth alloy flow rate in the primary loop of a lead-bismuth fast reactor according to the present invention includes the following steps:

[0048] 1) Obtain raw lead-bismuth alloy flow signals from the primary loop of several lead-based fast reactors, preprocess each raw lead-bismuth alloy flow signal to eliminate differences in dimensions and amplitude; divide all preprocessed lead-bismuth alloy flow signals into training and testing sets for model training and validation.

[0049] 2) The HOFEM module of NL-FCNN is used to decompose the lead-bismuth alloy flow signal into multiple scales to obtain the lead-bismuth alloy flow components at each scale. Then, DNAF is used to enhance the feature extraction capability of the lead-bismuth alloy flow components at each scale.

[0050] 3) The multi-scale lead-bismuth alloy flow components are input into the ANSA dynamic optimization network structure for feature extraction, and the parameters of the lead-bismuth alloy flow model are optimized based on the HOA algorithm to capture the nonlinear features in the lead-bismuth alloy flow signal.

[0051] 4) Based on the extracted multi-scale feature vectors, the NL-FCNN model is used to predict the lead-bismuth alloy flow signal. The parameters of the model are dynamically adjusted by RTLM to adapt to the changes in the lead-bismuth alloy flow signal. Error iterative correction is performed based on MEFA to obtain the optimal lead-bismuth alloy flow prediction value.

[0052] 5) Based on the DEM algorithm, the optimized lead-bismuth alloy flow prediction value is converted into an electrical signal, and then the electrical signal is input into the drive mechanism of the lead-bismuth main pump to drive the lead-bismuth main pump to operate at a variable frequency, thereby controlling the lead-bismuth alloy flow in the primary loop of the lead-based fast reactor.

[0053] The process of using the HOFEM module of NL-FCNN to perform multi-scale decomposition of the lead-bismuth alloy flow signal is as follows:

[0054] 11) Signal preprocessing;

[0055] 111) Standardized processing;

[0056] The original lead-bismuth alloy flow signal x(t) is standardized to eliminate dimensional and amplitude differences, enabling data to be modeled and analyzed on the same scale. This transforms the original lead-bismuth alloy flow signal into a distribution with a mean of 0 and a standard deviation of 1.

[0057]

[0058] Where μ is the mean of the signal, σ is the standard deviation of the signal, and x std (t) represents the standardized signal.

[0059] 112) Model training and validation

[0060] Determine the proportion α of the lead-bismuth alloy flow training set: usually α = 0.8.

[0061] Calculate the length of the training set T for the lead-bismuth alloy flow rate train :

[0062] The standardized lead-bismuth alloy flow signal was divided into training and testing sets for model training and validation.

[0063] Training set: x train (t)=x(t), t=1, 2,...,T train

[0064] Test set: x test (t) = x(t), t = T train +1, T train +2, ..., N

[0065] 12) Multiscale signal decomposition;

[0066] The preprocessed lead-bismuth alloy flow signal was decomposed into multiple scales using the High-Order Feature Extraction Module (HOFEM) of NL-FCNN.

[0067]

[0068] Among them, c i r(t) represents the lead-bismuth alloy flow signal component at the i-th scale, and r(t) represents the residual component.

[0069] The feature extraction capability of lead-bismuth alloy flow signal components at each scale is enhanced by using a dynamic nonlinear activation function (DNAF).

[0070] f(x;α,β)=α·tanh(β·x)+(1-α)·ReLU(x)

[0071] Where X is the input data, α is the dynamic weight parameter with a value range of [0, 1], used to control the ratio of the tanh function and the ReLU function, and β is the nonlinear adjustment parameter used to control the degree of nonlinearity of the tanh function.

[0072] The dynamic parameters are adjusted as follows:

[0073]

[0074] Gradient calculation:

[0075]

[0076] 13) Feature extraction and modeling;

[0077] 131) Feature extraction;

[0078] For each scale of the lead-bismuth alloy flow signal component c after multi-scale decomposition i (t) Perform feature extraction:

[0079] 1311) Dynamically optimize the network structure using the Adaptive Network Structure Adjustment Strategy (ANSA) of NL-FCNN:

[0080]

[0081] Among them, L old L represents the current network layer number. new The adjusted number of network layers is η, where η is the learning rate, controlling the step size for adjusting the network structure. loss function The gradient with respect to the number of network layers L.

[0082] 1312) The parameters of the model are optimized by the Hybrid Optimization Algorithm (HOA) to capture the nonlinear characteristics in the lead-bismuth alloy flow signal.

[0083] The process of optimizing the parameters of the lead-bismuth alloy flow model based on the HOA algorithm is as follows:

[0084] 21) Initialize parameters;

[0085] Initialize model parameters, lead-bismuth alloy flow signal, and genetic algorithm population.

[0086] 22) Gradient descent optimization;

[0087] Update the lead-bismuth alloy flow signal parameters using the gradient descent method:

[0088]

[0089] 23) Genetic algorithm optimization;

[0090] Selection: Select individuals with higher fitness from the population.

[0091] Crossover: Generate new individuals through crossover operations.

[0092] Mutation: Introducing new genes through mutation operations.

[0093]

[0094] Among them, GA(θ t ) is a genetic algorithm for the lead-bismuth alloy flow signal parameter θ t The optimization result; λ is the hybrid weight parameter used to control the ratio of gradient descent and genetic algorithm.

[0095] 24) Repeat steps 22) to 23) until the parameters converge or the maximum number of iterations is reached.

[0096] 25) The feature vector of the multi-scale lead-bismuth alloy flow signal is constructed as follows:

[0097] F = [f1, f2...f n ]

[0098] Among them, f i Let be the eigenvector of the i-th scale component.

[0099] 26) Signal prediction;

[0100] Based on the extracted multi-scale lead-bismuth alloy flow feature vector F, signal prediction is performed using NL-FCNN:

[0101]

[0102] in, This is the predicted value of the signal.

[0103] 27) Dynamic adjustment of model parameters;

[0104] The parameters of the lead-bismuth alloy flow model are dynamically adjusted through a real-time dynamic learning mechanism (RTLM) to adapt to changes in the signal.

[0105] The process of dynamically adjusting the model parameters using RTLM is as follows:

[0106] 31) Monitor signal changes in real time;

[0107] Monitoring input lead-bismuth alloy flow signal x t Calculate the change ||Δx t ||.

[0108] 32) Calculate the dynamic learning rate;

[0109] RTLM uses a dynamic learning rate ηt It adaptively adjusts based on changes in the input lead-bismuth alloy flow rate signal, and its mathematical expression is:

[0110] η t =η0·exp(-γ·||Δx t ||)

[0111] Where η0 is the initial learning rate, γ is the decay coefficient that controls the adjustment speed of the learning rate, and ||Δx t || represents the change in the input signal, i.e.:

[0112] ||Δx t ||=||x t -x t-1 ||

[0113] 33) Update model parameters;

[0114] RTLM dynamically updates the input lead-bismuth alloy flow model parameter θ using the gradient descent method. t ,Right now:

[0115]

[0116] Where, θ t Let η be the parameters of the lead-bismuth alloy flow model for the t-th iteration. t For dynamic learning rate, For the loss function L with respect to the parameter θ t The gradient, based on the current input lead-bismuth alloy flow rate signal x t calculate.

[0117] The loss function L is typically chosen as the mean squared error (MSE) or cross-entropy, i.e.:

[0118]

[0119] Among them, y i f(x) is the true value. i ;θ t ) represents the predicted value of the lead-bismuth alloy flow model.

[0120] 34) Repeated real-time learning;

[0121] Repeat steps 31) to 33) to continuously adapt to changes in the lead-bismuth alloy flow signal.

[0122] 35) Results evaluation and optimization.

[0123] The process of converting the optimized lead-bismuth alloy flow prediction value into an electrical signal based on the DEM algorithm (multi-dimensional error fusion algorithm) is as follows:

[0124] 41) Calculate the single-dimensional error;

[0125] Calculate the one-dimensional error e for the flow rate of each lead-bismuth alloy i，j :

[0126] e i，j =g j (y i f(x) i ;θ t )

[0127] Among them, g j (·) is the calculation function for the j-th error dimension, where the mean absolute error is taken;

[0128] 42) Multi-dimensional error fusion;

[0129] The errors from each dimension are weighted and fused to obtain the multi-dimensional error E. i :

[0130]

[0131] Where, β j is the weighting coefficient for the j-th lead-bismuth alloy flow error dimension.

[0132] Optimize the parameters and structure of the lead-bismuth alloy flow model based on the error results, and repeat the above operation until the optimal lead-bismuth alloy flow prediction value with the smallest error is obtained.

[0133] The process of converting the optimized lead-bismuth alloy flow prediction value into an electrical signal based on the DEM algorithm is as follows:

[0134] 51) Adaptive weight allocation, i.e.:

[0135]

[0136] in, Let be the dynamic characteristic function of the i-th input lead-bismuth alloy flow rate digital signal, used to measure the adaptability of the lead-bismuth alloy flow rate to the input signal; α is the adaptive coefficient, which controls the sensitivity of weight allocation.

[0137] 52) Multi-level mismatch compensation;

[0138] A multi-level mismatch compensation mechanism is introduced, dividing mismatch compensation into multiple levels {z1, z2, ..., z...}. L Each level corresponds to a different mismatch compensation strategy, which is then weighted and fused to obtain:

[0139]

[0140] Among them, z iFor the output of the digital signal of the i-th lead-bismuth alloy flow rate, ω l，i This represents the weight of the i-th element in the first level.

[0141]

[0142] 53) The output electrical signal is:

[0143]

[0144] Among them, v out(n) f is the simulated voltage of the lead-bismuth alloy flow rate output at the nth sampling time. i，n z is the conversion factor. l，n The input digital signal value of the lead-bismuth alloy flow rate at the nth sampling time.

[0145] Instance verification

[0146] To verify the reliability of this invention, the experimental data (i.e., the original lead-bismuth alloy flow rate) was obtained from a lead-based fast reactor demonstration unit. This lead-bismuth fast reactor had a load adjustment range of 60%-100%, and the testing period was one month. Data from the first 15 days was selected, and a total of 1252 data samples at 15-minute intervals were provided to train the prediction model. After the model's self-learning function was completed, data from the following 15 days were used to test the performance of the proposed model. Simultaneously, to compare the advancement of this invention, a multi-dimensional error fusion algorithm (MEFA) was used for evaluation, comparing it with traditional PSO, WEM, and RNN algorithms.

[0147] Using the error evaluation index SMAPE as the evaluation standard for each model, the experimental results are shown in Table 1.

[0148] Table 1

[0149] MEFA (%) 47.11 32.10 28.09 12.01

[0150] As shown in Table 1, this invention exhibits the smallest error coefficient compared to all other benchmark models. Compared to the PSO model, the error coefficients obtained by RNN, FCNN, and this invention are all significantly reduced, demonstrating that this invention can improve prediction accuracy better than the PSO model. Furthermore, compared to the RNN and FCNN models, the error coefficients of this invention are smaller than the other two models, proving that the algorithm model based on the NL-FCNN nonlinear fully connected neural network can improve the prediction accuracy of lead-bismuth alloy flow in the primary loop of lead-based fast reactors.

[0151] Example 2

[0152] The lead-bismuth fast reactor primary loop lead-bismuth alloy flow prediction and control system of the present invention includes:

[0153] The acquisition module is used to acquire the original lead-bismuth alloy flow signal of the primary loop of the lead-based fast reactor;

[0154] The decomposition module is used to perform multi-scale decomposition of the lead-bismuth alloy flow signal using the HOFEM module of NL-FCNN to obtain the lead-bismuth alloy flow components at each scale.

[0155] The extraction module is used to input the lead-bismuth alloy flow components at each scale into the ANSA dynamic optimization network structure for feature extraction, and obtain multi-scale feature vectors.

[0156] The prediction module is used to input the multi-scale feature vector into the NL-FCNN model to obtain the optimal lead-bismuth alloy flow prediction value;

[0157] The control module is used to control the lead-bismuth alloy flow rate in the primary loop of the lead-based fast reactor based on the optimized lead-bismuth alloy flow rate prediction value.

[0158] In this embodiment, the process of controlling the lead-bismuth alloy flow rate in the primary loop of the lead-based fast reactor based on the optimized lead-bismuth alloy flow rate prediction value is as follows:

[0159] The optimal lead-bismuth alloy flow prediction value is converted into an electrical signal based on the DEM algorithm. The electrical signal is then input into the drive mechanism of the lead-bismuth main pump to drive the lead-bismuth main pump to operate at a variable frequency, thereby controlling the lead-bismuth alloy flow in the primary loop of the lead-based fast reactor.

[0160] The module division in this embodiment is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in each embodiment of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0161] Example 3

[0162] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a method for predicting and controlling the lead-bismuth alloy flow rate in the primary loop of a lead-bismuth fast reactor. For example, the method includes: acquiring the original lead-bismuth alloy flow rate signal of the primary loop of a lead-based fast reactor; performing multi-scale decomposition on the lead-bismuth alloy flow rate signal using the HOFEM module of NL-FCNN to obtain lead-bismuth alloy flow rate components at each scale; inputting the lead-bismuth alloy flow rate components at each scale into an ANSA dynamic optimization network structure for feature extraction to obtain multi-scale feature vectors; inputting the multi-scale feature vectors into an NL-FCNN model to obtain an optimized predicted lead-bismuth alloy flow rate value; and controlling the lead-bismuth alloy flow rate of the primary loop of the lead-based fast reactor according to the optimized predicted lead-bismuth alloy flow rate value. The memory may include main memory, such as high-speed random access memory (RAM), or non-volatile memory, such as at least one disk storage device. The processor, network interface, and memory are interconnected via an internal bus, which may be an industry-standard architecture bus, a peripheral component interconnection standard bus, or an extended industry-standard architecture bus. The bus can be categorized as an address bus, data bus, or control bus. The memory stores programs; specifically, the program may include program code, which includes computer operation instructions. The memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0163] Example 4

[0164] A computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of a method for predicting and controlling the lead-bismuth alloy flow rate in the primary loop of a lead-bismuth fast reactor. For example, the method includes: acquiring the original lead-bismuth alloy flow rate signal of the primary loop of a lead-based fast reactor; performing multi-scale decomposition on the lead-bismuth alloy flow rate signal using the HOFEM module of NL-FCNN to obtain lead-bismuth alloy flow rate components at each scale; inputting the lead-bismuth alloy flow rate components at each scale into an ANSA dynamic optimization network structure for feature extraction to obtain multi-scale feature vectors; inputting the multi-scale feature vectors into an NL-FCNN model to obtain an optimized predicted lead-bismuth alloy flow rate value; and controlling the lead-bismuth alloy flow rate of the primary loop of the lead-based fast reactor according to the optimized predicted lead-bismuth alloy flow rate value. Specifically, the computer-readable storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. The volatile memory may include random access memory and / or cache memory, etc. The non-volatile memory may include read-only memory, hard disk, flash memory, optical disk, magnetic disk, etc.

[0165] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0166] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0167] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0168] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0169] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and disclosure of the invention. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0170] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

[0171] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any simple modifications, alterations, or equivalent structural changes made to the above embodiments based on the technical essence of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for predicting and controlling the primary loop lead-bismuth alloy flow rate of a lead-bismuth fast reactor, characterized in that, include: Obtain the original lead-bismuth alloy flow signal in the primary loop of the lead-bismuth fast reactor; The high-order feature extraction module of the nonlinear fully connected neural network model is used to perform multi-scale decomposition on the lead-bismuth alloy flow signal to obtain the lead-bismuth alloy flow components at each scale. The lead-bismuth alloy flow components at each scale are input into a dynamically optimized network structure based on an adaptive network structure adjustment strategy with gradient information feedback for feature extraction, resulting in multi-scale feature vectors. The multi-scale feature vectors are input into a nonlinear fully connected neural network model to obtain the optimal lead-bismuth alloy flow prediction value. The lead-bismuth alloy flow rate in the primary loop of the lead-bismuth fast reactor is controlled based on the optimized predicted flow rate. Optimization of nonlinear fully connected neural network models based on real-time dynamic learning mechanisms and multi-dimensional error fusion algorithms; The loss function for optimizing a nonlinear fully connected neural network model based on a real-time dynamic learning mechanism and a multi-dimensional error fusion algorithm is: in, y i For the true value, f (x i ; θ t () represents the predicted value from the lead-bismuth alloy flow model; The process of controlling the lead-bismuth alloy flow rate in the primary loop of the lead-bismuth fast reactor based on the optimized lead-bismuth alloy flow rate prediction value is as follows: The optimized lead-bismuth alloy flow prediction value is converted into an electrical signal based on the dynamic element matching algorithm. The electrical signal is then input into the drive mechanism of the lead-bismuth main pump to drive the lead-bismuth main pump to operate at a variable frequency, thereby controlling the lead-bismuth alloy flow rate in the primary loop of the lead-bismuth fast reactor. The electrical signal is represented as: in, The simulated voltage for lead-bismuth alloy flow rate output at the nth sampling time. For conversion factors, The input digital signal value of the lead-bismuth alloy flow rate at the nth sampling time.

2. The method for predicting and controlling the lead-bismuth alloy flow rate in the primary loop of a lead-bismuth fast reactor according to claim 1, characterized in that, Before obtaining the raw lead-bismuth alloy flow signal of the primary loop of the lead-bismuth fast reactor, the following steps are also required: The original lead-bismuth alloy flow signal is preprocessed to eliminate differences in dimensions and amplitude.

3. A lead-bismuth alloy flow prediction and control system for the primary loop of a lead-bismuth fast reactor, characterized in that, include: The acquisition module is used to acquire the original lead-bismuth alloy flow signal in the primary loop of the lead-bismuth fast reactor. The decomposition module is used to perform multi-scale decomposition of the lead-bismuth alloy flow signal using a high-order feature extraction module with a nonlinear fully connected neural network model to obtain lead-bismuth alloy flow components at each scale. The extraction module is used to input the lead-bismuth alloy flow components at each scale into a dynamically optimized network structure based on an adaptive network structure adjustment strategy with gradient information feedback for feature extraction, thereby obtaining multi-scale feature vectors. The prediction module is used to input the multi-scale feature vector into a nonlinear fully connected neural network model to obtain the optimal lead-bismuth alloy flow prediction value. The control module is used to control the lead-bismuth alloy flow rate in the primary loop of the lead-bismuth fast reactor based on the optimized lead-bismuth alloy flow rate prediction value; Optimization of nonlinear fully connected neural network models based on real-time dynamic learning mechanisms and multi-dimensional error fusion algorithms; The loss function for optimizing a nonlinear fully connected neural network model based on a real-time dynamic learning mechanism and a multi-dimensional error fusion algorithm is: in, y i For the true value, f (x i ; θ t () represents the predicted value from the lead-bismuth alloy flow model; The process of controlling the lead-bismuth alloy flow rate in the primary loop of the lead-bismuth fast reactor based on the optimized lead-bismuth alloy flow rate prediction value is as follows: The optimized lead-bismuth alloy flow prediction value is converted into an electrical signal based on the dynamic element matching algorithm. The electrical signal is then input into the drive mechanism of the lead-bismuth main pump to drive the lead-bismuth main pump to operate at a variable frequency, thereby controlling the lead-bismuth alloy flow rate in the primary loop of the lead-bismuth fast reactor. The electrical signal is represented as: in, The simulated voltage for lead-bismuth alloy flow rate output at the nth sampling time. For conversion factors, The input digital signal value of the lead-bismuth alloy flow rate at the nth sampling time.

4. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the lead-bismuth alloy flow prediction and control method for the primary loop of the lead-bismuth fast reactor as described in any one of claims 1-2.

5. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the lead-bismuth alloy flow prediction and control method for the primary loop of the lead-bismuth fast reactor as described in any one of claims 1-2.

Images