Pressurized water reactor primary loop boron concentration prediction and control method and system

Abstract

The invention discloses a pressurized water reactor primary loop boron concentration prediction and control method and system, and the method comprises the steps: obtaining an original helium flow sequence of a pressurized water reactor primary loop, and obtaining M boron concentration IMF components IMF1,..., IMFM; the M boron concentration IMF components IMF1,..., IMFM are put into a trained LSTM model for prediction, predicted values Y1, Y2,..., YM of the IMF components are obtained, and error coefficients delta Y1, delta Y2,..., delta YM corresponding to the boron concentration IMF components are calculated according to the predicted values Y1, Y2,..., YM of the IMF components; and inputting error coefficients delta Y1, delta Y2,..., delta YM corresponding to each boron concentration IMF component into the trained DenseNet prediction model to obtain an optimal IMF component predicted value # imgabs0 # of the boron concentration signal sequence, based on a DEM-DAC algorithm, controlling the boron concentration of the pressurized water reactor primary loop according to the optimal IMF component predicted value # imgabs1 # of the boron concentration signal sequence, and calculating the boron concentration of the pressurized water reactor primary loop according to the optimal IMF component predicted value # imgabs0 # of the boron concentration signal sequence. The method and the system can solve the technical problems that the boron concentration fluctuation of the primary loop is large and the unit load regulation performance is poor during the reactor variable load operation period.

Description

Technical Field

[0001] The invention belongs to the technical field of nuclear power engineering, and relates to a method and system for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor. Background Art

[0002] The control and regulation of the primary loop of a pressurized water reactor mainly involve the control of parameters such as reactor power, coolant temperature, and pressure to ensure the safe and stable operation of the reactor. By adding a soluble neutron absorber, such as boric acid, to the primary loop coolant, the reactivity is adjusted. By changing the boron concentration in the coolant, the neutron absorption degree in the core can be controlled. When the reactor is in the hot zero-power state, the boron concentration is relatively high. As the power increases, due to the negative reactivity feedback caused by the increase in core temperature, the boron concentration needs to be gradually reduced to compensate for the change in reactivity and maintain the reactor at the required power level. The boron concentration is mainly adjusted through the chemical and volume control system. When the boron concentration in the primary loop needs to be adjusted, the boric acid solution will first be transported to the volume control tank, mixed with the demineralized water in the tank and other components to form a coolant solution with an appropriate concentration, and then the charging pump sucks the coolant from the volume control tank and injects it into the primary loop system. The injection amount of the charging pump can be accurately controlled according to the specific requirements of the primary loop to effectively adjust parameters such as boron concentration, coolant volume, and reactor reactivity, ensuring the safe and stable operation of the reactor.

[0003] The existing technologies for predicting and adjusting the boron concentration in the primary loop of a pressurized water reactor have problems such as poor prediction accuracy and large fluctuations in the adjustment range, and cannot meet the requirements for the future operation of the pressurized water reactor to participate in deep peak shaving and frequency modulation of the power grid. First, in terms of signal processing, the currently used wavelet energy method (WEM) lacks self-adaptability, and wavelet transform is sensitive to noise. Especially when dealing with complex signals, phenomena such as edge blurring or loss may occur; local mean decomposition (LMD) also has some limitations. For example, when dealing with some extremely complex or severely noise-interfered signals, the decomposition results may be unstable or inaccurate. Second, in terms of parameter prediction processing, the existing particle swarm optimization algorithm (PSO) is prone to falling into local optima and has a fast convergence speed, resulting in low convergence accuracy; the recurrent neural network (RNN) is prone to problems such as gradient disappearance and gradient explosion, making it difficult for the model to learn long-distance dependencies in the sequence; finally, in terms of actuator control, traditional digital-to-analog conversion (DAC) is difficult to eliminate the non-linear errors caused by component mismatch and has low conversion accuracy. Summary of the Invention

[0004] The purpose of the present invention is to overcome the above-mentioned shortcomings of the existing technologies and provide a method and system for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor, which can solve the technical problems of large fluctuations in the boron concentration in the primary loop and poor unit load regulation performance during the variable load operation of the reactor.

[0005] To achieve the above object, the present invention discloses a method for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor, including:

[0006] Obtain the original helium gas flow rate sequence of the primary loop of the pressurized water reactor, perform SWT-LMD modal decomposition on the original helium gas flow rate sequence, reconstruct the decomposition result, and obtain M boron concentration IMF components IMF1,..., IMF M ;

[0007] Put the M boron concentration IMF components IMF1,..., IMF M into the trained LSTM model for prediction to obtain the predicted values Y1, Y2,..., Y of the IMF components M , and according to the predicted values Y1, Y2,..., Y of the IMF components M , calculate the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component M ;

[0008] Input the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component M into the trained DenseNet prediction model to obtain the optimized predicted value of the IMF component of the boron concentration signal sequence

[0009] Based on the DEM-DAC algorithm, control the boron concentration in the primary loop of the pressurized water reactor according to the optimized predicted value of the IMF component of the boron concentration signal sequence

[0010] A further improvement of the method for predicting and controlling the boron concentration in the primary loop of the pressurized water reactor according to the present invention is:

[0011] Further, the process of calculating the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component according to the predicted values Y1, Y2,..., Y of the IMF components M is as follows: M

[0012] Assume that the actual values of the IMF components corresponding to the predicted values Y1, Y2,..., Y M are M1, M2,..., M M , and based on the predicted values Y1, Y2,..., Y of each IMF component M and the actual values M1, M2,..., M of each IMF component M establish an error evaluation index D-SMAPE to obtain the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component M .

[0013] Further, based on the DEM-DAC algorithm, the predicted value of the optimized IMF component of the boron concentration signal sequence The process of controlling the boron concentration in the primary loop of a pressurized water reactor is as follows:

[0014] Based on the DEM-DAC algorithm, the predicted value of the optimized IMF component of the boron concentration signal sequence is converted into an electrical signal through a digital-to-analog conversion circuit, and then the electrical signal is input into the chemical and volume control system to control the injection amount of the charging pump, thereby controlling the boron concentration in the primary loop of the pressurized water reactor.

[0015] Further, the decomposition result is reconstructed to obtain a boron concentration reconstruction signal as:

[0016]

[0017] PF n (t) = a n (t)·cos(∫F n (t)dt)

[0018] where F(t) is a pure frequency modulation function and a(t) is an envelope estimation function.

[0019] Further, the predicted value of the optimized IMF component of the boron concentration signal sequence is:

[0020]

[0021] where w y and b y are the weight matrix and bias term of the fully connected layer, respectively.

[0022] The present invention discloses a prediction and control system for the boron concentration in the primary loop of a pressurized water reactor, including:

[0023] An acquisition module, configured to acquire the original helium flow rate sequence of the primary loop of the pressurized water reactor, perform SWT-LMD modal decomposition on the original helium flow rate sequence, and reconstruct the decomposition result to obtain M boron concentration IMF components IMF1,... IMF M ;

[0024] A first prediction module, configured to put the M boron concentration IMF components IMF1,... IMF M into a trained LSTM model for prediction to obtain the predicted values Y1, Y2,... Y M of the IMF components, and calculate the error coefficients ΔY1, ΔY2,... ΔY corresponding to each boron concentration IMF component according to the predicted values Y1, Y2,... Y M of the IMF componentsM ;

[0025] The second prediction module is used to calculate the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M Input into the trained DenseNet prediction model to obtain the optimized IMF component prediction value of the boron concentration signal sequence

[0026] Control module for optimizing IMF component prediction based on boron concentration signal sequence based on DEM-DAC algorithm Controlling the boron concentration in the primary circuit of a pressurized water reactor.

[0027] The further improvement of the PWR primary loop boron concentration prediction and control system of the present invention is:

[0028] Furthermore, the predicted values Y1, Y2, ..., Y M , calculate the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M The process is:

[0029] Assume that the predicted values Y1, Y2, ..., Y M The actual values of the corresponding IMF components are M1, M2, ..., M M , based on the predicted values of each IMF component Y1, Y2, ..., Y M and the actual values of each IMF component M1, M2, ..., M M The error evaluation index D-SMAPE is established to obtain the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component. M .

[0030] Furthermore, the DEM-DAC algorithm is used to predict the optimal IMF component of the boron concentration signal sequence. The process of controlling the boron concentration in the primary circuit of a pressurized water reactor is:

[0031] Based on the DEM-DAC algorithm, the optimized IMF component prediction value of the boron concentration signal sequence is The digital-to-analog conversion circuit converts the boron into an electrical signal, which is then input into the chemical and volume control system to control the injection volume of the charging pump and, in turn, the boron concentration in the primary circuit of the pressurized water reactor.

[0032] The present invention discloses a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for predicting and controlling the boron concentration in the primary circuit of a pressurized water reactor are implemented.

[0033] The present invention discloses a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps of the method for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor are realized.

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

[0035] When the method and system for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor according to the present invention are specifically operated, the SWT-LMD (non-local mean empirical mode decomposition method) is used to decompose the original flow rate sequence. The non-local mean denoising technology is introduced during the signal decomposition process to suppress the influence of noise on the decomposition result. The hierarchical progressive decomposition method couples the NLM and EMD technologies, overcomes the respective disadvantages of the two decomposition methods, and improves the decomposition accuracy and stability. Based on the LSTM (long short-term memory network), the predicted values of the component IMFs are obtained, and the boron concentration error coefficient is obtained by using the D-SMAPE (symmetric mean absolute percentage error algorithm) and introduced into the DenseNet (dense connection network) to obtain the optimized predicted value of the boron concentration, solving the problems of gradient disappearance and gradient explosion that are prone to occur during the training process of the traditional RNN (recurrent neural network), and effectively improving the robustness. At the same time, aiming at the problems that the traditional DAC (digital-to-analog conversion) is difficult to eliminate the non-linear error caused by component mismatch and has low conversion accuracy in the control of the actuator. The present invention is based on the DEM-DAC (dynamic error minimization digital-to-analog converter) conversion technology, which can reduce the non-linear error caused by component mismatch and effectively improve the control accuracy of the charging pump in the chemical and volume control system. Description of the Drawings

[0036] The specification drawings forming a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0037] Figure 1 It is a flowchart of the method of the present invention. Detailed Embodiments

[0038] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

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

[0040] It should also be understood that the terms used in the specification of the present invention are merely for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly dictates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0041] It should be further understood that the term "and / or" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations. For example, A and / or B can represent: the case where A exists alone, the case where A and B exist simultaneously, and the case where B exists alone. Additionally, in the present invention, the character " / " generally indicates that the objects before and after are in an "or" relationship.

[0042] It should be understood that although the terms first, second, third, etc. may be used in the embodiments of the present invention to describe preset ranges, etc., these preset ranges should not be limited to these terms. These terms are only used to distinguish the preset ranges from each other. 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 herein can be interpreted as "when" or "while" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if determined" or "if detected (stated condition or event)" can be interpreted as "when determined" or "in response to determining" or "when detected (stated condition or event)" or "in response to detecting (stated condition or event)".

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

[0045] Various structural schematic diagrams according to the disclosed embodiments of the present invention are shown in the drawings. These figures are not drawn to scale, where for the purpose of clear expression, some details are enlarged and some details may be omitted. The shapes of various regions and layers shown in the figures and their relative sizes and positional relationships are merely exemplary. In practice, there may be deviations due to manufacturing tolerances or technical limitations, and those skilled in the art can design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0046] Embodiment 1

[0047] Referring to Figure 1 , the method for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor according to the present invention includes the following steps:

[0048] 1) Obtain the original helium gas flow rate sequence of the primary loop of the pressurized water reactor, perform SWT-LMD modal decomposition on the original helium gas flow rate sequence, and reconstruct the decomposition result to obtain M boron concentration IMF components IMF1,..., IMF M ;

[0049] 2) Put the M boron concentration IMF components IMF1,..., IMF M into the trained LSTM model for prediction to obtain the predicted values Y1, Y2,..., Y M of the IMF components, where the actual values of the IMF components corresponding to the predicted values Y1, Y2,..., Y M are M1, M2,..., M M . Based on the predicted values Y1, Y2,..., Y M of the IMF components and the actual values M1, M2,..., M M of the IMF components, establish an error evaluation index D-SMAPE to obtain the error coefficients ΔY1, ΔY2,..., ΔY M

[0050] 3) Input the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M into the trained DenseNet prediction model to obtain the optimized IMF component prediction values of the boron concentration signal sequence

[0051] 4) Based on the DEM - DAC algorithm, the optimized IMF component prediction values of the boron concentration signal sequence are converted into electrical signals through a digital - to - analog conversion circuit, and then the electrical signals are input into the chemical and volume control system to control the injection amount of the charging pump, and then control the boron concentration of the primary loop of the pressurized water reactor.

[0052] The process of performing SWT - LMD modal decomposition on the original helium gas flow rate sequence in step 1) is as follows:

[0053] 11) SWT decomposition;

[0054] 111) Wavelet transform;

[0055] Perform continuous wavelet transform on the denoised primary loop boron concentration signal According to the characteristics of the boron concentration signal, select the Daubechies wavelet basis function method to obtain the wavelet coefficients W x (a, b) as:

[0056]

[0057] where a is the scale parameter, b is the translation parameter, ψ(t) is the wavelet basis function, and ψ* represents the complex conjugate.

[0058] 112) Synchrosqueezing;

[0059] Perform synchrosqueezing on the wavelet coefficients W x (a, b) to obtain a high - resolution time - frequency representation T x (a, b) as:

[0060]

[0061] where ω x (a,b) is the instantaneous frequency of the wavelet coefficient, and δ(·) is the Daubechies wavelet basis function.

[0062] 113) Mode extraction;

[0063] Extract the modal component Mk(t) of the boron concentration signal according to the time - frequency representation Tx(ω, b) as:

[0064]

[0065] where, ω k and ω k+1 are the frequency ranges of the k-th modal component.

[0066] 12) Steps for LMD decomposition of SWT decomposition components;

[0067] 121) Calculate the local mean and envelope function:

[0068] For the modal component M k (t) of the boron concentration signal, initialize the residual signal:

[0069] r(t) = M k (t)

[0070] Obtain all the local extreme points of the initialized residual boron concentration signal r(t), and then obtain the upper envelope max(r(t)) and the lower envelope min(r(t)) through linear interpolation between adjacent extreme points;

[0071] Define the local mean function m(t) as the average of the upper and lower envelopes:

[0072]

[0073] Define the envelope estimation function a(t) as half of the difference between the upper and lower envelopes:

[0074]

[0075] 122) Separate the amplitude modulation and frequency modulation components;

[0076] Subtract the local mean function m(t) from the boron concentration signal r(t), and then divide by the envelope estimation function a(t) to obtain the pure frequency modulation function F(t) as:

[0077]

[0078] 123) Construct the product function;

[0079] By integrating the pure frequency modulation function F(t), obtain the phase function, take the derivative of the phase function to obtain the instantaneous frequency, and finally construct the product function PF n (t) as:

[0080] PF n (t) = a n (t)·cos(∫F n (t)dt)

[0081] 124) Iterative decomposition:

[0082] The boron concentration signal Subtract the first product function PF1(t) to obtain the residual boron concentration signal r1(t); repeat the above steps for r i (t) to obtain the second product function PF2(t) and the remaining signal r2(t), and so on until the residual boron concentration signal r n (t) satisfies certain termination conditions, for example, the amplitude of the remaining signal is small enough or reaches the preset decomposition level.

[0083] 13) Modal reconstruction by combining SWT and LMD;

[0084] Perform the inverse wavelet transform on each component Ci after LMD decomposition and reconstruction to obtain the final reconstructed boron concentration signal as:

[0085]

[0086] The present invention provides a high-resolution time-frequency representation through synchronous compressed wavelet transform (SWT), combines the adaptive decomposition ability of local mean decomposition (LMD), and through the inverse wavelet transform, converts the signal from the time-frequency domain back to the time domain to complete the entire combination process, obtaining a signal decomposition result that has both the characteristics of wavelet transform multi-resolution analysis and the adaptive decomposition characteristics of LMD, and can more accurately extract the modal components of non-stationary signals.

[0087] In step 2), the process of putting the M boron concentration IMF components IMF1,..., IMF M into the trained LSTM model for prediction is as follows:

[0088] 21) Initialize the network boron concentration IMF parameters;

[0089] 211) Initialize the boron concentration IMF parameters of the LSTM network, including the weight matrices W f , W i , W o and W c and the bias terms b f , b i , b0, b c .

[0090] 212) Initialize the hidden state h0 and the memory cell state C0.

[0091] 22) Adaptive gating mechanism;

[0092] 221) Forget Gate:

[0093]

[0094] where, f tis the output of the forget gate at time t, σ is the Sigmoid function, and its output value ranges from 0 to 1; W f is the weight matrix of the forget gate; represents concatenating the hidden state h t-1 at the previous time step and the input x t at the current time step; b f is the bias term of the forget gate; g(t) is a time-dependent adaptive function used to dynamically adjust the weight of the forget gate; α f is the adaptive coefficient.

[0095] 222) Input Gate:

[0096] The activation value of the input gate is:

[0097]

[0098] where i t is the output of the input gate at time t, W i is the weight matrix of the input gate, b i is the corresponding bias term, α i is the adaptive coefficient.

[0099] 223) Memory cell state;

[0100] Calculate the candidate memory cell state value as:

[0101]

[0102] Calculate the candidate memory cell state value C t as:

[0103]

[0104] where the tanh function maps the output value to the range from -1 to 1, W C is the weight matrix of the candidate memory cell state, b C is the corresponding bias term.

[0105] 224) Output Gate:

[0106] Calculate the activation value of the output gate as:

[0107]

[0108] where o t is the output of the output gate at time t, W o is the weight matrix of the output gate, b o is the bias term, α ois the adaptive coefficient.

[0109] 225) Update the hidden state:

[0110] h t = o t ·tanh(C t )

[0111] Among them, the tanh function maps the output value to between -1 and 1.

[0112] 23) Multi-level memory unit;

[0113] For each level l, update the memory unit state as:

[0114]

[0115] Among them, and are the output of the forget gate and the input gate of the l-th layer respectively; is the candidate memory unit state of the l-th layer.

[0116] Calculate the multi-level hidden state as:

[0117]

[0118] Among them, is the output of the output gate of the l-th layer.

[0119] Determine the final hidden state h t as:

[0120]

[0121] Among them, β l is the weight coefficient of the l-th layer.

[0122] 24) Output the boron concentration prediction result;

[0123] Map the final hidden state h t to the boron concentration output space through the fully connected layer to obtain the boron concentration prediction result as:

[0124]

[0125] Among them, w y and b y are the weight matrix and the bias term of the fully connected layer respectively.

[0126] In step 2), based on the predicted values Y1, Y2,.., Y of each IMF component Mand the actual values M1, M2, ..., M of each IMF component M Establish the error evaluation index D - SMAPE, and obtain the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M The process is as follows:

[0127] 31) Calculate the dynamic error e of the boron concentration in the primary loop i It is:

[0128]

[0129] 32) Error aggregation;

[0130] Average all the errors in the boron concentration sequence to obtain:

[0131]

[0132] Segmented aggregation:

[0133]

[0134] 33) Output of the error coefficient;

[0135] Multiply the average value obtained in step 32) by 100% to obtain the final SMAPE value, and represent the error coefficients ΔY1, ΔY2, ..., ΔY of the boron concentration in percentage form n .

[0136] In step 3), the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M Are input into the trained DenseNet prediction model to obtain the optimized IMF component prediction values of the boron concentration signal sequence The process is as follows:

[0137] 41) Network construction;

[0138] 411) Initialize the dynamic dense block;

[0139] Set the sliding window size D;

[0140] Define the weight matrix of the gating function

[0141] 412) Configure the feature compensation path;

[0142] Add a 1×1 convolution compensator after each DenseBlock; Initialize the attenuation factor γ = 0.2;

[0143] 42) Training process;

[0144] 421) Warm-up stage (0 - 10 rounds);

[0145] fixed

[0146] Only the convolution weights in H1(·) are trained;

[0147] 422) Dynamic training phase (11-50 rounds);

[0148] Activation gating function learning;

[0149] Reduce the temperature coefficient τ according to a linear scheme:

[0150] τ=1.0-0.02×epoch

[0151] The update rules are:

[0152]

[0153] 423) Fine-tuning phase (rounds 51-100);

[0154] Freeze gate coefficient α K，1 for:

[0155] Remove the sparse regularization term (λ=0);

[0156] Fine-tune the convolution parameters;

[0157] 43) Inference deployment;

[0158] Gated pruning:

[0159] Delete α K，1 <0.1 connection path;

[0160] Quantization compression:

[0161] For the retained α K，1 Perform 8-bit fixed-point quantization:

[0162] α K，1 =round(127·α) / 127.

[0163] In step 4), based on the DEM-DAC algorithm, the optimized IMF component prediction value of the boron concentration signal sequence is The process of converting the digital-to-analog conversion circuit into an electrical signal is as follows:

[0164] 51) Adaptive weight allocation;

[0165] Dynamically assign weights to DAC components through an optimization formula to reduce the impact of component mismatch:

[0166]

[0167] Among them, ωj is the weight of the jth element, is the digital input signal of the i-th optimized boron concentration, s ij is the response of the jth element to the i-th signal.

[0168] 52) Establish a mathematical optimization model;

[0169] A mathematical model based on least squares method and stochastic optimization is established to dynamically adjust the parameters of DEM and DAC:

[0170]

[0171] Where Zi is the output electrical signal of the i-th optimized boron concentration, ω is the weight vector of the DAC element, is the actual output boron concentration electrical signal, and λ is the regularization parameter used to control the complexity of the model.

[0172] 53) Noise prediction and compensation;

[0173] The mathematical model is used to predict the quantization noise, and the compensation formula is used to reduce the effect of the noise on the boron concentration electrical signal:

[0174]

[0175] in, is the predicted quantization noise, and α is the compensation coefficient.

[0176] 54) Power consumption optimization;

[0177] Reducing power consumption of DEM and DAC through mathematical optimization:

[0178]

[0179] Among them, p j (ω J ) is the power consumption of the jth component, and M is the total number of components.

[0180] 55) Real-time correction;

[0181] Real-time correction algorithm based on mathematical model ensures the accuracy of output boron concentration electrical signal:

[0182]

[0183] Where η is the learning rate, is the gradient of the loss function.

[0184] Simulation experiment

[0185] To verify the reliability of the present invention, the obtained experimental data (i.e., the original boron concentration sequence) comes from a pressurized water reactor nuclear power plant participating in peak shaving, with a load regulation range of 60% - 100% and a test time of 1 month. The data of the first 15 days is selected, and a total of 3602 data samples at 15 - minute intervals are provided to train the prediction model. After the self - learning function of the model is completed, the data of the last 15 days is selected to test the performance of the proposed model. At the same time, to compare the advancement of the present invention, traditional PSO, WEM, LMD, and RNN are selected for comparison with the model proposed in this application.

[0186] Taking the error evaluation index SMAPE as the evaluation standard for each model, the experimental results are shown in Table 1.

[0187] Table 1

[0188]

[0189] As shown in Table 1, when compared with all other benchmark models, the present invention has the smallest error coefficient. Compared with the WEM and LMD models, the error coefficients obtained by the present invention have both decreased significantly, proving that the present invention can better improve the prediction accuracy compared with the WEM and LMD models; at the same time, when the present invention is compared with the PSO and RNN models, its error coefficients are smaller than those of the other two models, proving that the model of this application can effectively improve the prediction accuracy.

[0190] It should be noted that the present invention solves the technical problem that the primary loop boric acid flow rate is unstable and fluctuates greatly during the peak shaving and frequency modulation of the nuclear power plant, which often causes irregular changes in the prediction of the primary loop boric acid flow rate and seriously affects the safe and stable operation of the nuclear power plant. Based on the hierarchical progressive decomposition idea of complementary advantages and the combination method of automatically learning multi - scale features, a deep - learning network is established to build a prediction model, which improves the accuracy of boron concentration prediction.

[0191] Embodiment 2

[0192] The pressurized water reactor primary loop boron concentration prediction and control system of the present invention includes:

[0193] An acquisition module, configured to acquire the original helium flow rate sequence of the primary loop of the pressurized water reactor, perform SWT - LMD modal decomposition on the original helium flow rate sequence, and reconstruct the decomposition result to obtain M boron concentration IMF components IMF1,..., IMF M ;

[0194] A first prediction module, configured to put the M boron concentration IMF components IMF1,..., IMF M into the trained LSTM model for prediction to obtain the predicted values Y1, Y2,..., Y of the IMF components M, according to the predicted values of IMF components Y1, Y2, ..., Y M , calculate the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M ;

[0195] The second prediction module is used to calculate the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component. M Input into the trained DenseNet prediction model to obtain the optimized IMF component prediction value of the boron concentration signal sequence

[0196] Control module for optimizing IMF component prediction based on boron concentration signal sequence based on DEM-DAC algorithm Controlling the boron concentration in the primary circuit of a pressurized water reactor.

[0197] In this embodiment, the predicted values Y1, Y2, ..., Y M , calculate the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M The process is:

[0198] Assume that the predicted values Y1, Y2, ..., Y M The actual values of the corresponding IMF components are M1, M2, ..., M M , based on the predicted values of each IMF component Y1, Y2, ..., Y M and the actual values of each IMF component M1, M2, ..., M M The error evaluation index D-SMAPE is established to obtain the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component. M .

[0199] In this embodiment, the DEM-DAC algorithm is used to predict the optimal IMF component of the boron concentration signal sequence. The process of controlling the boron concentration in the primary circuit of a pressurized water reactor is:

[0200] Based on the DEM-DAC algorithm, the optimized IMF component prediction value of the boron concentration signal sequence is The digital-to-analog conversion circuit converts the boron into an electrical signal, which is then input into the chemical and volume control system to control the injection volume of the charging pump and, in turn, the boron concentration in the primary circuit of the pressurized water reactor.

[0201] In the embodiments of the present application, the division of modules is illustrative, merely a logical function division. In actual implementation, there may be other division methods. Additionally, in each embodiment of the present application, the various functional modules may be integrated in one processor, may exist separately physically, or two or more modules may be integrated in one module. The above integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0202] Embodiment III

[0203] 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, the steps of the method for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor are implemented. For example, it includes: obtaining the original helium gas flow rate sequence of the primary loop of a pressurized water reactor, performing SWT-LMD modal decomposition on the original helium gas flow rate sequence, reconstructing the decomposition result to obtain M boron concentration IMF components IMF1,..., IMF M ; putting the M boron concentration IMF components IMF1,..., IMF M into the trained LSTM model for prediction to obtain the predicted values Y1, Y2,..., Y of the IMF components M , and according to the predicted values Y1, Y2,..., Y of the IMF components M , calculating the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component M ; inputting the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component M into the trained DenseNet prediction model to obtain the optimized predicted value of the IMF component of the boron concentration signal sequence Based on the DEM-DAC algorithm, controlling the boron concentration in the primary loop of a pressurized water reactor according to the optimized predicted value of the IMF component of the boron concentration signal sequence Among them, the memory may include internal memory, such as high-speed random access memory, and may also include non-volatile memory, such as at least one disk memory, etc.; the processor, network interface, and memory are interconnected through an internal bus, and this internal bus can be an Industry Standard Architecture bus, a Peripheral Component Interconnect standard bus, an Extended Industry Standard Architecture bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. The memory is used to store programs. Specifically, the program may include program code, and the program code includes computer operation instructions. The memory may include internal memory and non-volatile memory and provide instructions and data to the processor.

[0204] Embodiment IV

[0205] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of the method for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor. For example, it includes: obtaining the original helium gas flow rate sequence of the primary loop of the pressurized water reactor, performing SWT-LMD modal decomposition on the original helium gas flow rate sequence, reconstructing the decomposition result to obtain M boron concentration IMF components IMF1,..., IMF M ; putting the M boron concentration IMF components IMF1,..., IMF M into a trained LSTM model for prediction to obtain the predicted values Y1, Y2,..., Y of the IMF components M , and according to the predicted values Y1, Y2,..., Y of the IMF components M , calculating the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component M ; inputting the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component M into a trained DenseNet prediction model to obtain the optimized predicted value of the boron concentration signal sequence Based on the DEM-DAC algorithm, control the boron concentration in the primary loop of the pressurized water reactor according to the optimized predicted value of the boron concentration signal sequence Specifically, the computer-readable storage medium includes but is not limited to, for example, 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 disc, magnetic disk, etc.

[0206] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code therein.

[0207] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate for implementing in the process Figure 1one or more processes and / or blocks Figure 1 means for the functions specified in one or more blocks

[0208] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction means that implements the functions in the process Figure 1 one or more processes and / or blocks Figure 1 in one or more blocks

[0209] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, such that a series of operational steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions in the process Figure 1 one or more processes and / or blocks Figure 1 in one or more blocks

[0210] Other embodiments of the present invention will be readily apparent to those skilled in the art upon consideration of the specification and the 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 known common general knowledge or conventional technical means in the technical field not disclosed by the present invention. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are pointed out by the following claims

[0211] It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes may be made without departing from its scope. The scope of the present invention is only limited by the appended claims

[0212] The above are only the preferred embodiments of the present invention, and do not limit the present invention in any way. Any simple modifications, changes, and equivalent structural changes made to the above embodiments according to the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention

Claims

1. A method for predicting and controlling the boron concentration in the primary circuit of a pressurized water reactor, characterized in that, Including: Obtain the original helium gas flow rate sequence of the primary loop of a pressurized water reactor, perform SWT-LMD modal decomposition on the original helium gas flow rate sequence, reconstruct the decomposition results, and obtain M boron concentration IMF components IMF1,... , IMF M ; Put the M boron concentration IMF components IMF1, ..., IMF M into the trained LSTM model for prediction to obtain the predicted values Y1, Y2, ..., Y of the IMF components M , and according to the predicted values Y1, Y2, ..., Y of the IMF components M , calculate the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M ; Input the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M into the trained DenseNet prediction model to obtain the optimized IMF component prediction values of the boron concentration signal sequence Based on the DEM-DAC algorithm, the predicted value of the optimized IMF component according to the boron concentration signal sequence Control the boron concentration in the primary loop of a pressurized water reactor.

2. The method for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor according to claim 1, characterized in that, The predicted values Y1, Y2,..., Y according to the IMF components M , calculate the error coefficients ΔY1, ΔY2,..., ΔY corresponding to the IMF components of each boron concentration M The process is as follows: Let the predicted values be Y1, Y2, ..., Y M The actual values of the corresponding IMF components are M1, M2, ..., M M , based on the predicted values Y1, Y2, ..., Y of each IMF component M and the actual values M1, M2, ..., M of each IMF component M Establish an error evaluation index D - SMAPE, and obtain the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M .

3. The method for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor according to claim 1, characterized in that, Based on the DEM-DAC algorithm, the predicted value of the optimized IMF component of the boron concentration signal sequence The process of controlling the boron concentration in the primary loop of a pressurized water reactor is as follows: Based on the DEM-DAC algorithm, the predicted value of the optimized IMF component of the boron concentration signal sequence is converted into an electrical signal through a digital-to-analog conversion circuit, and then the electrical signal is input into the chemical and volume control system to control the injection amount of the charging pump, thereby controlling the boron concentration in the primary loop of the pressurized water reactor.

4. The method for predicting and controlling the boron concentration in the primary circuit of a pressurized water reactor according to claim 1, wherein Reconstruct the decomposition result to obtain a boron concentration reconstruction signal which is PF n P(t) = a n P(t)·cos(∫F n (t)dt) Wherein, F(t) is a pure frequency modulation function, and a(t) is an envelope estimation function.

5. The method for predicting and controlling the boron concentration in the primary circuit of a pressurized water reactor according to claim 1, wherein The predicted value of the optimized IMF component of the boron concentration signal sequence is as follows: Among them, w y and b y are the weight matrix and bias term of the fully connected layer, respectively.

6. A primary loop boron concentration prediction and control system for a pressurized water reactor, characterized in that Including: An acquisition module, configured to acquire an original helium gas flow rate sequence of a primary loop of a pressurized water reactor, perform SWT-LMD modal decomposition on the original helium gas flow rate sequence, reconstruct the decomposition result, and obtain M boron concentration IMF components IMF1, ..., IMF M ; The first prediction module is used to put the M boron concentration IMF components IMF1,..., IMF M into the trained LSTM model for prediction, and obtain the predicted values Y1, Y2,..., Y of the IMF components M . According to the predicted values Y1, Y2,..., Y of the IMF components M , calculate the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component M ; The second prediction module is used to input the error coefficients ΔY1, ΔY2,..., ΔY corresponding to each boron concentration IMF component M into the trained DenseNet prediction model to obtain the optimized IMF component prediction values of the boron concentration signal sequence A control module, configured to predict, based on the DEM-DAC algorithm, a predicted value of an optimized IMF component of a boron concentration signal sequence, and control the boron concentration in the primary loop of a pressurized water reactor.

7. The pressurized water reactor primary loop boron concentration prediction and control system according to claim 6, characterized in that, Based on the predicted values Y1, Y2,..., Y of the IMF components M , calculate the error coefficients ΔY1, ΔY2,..., ΔY corresponding to the IMF components of each boron concentration M The process is as follows: Let the predicted values be Y1, Y2, ..., Y M The actual values of the corresponding IMF components are M1, M2, ..., M M , based on the predicted values Y1, Y2, ..., Y of each IMF component M and the actual values M1, M2, ..., M of each IMF component M Establish an error evaluation index D - SMAPE, and obtain the error coefficients ΔY1, ΔY2, ..., ΔY corresponding to each boron concentration IMF component M .

8. The pressurized water reactor primary loop boron concentration prediction and control system according to claim 6, characterized in that, Based on the DEM-DAC algorithm, the predicted value of the optimized IMF component according to the boron concentration signal sequence The process of controlling the boron concentration in the primary loop of a pressurized water reactor is as follows: Based on the DEM-DAC algorithm, the predicted value of the optimized IMF component of the boron concentration signal sequence is converted into an electrical signal through a digital-to-analog conversion circuit, and then the electrical signal is input into the chemical and volume control system to control the injection amount of the charging pump, thereby controlling the boron concentration in the primary loop of the pressurized water reactor.

9. 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, the steps of the method for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor according to any one of claims 1-5 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, the steps of the method for predicting and controlling the boron concentration in the primary loop of a pressurized water reactor according to any one of claims 1-5 are implemented.

Images