Method and system for obtaining an evaluation parameter of a numerical reactor

Abstract

The embodiment of the application discloses a kind of methods and systems for obtaining the evaluation parameter of numerical reactor, it is related to data processing technical field, method includes: based on the influence parameter of obtaining numerical reactor and its probability distribution type, determine the base function combination term of the influence parameter;According to sampling result and the base function combination term, determine polynomial base function matrix and order reduction coefficient, the sampling result is according to the probability distribution indicated by the probability distribution type to the influence parameter and is obtained by sampling;According to the polynomial base function matrix, the order reduction coefficient, the influence parameter and the probability distribution corresponding to the influence parameter, determine the evaluation parameter of the numerical reactor, by sampling and the order reduction coefficient is involved in the calculation of evaluation parameter, it can reduce intermediate calculation amount, to improve the efficiency of determining evaluation parameter in turn.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method and system for obtaining evaluation parameters of a numerical reactor. Background Technology

[0002] Numerical reactor multiphysics coupling simulation is an important research direction in the field of nuclear energy. The simulation process typically involves a large number of input parameters, and the uncertainty of these input parameters has a significant impact on the reliability and accuracy of the simulation results. In the field of engineering analysis, sensitivity analysis, as a part of uncertainty analysis, is a method used to assess and understand the degree of influence of uncertain factors, such as input parameters, on the output results of an engineering system or model. Related techniques use uncertainty analysis methods to obtain evaluation parameters that can be used to analyze the degree of influence of input parameters, but the computational efficiency is low.

[0003] Application content

[0004] In view of this, one of the objectives of this application is to provide a method and system for obtaining evaluation parameters of a numerical reactor, which can improve the efficiency of obtaining evaluation parameters of a numerical reactor.

[0005] To achieve the above objectives, the technical solution of this application is implemented as follows:

[0006] In a first aspect, embodiments of this application provide a method for obtaining evaluation parameters of a numerical reactor, including:

[0007] Obtain the influence parameters of the numerical reactor and their probability distribution types;

[0008] Based on the influencing parameters and their probability distribution types, determine the basis function combination terms of the influencing parameters;

[0009] Based on the sampling results and the basis function combination terms, the polynomial basis function matrix and the order reduction coefficients are determined. The sampling results are obtained by sampling the influencing parameters according to the probability distribution indicated by the probability distribution type.

[0010] The evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters.

[0011] In one possible implementation, after determining the basis function combination terms of the influencing parameters based on the influencing parameters and their probability distribution types, the method further includes:

[0012] Based on the comparison results of the first and second quantities, determine the target calculation strategy;

[0013] Based on the target computation strategy, determine the coefficients of each polynomial basis function in the polynomial basis function matrix;

[0014] The first quantity includes the number of samples in the sampling results, and the second quantity includes the number of terms in the basis function combination.

[0015] The evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, reduction coefficients, influence parameters, and their corresponding probability distributions. These parameters include:

[0016] The evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function in the polynomial basis function matrix, the reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters.

[0017] In one possible implementation, determining the target calculation strategy based on the comparison result between the first quantity and the second quantity includes:

[0018] If the comparison results indicate that the first quantity is greater than or equal to the second quantity, the least squares method is determined as the target calculation strategy.

[0019] Given that the least squares method is determined as the target computational strategy, the coefficients of the polynomial basis functions are determined according to the least squares method.

[0020] If the comparison results indicate that the first quantity is less than the second quantity, the sparse matrix method is determined as the target calculation strategy.

[0021] Given that the sparse matrix method is the target computational strategy, the coefficients of the polynomial basis functions are determined according to the sparse matrix method.

[0022] In one possible implementation, the evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function in the polynomial basis function matrix, the reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters, including:

[0023] The average value of the reduction coefficients is determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function, the influencing parameters, and the probability distributions corresponding to the influencing parameters.

[0024] Determine the variance of the reduction coefficients based on the average value and the reduction coefficients themselves.

[0025] The average value of the physical field of the numerical reactor is determined based on the average value of the reduced-order coefficients and the polynomial basis function matrix.

[0026] The variance of the physical field of the numerical reactor is determined based on the variance of the reduced-order coefficients and the polynomial basis function matrix.

[0027] The evaluation parameters include the mean value and variance of the physical fields of the numerical reactor.

[0028] In one possible implementation, after determining the polynomial basis functions and order reduction coefficients based on the sampling results and the basis function combination terms, the method further includes:

[0029] Based on the sampling results and the numerical reactor simulation model, the key physical quantity matrix was determined;

[0030] If the number of key physical quantities in the key physical quantity matrix is ​​greater than or equal to a preset number threshold, the key physical quantity matrix is ​​dimensionality reduced.

[0031] Given that the least squares method is chosen as the target computational strategy, the coefficients of the polynomial basis functions are determined using the least squares method, including:

[0032] For the key physical quantity matrix after dimensionality reduction, the coefficients of the polynomial basis functions are determined using the least squares method.

[0033] In one possible implementation, the key physical quantity matrix is ​​subjected to dimensionality reduction processing, including:

[0034] Based on the key physical quantity matrix, determine the relevant matrix;

[0035] Based on the non-zero eigenvalues ​​and eigenvectors of the correlation matrix, determine the eigenorthogonal decomposition modes of the correlation matrix;

[0036] Based on the intrinsic orthogonal decomposition modes and the key physical quantity matrix, the key physical quantity matrix after dimensionality reduction is determined.

[0037] In one possible implementation, the basis function combination term of the influencing parameters is determined based on the influencing parameters and the probability distribution type of the influencing parameters, including:

[0038] Given a defined dimension of the influencing parameters, the basis function combination terms are determined based on the dimension of the influencing parameters and the preset truncation order.

[0039] Secondly, embodiments of this application provide a system for obtaining evaluation parameters of a numerical reactor, the system comprising:

[0040] The acquisition module is used to acquire the influence parameters of a numerical reactor and their probability distribution types.

[0041] The first determining module is used to determine the basis function combination terms of the influencing parameters based on the influencing parameters and their probability distribution types.

[0042] The second determining module is used to determine the polynomial basis function matrix and the reduced order coefficients based on the sampling results and the basis function combination terms. The sampling results are obtained by sampling the influencing parameters according to the probability distribution indicated by the probability distribution type.

[0043] The third determination module is used to determine the evaluation parameters of the numerical reactor based on the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters.

[0044] Thirdly, embodiments of this application provide an electronic device, which includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it implements the method for obtaining evaluation parameters of a numerical reactor provided in the first aspect.

[0045] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program, which, when executed by one or more processors, implements the method for obtaining evaluation parameters of a numerical reactor provided in the first aspect.

[0046] The method for obtaining evaluation parameters of a numerical reactor provided in this application can determine the basis function combination terms of the influence parameters and their probability distribution types based on the obtained influence parameters of the numerical reactor. Then, based on the sampling results and the basis function combination terms, a polynomial basis function matrix and order reduction coefficients are determined. The sampling results are obtained by sampling the influence parameters according to the probability distribution indicated by the probability distribution type. Finally, the evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters. By sampling and incorporating the order reduction coefficients into the calculation of evaluation parameters, the amount of intermediate calculations can be reduced, thereby improving the efficiency of determining the evaluation parameters. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. It should be understood that the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A flowchart illustrating a method for obtaining evaluation parameters of a numerical reactor, provided as an embodiment of this application;

[0049] Figure 2 A graph showing the change of line power density over time, provided as an embodiment of this application;

[0050] Figure 3 A line graph of average values ​​obtained using Monte Carlo methods is provided as an embodiment of this application;

[0051] Figure 4 A variance line graph obtained using the Monte Carlo method is provided as an embodiment of this application;

[0052] Figure 5 An energy change diagram involved in a method for obtaining evaluation parameters of a numerical reactor provided in an embodiment of this application;

[0053] Figure 6 A probability density variation curve is provided as part of a method for obtaining evaluation parameters of a numerical reactor, as described in an embodiment of this application.

[0054] Figure 7 Another probability density variation curve is provided for a method of obtaining evaluation parameters of a numerical reactor according to an embodiment of this application.

[0055] Figure 8 Comparative diagrams related to a method for obtaining evaluation parameters of a numerical reactor provided in an embodiment of this application;

[0056] Figure 9 A schematic diagram of the functional modules of a system for obtaining evaluation parameters of a numerical reactor, provided for an embodiment of this application;

[0057] Figure 10 This is a diagram illustrating the internal structure of an electronic device as provided in an embodiment of this application.

[0058] Explanation of reference numerals in the attached figures:

[0059] The system 900 for acquiring evaluation parameters of a numerical reactor includes an acquisition module 910, a first determination module 920, a second determination module 930, and a third determination module 940. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0061] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0062] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0063] In various embodiments of this application, the expression "or" or "at least one of A and / or B" includes any combination or all combinations of the words listed simultaneously. For example, the expression "A or B" or "at least one of A and / or B" may include A, may include B, or may include both A and B.

[0064] In the description of this application, it should be noted that if terms such as "upper," "lower," "inner," or "outer" are used to indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of the invention is usually placed during use, they are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0065] Furthermore, the terms "first" and "second" are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0066] It should be noted that, where there is no conflict, the features in the embodiments of this application can be combined with each other.

[0067] Furthermore, in the embodiments of this application, the term "connection" can refer to "electrical connection" or "direct connection." "Electrical connection" can refer to a direct electrical connection between two components, or it can refer to an electrical connection between two components via one or more normally open tubes or other components.

[0068] To facilitate a better understanding of the solutions in the embodiments of this application, the relevant technical terms will be introduced first below.

[0069] A numerical reactor is a tool that uses computer simulation technology to perform detailed numerical simulations of the physical, thermal, and hydraulic processes of a nuclear reactor. A numerical reactor can be regarded as a "virtual reactor" that reproduces various phenomena inside a nuclear reactor through complex mathematical models and calculation algorithms.

[0070] The Monte Carlo method, also known as the statistical simulation method, is a numerical computation method guided by probability and statistics theory. It solves various mathematical and physical problems through random sampling, and is particularly widely used in solving complex integrals, solving differential equations, and simulating systems with uncertainties.

[0071] To address the technical problems in the background art, embodiments of this application provide a method and system for obtaining evaluation parameters of a numerical reactor. The method for obtaining evaluation parameters of a numerical reactor provided in this application will be described first.

[0072] Please see Figure 1, Figure 1 This application provides a flowchart of a method for obtaining evaluation parameters of a numerical reactor. This method can be applied to systems and electronic devices used in the following embodiments for obtaining evaluation parameters of a numerical reactor. These electronic devices include personal computers, servers, mobile devices, cloud computing platforms, and supercomputers. The method for obtaining evaluation parameters of a numerical reactor will be described below from the perspective of its application in electronic devices, and specifically includes the following steps:

[0073] Step 110: Obtain the influence parameters of the numerical reactor and their probability distribution types.

[0074] Step 120: Determine the basis function combination terms of the influencing parameters based on the influencing parameters and their probability distribution types.

[0075] Step 130: Based on the sampling results and the basis function combination terms, determine the polynomial basis function matrix and the reduced order coefficients. The sampling results are obtained by sampling the influencing parameters according to the probability distribution indicated by the probability distribution type.

[0076] Step 140: Determine the evaluation parameters of the numerical reactor based on the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters.

[0077] The method for obtaining evaluation parameters of a numerical reactor provided in this application can determine the basis function combination terms of the influence parameters and their probability distribution types based on the obtained influence parameters of the numerical reactor. Then, based on the sampling results and the basis function combination terms, a polynomial basis function matrix and order reduction coefficients are determined. The sampling results are obtained by sampling the influence parameters according to the probability distribution indicated by the probability distribution type. Finally, the evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters. By sampling and incorporating the order reduction coefficients into the calculation of evaluation parameters, the amount of intermediate calculations can be reduced, thereby improving the efficiency of determining the evaluation parameters.

[0078] The following will discuss how Figure 1 The steps of the Chinese method are explained in detail.

[0079] In step 110, the electronic device can acquire the influence parameters of the numerical reactor and the probability distribution type of the influence parameters.

[0080] Influencing parameters include uncertainties in the reactor, which have varying degrees of impact on the reactor's reactivity and energy output. Influencing parameters can include geometric parameters, nuclear physics parameters, thermal-hydraulic parameters, material property parameters, and control system parameters. If X represents an influencing parameter, then X = (x1, x2, ..., x...).d ), where d represents the number of parameters that affect the input.

[0081] Specifically, geometric parameters may include the geometry and arrangement of nuclear reactor fuel rods, fuel assemblies, core, and other related parameters.

[0082] Nuclear physics parameters may include parameters such as the cross-section and energy deposition of various types of nuclear reaction channels (such as absorption, fission, scattering, etc.). Their uncertainty may come from measurement errors in experimental data, simplification of some theoretical models, etc.

[0083] Thermal hydraulic parameters may include parameters such as coolant flow rate, temperature and pressure, and their uncertainty may come from factors such as measurement error, equipment failure or operator error.

[0084] Material property parameters may include those of reactor components, such as strength, plasticity, and fatigue life. These parameters may be subject to uncertainty, which may arise from material defects, environmental factors (such as temperature and pressure), and aging.

[0085] Control system parameters may include gain, hysteresis, and time constant, and may be subject to uncertainty. This uncertainty may arise from factors such as measurement error, equipment failure, or operator error.

[0086] The probability distribution type can be used to indicate the probability distribution of the corresponding influencing parameter. The probability distribution of the influencing parameter can be used to quantify the influencing parameter with uncertainty. For example, for the influencing parameter of coolant flow rate, a normal distribution can be used to describe the fluctuation of coolant flow rate around a certain average value.

[0087] In some embodiments, different probability distribution types can be distinguished by different identifiers, such as identifier 001 representing probability distribution A and identifier 002 representing probability distribution B. In addition to Arabic numerals as shown in the example, the identifiers can also be English letters, punctuation marks, or combinations of at least two of Arabic numerals, English letters, and punctuation marks; these will not be elaborated further here.

[0088] The probability distribution of different influencing parameters may vary depending on the type and source of the parameters. For example, some influencing parameters may be assumed to follow a normal, uniform, or other common probability distribution. For other types of probability distributions, separation transformations can be used. For instance, if a probability distribution needs to be converted to a Gaussian distribution, appropriate transformation methods such as the Box-Cox transform or Yeo-Johnson transform can be used. These transformations convert the original data to data that satisfies the Gaussian distribution assumption. If a probability distribution needs to be converted to a Beta distribution, the inverse transformation method of the Beta distribution can be used, which maps the original data to the domain of the Beta distribution.

[0089] In some embodiments, the electronic device may, in response to an acquisition command, acquire the influence parameters of the numerical reactor and the probability distribution type of the influence parameters. The acquisition command may be used to instruct the electronic device to acquire the influence parameters and the probability distribution type; it may be generated manually by the user or automatically at a preset time, and this is not limited here.

[0090] In step 120, when the electronic device determines the influencing parameters and their probability distribution types based on the aforementioned step 110, it can further determine the basis function combination terms of the influencing parameters, which is convenient for subsequent calculation of the evaluation parameters of the numerical reactor.

[0091] Specifically, for influence parameters with uniform probability distributions, Legendre polynomials can be used as an orthogonal basis, and the p-order polynomials can be obtained through the `special.eval_legendre` function in the open-source scipy library. p is the order.

[0092] For probability distributions with a beta distribution, Jacobi polynomials can be used as an orthogonal basis. The p-th order polynomial can be obtained using the `scipy.special.jacobi` function in the open-source scipy library. p is the order.

[0093] For probability distributions that are Gaussian, Hermite polynomials can be used as an orthogonal basis, and the p-order polynomial can be obtained using the `special.hermite` function in the open-source scipy library. p is the order.

[0094] For the i-th variable x in the d-dimensional variable space i (i.e., the i-th influencing parameter), its entire p-order polynomial expression is: p i The order polynomial is represented as

[0095] If each influencing parameter is independent, then their orders are (p1,…,p) d The basis function combination terms can be represented as the product of independent basis functions for each variable:

[0096]

[0097] Therefore, the total number of basis function combinations, Nc, can be expressed as follows:

[0098] N C =(P+1)d (2)

[0099] In step 130, the electronic device may further determine the polynomial basis function matrix and the reduced-order coefficients, wherein the polynomial basis function matrix includes multiple polynomial basis functions.

[0100] The sampling result can be composed of multiple sampled samples. Specifically, the electronic device can sample the influencing parameters according to the probability distribution function corresponding to the probability distribution of the influencing parameters, and the sampling result can be composed of multiple sampled samples.

[0101] The pre-defined sampling method includes at least one of random sampling and Latin hypercube sampling.

[0102] In some embodiments, the function for Latin hypercube sampling using the scipy open-source library is scipy.stats.qmc.LatinHypercube.

[0103] In some embodiments, a preset number of samplings can be set to sample the influencing parameters according to a preset sampling method, including sampling the influencing parameters according to a preset number of samplings based on the preset sampling method.

[0104] Specifically, the electronic device can substitute the sampled sample Sx (with dimension Ng*d, where Ng represents the number of sampled samples) from the sampling results into Nc to obtain the polynomial basis function matrix. (with dimensions Ng*Nc), then the electronic device can determine the order reduction coefficients based on the polynomial basis function matrix.

[0105] In step 140, the electronic device, based on the relevant parameters obtained in steps 110 to 130, such as the polynomial basis function matrix, reduction coefficients, influencing parameters, and the probability distributions corresponding to the influencing parameters, can quickly determine the evaluation parameters of the numerical reactor. The evaluation parameters of the numerical reactor may include the mean and variance of the physical fields of the numerical reactor. Based on these two evaluation parameters, the electronic device can further determine the degree of influence of the influencing parameters on the output results of the engineering system or model simulated by the numerical reactor, thus completing a sensitivity analysis of the influencing parameters.

[0106] In one possible implementation, after determining the polynomial basis functions and order reduction coefficients based on the sampling results and the basis function combination terms, the method further includes:

[0107] Based on the sampling results and the numerical reactor simulation model, the key physical quantity matrix was determined;

[0108] If the number of key physical quantities in the key physical quantity matrix is ​​greater than or equal to a preset number threshold, the key physical quantity matrix is ​​dimensionality reduced.

[0109] Given that the least squares method is chosen as the target computational strategy, the coefficients of the polynomial basis functions are determined using the least squares method, including:

[0110] For the key physical quantity matrix after dimensionality reduction, the coefficients of the polynomial basis functions are determined using the least squares method.

[0111] This application embodiment reduces the amount of data to be processed by performing dimensionality reduction on the key physical quantity matrix when it is determined that the key physical quantity matrix is ​​too large, that is, greater than or equal to a preset number, thereby improving the efficiency of determining the evaluation parameters of the numerical reactor.

[0112] Specifically, the electronic device can input the sampled sample Sx (with dimensions Ng*d) from the sampling results into the simulation model of the numerical reactor, such as M(x), to obtain the key physical quantity matrix Q (with dimensions Ng*Nq).

[0113] It should be noted that by inputting each sample (which can be considered a design scheme in the design space) into the numerical reactor simulation model M(x), the physical field (physical quantity of interest) under each design scheme can be obtained. The numerical reactor simulation model can output Nq physical quantities of interest. Considering that the mesh of the physical field is very fine, such as exceeding tens of thousands or millions of meshes, and that there are multiple physical fields such as nucleus (neutron field)-thermal (temperature, pressure, flow rate)-force (geometric deformation, stress field)-chemical (distribution of corrosion products, distribution of sediments), etc., the physical quantities of interest increase dramatically, so the value of Nq can be very large.

[0114] In M(x), x represents the sampled sample, that is, the true value obtained by sampling the influencing parameter.

[0115] In a specific embodiment, the number of samples Ng = 10000, and each sample has 7 influencing parameters. The 10000 samples are input into the numerical reactor simulation model M(x) to obtain the key physical quantity matrix Q. Taking reactor transient behavior as an example, the i-th physical quantity Q... i The dimension is 5024*100, representing the neutron flux (or line power density) at 5024 discrete points in space over 100 time steps. The unit of line power density is watts per centimeter (W / cm). The maximum line power density of the entire stack can be represented as a curve over time. Figure 2 , Figure 2 This is a graph showing the change of line power density over time, provided as an embodiment of this application.

[0116] exist Figure 2In the graph, the horizontal axis represents time (in seconds), and the vertical axis represents line power density (in watts per centimeter). The graph shows the line power density of the first 300 samples (represented by curves of different colors) over time. It can be observed that near time 0, the line power density of all samples remains at a relatively low value, meaning that initially, the line power density of all samples is at a low level. As time increases, the line power density of each sample increases rapidly, reaches a peak, then decreases and stabilizes at a relatively low value. Although the overall trend of the first 300 samples is similar, there are still some differences in the height of the peak and the process of decline, reflecting the differences in the characteristics of line power density change among different samples.

[0117] The peak line power density at all times is statistically analyzed, and its average value is determined as follows: Figure 3 and variance, Figure 4 . Figure 3 This application provides an embodiment of a line graph of average values ​​obtained using Monte Carlo methods. Figure 4 This application provides an embodiment of a variance line graph obtained using the Monte Carlo method. Figure 3 and Figure 4 The horizontal axis of both graphs represents the number of samples (in units), and the vertical axis represents the peak line power density (in watts per centimeter).

[0118] It can be observed that, Figure 3 and Figure 4 The mean and variance of the line power density peak value are obtained by Monte Carlo random sampling. A certain number of samples, such as more than 4000, are required to obtain a relatively stable and accurate mean and variance.

[0119] The preset quantity threshold can be set according to actual needs; no limit is set here.

[0120] In this embodiment of the application, after dimensionality reduction of the key physical quantity matrix, the required number of samples is less than 4000. By reducing the number of samples, the computation time can be reduced, thereby improving the efficiency of determining the evaluation parameters of the numerical reactor.

[0121] In one possible implementation, the key physical quantity matrix is ​​subjected to dimensionality reduction processing, including:

[0122] Based on the key physical quantity matrix, determine the relevant matrix;

[0123] Based on the non-zero eigenvalues ​​and eigenvectors of the correlation matrix, determine the eigenorthogonal decomposition modes of the correlation matrix;

[0124] Based on the intrinsic orthogonal decomposition modes and the key physical quantity matrix, the key physical quantity matrix after dimensionality reduction is determined.

[0125] Specifically, the correlation matrix R can be represented as follows:

[0126] R = Q T Q, (3)

[0127] in,

[0128] Determine the non-zero eigenvalues ​​λ of the correlation matrix R j and eigenvector φ j And determine the effect of the correlation matrix R on the eigenvector φ j The results are expressed as follows:

[0129] Rφ j =λ j φ j (4)

[0130] Among them, j=1, 2,…,Nq, λ1>>λ2>>…>>λ Nq >0.

[0131] The orthogonal eigenvalue decomposition modes (POD modes) that form the correlation matrix can be represented as follows:

[0132]

[0133] Where j = 1, 2, ..., Nq, and S represents the snapshot matrix in the POD method.

[0134] Then only the first r-order modes are selected: The selection principle is to find the minimum r that satisfies the following conditions:

[0135]

[0136] In some embodiments, τ is 0.999.

[0137] Next, the electronic device can construct the mode matrix based on the selected first r modes as follows:

[0138]

[0139] in,

[0140] According to the definition of modality, the multiphysics field Q under the i-th sample of the numerical reactor simulation model is... i It can be represented as follows:

[0141]

[0142] We can take α = [α1 … α] r ], and define the modal coefficient matrix:

[0143]

[0144] Calculate the projected matrix Y:

[0145] Y = Qφ, (9)

[0146] Based on the intrinsic orthogonal decomposition modes and the key physical quantity matrix, the dimension-reduced key physical quantity matrix A is determined, including using the least squares method, or optionally using the open-source mathematical library NumPy to solve it. The solution formula is as follows:

[0147] A=np.linalg.lstsq(φ.T,YT), (10)

[0148] Here, ".T" represents the matrix transpose operation, and the dimension of A is Ng*r.

[0149] For example, if the key physical quantity matrix Q in this embodiment is determined based on the sampling results of only 1000 samples and the simulation model of the numerical reactor, the spatiotemporal distribution field of the reactor core (dimension 5024100) can be obtained for the 1000 samples, and its reduced-order mode corresponds to λ. i Please see Figure 5 , Figure 5 This is an energy change diagram related to a method for obtaining evaluation parameters of a numerical reactor, provided in an embodiment of this application.

[0150] exist Figure 5 In the diagram, the horizontal axis represents the number of modes, and the two vertical axes represent the modal energies (10T, ... 1 Up to 10 6 The cumulative energy contribution (from 0.5 to 1.0) is shown by the red dashed line. As the number of modes increases, the cumulative energy contribution gradually approaches 1.0, indicating that the first few modes contribute most of the energy, while the cumulative energy contribution of later modes gradually decreases. In practical applications, only the first few modes need to be considered to capture most of the system's energy response. With τ = 0.99, there are 11 reduced-order modes, corresponding to 11 reduction-order coefficients, i.e., Ng = 11.

[0151] In one possible implementation, after determining the basis function combination terms of the influencing parameters based on the influencing parameters and their probability distribution types, the method further includes:

[0152] Based on the comparison results of the first and second quantities, determine the target calculation strategy;

[0153] Based on the target computation strategy, determine the coefficients of each polynomial basis function in the polynomial basis function matrix;

[0154] The first quantity includes the number of samples in the sampling results, and the second quantity includes the number of terms in the basis function combination.

[0155] The evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, reduction coefficients, influence parameters, and their corresponding probability distributions. These parameters include:

[0156] The evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function in the polynomial basis function matrix, the reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters.

[0157] This application embodiment determines the coefficients of each polynomial basis function in the polynomial basis function matrix by comparing the number of sampled samples and the number of terms in the basis function combination. Then, the coefficients of each polynomial basis function are used in the calculation of the evaluation parameters of the numerical reactor. By determining a suitable target calculation strategy, the efficiency of determining the evaluation parameters of the numerical reactor can be further improved.

[0158] Specifically, the number of terms in the basis function combination is Nc in the aforementioned embodiment. The coefficients of the polynomial basis functions can determine the specific form and properties of the polynomial basis functions. By determining a suitable target calculation strategy, the electronic equipment can accurately and quickly calculate the coefficients of the polynomial basis functions, thereby improving the accuracy and reliability of determining the evaluation parameters of the numerical reactor.

[0159] In one possible implementation, determining the target calculation strategy based on the comparison result between the first quantity and the second quantity includes:

[0160] If the comparison results indicate that the first quantity is greater than or equal to the second quantity, the least squares method is determined as the target calculation strategy.

[0161] Given that the least squares method is determined as the target computational strategy, the coefficients of the polynomial basis functions are determined according to the least squares method.

[0162] If the comparison results indicate that the first quantity is less than the second quantity, the sparse matrix method is determined as the target calculation strategy.

[0163] Given that the sparse matrix method is the target computational strategy, the coefficients of the polynomial basis functions are determined according to the sparse matrix method.

[0164] Specifically, the comparison result indicates that the first quantity is greater than or equal to the second quantity, that is, Ng≥Nc, which can indicate that the sampling sample is sufficient. In the case of a sufficient sampling sample, the electronic device can use the least squares method to determine the coefficients of the polynomial basis function. It should be noted that in the case of a sufficient sampling sample, using the least squares method can effectively average out the random noise in the data and improve the reliability of solving the coefficients of the polynomial basis function.

[0165] The comparison result indicates that the first quantity is less than the second quantity, that is, Ng<Nc, which can indicate that the sampling sample is insufficient. In the case of an insufficient sampling sample, the electronic device can use the sparse matrix method to determine the coefficients of the polynomial basis function. It should be noted that considering that there are multiple influencing parameters in the numerical reactor and the multiple influencing parameters are not equally important. Generally, only a few influencing parameters play a decisive role in the simulation system of the numerical reactor. At this time, from the perspective of the coefficient matrix composed of the coefficients of the polynomial basis function to be solved, there are a large number of coefficient characteristics (that is, the coefficients of some polynomial basis functions are very small, such as close to zero). Therefore, by relying on a small number of sampling calculations, a high-order polynomial fitting accuracy can be obtained. However, the computational amount and computational time of a single sampling in the simulation system of the numerical reactor are very large. When Ng<Nc, the sparse matrix method can be used to estimate the coefficient matrix U composed of the coefficients of the polynomial basis function to be solved, which can reduce the computational amount and computational time.

[0166] Assume that the i-th reduced-dimensional coefficient of the physical field of interest is expressed as follows:

[0167]

[0168] where, Ψ i,j is the j-th polynomial basis function, u i,j is the coefficient of the j-th polynomial basis function, which is the coefficient to be solved. The coefficients of all polynomial basis functions can form a coefficient matrix U, then there is:

[0169] Q = ΨU, (12)

[0170] U = Ψ + Q, (13)

[0171] where, the dimension of Ψ is N g ×N C , the dimension of U is N c ×r, and the dimension of Q is N g ×r. Ψ + represents the pseudo-inverse (or generalized inverse) of Ψ. In some embodiments, the pseudo-inverse can be solved using the linalg.pinv function in the open-source numpy library, or solved using the open-source scipy.linalg.pinv.

[0172] In the aforementioned embodiments, for the key physical quantity matrix after dimensionality reduction, the coefficients of the polynomial basis functions are determined using the least squares method. Only Q in formula (12) needs to be replaced with the dimensionality-reduced key physical quantity matrix A, i.e.:

[0173] A = ΨU, (14)

[0174] The key physical quantity matrix obtained after order reduction is represented by the modal coefficient matrix A, which is composed of α. Ng It consists of modal coefficients.

[0175] Furthermore, targeting Figure 5 The 11 reduced-order coefficients are used as the fitting objects, and they are expanded according to the Polynomial Chaos Expansion (PCE) surrogate model to obtain a chaotic polynomial with 11 reduced-order coefficients. The corresponding probability density variation curves are shown below. Figure 6 As shown, Figure 6 This is a probability density variation curve related to a method for obtaining evaluation parameters of a numerical reactor provided in an embodiment of this application. Figure 6 In the diagram, the horizontal axis represents the coefficient values, and the vertical axis represents the probability density, which is obtained based on kernel density estimation (KDE). p represents the degree of the polynomial (e.g., p=2 is a quadratic polynomial expansion). It can be observed that the probability density changes significantly under polynomial expansions of different degrees.

[0176] If sampling using a 5th-order polynomial yields a probability density corresponding to 1000 samples, the change is quite similar to the probability density corresponding to 10000 samples obtained using Monte Carlo sampling. For details, please refer to [link to relevant documentation]. Figure 7 , Figure 7 This is another probability density variation curve related to a method for obtaining evaluation parameters of a numerical reactor provided in an embodiment of this application. It can be observed that... Figure 7 The two curves in (a), (b), (c), (d), (i), and (k) are quite similar, meaning that the embodiments of this application can obtain convergent and highly accurate average and variance even with a small sample size, such as 1000.

[0177] Where d represents the order of the polynomial, and the curve corresponding to KDE MC represents the change curve of the probability density of 10,000 samples obtained by Monte Carlo sampling.

[0178] In one possible implementation, the evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function in the polynomial basis function matrix, the reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters, including:

[0179] The average value of the reduction coefficients is determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function, the influencing parameters, and the probability distributions corresponding to the influencing parameters.

[0180] Determine the variance of the reduction coefficients based on the average value and the reduction coefficients themselves.

[0181] The average value of the physical field of the numerical reactor is determined based on the average value of the reduced-order coefficients and the polynomial basis function matrix.

[0182] The variance of the physical field of the numerical reactor is determined based on the variance of the reduced-order coefficients and the polynomial basis function matrix.

[0183] The evaluation parameters include the mean value and variance of the physical fields of the numerical reactor.

[0184] Specifically, determine the average value E(a) of the i-th order reduction coefficient. i The following formula can be used to calculate:

[0185]

[0186] Where x represents the influencing parameter, p(x) represents the probability distribution function of the influencing parameter x, and u i,j Let Ψ represent the coefficients of the j-th polynomial basis function. i,j Let u represent the basis function of the j-th polynomial. i,1 The constant term represents the terms in a polynomial expansion.

[0187] Method for reducing order coefficients D(a) i The following formula can be used to calculate:

[0188]

[0189] The average physical field E(Q) of a numerical reactor can be calculated using the following formula:

[0190]

[0191] in, Both represent vectors of size 1*Nq.

[0192] The variance of the physical fields of a numerical reactor can be calculated using the following formula:

[0193]

[0194] in, Represents a vector of size 1*Nq.

[0195] To clearly demonstrate that the embodiments of this application can obtain relatively stable average and variance using a small sample size, such as 1000, please refer to [link to relevant documentation]. Figure 8 , Figure 8 Comparative diagrams related to a method for obtaining evaluation parameters of a numerical reactor provided in an embodiment of this application. Figure 8 In the diagram, (a) represents a broken line representing the average change of peak line power density with different sample sizes. Figure 8 In Figure (b), the line represents the variance variation of the peak power density with different sample sizes. Specifically, the line corresponding to MC represents the mean / variance variation when simulated using the Monte Carlo method; MC(10000) represents the fitted mean / variance when simulated using 1000 samples and the Monte Carlo method; and the line corresponding to PCE p=4 represents the mean / variance variation when simulated using the method described in this embodiment. It can be observed that the Monte Carlo method generally requires more than 3000 samples to obtain a relatively stable mean and variance, while the method corresponding to this embodiment only requires about 1000 samples to obtain a relatively stable mean and variance.

[0196] In one possible implementation, the basis function combination term of the influencing parameters is determined based on the influencing parameters and the probability distribution type of the influencing parameters, including:

[0197] Given a defined dimension of the influencing parameters, the basis function combination terms are determined based on the dimension of the influencing parameters and the preset truncation order.

[0198] The embodiments of this application can indirectly improve the efficiency of obtaining evaluation parameters of numerical reactors by reducing the number of terms in the basis function combination.

[0199] As can be seen from the aforementioned formula (2), when the parameter d is large, the term explosion phenomenon of basis function combination type is likely to occur. This application embodiment sets the p-order polynomial in the aforementioned embodiment... The term 'p', or the preset cutoff order, is used to reduce the number of basis function combinations, thereby indirectly improving the efficiency of obtaining evaluation parameters for the numerical reactor. This application does not specifically limit the preset cutoff order 'p' in its embodiments.

[0200] The number of basis function cross-combination terms after truncation is:

[0201]

[0202] Where, p limit It can represent the preset truncation order.

[0203] In some embodiments, the electronic device may also truncate higher-order terms of certain influencing parameters, such as univariate influencing parameters, based on a first preset parameter, thereby reducing the basis function combination terms.

[0204] In some embodiments, the electronic device may also limit the number of variable crosses in the basis function combination terms based on a preset cross threshold. For example, a preset cross threshold of 2 may indicate that in the process of determining the basis function combination terms, only crosses between pairs of variables are considered, and crosses between three or more variables are not considered.

[0205] In some embodiments, the electronic device may limit the total order of the interaction terms based on a second preset parameter. For example, for interaction terms between two variables, only lower-order interaction terms are considered.

[0206] Corresponding to the above method embodiments, this application also provides a system for obtaining evaluation parameters of a numerical reactor. Please refer to [link to relevant documentation]. Figure 9 , Figure 9 A functional module diagram of a system for acquiring evaluation parameters of a numerical reactor, provided for an embodiment of this application, wherein the system 900 for acquiring evaluation parameters of a numerical reactor includes:

[0207] Module 910 is used to acquire the influence parameters of the numerical reactor and their probability distribution types.

[0208] The first determining module 920 is used to determine the basis function combination terms of the influencing parameters based on the influencing parameters and their probability distribution types.

[0209] The second determining module 930 is used to determine the polynomial basis function matrix and the reduced order coefficients based on the sampling results and the basis function combination terms. The sampling results are obtained by sampling the influencing parameters according to the probability distribution indicated by the probability distribution type.

[0210] The third determination module 940 is used to determine the evaluation parameters of the numerical reactor based on the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters.

[0211] The system for obtaining evaluation parameters of a numerical reactor provided in this application embodiment can achieve, for example: Figure 1 The various processes implemented in the Chinese method embodiments can achieve similar or the same technical effects, and will not be described again here to avoid repetition.

[0212] In one possible implementation, the system 900 for acquiring evaluation parameters of a numerical reactor further includes a fourth determining module, which is used for:

[0213] Based on the comparison results of the first and second quantities, determine the target calculation strategy;

[0214] Based on the target computation strategy, determine the coefficients of each polynomial basis function in the polynomial basis function matrix;

[0215] The first quantity includes the number of samples in the sampling results, and the second quantity includes the number of terms in the basis function combination.

[0216] The third determining module 940 is also specifically used for:

[0217] The evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function in the polynomial basis function matrix, the reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters.

[0218] In one possible implementation, the fourth determining module includes a first determining submodule, which is used for:

[0219] Based on the comparison between the first and second quantities, the target calculation strategy is determined, including:

[0220] If the comparison results indicate that the first quantity is greater than or equal to the second quantity, the least squares method is determined as the target calculation strategy.

[0221] Given that the least squares method is determined as the target computational strategy, the coefficients of the polynomial basis functions are determined according to the least squares method.

[0222] If the comparison results indicate that the first quantity is less than the second quantity, the sparse matrix method is determined as the target calculation strategy.

[0223] Given that the sparse matrix method is the target computational strategy, the coefficients of the polynomial basis functions are determined according to the sparse matrix method.

[0224] In one possible implementation, the third determining module 940 is further specifically used for:

[0225] The average value of the reduction coefficients is determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function, the influencing parameters, and the probability distributions corresponding to the influencing parameters.

[0226] Determine the variance of the reduction coefficients based on the average value and the reduction coefficients themselves.

[0227] The average value of the physical field of the numerical reactor is determined based on the average value of the reduced-order coefficients and the polynomial basis function matrix.

[0228] The variance of the physical field of the numerical reactor is determined based on the variance of the reduced-order coefficients and the polynomial basis function matrix.

[0229] The evaluation parameters include the mean value and variance of the physical fields of the numerical reactor.

[0230] In one possible implementation, the system 900 for acquiring evaluation parameters of the numerical reactor further includes a fifth determining module, which is used for:

[0231] Based on the sampling results and the numerical reactor simulation model, the key physical quantity matrix was determined;

[0232] If the number of key physical quantities in the key physical quantity matrix is ​​greater than or equal to a preset number threshold, the key physical quantity matrix is ​​dimensionality reduced.

[0233] Given that the least squares method is chosen as the target computational strategy, the coefficients of the polynomial basis functions are determined using the least squares method, including:

[0234] For the key physical quantity matrix after dimensionality reduction, the coefficients of the polynomial basis functions are determined using the least squares method.

[0235] In one possible implementation, the fifth determining module includes a second determining submodule, the second determining submodule being used for:

[0236] Based on the key physical quantity matrix, determine the relevant matrix;

[0237] Based on the non-zero eigenvalues ​​and eigenvectors of the correlation matrix, determine the eigenorthogonal decomposition modes of the correlation matrix;

[0238] Based on the intrinsic orthogonal decomposition modes and the key physical quantity matrix, the key physical quantity matrix after dimensionality reduction is determined.

[0239] In one possible implementation, the first determining module 920 is further specifically used for:

[0240] Given a defined dimension of the influencing parameters, the basis function combination terms are determined based on the dimension of the influencing parameters and the preset truncation order.

[0241] This application also provides an electronic device. Please refer to [link to previous application]. Figure 10 , Figure 10 This is a diagram illustrating the internal structure of an electronic device according to an embodiment of this application. The electronic device includes a processor, a memory, and a network interface connected via a system bus. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and may also store a computer program. When executed by the processor, this computer program enables the processor to implement the method for obtaining evaluation parameters of a numerical reactor, as described in the above embodiment. The internal memory may also store a computer program, which, when executed by the processor, enables the processor to execute the method for obtaining evaluation parameters of a numerical reactor. Those skilled in the art will understand that... Figure 10The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0242] This application also discloses a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the method for obtaining evaluation parameters of a numerical reactor as described in the method embodiment.

[0243] This application provides a computer program product stored in a storage medium. The program product is executed by at least one processor to implement the various processes of the embodiments of the method for obtaining evaluation parameters of a numerical reactor as described above, and can achieve similar or the same technical effects. To avoid repetition, it will not be described again here.

[0244] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0245] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A method for obtaining evaluation parameters of a numerical reactor, characterized in that, The method includes: The influence parameters of a numerical reactor and their probability distribution types are obtained, wherein the influence parameters include at least one of geometric parameters, nuclear physics parameters, thermal-hydraulic parameters, material property parameters, and control system parameters; Based on the probability distribution type corresponding to the influencing parameter, an orthogonal basis function is matched for the influencing parameter, and the basis function combination term of the influencing parameter is determined based on the orthogonal basis function; wherein, when the probability distribution type is uniform, the Legendre polynomial is determined as the orthogonal basis function of the influencing parameter; when the probability distribution type is beta, the Jacobi polynomial is determined as the orthogonal basis function of the influencing parameter; when the probability distribution type is Gaussian, the Hermitian polynomial is determined as the orthogonal basis function of the influencing parameter. Based on the sampling results and the basis function combination terms, the polynomial basis function matrix and the order reduction coefficients are determined. The sampling results are obtained by sampling the influence parameters according to the probability distribution indicated by the probability distribution type. The evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters; wherein the evaluation parameters include the average value and the variance of the physical field of the numerical reactor, and the physical field includes at least one of the nuclear physical field, thermophysical field, mechanical field, and chemical physical field of the numerical reactor.

2. The method as described in claim 1, characterized in that, After determining the basis function combination term of the influence parameters based on the influence parameters and their probability distribution type, the method further includes: Based on the comparison results of the first and second quantities, determine the target calculation strategy; Based on the target calculation strategy, determine the coefficients of each polynomial basis function in the polynomial basis function matrix; Wherein, the first quantity includes the number of sampled samples in the sampling results, and the second quantity includes the number of terms in the basis function combination terms; The step of determining the evaluation parameters of the numerical reactor based on the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters includes: The evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function in the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters.

3. The method as described in claim 2, characterized in that, The step of determining the target calculation strategy based on the comparison result between the first quantity and the second quantity includes: If the comparison result indicates that the first quantity is greater than or equal to the second quantity, then the least squares method is determined as the target calculation strategy. When the least squares method is determined as the target computation strategy, the coefficients of the polynomial basis functions are determined according to the least squares method. If the comparison result indicates that the first quantity is less than the second quantity, the sparse matrix method is determined as the target calculation strategy. When the sparse matrix method is determined as the target computation strategy, the coefficients of the polynomial basis function are determined according to the sparse matrix method.

4. The method as described in claim 3, characterized in that, The step of determining the evaluation parameters of the numerical reactor based on the polynomial basis function matrix, the coefficients of each polynomial basis function in the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters includes: The average value of the order reduction coefficients is determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function, the influence parameters, and the probability distributions corresponding to the influence parameters. The variance of the reduction coefficients is determined based on the average value of the reduction coefficients and the reduction coefficients themselves. The average value of the physical field of the numerical reactor is determined based on the average value of the reduction coefficients and the polynomial basis function matrix. The variance of the physical field of the numerical reactor is determined based on the variance of the reduced-order coefficients and the polynomial basis function matrix.

5. The method as described in claim 3, characterized in that, After determining the polynomial basis functions and order reduction coefficients based on the sampling results and the basis function combination terms, the method further includes: Based on the sampling results and the simulation model of the numerical reactor, the key physical quantity matrix is ​​determined; If the number of key physical quantities in the key physical quantity matrix is ​​greater than or equal to a preset number threshold, the key physical quantity matrix is ​​subjected to dimensionality reduction processing. When the least squares method is determined to be the target computation strategy, determining the coefficients of the polynomial basis functions according to the least squares method includes: For the key physical quantity matrix after dimensionality reduction, the coefficients of the polynomial basis functions are determined according to the least squares method.

6. The method as described in claim 5, characterized in that, The dimensionality reduction process for the key physical quantity matrix includes: Based on the aforementioned key physical quantity matrix, determine the relevant matrix; Based on the non-zero eigenvalues ​​and eigenvectors of the correlation matrix, determine the intrinsic orthogonal decomposition modes of the correlation matrix; Based on the intrinsic orthogonal decomposition modes and the key physical quantity matrix, the key physical quantity matrix after dimensionality reduction is determined.

7. The method as described in claim 1, characterized in that, The step of determining the basis function combination term of the influence parameter based on the influence parameter and the probability distribution type of the influence parameter includes: Given the dimension of the influencing parameter, the basis function combination term is determined based on the dimension of the influencing parameter and the preset truncation order.

8. A system for acquiring evaluation parameters of a numerical reactor, characterized in that, The system includes: The acquisition module is used to acquire the influence parameters of the numerical reactor and their probability distribution types. The influence parameters include at least one of geometric parameters, nuclear physics parameters, thermal-hydraulic parameters, material property parameters, and control system parameters. The first determining module is used to match corresponding orthogonal basis functions for the influence parameters according to the probability distribution type corresponding to the influence parameters, and to determine the basis function combination terms of the influence parameters according to the orthogonal basis functions; wherein, when the probability distribution type is a uniform distribution, Legendre polynomials are determined as orthogonal basis functions of the influence parameters; when the probability distribution type is a beta distribution, Jacobi polynomials are determined as orthogonal basis functions of the influence parameters; and when the probability distribution type is a Gaussian distribution, Hermitian polynomials are determined as orthogonal basis functions of the influence parameters. The second determining module is used to determine the polynomial basis function matrix and the order reduction coefficients based on the sampling results and the basis function combination terms. The sampling results are obtained by sampling the influence parameters according to the probability distribution indicated by the probability distribution type. The third determining module is used to determine the evaluation parameters of the numerical reactor based on the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distribution corresponding to the influence parameters; wherein the evaluation parameters include the average value of the physical field of the numerical reactor and the variance of the physical field of the numerical reactor, and the physical field includes at least one of the nuclear physical field, thermophysical field, mechanical field, and chemical physical field of the numerical reactor.

9. The system as described in claim 8, characterized in that, The system also includes: The fourth determining module is used to determine a target calculation strategy based on the comparison result of the first quantity and the second quantity; and to determine the coefficients of each polynomial basis function in the polynomial basis function matrix based on the target calculation strategy; wherein, the first quantity includes the number of sampled samples in the sampling result, and the second quantity includes the number of terms in the basis function combination terms; The third determining module is also used for: The evaluation parameters of the numerical reactor are determined based on the polynomial basis function matrix, the coefficients of each polynomial basis function in the polynomial basis function matrix, the order reduction coefficients, the influence parameters, and the probability distributions corresponding to the influence parameters.

10. The system as described in claim 8, characterized in that, The first determining module is further configured to: Given the dimension of the influencing parameter, the basis function combination term is determined based on the dimension of the influencing parameter and the preset truncation order.

Images