Neutron transport optimization simulation method with negative scattering cross section
By using hybrid finite element method and integral method in neutron transport simulation, the self-scattering term is integrated into the response matrix, and the problems of high computing resource consumption and slow iteration convergence caused by anisotropic scattering in the prior art are solved, and the problem of high computing resource consumption and slow iteration convergence caused by anisotropic scattering in the prior art is realized. Efficient and stable solution to neutron transport equations.
Patent Information
- Application Number
- CN202410466809.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-18
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-04-18
AI Technical Summary
In the prior art, when dealing with the neutron flux distribution of pressurized water reactor core, there are problems such as high computing resource consumption and slow iteration of iterative computing convergence caused by anisotropic scattering. Especially when there is a strong forward scattering material, negative scattering may be introduced. Cross-section, reduces the stability and convergence speed of iterative calculations, and even leads to divergence of calculations.
Using hybrid finite element method and integral method, through matrix transformation and angle integration, the self-scattering term is moved from the right side of the equation to the left side, construct the response matrix and solve it using the red and black Gauss Seidel method to avoid the efficiency of iterative calculation of negative scattering cross-section and stability impact.
An efficient and stable iterative format is realized, which significantly accelerates the solution process of neutron transport equations, avoids the problems of slow iteration convergence and numerical stability, and improves the stability and efficiency of the calculation.
Smart Images

Figure CN118364672B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a technology in the field of nuclear reactor control, in particular to a neutron transport optimization simulation method of a negative scattering cross section. Background Art
[0002] The neutron flux distribution in the core of a pressurized water reactor is obtained by solving the neutron transport equation. A key numerical problem in the solution process comes from anisotropic scattering. In order to reduce the consumption of computing resources, the commonly used method is transport correction, which can ensure high accuracy when only isotropic scattering calculations are performed. In the case of problems with strong forward scattering materials, transport correction may introduce negative scattering cross sections, reduce the convergence speed of iterative calculations, and even cause calculation divergence. Summary of the invention
[0003] In view of the problem that the existing technology requires users to select correction coefficients based on experience, which leads to large deviations in simulation results and low stability, the present invention proposes a neutron transport optimization simulation method with negative scattering cross section. Based on the hybrid finite element method and the integral method, matrix transformation and angle integration are used to avoid the influence of the negative scattering cross section on the efficiency and stability of iterative calculation, and a stable iterative format is obtained for application in solving the neutron transport equation of the reactor core.
[0004] The present invention is achieved through the following technical solutions:
[0005] The present invention relates to a neutron transport optimization simulation method for negative scattering cross section. The method comprises inputting self-scattering cross section data required for neutron transport simulation into a hybrid finite element neutron transport simulation solver, using matrix transformation and angle integration, moving the self-scattering term from the left side of an equation to a response matrix of the equation, completing the construction of the response matrix, and obtaining a response matrix equation. The response matrix equation is then directly solved by a red-black Gauss-Seidel method.
[0006] The present invention relates to a system for implementing the above method, comprising: a response matrix construction unit and a response matrix equation solving unit, wherein: the response matrix construction unit performs response matrix construction processing according to input material and finite element mesh information to obtain a response matrix, and the response matrix equation solving unit performs equation iterative solution according to the response matrix to obtain a physical quantity to be solved.
[0007] Technical Effects
[0008] The present invention is based on the hybrid finite element method and the integral method, and uses matrix transformation and angle integration to move the self-scattering term to the response matrix on the left side of the equation. Compared with the prior art, the present invention can avoid the self-scattering term with a negative cross section on the right side of the equation, which brings about the problem of slow iterative convergence and numerical stability, and obtain an efficient and stable iterative format. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] Figure 1 It is a schematic diagram of the principle of the present invention;
[0010] Figure 2 It is a calculation flow chart of the present invention;
[0011] Figure 3 This is a geometric diagram of the PWR grid element problem;
[0012] Figure 4 This is a geometric diagram of the small pressurized water reactor problem.
[0013] Figure 5 This is the result diagram of eigenvalue iteration of the PWR grid element eigenvalue problem;
[0014] Figure 6 This is the eigenvalue iteration result diagram for the small reactor eigenvalue problem. DETAILED DESCRIPTION
[0015] like Figure 1 As shown, this embodiment relates to an isotropic second-order even-parity neutron transport optimization simulation method with a negative scattering cross section, comprising:
[0016] Step 1: Decompose the problem to be solved into several adjacent but non-overlapping sub-regions in space, i.e., nodes. Perform variation of the discrete functional F of the transport equation in each node with respect to ψ to obtain the discrete equation: A(Ω)ψ(Ω)=Σ s0 I V φ+qE(Ω)χ, where: F[ψ,χ γ ]=∫dΩψ T (Ω)A(Ω)ψ(Ω)-φ T Σ s0 I V φ-2φ T q+2∫dΩψ T (Ω)E(Ω)χ, Ω is the angle; ψ is the even-order angular flux moment defined inside the node, which is discretized using the hemispherical angle quadrature group; χ is the odd-order angular flux moment defined on the node surface, which is discretized using the odd-order spherical harmonic function; A(Ω) is the spatial discrete matrix, which is a symmetric matrix obtained by numerical integration; Σ s0 is the self-scattering cross section; I V is the matrix obtained by multiplying the unit matrix and the block volume; φ is the scalar flux moment, defined as φ = ∫dΩψ(Ω); q is the source term moment inside the block; E(Ω) is the surface discrete matrix defined on the block surface, obtained by numerical integration.
[0017] Step 2: Transform the scattering term Σ by matrix transformation and angle integration. s0 I V Move φ from the right side of the equation to the left side and multiply both sides by A -1(Ω), we get ψ(Ω)=A -1 (Ω)Σ s0 I V φ+A -1 (Ω)qA -1 (Ω)E(Ω)χ, and then integrate the angle Ω variable to obtain: φ=HΣ s0 I V φ+Hq-Mχ, where: H=∫dΩA -1 (Ω), obtained by numerical integration, matrix Z = I-HΣ S0 I V , the expression of the scalar flux moment is: φ=Z -1 Hq-Z -1 Mχ, so that the right side of the equation no longer contains self-scattering terms, such as Figure 1 shown.
[0018] Step 3: directly solve the linear equations obtained in step 2 by the red-black Gauss-Seidel method to obtain the neutron scalar flux moment without self-scattering iteration, so it has the unconditional stability of self-scattering iteration. This method can also be extended to the neutron transport equation of anisotropic scattering.
[0019] After specific practical experiments, in the eigenvalue problem of the pressurized water reactor grid element, such as Figure 3 As shown in the figure, the grid element spacing is set to 2 cm, and the neutron cross section of the 33 energy group and the hybrid finite element method solver are used. The calculation results are as follows: This method accelerates the convergence of the power iteration by shifting the self-scattering cross section to the left. After 12 iterations, the error of the eigenvalue is reduced to 10 -6 The following, such as Figure 5 As shown; the self-scattering cross section of the prior art is on the right side of the equation, and without the additional introduction of a stability coefficient, it takes 155 iterations for the eigenvalue error to drop to 10 -6 The following, such as Figure 5 As shown in Figure 2, a speedup of more than 10 times was achieved.
[0020] In the eigenvalue problem of a small reactor, such as Figure 4 As shown in the figure, using the neutron cross section of 33 energy groups and the hybrid finite element method solver, the calculation results can be obtained as follows: In this method, when using the P0 scattering approximation or the P0 transport correction, 23 power iterations are used to converge the eigenvalue error to 10 -6 The following, such as Figure 5 As shown; in the prior art, without introducing an additional stability coefficient, it takes 449 steps to reach convergence under the calculation using the P0 scattering approximation, and when the P0 transport correction is used, the eigenvalue calculation diverges, as shown in Figure 6 shown.
[0021] Compared with the prior art, the present invention integrates the self-scattering term into the response matrix through matrix change and angle integration, so that the spectral radius of the intra-group iteration is reduced to 0, and the iteration is unconditionally stable; the spectral radius is 0 and the intra-group iteration can converge in one step, reducing the number of total iterations required.
[0022] The above-mentioned specific implementation can be partially adjusted in different ways by those skilled in the art without departing from the principle and purpose of the present invention. The protection scope of the present invention shall be based on the claims and shall not be limited by the above-mentioned specific implementation. Each implementation scheme within its scope shall be subject to the constraints of the present invention.
Claims
1. A neutron transport optimization simulation method with negative scattering cross section, characterized in that: By inputting the self-scattering cross-section data required for neutron transport simulation into the hybrid finite element neutron transport simulation solver, the self-scattering term is moved from the left side of the equation to the response matrix of the equation by using matrix transformation and angle integration, the response matrix is constructed and the response matrix equation is obtained; then the response matrix equation is directly solved by the red-black Gauss-Seidel method, which includes: Step 1: Decompose the problem to be solved into several adjacent but non-overlapping sub-regions in space, i.e., nodes. Perform variation of the discrete functional F of the transport equation in each node with respect to ψ to obtain the discrete equation: A(Ω)ψ(Ω)=Σ s0 I V φ+qE(Ω)χ, where: F[ψ,χ γ ]=ρdΩψ T (Ω)A(Ω)ψ(Ω)-φ T Σ s0 I V φ-2φ T q+2ρdΩψ T (Ω)E(Ω)χ, Ω is the angle; ψ is the even-order angular flux moment defined inside the node, which is discretized using the hemispherical angle quadrature group; χ is the odd-order angular flux moment defined on the node surface, which is discretized using the odd-order spherical harmonic function; A(Ω) is the spatial discrete matrix, which is a symmetric matrix obtained by numerical integration; Σ s0 is the self-scattering cross section; I V is the matrix obtained by multiplying the unit matrix and the block volume; the scalar flux moment φ=ρdΩψ(Ω); q is the source term moment inside the block; E(Ω) is the surface discrete matrix defined on the block surface, obtained by numerical integration; Step 2: Transform the scattering term Σ by matrix transformation and angle integration. s0 I V Move φ from the right side of the equation to the left side and multiply both sides by A -1 (Ω), we get ψ(Ω)=A -1 (Ω)Σ s0 I V φ+A -1 (Ω)qA -1 (Ω)E(Ω)χ, and then integrate the angle Ω variable to obtain: φ=HΣ s0 I V φ+Hq-Mχ, where: H=∫dΩA -1 (Ω), obtained by numerical integration, matrix Z = I-HΣ S0 I V , the expression of the scalar flux moment is: φ=Z -1 Hq-Z -1 Mχ, so that the right side of the equation no longer contains the self-scattering term; Step 3: directly solve the linear equations obtained in step 2 by the red-black Gauss-Seidel method to obtain the neutron flux moment.
2. A neutron transport optimization simulation system for implementing the negative scattering cross section of the method of claim 1, characterized in that: include: A response matrix construction unit and a response matrix equation solving unit, wherein: the response matrix construction unit performs response matrix construction processing according to input material and finite element mesh information to obtain a response matrix, and the response matrix equation solving unit performs equation iterative solution according to the response matrix to obtain a physical quantity to be solved.
