CN106484991A - A kind of Monte Carlo geometric manipulations method of coupling spline surface and analytic surface in nuclear mockup analysis - Google Patents

Abstract

The invention discloses a Monte Carlo geometric processing method for coupling a spline surface and an analytic surface in nuclear simulation analysis. When calculating the Monte Carlo particle transport of reactor building or fusion reactor, due to the complex spline surface in the geometric model, the traditional Monte Carlo program cannot handle it, and the model needs to be simplified. The invention classifies and processes the curved surfaces of the geometry in the CAD model. The analytic curved surfaces are described by algebraic equations, and the spline curved surfaces are described by parametric equations, which realizes the geometric processing of arbitrary curved surfaces by the Monte Carlo program, and at the same time utilizes the limited parameters of the spline curved surface parametric equations. Spatial bounding box to accelerate the processing of geometries containing spline surfaces. Compared with the Monte Carlo analytical surface geometry processing method that only uses algebraic equations, the present invention does not simplify the original geometric model, thereby ensuring the calculation accuracy of the model; compared with the Monte Carlo spline surface geometry processing method that only uses parametric equations method to improve the computational efficiency.

Description

A Monte Carlo Geometry of Coupling Spline Surfaces and Analytical Surfaces in Nuclear Simulation Analysis Approach

technical field

The invention relates to a geometric processing method for arbitrary curved surfaces in Monte Carlo particle transport in the fields of reactor physics, neutronics analysis and the like.

Background technique

Fusion energy is one of the dominant forms of human energy in the future, and it is also one of the important ways to finally solve the energy and environmental problems of human society and promote the sustainable development of human society. Nuclear simulation analysis is a necessary means for the safe and effective application of fusion energy. Since the fusion device has various fine parts with complex shapes and a large number of complex surfaces (taking the International Thermonuclear Experimental Reactor ITER as an example, there are a large number of fine structures that are radiation-sensitive or contain holes for different purposes that need to be modeled), the traditional determination The theoretical method can no longer meet the requirements of accurate simulation analysis of fusion devices. The biggest advantage of the Monte Carlo particle transport method is that it can describe the characteristics and processes of things with random properties very realistically, and has little restrictions on geometric models and materials, and can accurately simulate particle transport under complex sources, complex materials, and complex geometries. Therefore, it is widely used in the nuclear simulation analysis of fusion reactors.

The radiation dose calculation in the reactor building, the purpose of the calculation is to evaluate the radiation dose in the building, to optimize the shielding design of the building (wall thickness, pipe thickness, etc.) time. In the past, the simple point-core calculation method was used, which was fast and suitable for fewer pipes in the plant (or fewer pipes that played a major role), but when the plant contained a large number of pipes with high source strength, the calculation The accuracy is low, so it needs to be processed by Monte Carlo method.

Before using the Monte Cal program to perform particle transport simulation calculations, it is necessary to use specific format identifiers, keywords, and geometric parameters to describe the spatial position, size, shape type, and other relationships of the geometric model, so as to establish the physical space to be calculated. Complete spatial geometric relationship, and ensure that each point in the physical space to be calculated is uniquely defined, and there are no undefined or repeatedly defined spatial points.

The geometric modeling system based on the Monte Carlo method also has certain limitations and cannot accurately describe the spline surface. For example, the outer wall of the plasma in the middle of the ITER model, the surface of the stellarator device, and the joints of the pipeline are all typical splines. surface. Although some Monte Carlo programs have realized the geometric processing of spline surfaces directly based on the CAD underlying function library, such as MCBEND and McCARD, all surfaces of the geometry are described by the parametric equations of spline surfaces. The calculation efficiency is very low and cannot meet the design requirements. Therefore, in the nuclear design and analysis of the current fusion device, the spline surface is still simplified to an ordinary analytical surface for Monte Carlo particle transport, and it is difficult to achieve high-precision numerical simulation.

In order to solve the problem of geometric processing of arbitrary curved surfaces in nuclear simulation analysis of reactor buildings and fusion reactors, the invention discloses a Monte Carlo geometric processing method for coupling spline curved surfaces and analytic curved surfaces, which classifies and processes the curved surfaces of geometric bodies in CAD models. The analytic surface is described by algebraic equations, and the spline surface is described by parametric equations, which realizes the geometric processing of arbitrary surfaces by the Monte Carlo program. At the same time, the geometric processing of spline surfaces is carried out by using the bounding box of the limited space of the parametric equations of spline surfaces. accelerate.

Contents of the invention

The purpose of the invention is to provide a geometric processing method for arbitrary curved surfaces in Monte Carlo particle transport. Classify and process the surfaces of the geometry in the CAD model. Analytical surfaces are described by algebraic equations, and spline surfaces are described by parametric equations, which realizes the geometric processing of arbitrary surfaces by Monte Carlo programs, and at the same time utilizes the limited space encircled by the parametric equations of spline surfaces box, to accelerate the processing of geometries containing spline surfaces. Compared with the Monte Carlo analytical surface geometry processing method that only uses algebraic equations, the present invention does not simplify the original geometric model, thereby ensuring the calculation accuracy of the model; compared with the Monte Carlo spline surface geometry processing method that only uses parametric equations method to improve the computational efficiency.

The technical solution adopted in the present invention is realized as follows: a Monte Carlo geometric processing method for coupling spline surfaces and analytical surfaces in nuclear simulation analysis, when performing Monte Carlo particle transport calculations for reactor buildings or fusion reactors, due to the geometric model There are complex spline surfaces in the CAD model, which cannot be handled by the traditional Monte Carlo program, and the model needs to be simplified. This method classifies the surfaces of the geometry in the CAD model. The analytical surface is described by algebraic equations, and the spline surface is described by parametric equations. , to realize the geometric processing of any surface by Monte Carlo program, and at the same time use the bounding box of the limited space of the parametric equation of the spline surface to accelerate the geometric processing of the spline surface, which specifically includes the following steps:

Step (1), identify the spline surface in the CAD model:

Traverse all the geometry in the CAD model, for any geometry, obtain all the surfaces that make up the geometry according to its topology, and then judge the type of each surface, if the type of surface does not belong to the analytical surface (plane, sphere, cylinder, cone surface, torus, quadratic surface), mark it as a spline surface and record it.

Step (2), conversion CAD model generates Monte Carlo calculation model:

All the geometry in the CAD model is described by the surface and the Boolean operation between the surfaces. When the surface type is an analytical surface, it is expressed in the form of a surface equation, and the type of the analytical surface and the corresponding parameters are recorded. When the surface type is For a spline surface, record all the geometry corresponding to the spline surface and the serial number in the topology surface list of the corresponding geometry, and generate the bounding box information corresponding to the spline surface in the calculation model.

Step (3), Monte Carlo calculation model geometric processing of coupling spline surface and analytical surface:

When performing Monte Carlo particle transport, it is necessary to calculate the distance from the particle to each surface of the geometry along the ray direction, so as to obtain the shortest distance from the particle to the boundary of the geometry along the ray direction. Traverse all the surfaces in the geometry, if it is an analytic surface, directly use the ray and surface equation to perform intersection operation, if it is a spline surface, find the spline surface on the corresponding geometry topology surface according to the serial number in step (2) The position in the list, and then use the CAD underlying library function to obtain the spline surface, and use the Newton iterative method to solve the intersection point of the particle trajectory and the spline surface, and then calculate the distance from the particle (along the ray direction) to the spline surface. Before solving the intersection point of the ray and the spline surface, according to the space bounding box of the spline surface, perform the intersection operation of the ray and the bounding box, if there is no intersection point between the ray and the bounding box, then there is no need to calculate the ray and the spline surface cross operation.

The advantage of the present invention compared with prior art is:

(1) Compared with the Monte Carlo analytical surface geometry processing method that only uses algebraic equations, the present invention does not do any simplification to the geometric model of the original reactor building or fusion reactor, which ensures the calculation accuracy of the model and saves a large amount of models at the same time preprocessing time.

(2) Compared with the Monte Carlo spline surface geometry processing method that only uses parametric equations, the present invention improves the efficiency of complex model calculations when calculating particle transport in reactor buildings or fusion reactors.

Description of drawings

Figure 1 is a schematic diagram of the Monte Carlo calculation model generation of coupled spline surfaces and analytical surfaces;

Fig. 2 is a schematic diagram of Monte Carlo geometric processing of a coupled spline surface and an analytical surface;

Fig. 3 is the model schematic diagram of computing example 1;

FIG. 4 is a schematic diagram of the model of Calculation Example 2.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

The present invention provides a Monte Carlo geometric processing method for coupling spline surfaces and analytical surfaces in nuclear simulation analysis. When performing Monte Carlo particle transport calculations in reactor buildings or fusion reactors, due to complex spline surfaces in the geometric model, The traditional Monte Carlo program cannot handle it, and the model needs to be simplified. This method classifies the surface of the geometry in the CAD model. The analytical surface is described by algebraic equations, and the spline surface is described by parametric equations, which realizes the Monte Carlo program. Geometry processing of arbitrary surfaces, while using the limited space bounding box of the spline surface parametric equation to accelerate the geometry processing containing spline surfaces, the specific steps are as follows:

1. Import/create CAD geometry model

Import the CAD model containing all the geometry into the software interface, if there is no ready-made CAD model, create the CAD model required for calculation through the software interface. After completing the import/creation of the model, the entire calculation area is defined as R. If it is a created CAD model, it also needs to be exported to a CAD file in a common format (such as SAT, STP, etc.) for use in the Monte Carlo calculation model that couples CAD and CSG.

2. Identify spline surfaces in CAD models

Traversing all N geometric bodies in the CAD model, for any geometric body Solid _i (Solid _i ∈ R, i=1, N), obtain M _i faces Face _i,j (Face _{i, j} ∈Solid _i , j=1,M _i ), then traverse F _i,j and judge the type of each surface. If the type of F _{i1, j1} in Solid _i1 does not belong to the analytical surface (F _{i2, j2} ∈ S), where S = (plane, sphere, cylinder, cone, torus, quadratic surface), then Solid _i1 Mark as a geometry containing a spline surface, mark F _{i1, j1} as a spline surface, and record the serial number j1 of the spline surface in the geometry topology surface list; if the types of F _{i2, j2} in Soli _di2 belong to the analytical surface (F _{i2, j2} ∈ S), then it will not be processed. The specific process is shown in Figure 1.

Repeat the above operations until all the geometry and their surfaces in the model have been traversed to form the following collection,

Geometry collection with spline surface: Solid _N = {Solid _t1 }

Geometry collection containing only analytical surfaces: Solid _A = {Solid _t2 }

Set of spline surfaces: Face _N = {Face _{i, j} }, (i=i1, f=f1)

Analytical surface set: Face _A = {Face _{i, j} }, (i=i1&j≠j1) or (i=i2&j=1, Mi)

3. Convert the CAD model to generate a Monte Carlo calculation model:

1) Generate an expression for an analytic surface. Traverse all the analytic surfaces in the collection Face _A , including the type of analytic surface and the necessary parameter values, and also need to specify a unique string (surface type + corresponding serial number) for each analytic surface, for example, <Analytic_surface id= "1"type="plane"para="1 2 3 4"/>

2) Generate the expression of the spline surface. Traverse all the spline surfaces in the collection Face _N , including the serial number of the corresponding geometry, the serial number in the geometry face list and the bounding box information of the spline surface, and also need to formulate a unique string for each spline surface (surface type + corresponding serial number), for example,

<Nurbs_surface id="2"solidnum="2"indexnum="3"boundingbox="-10 -10-10 10 10 10"/>

3) Expressions for generating geometry. Boolean operations between surfaces are represented by specific characters, Boolean intersection (space), Boolean union (:), and Boolean subtraction (#). Iterates over all geometries and generates geometric expressions based on Boolean relationships between the surfaces of the current geometry. At the same time, each geometry also needs to specify a unique serial number (integer type), for example,

<Solid id="2" value="1 2 3 4"/>

4. Geometry processing of the Monte Carlo calculation model coupled with CAD and CSG:

When performing Monte Carlo particle transport, it is necessary to calculate the distance from the particle (along the ray direction) to each surface of the geometry, so as to obtain the shortest distance from the particle (along the ray direction) to the boundary of the geometry. Traverse all the surfaces in the geometry, if it is an analytic surface, directly use the ray and surface equation to perform intersection operation, if it is a spline surface, find the position of the spline surface in the corresponding geometry topology surface list according to the serial number, and then The spline surface is obtained by using the CAD underlying library function, and the intersection point between the particle trajectory and the spline surface is solved by Newton iterative method, and the distance from the particle (along the ray direction) to the spline surface is calculated accordingly. Before solving the intersection point of the ray and the spline surface, according to the space bounding box of the spline surface, perform the intersection operation of the ray and the bounding box, if there is no intersection point between the ray and the bounding box, then there is no need to calculate the ray and the spline surface cross operation. The specific process is shown in Figure 2.

5. Application examples, including the following:

Application example 1:

As shown in Figure 3, it is a geometry similar to a pipe joint, which contains a spline surface. When performing Monte Carlo particle transport, it is assumed that a space point (5.0, 5.0, 5.0) and a ray direction (0.707, 0.707, 0.0) are given, and it is required to implement Monte Carlo geometric processing based on coupled spline surfaces and analytical surfaces. It mainly includes the following steps:

1) Identify the spline surface in the CAD model. Since the model has only one geometry Solid ₁ , the three faces Face _1,1 , Face _1,2 , and Face _1,3 that make up the geometry are obtained according to its topology; then the type of each face is judged: Face _1,1 is analytical Surface (plane), Face _1,2 are spline surfaces, Face _1,1 are analytical surfaces (plane). In this way, Face _1,1 and Face _1,3 will be added to the collection of analytical surfaces Face _x ={Face _1,1 ,Face _1,3 }, Face _1,2 will be added to the collection of spline surfaces Face _N ={Face _{1, 2} }. The elements Face ₁ , 1, Face 1, ₃ in the Face _A collection need to store the type "p" (representing the plane) and parameters of the corresponding analytical surface, and the elements Face _{1 and 2} in the Face _N collection store the serial number of the corresponding geometry 1. The serial number 2 in the geometry surface list and the bounding box information of the spline surface (0 to 25.0 in the X direction; -25.0 to 25.0 in the Y direction; -4.000495 to 6.000495 in the Z direction). At the same time, the set of surface types that make up the geometry Solid ₁ is SolidSurface _Type = {A, S, A}, and the set of corresponding surface numbers is SolidSurface _Index = {1, 1, 2}.

2) Convert the CAD model to generate a Monte Carlo calculation model:

a) Generate an expression for the analytic surface. Traverse all analytical surfaces in the collection F, including the type of the analytical surface and the necessary parameter values, and also need to specify a unique string (surface type + corresponding serial number) for each analytical surface, as follows:

<Analytic_surface id="A1" type="p" para="0 1 0 12"/>

<Analytic_surface id="A2" type="p" para="1 0 0 0"/>

b) Expressions for generating spline surfaces. Traverse all the spline surfaces in the set F, including the serial number of the corresponding geometry, the serial number in the geometry surface list and the bounding box information of the spline surface, and also need to formulate a unique string for each spline surface ( Surface type + corresponding serial number), as follows:

<Nurbs_surface id="S1"solidnum="1"indexnum="2"

boundingbox="0 -25.0 -4.000495 25.0 25.0 6.000495"/>

c) Expressions for generating geometry. Boolean operations between surfaces are represented by specific characters, Boolean intersection (space), Boolean union (:), and Boolean subtraction (#). Traverse all the geometry, and generate the expression of the geometry according to the Boolean relationship between the surfaces of the current geometry, as follows:

<Solid id="1" value="A1 A2 -S1"/>

3) Geometry processing of the Monte Carlo calculation model coupling CAD and CSG: traverse all surfaces in the geometry, for Face ₁ , 1 , Face 1, ₃ , directly calculate the spatial point (5.0, 5.0, 5.0) along the ray direction (0.707, 0.707, 0.0) to the two planes; for Face ₁ , 2, first calculate whether the space point (5.0, 5.0, 5.0) intersects with the bounding box of the spline surface along the ray direction (0.707, 0.707, 0.0), after the intersection Then use the CAD underlying library function to calculate the distance from the spatial point (5.0, 5.0, 5.0) along the ray direction (0.707, 0.707, 0.0) to the spline plane.

Application example 2:

As shown in Figure 4, it is a typical superconducting stellarator-shaped fusion reactor. The external device of the fusion reactor is constructed by some regular geometric bodies or ordinary analytical surfaces, while the internal parts are mainly spliced by spline surfaces. become. When performing Monte Carlo particle transport, it is assumed that the source particles are emitted isotropically from the cylinder in the middle of the model in Figure 4, and it is required to implement Monte Carlo geometric processing based on the method of coupled spline surface and analytical surface. It mainly includes the following steps:

1) Identify the spline surface in the CAD model. Traverse the N geometries in the CAD model, for any solid _i , obtain the M _i faces that make up the geometry according to its topology, then traverse F _i,j and judge the type of each face. If the type of F _i1,j1 in Solid _i1 does not belong to the analytic surface, then mark Solid _i1 as a geometry containing a spline surface, mark F _i1,j1 as a spline surface, and record the spline surface in the geometry topology surface list The serial number j1 in ; if the types of F _{i2 and j2} in Solid _i2 belong to analytic surfaces, it will not be processed.

Repeat the above operations until all the geometry and their surfaces in the model have been traversed to form the following collection,

Geometry set with spline surface: Solid _N = {V5, V6, V7, V8, V9}

Geometry collection containing only analytical surfaces: Solid _A = {V1, V2, V3, V4, V10, V11, V12, V13}

Set of spline surfaces: Face _N ＝{F _5,1 ，…，F _9,1 }

Analytical surface set: Face _A = {F _1,1 , F _1,2 , F _1,3 , ..., F _13,1 , F _13,2 , F _13,3 , F _5,2 , F _5,3 ,………}

2) Convert the CAD model to generate a Monte Carlo calculation model:

a) Generate an expression for the analytic surface. Traverse all analytical surfaces in the collection Face _A , including the type of the analytical surface and the necessary parameter values, and also need to specify a unique string (surface type + corresponding serial number) for each analytical surface, as follows:

<Analytic_surface id="A11" type="tz" para="0 0-200 1400 1000 1000"/>

<Analytic_surface id="A12" type="px" para="700.0"/>

<Analytic_surface id="A13" type="px" para="2000.0"/>

...

<Analytic_surface id="A72" type="p" para="-0.31 0.95 0 0"/>

<Analytic_surface id="A73" type="p" para="0.30 0.95 0 0"/>

...

b) Expressions for generating spline surfaces. Traverse all spline surfaces in the collection Face _N , including the corresponding geometry

The serial number of the spline surface, the serial number in the geometric surface list and the bounding box information of the spline surface, and also need

Make a unique string (surface type + corresponding serial number) for each spline surface, as follows:

...

<Nurbs_surface id="S71"solidnum="7"indexnum="1"

boundingbox="1116.15 -592.45 -268.18 1651.49 592.45 268.18"/>

...

c) Expressions for generating geometry. Traverse all geometry, generate

The expression into a geometry is as follows:

<Solid id="1"value="-A11 A12 -A13"/>

...

<Solid id="7"value="-A71 A72 -A73"/>

...

3) Geometry processing of the Monte Carlo calculation model coupled with CAD and CSG: After sampling the position and direction of the source particle, the transport calculation is performed. When the particle is in the geometry without spline surface, use the original method to calculate the distance; when the particle is in the geometry with spline surface, use whether the particle ray intersects with the bounding box of the spline surface to filter the effective samples Strip surface, and use the CAD underlying library to perform intersection operation, and finally calculate the shortest effective distance of the particle knife geometry boundary.

The parts not described in detail in the present invention belong to the well-known technology in the art.

Although the illustrative specific embodiments of the present invention have been described above, so that those skilled in the art can understand the present invention, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art, As long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are included in the protection list.

Claims (2)

1. A Monte Carlo geometric processing method for coupled spline surfaces and analytical surfaces in nuclear simulation analysis, characterized in that: when carrying out reactor building or fusion reactor Monte Carlo particle transport calculations, due to complex geometric models The spline surface cannot be processed by the traditional Monte Carlo program, and the model needs to be simplified. This method classifies the surface of the geometry in the CAD model. The analytical surface is described by algebraic equations, and the spline surface is described by parametric equations. The Carlo program can process the geometry of any surface, and at the same time use the bounding box of the limited space of the parametric equation of the spline surface to accelerate the geometric processing of the spline surface, which specifically includes the following steps: Step (1), identify the spline surface in the CAD model: Traversing all the geometry in the CAD model of the reactor building or fusion reactor geometric model, for any geometry, obtain all the surfaces that make up the geometry according to its topology, and then judge the type of each surface, if the type of surface does not belong to the analytical surface, If the analytic surface is a plane, sphere, cylinder, cone, torus or quadratic surface, it is marked as a spline surface and recorded; Step (2), conversion CAD model generates Monte Carlo calculation model: Convert all the geometries in the CAD model, and use the combined entity representation (Construct Solid Geometry, CSG) to describe, that is, use the surface and the Boolean operation between the surfaces. When the surface type is an analytical surface, it is expressed in the form of a surface equation and recorded. The type of the analytical surface and the corresponding parameters, and when the surface type is a spline surface, record all the geometries corresponding to the spline surface and the serial numbers in the topological surface list of the corresponding geometry; Step (3), Monte Carlo calculation model geometric processing of coupling spline surface and analytical surface: When performing Monte Carlo particle transport, it is necessary to calculate the distance from the particle to each surface of the geometry along the ray direction, so as to obtain the shortest distance from the particle to the boundary of the geometry along the ray direction; Traverse all the surfaces in the geometry, if it is an analytic surface, directly use the ray and surface equation to perform intersection operation, if it is a spline surface, find the spline surface on the corresponding geometry topology surface according to the serial number in step (2) The position in the list, and then use the CAD underlying library function to obtain the spline surface, and use the Newton iterative method to solve the intersection point of the particle trajectory and the spline surface, and then calculate the distance from the particle along the ray direction to the spline surface. 2. the Monte Carlo geometric processing method of coupling spline surface and analytical surface in a kind of nuclear simulation analysis according to claim 1, it is characterized in that: when described step (2) generates Monte Carlo calculation model, for spline surface, generate its corresponding bounding box information; said step (3) before solving the intersection point of ray and spline surface, according to the space bounding box of spline surface, carry out the intersection operation of ray and bounding box, if ray If there is no intersection point with the bounding box, there is no need to perform the intersection operation between the ray and the spline surface.