Hexahedral mesh parallel generation method of fast reactor core assembly based on block structure method

Abstract

The invention relates to the technical field of finite element grids, and particularly discloses a block structure method-based hexahedral grid parallel generation method for a fast reactor core component, which comprises the following steps of: performing Delay inner triangulation in an area on the same side of a winding wire, solving a Laplace equation to generate an isoline, and generating an orthogonal grid according to the isoline; obtaining a triangular surface grid template on the same side of the winding wire; dividing the fuel rod assembly into six sections, dividing each section into a plurality of layers, taking the first section as a basic section, dividing block grids in each layer on the basis of the triangular surface grid template on the same side of the winding wire, generating quadrilateral surface grids in the block grids, rotating the grids on each layer at a set angle once to obtain grids on each layer of other sections, and forming the quadrilateral surface grids on the same side of the winding wire; the grids of all the layers are connected to form a hexahedral grid model; the hexahedral mesh model is expanded; and carrying out parallel optimization on the loop call function. According to the method, the problems that the hexahedral mesh of the fast reactor core assembly cannot be correctly generated and the mesh division efficiency is low are solved.

Description

Technical Field

[0001] The invention relates to the technical field of finite element meshes, and in particular to a method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method. Background Art

[0002] In the thermal-hydraulic simulation of fast neutron reactors, the mesh type has a significant impact on the accuracy of the simulation results. At present, the commonly used mesh types mainly include tetrahedral meshes and hexahedral meshes. Compared with tetrahedral meshes, hexahedral meshes are closer to orthogonality, which effectively reduces the condition number of numerical calculations, and thus have higher numerical accuracy, lower spatial complexity and higher efficiency.

[0003] At present, according to the reference "Pietroni N, Campen M, Sheffer A, et al. Hex-mesh generation and processing: a survey [J]. ACM transactions on graphics, 2022, 42 (2): 1-44.", common hexahedral mesh generation algorithms include sweeping method, mapping method, etc. The sweeping method is a method of connecting a two-dimensional quadrilateral mesh into a hexahedral mesh by geometric transformations such as rotation, stretching and translation along a path. Among them, the initial surface is called the "source" and the surface finally reached along the path is called the "target". This method can easily divide the mesh by identifying the source and the target. However, for more complex shapes, it may be difficult to find a suitable sweeping path using the sweeping method, or it may need to be decomposed into multiple sources or targets. The mapping method maps the complex geometry of the physical space to the regular computational space through coordinate transformation to form a regular spatial geometry, then divides the mesh in the computational space, and finally maps it back to the geometry of the physical space. However, for complex geometric shapes, the use of the mapping method has the disadvantage of low approximation accuracy and difficulty in automation.

[0004] The hexahedral mesh generation algorithms of mainstream commercial meshing software mainly use the above two methods. For example, ICEM provides three hexahedral mesh generation algorithms: sweeping method, sweep-based decomposition method and hexahedral block-based decomposition method. Among them, the method based on hexahedral block decomposition means first decomposing the model into a set of hexahedral blocks, and then using the mapping method to generate a hexahedral mesh for each sub-block. However, the hexahedral mesh generation algorithm provided by ICEM cannot generate a correct hexagonal mesh model when processing the fast reactor core component model. There are problems such as the outer boundary not matching the geometry and not matching the mesh nodes of the internal flow channel (such as the attached Figure 1 GMSH can generate hexahedral meshes by sweeping method. However, since the winding wire increases the geometric complexity of the model, simply using the sweeping method cannot generate a correct mesh model.

[0005] Therefore, the technical problem to be solved by the present invention is the problem that the hexahedral mesh of the fast reactor core assembly cannot be correctly generated and the mesh division efficiency is low. Summary of the invention

[0006] In order to solve the above problems, the present invention provides a method for parallel generation of hexahedral meshes of fast reactor core assemblies based on the block structure method, so as to solve the problem that the hexahedral meshes of fast reactor core assemblies cannot be correctly generated. Taking the assembly of 19 fuel rods as a benchmark and utilizing the central symmetry characteristics of the hexagon, the block structure method is first used to generate a quadrilateral mesh with one-sixth of the winding pitch as the basic segment, and a complete assembly mesh is constructed through a 60° rotation operation, which is finally extended to different rod bundle assembly models.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0008] A method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method, the method comprising:

[0009] Delaunay triangulation is performed on the area on the same side of the wire winding, the Laplace equation is solved to generate contour lines, and an orthogonal grid is generated based on the contour lines to obtain a triangular surface mesh template on the same side of the wire winding;

[0010] The fuel rod assembly is divided into six sections, each section is further divided into several layers, the first section is used as the basic section, a block grid is divided in each layer based on the triangular surface grid template on the same side of the winding wire, and a quadrilateral surface grid is generated in the block grid, each layer grid is rotated once by a set angle to obtain each layer grid of other sections, and each layer grid is connected to form a hexahedral grid model;

[0011] Expanding the hexahedral grid model;

[0012] The loop calling function is optimized in parallel; wherein the loop calling function is used to generate quadrilateral meshes of each layer.

[0013] Furthermore, Delaunay triangulation is performed on the area on the same side of the wire winding, the Laplace equation is solved to generate contour lines, and an orthogonal grid is generated according to the contour lines to obtain a triangular surface mesh template on the same side of the wire winding, including:

[0014] According to the wire winding information, Delaunay triangulation is used in the area on the same side of the wire winding to obtain Delaunay triangulated grid points;

[0015] The generated Delaunay triangle mesh points are used to solve the Laplace equation using the finite element method to obtain the solutions for all mesh points;

[0016] At the boundary of the template on the same side of the wire winding, spline interpolation is used to make the points on the boundary distributed at a uniform angle; linear interpolation is performed between the internal grid points to find the points with the same function value as the corresponding angle, and the points with the same function value are connected to form contour lines. Spline interpolation is performed in the radial direction according to the contour lines to obtain a preliminary orthogonal grid, and then the corresponding grid points are obtained through spline interpolation to form the final orthogonal grid.

[0017] Furthermore, the wire winding information includes the center spacing of the fuel rods, the radius of the fuel rods, the wire winding radius, the wire filling radius, the radius of the grid template, the center coordinates of the grid template, the rotation angle of the wire winding, the half-angle angle of the grid template, the calibration point of the grid template, the truncation length of the wire winding and the embedding depth of the wire winding.

[0018] Furthermore, based on the triangular surface mesh template on the same side of the wire wrap, a block mesh is divided in each layer, and a quadrilateral surface mesh is generated in the block mesh, including:

[0019] Calculate the center position of each fuel rod, connect the center point, the points on the edge and the points on the corner of the fuel rod to form a unit corresponding to the fuel rod, calculate the center point of each unit, connect the center point of each unit, the points on the boundary of the fuel rod and the points on the boundary of the assembly to divide the block grid;

[0020] Generate grid points on the curved boundary of the block network based on the projection method;

[0021] According to the location of the point, the grid points on the straight boundary of the block network are generated by using the Newton search and spline interpolation methods and the projection method respectively; the grid points on the straight boundary of the block network include the grid points on the opposite side of the wire winding and the grid points on the same side of the wire winding, and the grid points on the opposite side of the wire winding are the grid points in the area where the angle with the wire winding direction is greater than 90°;

[0022] Generate grid points in the block network near the fuel rods based on the projection method;

[0023] The grid points within the block network located at the component boundaries are generated by linear interpolation.

[0024] Furthermore, the grid points on the curve boundary of the block network are generated based on the projection method, including:

[0025] Points are added at equal distances along the line connecting the two end points of the curve, and the projection points of the added points on the fuel rod and the winding wire are found by Newton's method along the normal direction to generate points on the boundary of the block curve; the iterative method of Newton's method is as follows:

[0026]

[0027] Where x0 is the point of the previous iteration, x1 is the point of the current iteration, dx is the distance between the points of the previous two iterations, z1 is the shortest distance to the fuel rod and the winding in the previous iteration, and dz is the difference between the shortest distances of the previous two iterations.

[0028] Furthermore, according to the location of the point, the Newton search and spline interpolation methods and the projection method are used to generate grid points on the straight line boundary of the block network, including:

[0029] The grid points on the opposite side of the wire winding are found by the gradient descent method. The projection point of the starting point on the fuel rod is obtained. After the projection point is obtained, evenly distributed points are generated along the direction of the projection point to obtain the grid points on the opposite side of the wire winding. The iterative formula of the gradient descent is:

[0030]

[0031] Where x0 is the point of the previous iteration, x1 is the point of the current iteration, h is the step size, grd is the gradient of the current point relative to the fuel rod, that is, the direction of the shortest path from the current point to the fuel rod, and norm(grd) is the norm of the gradient vector;

[0032] For the grid points on the same side of the wire wrap, the position of the starting point in the triangular mesh template is determined by Newton search; in the process of finding the position of the grid point, the Newton search process is regarded as finding the solution of the function (x, y) = f(s, t) at x = x0, y = y0; wherein x0, y0 are the coordinates of the starting point, s is the radial distribution position of the mesh template, and t is the angle of the grid point in the template. After obtaining the radial distribution position and angle of the starting point, the distribution position s′ of the point on the row where the starting point is located is calculated according to the radial distribution position of the starting point, and then the grid point on the same side of the wire wrap is obtained by interpolation based on the distribution position s′ and the angle t.

[0033] Furthermore, each layer grid is rotated once by a set angle to obtain each layer grid of other segments, and each layer grid is connected to form a hexahedral grid model, including:

[0034] The base segment is rotated 5 times, each time by 60°, to obtain a layer mesh of the wire length assembly. The meshes of each layer are connected, and then the mesh around the central fuel rod is translated into the block that has not been divided into meshes to obtain a hexahedral mesh model.

[0035] Furthermore, the hexahedral mesh model is a hexahedral mesh model of a 19-pin component, and the hexahedral mesh model is expanded by the following steps to obtain a hexahedral mesh model of a 7-pin component:

[0036] Number the fuel rods and blocks of the 19-pin assembly by position and type;

[0037] Match the numbers on the 19-pin assembly to the 7-pin assembly according to the positions of the fuel rods and blocks;

[0038] According to the positions of the blocks in the 19-pin assembly and the 7-pin assembly, the distance to be translated of the fuel rod is calculated, and based on the distance to be translated of the fuel rod, the blocks of the 19-pin assembly are translated to the corresponding positions of the 7-pin assembly.

[0039] Furthermore, the loop calling function is optimized in parallel, including:

[0040] The parfor tool provided by Matlab is used to perform parallel processing on the cyclically called functions.

[0041] The present invention has at least the following beneficial effects:

[0042] (1) The difficulty in generating hexahedral meshes for fast reactor core components is solved. The hexahedral meshes generated by the present invention can be well matched to the geometry and can meet the computational requirements of thermal fluid simulation programs;

[0043] (2) The problem of low efficiency in hexahedral mesh generation was solved. The program was optimized in parallel using the parfor tool, which significantly improved the program efficiency. In the case of quad-core parallelism, the speedup ratio of the parallel program can reach 3.51, which is close to the linear speedup ratio. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The figure shows a single-channel hexahedral mesh model generated by CEM according to the prior art, whose outer boundary is obviously inconsistent with the geometry.

[0045] Figure 2 A flow chart of a method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method according to an embodiment of the present invention is shown.

[0046] Figure 3 Schematic diagrams of the components of a block grid according to an embodiment of the present invention are shown, wherein (a) shows the components of the blocks around the wire, and (b) shows the components of the blocks at the boundary of the component.

[0047] Figure 4 FIG. 4 shows a schematic diagram of some parameters of a grid template of a block grid according to an embodiment of the present invention.

[0048] Figure 5 A schematic diagram of units and blocks of a 19-pin component according to an embodiment of the present invention is shown; wherein (a) shows the units divided on the 19-pin component, and (b) shows the blocks divided on the 19-pin component.

[0049] Figure 6A schematic diagram of the 19-pin assembly fuel rods and type numbers according to an embodiment of the present invention is shown; wherein (a) shows the numbering of the 19-pin assembly fuel rods, with the serial numbers increasing from bottom to top and from left to right, and (b) shows the type numbers of the 19-pin assembly fuel rods, which are classified and numbered according to the location of the fuel.

[0050] Figure 7 A schematic diagram of different Pennisetum 7-pin assembly fuel rods and type numbers according to an embodiment of the invention is shown; (a) shows the numbering of the 7-pin assembly fuel rods, from bottom to top and from left to right, with the serial numbers increasing; (b) shows the type numbering of the 7-pin assembly fuel rods, which are classified and numbered according to the location of the fuel. DETAILED DESCRIPTION

[0051] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0052] The specific implementation of the present invention is further described in detail below in conjunction with the drawings and examples.

[0053] The embodiment of the present invention provides a method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method. This embodiment takes a hexahedral mesh model of a 7-pin assembly with a 100 mm (i.e., one winding pitch) as an example to further illustrate the present invention. Among them, each sixth pitch of the model is divided into 48 layers, and each layer of blocks is divided into 3 rows and 50 columns. The program uses 4 cores for parallel optimization on an Intel i-9300HF processor.

[0054] See also Figure 2 , is a flowchart of the parallel generation method of hexahedral meshes for fast reactor core components based on the block structure method. Figure 2 As shown, the method comprises the following steps:

[0055] Step S1, generating a triangular surface mesh template on the same side of the wire winding: performing Delaunay triangulation on the area on the same side of the wire winding, solving the Laplace equation to generate contour lines, and generating an orthogonal mesh based on the contour lines.

[0056] Due to the complex geometry on the same side of the wire (i.e. the area where the angle between the grid point and the wire direction does not exceed 90°), it is difficult to directly divide the orthogonal quadrilateral mesh in the block. Therefore, it is necessary to generate an orthogonal mesh template on the same side of the wire before dividing the quadrilateral mesh to assist in the subsequent generation of the quadrilateral mesh.

[0057] In some embodiments, the specific steps of generating the triangular face mesh template are as follows:

[0058] Step S11, according to the basic information of the wire winding, Delaunay triangulation is used in the area on the same side of the wire winding, wherein the wire winding information includes the center distance of the fuel rods, the radius of the fuel rods, the wire winding radius, the wire winding corner radius, the radius of the grid template, the center coordinates of the grid template, the rotation angle of the wire winding, the half angle of the grid template, the calibration point of the grid template, the truncation length of the wire winding and the embedding depth of the wire winding (such as Figure 4 The obtained Delaunay triangulated mesh points are used for subsequent finite element method calculations.

[0059] Step S12, using the generated Delaunay triangular grid points to solve the Laplace equation using the finite element method to obtain solutions for all grid points. Since the solutions of the Laplace equation have good orthogonality, these solutions can be used to generate contour lines to assist in the generation of orthogonal grids.

[0060] Step S13, at the boundary of the template on the same side of the wire winding, use spline interpolation to make the points on the boundary distributed at a more uniform angle. Perform linear interpolation between the internal grid points to find the points with the same function value as the corresponding angle. These points with the same function value are connected to form isovalue lines. Spline interpolation is performed in the radial direction according to the isovalue lines to obtain a preliminary orthogonal grid.

[0061] Step S14, in order to improve the quality of the grid, the distribution of the grid points on the boundary is readjusted so that the points near the wire fillet are more concentrated, and then the corresponding grid points are obtained by spline interpolation to form the final orthogonal grid.

[0062] Step S2, hexahedral meshing based on block structure: the 19-pin fuel rod assembly is divided into six sections, each of which is further divided into several layers. The first section is used as the basic section, and block meshes are divided in each layer. Then, quadrilateral surface meshes are generated in the block meshes (such as Figure 3 ). Rotate the mesh of each layer by 60° to get the mesh of each layer of other segments. Finally, connect the mesh of each layer to form a hexahedral mesh.

[0063] In this embodiment, step S2 takes the assembly of 19 fuel rods as the reference, subdivides the basic segment of the assembly with a pitch of one-sixth into several layers, and first divides the large block grid in each layer, and then divides the quadrilateral grid in the block grid. The basic segment is rotated to obtain a layer grid of one pitch. Finally, the layer grids are connected to form a hexahedral grid model. When generating the quadrilateral grid, only the grid points of the boundary block, the block around the outermost fuel rod, and the block around the central fuel rod are generated. The grids of other blocks can be obtained by translating the grids of the block around the central fuel rod. It is specifically implemented by the following steps:

[0064] Step S21, calculate the center position of each fuel rod, connect the center point, the edge points and the corner points of the fuel rod to form a large unit (such as Figure 5 ), calculate the center point of each unit. Connect the center point of the unit, the points on the fuel rod boundary, and the points on the assembly boundary to divide the block grid. Each block will be divided into a quadrilateral grid with a specified number of rows and columns.

[0065] Step S22, the grid points on the block curve boundary are obtained by projection. First, points are added at equal distances along the direction of the line connecting the two end points of the curve, and then these points are projected on the fuel rod and the winding wire using Newton's method along the normal direction, thereby generating points on the block curve boundary. The iterative method of Newton's method is as follows:

[0066]

[0067] Among them, x0 is the point of the previous iteration, x1 is the point of the current iteration, h is the step size, grd is the gradient of the current point relative to the fuel rod, that is, the direction of the shortest path from the current point to the fuel rod, and norm(grd) is the norm of the gradient vector.

[0068] Step S23 , grid points on the straight boundary of the block are generated by using Newton search and spline interpolation methods and projection methods according to the locations of the points.

[0069] The grid points on the opposite side of the wire winding (i.e., the area where the angle between the grid points and the wire winding direction is greater than 90°) are generated by projection. The projection point of the starting point (the vertex of the block or the point on the block boundary) on the fuel rod is found by the gradient descent method. The iterative formula is:

[0070]

[0071] Where h is the step size, grd is the gradient of the current point relative to the fuel rod, that is, the direction of the shortest path from the current point to the fuel rod, and norm(grd) is the norm of the gradient vector. After obtaining the projection point, evenly distributed points are generated along the direction of the projection point to form a quadrilateral grid in the block.

[0072] Since the geometry on the same side of the wire winding is more complex, a grid template on the same side of the wire winding is needed to assist in generating grid points. In the process of generating grid points, it is necessary to find the position of the starting point in the grid template. This process is implemented through Newton search. In the process of finding the position of the grid point, the process of Newton search is regarded as finding the solution of the function (x, y) = f(s, t) at x = x0, y = y0. Among them, x0, y0 are the coordinates of the starting point, s is the radial distribution position of the grid template, and t is the angle of the grid point in the template. After obtaining the radial distribution position and angle of the starting point, the distribution position s′ of the point on the row where the point is located is calculated according to the radial distribution position of the point, and then the grid point is interpolated according to the distribution position s′ and the angle t.

[0073] Step S24: The grid points in the block near the fuel rod are generated in a similar manner to the grid points on the curved boundary of the block, and are generated by using Newton search and spline interpolation methods and projection methods according to the locations of the points.

[0074] Step S25 , the block grid points located at the component boundary are directly generated by linear interpolation.

[0075] Step S26, rotate the base segment 5 times, 60° each time, to obtain a layer mesh of the wire length assembly. Connect the layer meshes, and then translate the mesh around the central fuel rod to the block that has not been meshed, to complete the hexahedral meshing of the 19pin assembly.

[0076] Step S3, mesh model expansion: expanding the mesh from the 19-pin model to component models of other rod bundles.

[0077] In some embodiments, the hexahedral mesh model of the 19-pin component can be extended to the component models of other rod bundles through mesh model expansion. The specific steps are as follows, taking the 7-pin component as an example:

[0078] Step S31, number the fuel rods and blocks of the 19-pin assembly by position and type (see attached Figure 6 ).

[0079] Step S32, according to the position of the fuel rod and the block, the number on the 19 pin is matched to the 7 pin (as shown in the attached Figure 7 ).

[0080] Step S33, according to the position of the block in the 19-pin assembly and the 7-pin assembly, the distance that the fuel rod needs to be translated is calculated, and then the 19-pin block is translated to the corresponding position of the 7-pin assembly.

[0081] Step S4, parallel optimization: Use the parfor tool provided by Matlab to perform parallel processing on the functions of dividing each layer of quadrilateral meshes.

[0082] In this embodiment, as shown in Table 1, the parfor tool provided by Matlab is used to parallel optimize the program. When using the parfor loop to execute parallel tasks, MATLAB will automatically assign independent loop iterations to multiple Matlab workers in the parallel pool. After each worker is assigned a task, it will exclusively use a process to execute the code in the loop body.

[0083] Table 1 Parallel optimization process

[0084]

[0085] In the process of serially executing the hexahedral meshing program, since the quadrilateral mesh needs to be divided into layers, the program needs to call the function gen_ess_blocks_L cyclically to generate each layer of quadrilateral mesh, and the amount of calculation data of the function gen_ess_blocks_L is large. This causes the time required to execute this loop to account for a large proportion of the running time of the entire program. Taking the model of a 7-fuel rod assembly with a length of 100mm (1 winding pitch) as an example, on a Windows 11 operating system with 16.0GB of memory, the execution time of serially generating about 1 million hexahedral meshes using an Intel i-9300HF processor is about 2915 seconds, and the execution time of the function gen_ess_blocks_L is about 2746 seconds, accounting for 94.2% of the entire program execution time.

[0086] Therefore, the present invention uses parfor to execute the function in parallel according to Table 1. When four threads are executed in parallel, it takes about 831 seconds to generate the same hexahedral mesh, and the speedup ratio is about 3.51, which is basically close to the linear speedup ratio, and the efficiency of mesh generation is significantly improved.

[0087] The above implementation modes are only used to illustrate the present invention, but not to limit the present invention. Ordinary technicians in the relevant technical field can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, all equivalent technical solutions also belong to the scope of the present invention. The patent protection scope of the present invention should be defined by the claims.

Claims

1. A method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method, characterized in that: The method comprises: Delaunay triangulation is performed on the area on the same side of the wire winding, the Laplace equation is solved to generate contour lines, and an orthogonal grid is generated based on the contour lines to obtain a triangular surface mesh template on the same side of the wire winding; The fuel rod assembly is divided into six sections, each section is further divided into several layers, the first section is used as the basic section, a block grid is divided in each layer based on the triangular surface grid template on the same side of the winding wire, and a quadrilateral surface grid is generated in the block grid, each layer grid is rotated once by a set angle to obtain each layer grid of other sections, and each layer grid is connected to form a hexahedral grid model; Expanding the hexahedral grid model; The loop calling function is optimized in parallel; wherein the loop calling function is used to generate quadrilateral meshes of each layer.

2. The method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method according to claim 1, characterized in that: Delaunay triangulation is performed on the area on the same side of the wire winding, the Laplace equation is solved to generate contour lines, and an orthogonal grid is generated based on the contour lines to obtain a triangular surface mesh template on the same side of the wire winding, including: According to the wire winding information, Delaunay triangulation is used in the area on the same side of the wire winding to obtain Delaunay triangulated grid points; The generated Delaunay triangle mesh points are used to solve the Laplace equation using the finite element method to obtain the solutions for all mesh points; At the boundary of the template on the same side of the wire winding, spline interpolation is used to make the points on the boundary distributed at a uniform angle; linear interpolation is performed between the internal grid points to find the points with the same function value as the corresponding angle, and the points with the same function value are connected to form contour lines. Spline interpolation is performed radially according to the contour lines to obtain a preliminary orthogonal grid, and then the corresponding grid points are obtained through spline interpolation to form the final orthogonal grid.

3. The method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method according to claim 2, characterized in that: The wire winding information includes the fuel rod center spacing, fuel rod radius, wire winding radius, wire winding corner radius, grid template radius, grid template center coordinates, wire winding rotation angle, grid template half angle, grid template calibration point, wire winding cutoff length and wire winding embedding depth.

4. The method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method according to claim 1, characterized in that: Based on the triangular surface mesh template on the same side of the wire wrap, a block mesh is divided in each layer, and a quadrilateral surface mesh is generated in the block mesh, including: Calculate the center position of each fuel rod, connect the center point, the points on the edge and the points on the corner of the fuel rod to form a unit corresponding to the fuel rod, calculate the center point of each unit, connect the center point of each unit, the points on the boundary of the fuel rod and the points on the boundary of the assembly to divide the block grid; Generate grid points on the curved boundary of the block network based on the projection method; According to the location of the point, the grid points on the straight boundary of the block network are generated by using the Newton search and spline interpolation methods and the projection method respectively; the grid points on the straight boundary of the block network include the grid points on the opposite side of the wire winding and the grid points on the same side of the wire winding, and the grid points on the opposite side of the wire winding are the grid points in the area where the angle with the wire winding direction is greater than 90°; Generate grid points in the block network near the fuel rods based on the projection method; The grid points within the block network located at the component boundaries are generated by linear interpolation.

5. The method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method according to claim 4, characterized in that: Generate mesh points on the curve boundary of the block network based on the projection method, including: Points are added at equal distances along the line connecting the two end points of the curve, and the projection points of the added points on the fuel rod and the winding wire are found by Newton's method along the normal direction to generate points on the boundary of the block curve; the iterative method of Newton's method is as follows: Where x0 is the point of the previous iteration, x1 is the point of the current iteration, dx is the distance between the points of the previous two iterations, z1 is the shortest distance to the fuel rod and the winding in the previous iteration, and dz is the difference between the shortest distances of the previous two iterations.

6. The method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method according to claim 4, characterized in that: According to the location of the point, the grid points on the straight line boundary of the block network are generated by using Newton search and spline interpolation methods and projection methods, including: The grid points on the opposite side of the wire winding are found by the gradient descent method. The projection point of the starting point on the fuel rod is obtained. After the projection point is obtained, evenly distributed points are generated along the direction of the projection point to obtain the grid points on the opposite side of the wire winding. The iterative formula of the gradient descent is: Where x0 is the point of the previous iteration, x1 is the point of the current iteration, h is the step size, grd is the gradient of the current point relative to the fuel rod, that is, the direction of the shortest path from the current point to the fuel rod, and norm(grd) is the norm of the gradient vector; For the grid points on the same side of the wire wrap, the position of the starting point in the triangular mesh template is determined by Newton search; in the process of finding the position of the grid point, the Newton search process is regarded as finding the solution of the function (x, y) = f(s, t) at x = x0, y = y0; wherein x0, y0 are the coordinates of the starting point, s is the radial distribution position of the mesh template, and t is the angle of the grid point in the template. After obtaining the radial distribution position and angle of the starting point, the distribution position s′ of the point on the row where the starting point is located is calculated according to the radial distribution position of the starting point, and then the grid point on the same side of the wire wrap is obtained by interpolation based on the distribution position s′ and the angle t.

7. The method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method according to claim 4, characterized in that: The meshes of each layer are rotated once by a set angle to obtain the meshes of each layer of other segments, and the meshes of each layer are connected to form a hexahedral mesh model, including: The base segment is rotated 5 times, each time by 60°, to obtain a layer mesh of the wire length assembly. The meshes of each layer are connected, and then the mesh around the central fuel rod is translated into the block that has not been divided into meshes to obtain a hexahedral mesh model.

8. The method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method according to claim 1, characterized in that: The hexahedral mesh model is a hexahedral mesh model of a 19-pin component. The hexahedral mesh model is expanded by the following steps to obtain a hexahedral mesh model of a 7-pin component: Number the fuel rods and blocks of the 19-pin assembly by position and type; Match the numbers on the 19-pin assembly to the 7-pin assembly according to the positions of the fuel rods and blocks; According to the positions of the blocks in the 19-pin assembly and the 7-pin assembly, the distance to be translated of the fuel rod is calculated, and based on the distance to be translated of the fuel rod, the blocks of the 19-pin assembly are translated to the corresponding positions of the 7-pin assembly.

9. The method for parallel generation of hexahedral meshes of a fast reactor core assembly based on a block structure method according to claim 1, characterized in that: Parallel optimization of loop-calling functions, including: The parfor tool provided by Matlab is used to perform parallel processing on the cyclically called functions.

Images