Improved non-uniform radiation field grid division method and system
By employing a method of first sparsification and then refinement in the radiation field calculation, and adjusting the grid weights according to the distance of the source term enclosing the ellipsoid, the problem of difficulty in balancing computation time and accuracy in existing technologies is solved, and efficient and accurate radiation field calculation is achieved.
Patent Information
- Application Number
- CN202511379041.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-25
- Publication Date
- 2026-01-09
AI Technical Summary
Existing technologies struggle to balance computational efficiency and accuracy in radiation field calculations. Initial mesh parameter settings are difficult to control, and the impact of source size on mesh generation is not considered, leading to increased computation time or reduced accuracy.
The method of first sparsifying and then refining is adopted. By performing N-fold sparsification on the initial uniform grid, the weight multiple is determined according to the distance from the grid center point to the ellipsoid enclosed by the source term, and the refining process is only performed in the region near the source term to form an adaptive non-uniform grid partition.
This approach reduces the number of calculation points while maintaining computational accuracy, improves the efficiency of radiation field calculation, and enhances the visualization of radiation fields and the accuracy of dose assessment.
Smart Images
Figure CN121301705A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of radiation field calculation technology, and in particular to an improved method and system for non-uniform radiation field mesh generation. Background Technology
[0002] To visualize the radiation dose rate distribution at nuclear facilities and conduct subsequent dose assessments for workers, it is necessary to first calculate the radiation field. The radiation field consists of a series of three-dimensional lattices containing coordinate and dose rate information, and the lattice coordinates are usually obtained by meshing. The traditional method for meshing the radiation field is the uniform meshing method. This method is simple in logic and easy to implement. It only requires specifying the number of divisions in the X, Y, and Z directions to complete the meshing of the radiation field. The disadvantage is that it cannot balance the calculation efficiency and calculation accuracy of the radiation field: that is, the finer the radiation field meshing, the higher the calculation accuracy of the radiation field, and the better the visualization effect and the accuracy of dose assessment. However, it will also lead to an increase in the number of calculation points in the lattice, thereby increasing the calculation time of the radiation field. To solve this problem, the existing technologies propose a fast calculation and source term inversion algorithm for γ radiation field and its preliminary application [D]. Tsinghua University, 2016; a variable weight meshing method based on the gradient change of the radiation field [J]. Journal of Tsinghua University (Natural Science Edition), 2019, 59(10): 861-865. By using different weighting factors for different types of source terms, the original 1-3 grids extending near the source term are refined and densified, effectively shortening the calculation time of the radiation field.
[0003] While this non-uniform mesh generation method improves the computational efficiency of the radiation field, it still requires improvement in practical applications. For example, because this method refines a custom initial mesh, setting the initial mesh parameters is difficult to control. If the initial mesh spacing is too small, the number of 3D computation points in the final output will exceed expectations, significantly increasing computation time while offering only limited accuracy improvement. If the initial mesh spacing is too large, the density of 3D computation points in the vicinity of the source term may be low after mesh refinement, affecting the accuracy of the radiation field calculation. Furthermore, this method does not consider the influence of the source size on the radiation field mesh generation, with a fixed number of refined meshes extending outward in each direction (1-3 original meshes). When the radiation source is large and the mesh spacing is relatively small, the number of meshes requiring refinement in its vicinity may be greater than the method anticipates, compromising the accuracy of the radiation field calculation and consequently reducing the visualization effect of the radiation field and the accuracy of dose assessment. Summary of the Invention
[0004] The purpose of this invention is to address the problems existing in the background technology by proposing an improved method for non-uniform radiation field mesh generation. This method can achieve adaptive non-uniform mesh generation based on an initially set uniform mesh, and compared with existing technologies, it can better balance the computational efficiency and accuracy of the radiation field.
[0005] The technical solution of the present invention, in its first aspect, provides an improved method for mesh generation of a non-uniform radiation field, comprising the following specific steps:
[0006] S1. Set the initial uniform grid for the radiation field calculation region;
[0007] S2. Perform N-fold sparsification on the initially set uniform grid;
[0008] S3. Calculate the distance from the center point of each sparsified grid to the ellipsoid enclosed by the source term, and determine the weight multiple of the grid based on the judgment distance;
[0009] S4. Complete the traversal of all source items and grids in sequence;
[0010] S5. Refine all grids with a weight multiple greater than 1 according to the corresponding multiple;
[0011] S6. Summarize and output all grid intersections, which is the final radiation field calculation lattice.
[0012] Preferably, in step S1, a uniform grid is set for the calculation region of the radiation field according to the boundary conditions of the radiation field, and the number of divisions in the X, Y, and Z directions of the radiation field calculation region is determined.
[0013] Preferably, in step S2, after the uniform grid is sparsified by N times, the grid spacing in each direction is increased to N times the original size; N is an integer and N>1; the specific value needs to be determined according to the size of the radiation scene and the radiation field calculation accuracy and efficiency that the user wants to achieve.
[0014] Preferably, a judgment threshold is set in step S3;
[0015] When the distance is less than the preset threshold, the weight multiplier of the grid is set to N; otherwise, it is set to 1.
[0016] Preferably, in step S3, if the source item is a point source, the determination threshold is set to 100cm; if it is less than 100cm, the weight multiple of the grid is set to N; otherwise, it is set to 1.
[0017] Preferably, in step S3, the equation of the enclosing ellipsoid of the cylindrical source is:
[0018]
[0019] The equation of the enclosing ellipsoid of the circular tube source is:
[0020]
[0021] The equation of the enclosing ellipsoid of the conical source is:
[0022]
[0023] The equation of the enclosing ellipsoid of the cuboid source is:
[0024]
[0025] The equation of the ellipsoid surrounding the spherical source is the same as its spherical equation:
[0026] (x-x0) 2 +(y-y0) 2 +(z-z0) 2 =R 2
[0027] In the above formula, R is the radius; H is the height; R 外 L is the outer radius; W is the length; x0, y0, z0 are the center coordinates of the source term; if the source term is rotated, the coordinates of the grid center point need to be transformed to the coordinate system of the source term using a rotation matrix before calculating the distance from the center point of the grid to the enclosing ellipsoid.
[0028] Preferably, in step S5, the mesh is densified, and the spacing of the resulting sub-mesh is 1 / N times the spacing of the mesh before refinement.
[0029] A second aspect of the present invention provides an improved non-uniform radiation field gridding system, which uses the above-described method for gridding, including:
[0030] The input module is used to input the three-dimensional model of the radiation field, as well as the type and coordinates of the source terms;
[0031] The initial grid setting module is used to set up an initial uniform grid for the radiation field calculation region;
[0032] The mesh sparsification module is used to sparsify the initial uniform mesh by a factor of N.
[0033] The weight calculation module is used to calculate the distance from the center point of each grid after sparsification to at least one source term enclosing ellipsoid, and determine the weight multiple of the corresponding grid based on the judgment relationship between the distance and the preset threshold.
[0034] The mesh refinement module is used to refine all meshes with a weight multiple greater than 1 according to their corresponding weight multiples.
[0035] The computation module is used to control the sequential traversal calculation of all source terms and all sparsified meshes;
[0036] The lattice output module is used to summarize and output the intersection points of all refined and unrefined meshes to generate the final radiation field calculation lattice and output it.
[0037] A third aspect of the present invention provides a computing device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; the memory storing instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the method described above.
[0038] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method.
[0039] Compared with the prior art, the present invention has the following beneficial technical effects:
[0040] 1. Existing non-uniform mesh generation methods directly refine the initial mesh, making it difficult to control the initial mesh parameters. If the initial mesh spacing is too small, the number of calculated radiation field points in the final output will exceed expectations, resulting in limited accuracy improvement but a significant increase in computation time. If the initial mesh spacing is too large, the density of calculated points near the source term may be low after mesh refinement, affecting the accuracy of the radiation field calculation. This invention employs a method of first sparsifying and then re-refining. First, according to the actual needs of radiation field visualization and dose assessment, the mesh parameters are initially set to obtain a relatively dense radiation field mesh. Then, the mesh of the entire region is sparsified, increasing the mesh spacing to N times the original value. Finally, the mesh near the source is refined and densified again by the same multiple. This achieves the mesh density near the source term being the density previously set according to actual needs, while the mesh density in areas farther from the source is relatively low. The total number of calculated radiation field points is less than the number corresponding to the initial mesh, but the calculation accuracy is similar.
[0041] 2. Existing non-uniform mesh generation methods do not consider the influence of source size on radiation field mesh generation, resulting in a fixed number of refined meshes extending outward in each direction (1-3 original meshes). When the radiation source is large and the mesh spacing is relatively small, the number of meshes requiring refinement in its vicinity may exceed the method's expectations. This invention improves the criterion for determining whether mesh refinement is necessary based on the radiation field variation patterns in the vicinity of the source term. The criterion has been changed from whether it is an extended mesh to whether the ratio of the distance from the mesh center point to the ellipsoid surrounding the source term to the source dimension is less than 1.5 (or less than 100cm for point sources). This makes the mesh generation near the source term more reasonable, better balancing computational efficiency and accuracy. Attached Figure Description
[0042] Figure 1 This is a flowchart of the improved non-uniform radiation field mesh generation method in an embodiment of the present invention;
[0043] Figure 2 This is a schematic diagram of the process of first thinning and then re-refining the mesh in an embodiment of the present invention;
[0044] Figure 3 This is a schematic diagram illustrating the improved mesh refinement judgment criteria in an embodiment of the present invention;
[0045] Figure 4 This is the distribution of the calculated point locations in an embodiment of the present invention. Detailed Implementation
[0046] Example 1
[0047] The present invention will now be further described with reference to the accompanying drawings and specific embodiments.
[0048] like Figure 1 As shown, an improved method for mesh generation of non-uniform radiation fields includes:
[0049] S1. Initially set a uniform grid for the radiation field calculation region, that is, set the number of divisions in the X, Y, and Z directions of the radiation field calculation region, as shown in the schematic diagram. Figure 2 As shown on the left.
[0050] S2. Perform N-fold sparsification on the initially set uniform grid (that is, increase the grid spacing in each direction to N times the original value), as shown in the diagram. Figure 2 As shown in (middle).
[0051] S3. Calculate the distance from the center point of each sparsified grid to the ellipsoid enclosed by the source term. Determine if the distance is less than 1.5 times the linear dimension of the source (or less than 100cm if the source term is a point source). If less, set the grid weight multiplier to N; otherwise, set it to 1. A schematic diagram is shown below. Figure 3As shown. In this embodiment, the threshold is determined by simulating the gradient of the radiation field near the radiation source, and the threshold can be between 1.2 and 1.5 times the linear dimension or within the range of 80-100 cm.
[0052] The equations for the enclosing ellipsoids differ for each type of source term. Taking some common types of source terms as examples, the equation for the enclosing ellipsoid of a cylindrical source is as follows:
[0053]
[0054] The equation of the enclosing ellipsoid of the circular tube source is:
[0055]
[0056] The equation of the enclosing ellipsoid of the conical source is:
[0057]
[0058] The equation of the enclosing ellipsoid of the cuboid source is:
[0059]
[0060] The equation of the ellipsoid surrounding the spherical source is the same as its spherical equation:
[0061] (x-x0) 2 +(y-y0) 2 +(z-z0) 2 =R 2
[0062] In the above formula, R is the radius; H is the height; R 外 L is the outer radius; W is the length; and x0, y0, and z0 are the center coordinates of the source term. If the source term is rotated, the coordinates of the grid center point need to be transformed to the coordinate system of the source term using a rotation matrix before calculating the distance from the center point of the grid to the enclosing ellipsoid.
[0063] Furthermore, the linearity of a source refers to the maximum length of the source term when measured from all directions; for example, the linearity of a cylindrical source and a conical source is... The dimensions of the circular tube source are The dimensions of the cuboid source are The dimensions of the spherical source are 2R.
[0064] S4. Complete the traversal of all source items and grids in sequence.
[0065] S5. Refine all grids with a weight multiplier greater than 1 according to the corresponding multiplier (i.e., refine the internal mesh, and the spacing of the resulting sub-mesh is 1 / N times the original mesh spacing). A schematic diagram is shown below. Figure 2 As shown on the right.
[0066] S6. Summarize and output all grid intersections, which is the final radiation field calculation lattice.
[0067] The following uses a specific example to describe the solution of the present invention in detail:
[0068] Taking the calculation of the radiation field in a room as an example, the room is 15m long and wide, and 4m high. Four iron cylinders with an outer diameter of 0.5m and a height of 1.2m are placed in the room, each storing cylindrical source terms. The entire room is selected as the calculation area for the radiation field calculation. For ease of comparison, the initial mesh parameters of the uniform mesh generation method are set to 60×60×16; the initial mesh parameters of the existing non-uniform mesh generation method are set to 30×30×8, with a 2x refinement; the improved non-uniform radiation field mesh generation method provided by this invention has an initial mesh parameter of 60×60×16, with a 2x sparsification and refinement factor. The final calculation time results are shown in Table 1.
[0069] Table 1 Comparison of radiation field calculation time for each method
[0070]
[0071] In addition, to compare the accuracy of the radiation field calculations, some calculation points were selected in the room, and their location distribution is as follows: Figure 4 As shown in Table 2, the interpolation results of the radiation field at the calculation point obtained by each method are shown in Table 2.
[0072] Table 2 Comparison of Radiation Field Calculation Accuracy of Various Methods
[0073]
[0074] The results show that for calculation points P1 and P3, which are far from the source term, both the existing and improved non-uniform mesh generation methods achieve computational accuracy close to that of the traditional uniform mesh generation method. However, for points P2, P4, and P5, which are closer to the source, the former shows a significant deviation, while the improved non-uniform radiation field mesh generation method maintains its accuracy. This indicates that although the existing non-uniform mesh generation method has the lowest computational time, it has a significant impact on the computational accuracy of the radiation field. The improved non-uniform radiation field mesh generation method proposed in this invention can better balance computational efficiency and computational accuracy.
[0075] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.
Claims
1. An improved method for mesh generation of non-uniform radiation fields, characterized in that, The specific steps include the following: S1. Set an initial uniform grid for the radiation field calculation region and determine the boundary conditions of the radiation field; S2. Perform N-fold sparsification on the initially set uniform grid, where N is an integer and N>1; S3. Calculate the distance from the center point of each sparsified grid to the ellipsoid enclosed by the source term, set a judgment threshold, and determine the weight multiple of the grid based on the distance. S4. Complete the traversal of all source items and grids in sequence; S5. Refine all grids with a weight multiple greater than 1 according to the corresponding multiple; S6. Summarize and output all grid intersections, which is the final radiation field calculation lattice.
2. The improved non-uniform radiation field mesh generation method according to claim 1, characterized in that, In step S1, a uniform grid is set for the calculation region of the radiation field according to the boundary conditions of the radiation field, and the number of divisions in the X, Y, and Z directions of the radiation field calculation region is determined.
3. The improved non-uniform radiation field gridding method according to claim 1, characterized in that, In step S2, after the uniform grid is thinned by N times, the grid spacing in each direction is increased to N times the original size.
4. The improved non-uniform radiation field mesh generation method according to claim 1, characterized in that, In step S3, a judgment threshold is set; When the distance is less than the preset threshold, the weight multiplier of the grid is set to N; otherwise, it is set to 1.
5. The improved non-uniform radiation field mesh generation method according to claim 1, characterized in that, In step S3, if the source is a point source, the threshold is set to 100cm. If it is less than 100cm, the weight of the grid is set to N. Otherwise, it is set to 1.
6. The improved non-uniform radiation field mesh generation method according to claim 1, characterized in that, In step S3, the equation of the enclosing ellipsoid of the cylindrical source is: The equation of the enclosing ellipsoid of the circular tube source is: The equation of the enclosing ellipsoid of the conical source is: The equation of the enclosing ellipsoid of the cuboid source is: The equation of the ellipsoid surrounding the spherical source is the same as its spherical equation: (x-x0) 2 +(y-y0) 2 +(z-z0) 2 =R 2 In the above formula, R is the radius; H is the height; R 外 L is the outer radius; W is the length; x0, y0, z0 are the center coordinates of the source term; if the source term is rotated, the coordinates of the grid center point need to be transformed to the coordinate system of the source term using a rotation matrix before calculating the distance from the center point of the grid to the enclosing ellipsoid.
7. The improved non-uniform radiation field mesh generation method according to claim 1, characterized in that, In step S5, the mesh is densified, and the spacing of the resulting sub-mesh is 1 / N times the spacing of the original mesh.
8. An improved non-uniform radiation field meshing system, wherein the meshing is performed using the method described in any one of claims 1-7, characterized in that, include: The input module is used to input the three-dimensional model of the radiation field, as well as the type and coordinates of the source terms; The initial grid setting module is used to set up an initial uniform grid for the radiation field calculation region; The mesh sparsification module is used to sparsify the initial uniform mesh by a factor of N. The weight calculation module is used to calculate the distance from the center point of each grid after sparsification to at least one source term enclosing ellipsoid, and determine the weight multiple of the corresponding grid based on the judgment relationship between the distance and the preset threshold. The mesh refinement module is used to refine all meshes with a weight multiple greater than 1 according to their corresponding weight multiples. The computation module is used to control the sequential traversal calculation of all source terms and all sparsified meshes; The lattice output module is used to summarize and output the intersection points of all refined and unrefined meshes to generate the final radiation field calculation lattice and output it.
9. A computing device, comprising: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the steps of the method as described in any one of claims 1-7.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-7.