Method and device for automatically generating unstructured grid based on two-dimensional grid model

Abstract

The invention provides an unstructured grid automatic generation method and device based on a two-dimensional grid model, and belongs to the technical field of seismic wave numerical simulation of seismic prospecting, and the technology comprises the steps: firstly, automatically carrying out the clustering analysis of the attribute value of a grid model according to the imported grid model, and forming a discretely segmented attribute block; and on the basis, extracting an attribute body boundary, carrying out boundary deduplication and merging, and reconstructing a two-dimensional profile topological relation. And then resampling constraint conditions according to a grid size requirement to obtain a grid size, and finally generating an unstructured grid by adopting limited Delaunay triangular grid subdivision. According to the method, the artificial horizon and fault plane interpretation process in the finite element grid generated by the grid model can be reduced to the greatest extent, so that the grid generation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of seismic wave numerical simulation of seismic exploration, and more specifically, to a method and device for automatically generating an unstructured grid based on a two-dimensional grid model. Background Art

[0002] The research and development of seismic wavefield simulation technology based on finite element methods is deepening and maturing. However, the geological models in the field of oil and gas seismic exploration are very complex and mostly three-dimensional regular grid data bodies. How to quickly generate unstructured grids that can be used for finite element seismic simulation from two-dimensional regular grid model data and meet the requirements of adaptive changes in grid size with model attributes has become a bottleneck problem in the application and in-depth development of finite element seismic wavefield simulation technology.

[0003] Since there are often a large number of faults, pinch-outs and erosion layers in geological structures, large geological structure models usually show discontinuity and high complexity in terms of geometric shape and physical parameters. When simulating seismic wave propagation, it is necessary to maintain a balance between simulation accuracy and operation efficiency. Therefore, it is of great practical significance to develop simulation methods with low numerical dispersion, high stability and high computational efficiency. As a mature and high-precision numerical simulation method, finite element method has strong adaptability to the description of complex geological areas, so it is very suitable for simulating the propagation of seismic wave fields in complex structures. Since the finite element method has a high degree of programmability and is easy to handle complex structural problems, it is widely used in various numerical simulation fields. The accuracy, efficiency, numerical dispersion and stability of the finite element method simulation are closely related to the unit interpolation order, mesh shape, unit stiffness matrix synthesis method and mass matrix composition involved. The finite element method can adapt to complex geological structures and free boundary conditions in space. Its simulation results are high in accuracy and small in numerical dispersion. It has strong applicability to different problems, but it will greatly increase the number of grids for complex and discontinuous problems, and the computational efficiency also needs to be improved. At present, the finite element method still needs to be further improved in dealing with discontinuous media containing cracks or pores, problems containing multiphase media, cross-scale problems, and engineering problems containing variable media, especially in the mesh generation method of finite element solution. The quality of the mesh directly affects the accuracy, stability, and correctness of numerical operations such as finite element. At the same time, due to the limitations of computer software and hardware, numerical calculations such as finite element analysis and numerical simulation have great restrictions on the number of computational grids. Summary of the invention

[0004] In view of this, the present invention discloses an automatic generation method of unstructured grids based on a two-dimensional grid model. The method imports a grid model, automatically clusters the grid model attribute values, forms discretely segmented attribute blocks, extracts the attribute body boundaries, performs boundary de-coincidence merging, reconstructs the topological relationship, and then performs constrained resampling according to the grid size requirements to obtain the grid size, and finally uses limited Delaunay grid partitioning to generate an unstructured grid. This technology can minimize the artificial horizon and fault plane interpretation process in generating finite element grids from grid models, thereby improving the efficiency of grid generation.

[0005] According to one aspect of the present invention, a method for automatically generating an unstructured grid based on a two-dimensional grid model is proposed, the method comprising:

[0006] Step 1, importing a two-dimensional seismic grid data model;

[0007] Step 2, analyzing the grid attributes of the two-dimensional seismic grid data model to obtain discrete attribute blocks;

[0008] Step 3, extracting the boundary of the attribute block;

[0009] Step 4, de-overlapping and merging the boundaries of the extracted attribute blocks to obtain topological separation points;

[0010] Step 5, establishing a two-dimensional cross-section topology according to the topological separation points;

[0011] Step 6, calculating the triangle resolution of the two-dimensional section topology;

[0012] Step 7, performing constraint resampling of the two-dimensional profile topology to obtain a sampling result;

[0013] Step 8, triangulating the sampling results to obtain a triangulated network;

[0014] Step 9, sampling the grid attributes onto the triangulated network to obtain triangulated network information;

[0015] Step 10: output the geometric information, topological information and attribute information of the triangulated network information according to the requirements of numerical simulation.

[0016] In some implementations, in step 2, analyzing the grid attributes of the two-dimensional seismic grid data model specifically includes: performing data statistics on the distribution characteristics of the grid attributes, and merging or clustering and truncating the attribute value range according to the statistical results.

[0017] In some embodiments, the boundary extraction of the attribute block in step 3 specifically includes: according to the connectivity of the raster attributes analyzed in step 2, extracting the boundaries of the same type of attributes, and using the grid lines of the raster to record the boundaries of the attributes.

[0018] In some embodiments, step 5 specifically includes:

[0019] Step 5.1, obtain any node, such as point A;

[0020] Step 5.2, obtain all the arc segments sharing node A, and calculate the angles of the straight line segments with A as the starting point and other adjacent points as the ending points, such as the angles of AB, AC, and AD;

[0021] Step 5.3, according to the magnitude of the angles, search for the line segments along the counterclockwise direction starting from the positive X-axis, such as AD, and continue to search for the adjacent side closest to AD along the counterclockwise direction, such as AC, to obtain another node C of the arc segment AC. Repeat step 5.2 with C as the center, start from the CA line segment along the counterclockwise direction to obtain the closest adjacent arc segment CF, and then calculate the angles of the line segments related to point F using the steps of 5.2.

[0022] Step 5.4, repeat steps 5.2 and 5.3 to complete the tracing of a single polygon, such as obtaining ACFDA;

[0023] Step 5.5, after obtaining the polygon ACFDA from AD to AC, continue to search for the arc segment along the counterclockwise direction starting from the AC side, such as AB, and repeat steps 5.2, 5.3, and 5.4 to trace out the polygon ABFECA;

[0024] Step 5.6, repeat steps 5.2, 5.3, 5.4, and 5.5 to complete the tracing of all polygons related to point A;

[0025] Step 5.7, repeat steps 5.1, 5.2, 5.3, 5.4, 5.5, and 5.6 to complete the tracing of all polygons, thereby establishing the topological relationship.

[0026] In some embodiments, the resampling of the constraint conditions for the two-dimensional profile topology in step 7 specifically includes:

[0027] Step 7.1, based on the two-dimensional profile topology established in step 5, virtually connect a straight line between the start and end points of the curve segments therein, calculate the distances from all points on the curve to the straight line, and find the maximum distance value dmax. Compare dmax with a pre-given threshold D. If dmax < D, then execute step 7.2; otherwise, execute step 7.3;

[0028] Step 7.2, sample the midpoints on the curve at triangular resolution, then the line segment is used as an approximation of the curve, and this section of the curve is processed;

[0029] Step 7.3, retain the coordinate points corresponding to dmax, and divide the curve into two parts with these coordinate points as the boundary, and repeat Step 7.1 for these two parts until all dmax are < D, that is, the thinning of the curve is completed.

[0030] In some embodiments, Step 8 specifically includes: calling the restricted Delaunay triangulation algorithm, adding points in the region according to the Delaunay triangulation criterion at the triangular mesh resolution dresolutioan to complete the Delaunay triangulation.

[0031] In some embodiments, the sampling of the grid attributes to the triangular mesh in Step 9 specifically includes: setting the search radius of the sampling step according to the distribution of the grid attributes, the size of the triangular mesh, and the range of the attribute block, and sampling the grid attributes to the corresponding triangles of the triangular mesh through the sampling step.

[0032] According to one aspect of the present invention, there is also provided an automatic unstructured grid generation device based on a two-dimensional grid model, including:

[0033] A model import unit for importing a two-dimensional seismic grid data model;

[0034] A model analysis unit for analyzing the grid attributes of the two-dimensional seismic grid data model to obtain discrete attribute blocks;

[0035] A model processing unit for extracting the boundaries of the attribute blocks, removing duplicates and merging the boundaries to obtain topological separation points;

[0036] A topology generation unit for establishing a two-dimensional profile topology according to the topological separation points and calculating the triangular resolution of the two-dimensional profile topology;

[0037] A constrained resampling unit for performing constrained condition resampling of the two-dimensional profile topology to obtain a sampling result and performing triangulation on the sampling result to obtain a triangular mesh;

[0038] A grid generation unit; for sampling the grid attributes to the triangular mesh to obtain triangular mesh information, and outputting the geometric information, topological information, and attribute information of the triangular mesh information according to the requirements of numerical simulation.

[0039] According to another aspect of the present invention, there is also provided an electronic device, the electronic device includes:

[0040] A memory storing executable instructions;

[0041] A processor runs the executable instructions in the memory to implement the above-mentioned method for automatically generating an unstructured grid based on a two-dimensional grid model.

[0042] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for automatically generating an unstructured grid based on a two-dimensional grid model as described above is implemented.

[0043] This technical solution has at least the following advantages:

[0044] The method and device for automatically generating an unstructured grid based on a two-dimensional grid model of the present invention first automatically clusters the attribute values ​​of the grid model according to the imported grid model to form a discretely segmented attribute block. On this basis, the attribute body boundary is extracted, and the boundary de-coincidence and merging are performed to reconstruct the topological relationship. Then, constrained resampling is performed according to the grid size requirement to obtain the grid size, and finally the unstructured grid is generated by limited Delaunay grid partitioning. There is basically no need or very little need to re-interpret the grid model geologically. The technology can automatically cluster the attribute values ​​of the grid model, segment the block, reconstruct the topological relationship, and then resample the grid size according to the grid size requirement to finally generate an unstructured grid. The grid quality of geophysical numerical simulation is guaranteed under the constraint conditions, including resolution and grid unit shape, and the number of grids is controlled at the same time to ensure the robustness of the calculation model and the accuracy of the results under the condition of meeting certain computing resources and efficiency. At the same time, the technology can minimize the manual interpretation process from the grid model to the generated finite element grid, thereby improving the grid generation efficiency.

[0045] The methods and apparatus of the present invention have other features and advantages that will be apparent from or will be described in detail in the accompanying drawings and subsequent detailed descriptions incorporated herein, which together serve to explain the specific principles of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The above and other objects, features and advantages of the present invention will become more apparent through a more detailed description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components throughout the exemplary embodiments of the present invention.

[0047] Figure 1 A flow chart of a method for automatically generating an unstructured grid based on a two-dimensional grid model according to an embodiment of the present invention is shown.

[0048] Figure 2 A schematic diagram of a two-dimensional grid model according to an embodiment of the present invention is shown;

[0049] Figure 3 A schematic diagram of grid model attribute analysis and clustering according to an embodiment of the present invention is shown;

[0050] Figure 4 A schematic diagram showing the effect of boundary extraction of attribute blocks of a two-dimensional grid model according to an embodiment of the present invention is shown;

[0051] Figure 5 A schematic diagram of generating a polygonal topological relationship according to an embodiment of the present invention is shown;

[0052] Figure 6 A schematic diagram showing the boundary editing and preprocessing effect of a two-dimensional grid model attribute block according to an embodiment of the present invention is shown;

[0053] Figure 7 A schematic diagram showing the effect of unstructured grid generation based on boundary constraints of a two-dimensional grid model attribute block according to an embodiment of the present invention is shown;

[0054] Figure 8 A schematic diagram of sampling steps of sampling grid attributes onto a triangulated network according to an embodiment of the present invention is shown;

[0055] Fig. 9 A diagram showing the effect of a sampling step according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0056] The preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0057] Example 1

[0058] Figure 1 A flow chart of a method for automatically generating an unstructured grid based on a two-dimensional grid model according to an embodiment of the present invention is shown. As shown in the figure, the method includes the following steps:

[0059] Step 1, importing a two-dimensional seismic grid data model, specifically importing a two-dimensional seismic grid data model or extracting a required two-dimensional seismic grid data model from a three-dimensional seismic grid data model, such as Figure 2 Take the model shown as an example;

[0060] Step 2: Analyze the grid attributes of the two-dimensional seismic grid data model to obtain discrete attribute blocks. Specifically, perform data statistics on the attribute distribution characteristics, merge or cluster the attribute value range according to the statistical results, and form discrete attribute blocks, such as Figure 3 As shown;

[0061] Step 3: Extract the boundaries of the attribute blocks. Specifically, according to the connectivity of the clustered grid attributes, extract the boundaries of the same attributes. Specifically, use the grid lines of the grid to record the boundaries of the attributes, such as Figure 4 As shown;

[0062] Step 4: Remove duplicates and merge the boundaries of the extracted attribute blocks to obtain topological separation points. Different attribute blocks have their boundaries extracted, and there will inevitably be a large number of repeated boundary lines. The boundary lines of different attribute blocks extracted in step 3 are merged and removed, and then the bifurcation points are marked and recorded as topological separation points.

[0063] Step 5, establishing a two-dimensional section topology according to the topological separation points;

[0064] Step 6, calculating the triangle resolution of the two-dimensional profile topology, setting the triangle resolution or automatically setting the triangle resolution dresolutioan according to the characteristics of the seismic signal;

[0065] Step 7, performing constraint resampling of the two-dimensional profile topology to obtain a sampling result;

[0066] Step 8: Triangulate the sampling results to obtain a triangulated network. Specifically, for the results from steps 1-7, Figure 6 As shown, the limited Delaunay triangulation algorithm is called, and points are added in the area according to the Delaunay criterion according to the triangulation resolution dresolutioan to complete the Delaunay triangulation. The triangulation effect is as follows Figure 7 As shown;

[0067] Step 9: Sample the grid attributes onto the triangulated network to obtain triangulated network information. Specifically, the search radius of the sampling step is set according to the grid attribute distribution, the size of the triangulated network, and the range of the attribute body. Take reasonable sampling steps to sample the grid attributes onto the triangles corresponding to the triangulated network. The specific steps are as follows: Figure 8 The specific steps are as follows: Fig. 9 As shown;

[0068] Step 10: Output the geometric information, topological information and attribute information of the triangulated network according to the requirements of the numerical simulation.

[0069] In one embodiment of the present invention, in step 2, analyzing the grid attributes of the two-dimensional seismic grid data model specifically includes: performing data statistics on the distribution characteristics of the grid attributes, and merging or clustering and truncating the attribute value range according to the statistical results.

[0070] In one embodiment of the present invention, the step 3 of extracting the boundaries of the attribute blocks specifically includes: extracting the boundaries of the same attributes according to the connectivity of the grid attributes analyzed in step 2, and recording the boundaries of the attributes using grid lines of the grid.

[0071] In one embodiment of the present invention, step 5 specifically includes:

[0072] Step 5.1, obtain any node, such as point A;

[0073] Step 5.2, obtain all arcs that share the node A, and calculate the angles of all straight line segments that start at A and end at other adjacent points, such as the angles of AB, AC, and AD;

[0074] Step 5.3, according to the size of the angle, start from the positive X-axis and search for the line segment in the counterclockwise direction, such as AD, and continue to search for the nearest adjacent edge of AD in the counterclockwise direction, such as AC, to obtain the other node C of arc segment AC, and repeat step 5.2 with C as the center, starting from the CA line segment and going counterclockwise to obtain the nearest adjacent arc segment CF, and then use the steps in 5.2 to calculate the angle of the line segment related to point F.

[0075] Step 5.4, repeat steps 5.2 and 5.3 to complete the tracking of a single polygon, such as obtaining ACFDA;

[0076] Step 5.5, after obtaining the polygon ACFDA from AD to AC, continue to search for arc segments in the counterclockwise direction starting from the AC side, such as AB, and repeat steps 5.2, 5.3, and 5.4 to trace out the polygon ABFECA;

[0077] Step 5.6, repeat steps 5.2, 5.3, 5.4 and 5.5 to complete the tracking of all polygons related to point A;

[0078] Step 5.7, repeat step 5.1, step 5.2, step 5.3, step 5.4, step 5.5 and step 5.6 to complete the tracing of all polygons, thereby establishing a topological relationship.

[0079] In one embodiment of the present invention, the constraint resampling of the two-dimensional cross-section topology in step 7 specifically includes:

[0080] Step 7.1: Based on the two-dimensional profile topology established in Step 5, virtually connect a straight line between the start and end points of the curve segments therein, calculate the distances from all points on the curve to the straight line, and find the maximum distance value dmax. Compare dmax with a pre-given threshold D. If dmax < D, execute Step 7.2; otherwise, execute Step 7.3.

[0081] Step 7.2: Sample the intermediate points on the curve at the triangle resolution, then the straight line segment serves as an approximation of the curve, and this segment of the curve is processed.

[0082] Step 7.3: Retain the coordinate points corresponding to dmax, and divide the curve into two parts with the coordinate points as the boundary. Repeat Step 7.1 for these two parts until all dmax values are < D, that is, the thinning of the curve is completed.

[0083] In an embodiment of the present invention, Step 8 specifically includes: calling the restricted Delaunay triangulation algorithm, adding points in the region according to the Delaunay triangulation criterion at the triangle network resolution dresolutioan to complete the Delaunay triangulation.

[0084] In an embodiment of the present invention, sampling the grid attributes onto the triangular network in Step 9 specifically includes: setting the search radius of the sampling step according to the distribution of the grid attributes, the size of the triangular network, and the range of the attribute block body, and sampling the grid attributes onto the corresponding triangles of the triangular network through the sampling step.

[0085] A method for automatically generating an unstructured grid based on a two-dimensional grid model proposed in an embodiment of the present invention ensures the grid quality of geophysical numerical simulation while taking into account the constraint conditions, including the resolution and the shape of the grid cells, and at the same time controls the number of grids to ensure the robustness of the calculation model and the accuracy of the results under the condition of meeting certain computing resources and efficiency.

[0086] Example 2

[0087] According to an embodiment of the present invention, there is provided an apparatus for automatically generating an unstructured grid based on a two-dimensional grid model, and the apparatus includes:

[0088] A model import unit for importing a two-dimensional seismic grid data model, wherein the imported two-dimensional seismic grid data model further includes the required two-dimensional seismic grid data model extracted from a three-dimensional seismic grid data model;

[0089] A model analysis unit for analyzing the grid attributes of the two-dimensional seismic grid data model to obtain discrete attribute blocks. Specifically, the analysis includes performing data statistics on the attribute distribution characteristics, merging or clustering and truncating the attribute value range according to the statistical results to form discrete attribute blocks.

[0090] The model processing unit is used to extract the boundaries of the attribute blocks, remove duplicates and merge the boundaries, and obtain topological separation points. Specifically, according to the connectivity of the grid attributes after clustering, the boundaries of the same attributes are extracted, and the grid lines of the grid are used to record the boundaries of the attributes. The extracted boundary lines of different attribute bodies are merged and removed, and then the bifurcation points are marked and recorded as topological separation points.

[0091] A topology generation unit is used to establish a two-dimensional section topology according to the topological separation point, and calculate the triangle resolution of the two-dimensional section topology, such as Figure 5 As shown, the specific operations are as follows: Since the same curve segment can be shared by at most two polygons, the corresponding steps are as follows:

[0092] 1) Get any node, such as point A;

[0093] 2) Get all arcs that share the same node A, and calculate the angles of all straight line segments that start at A and end at other adjacent points, such as the angles of AB, AC, and AD, where B, D, and C may also be intermediate points;

[0094] 3) Based on the angle, start from the positive X axis and search for line segments in the counterclockwise direction, such as AD, and continue to search for the nearest adjacent edge of AD in the counterclockwise direction, such as AC. Get another node C of arc segment AC, and repeat step 2 with C as the center). Start from line segment CA and go counterclockwise to get the nearest adjacent arc segment CF. Then use step 2) to calculate the angle of the line segment related to point F.

[0095] 4) Repeat steps 2) and 3) to complete the tracking of a single polygon, such as obtaining ACFDA;

[0096] 5) After obtaining the polygon ACFDA from AD to AC, continue to search for arc segments in the counterclockwise direction starting from the AC side, such as AB. Repeat steps 2), 3), and 4) to trace out the polygon ABFECA;

[0097] 6) Repeat steps 2), 3), 4), and 5) to complete the tracking of all polygons related to point A;

[0098] 7) Repeat steps 1), 2), 3), 4), 5), and 6) to complete the tracking of all polygons;

[0099] Of course, the polygon can also be searched in a clockwise direction, that is, the adjacent arc segments of a certain arc segment can be obtained completely in a clockwise direction, thereby completing the tracking of the entire polygon and establishing a topological relationship.

[0100] Then, set the triangle resolution or automatically set the triangle resolution dresolutioan according to the characteristics of the seismic signal.

[0101] A constrained resampling unit is used to perform constrained resampling on the constraints of the two-dimensional profile topology, obtain a sampling result, and perform triangulation on the sampling result to obtain a triangular mesh. The specific steps are as follows:

[0102] (1) Based on the macroscopic map established in step 5, connect a virtual straight line between the start and end points of the curve segment, calculate the distances from all points on the curve to the straight line, and find the maximum distance value dmax. Compare dmax with a pre-given threshold D (usually defaulting to 3-4 times the grid diagonal length or dresolutioan / 2).

[0103] (2) If dmax < D, sample the intermediate points on this curve segment with dresolutioan; then this straight line segment is used as an approximation of the curve, and this segment of the curve is processed.

[0104] (3) If dmax ≥ D, retain the coordinate points corresponding to dmax, and divide the curve into two parts with this point as the boundary. Repeat using this method for these two parts, that is, repeat steps (1) and (2) until all dmax are < D, which means the thinning of the curve is completed.

[0105] Obviously, the thinning accuracy of this step is also related to dresolutioan. The larger dresolutioan is, the greater the degree of simplification, the more points are reduced. On the contrary, the lower the degree of simplification, the more points are retained, and the shape is more similar to the original curve. The specific sampling effect is as Figure 6 shown.

[0106] A grid generation unit; sets the search radius of the sampling step according to the grid attribute distribution, the size of the triangular mesh, and the range of the attribute body, and is used to sample the grid attributes onto the triangular mesh to obtain triangular mesh information, and output the geometric information, topological information, and attribute information of the triangular mesh information according to the requirements of numerical simulation. The specific steps are as Figure 8 shown, and the specific step effect is as Fig. 9 shown.

[0107] An embodiment of the present invention provides a device for automatically generating an unstructured grid based on a grid model. This device basically does not require or requires very little re-interpretation of the geological structure of the grid model, can automatically cluster the attribute values of the grid model, segment the blocks, reconstruct the topological relationship, then resample the grid size according to the grid size requirements, and finally generate an unstructured grid. It can minimize the process of manually interpreting the horizons and fault planes in generating a finite element grid from the grid model, thereby improving the grid generation efficiency.

[0108] Example 3

[0109] According to another aspect of the present invention, an electronic device is provided. The electronic device comprises:

[0110] Memory, which stores executable instructions:

[0111] A processor runs the executable instructions in the memory to implement the method for automatically generating an unstructured grid based on a two-dimensional grid model according to the present invention.

[0112] The method comprises the following steps:

[0113] A method for automatically generating an unstructured grid based on a two-dimensional grid model, characterized in that the method comprises:

[0114] Step 1, importing a two-dimensional seismic grid data model;

[0115] Step 2, analyzing the grid attributes of the two-dimensional seismic grid data model to obtain discrete attribute blocks, specifically, performing data statistics on the distribution characteristics of the grid attributes, and merging or clustering to truncate the attribute value range according to the statistical results;

[0116] Step 3, extracting the boundaries of the attribute blocks, specifically, extracting the boundaries of the same attributes according to the connectivity of the grid attributes analyzed in step 2, and recording the boundaries of the attributes using grid lines of the grid;

[0117] Step 4, de-overlapping and merging the boundaries of the extracted attribute blocks to obtain topological separation points;

[0118] Step 5, establishing a two-dimensional cross-section topology according to the topological separation points, specifically includes the following steps:

[0119] Step 5.1, obtain any node, such as point A;

[0120] Step 5.2, obtain all arcs that share the node A, and calculate the angles of all straight line segments that start at A and end at other adjacent points, such as the angles of AB, AC, and AD;

[0121] Step 5.3, based on the angle, search for a line segment in the counterclockwise direction starting from the positive X axis, such as AD, and continue to search for the nearest adjacent edge of AD in the counterclockwise direction, such as AC, to obtain another node C of arc segment AC, and repeat step 5.2 with C as the center, starting from line segment CA and going counterclockwise to obtain the nearest adjacent arc segment CF, and then use step 5.2 to calculate the angle of the line segment related to point F;

[0122] Step 5.4, repeat Step 5.2 and Step 5.3 to complete the tracing of a single polygon, such as obtaining ACFDA;

[0123] Step 5.5, after obtaining the polygon ACFDA from AD to AC, continue to search for arc segments in the counterclockwise direction starting from side AC, such as AB, and repeat Step 5.2, Step 5.3, and Step 5.4 to trace out the polygon ABFECA;

[0124] Step 5.6, repeat Step 5.2, Step 5.3, Step 5.4, and Step 5.5 to complete the tracing of all polygons related to point A;

[0125] Step 5.7, repeat Step 5.1, Step 5.2, Step 5.3, Step 5.4, Step 5.5, and Step 5.6 to complete the tracing of all polygons, thereby establishing the topological relationship;

[0126] Step 6, calculate the triangle resolution of the two-dimensional profile topology;

[0127] Step 7, perform resampling of the constraint conditions of the two-dimensional profile topology to obtain the sampling result. The specific steps of Step 7 are as follows:

[0128] Step 7.1, based on the two-dimensional profile topology established in Step 5, virtually connect a straight line between the start and end points of the curve segment therein, calculate the distances from all points on the curve to the straight line, and find the maximum distance value dmax. Compare dmax with a pre-given threshold D. If dmax < D, then execute Step 7.2; otherwise, execute Step 7.3;

[0129] Step 7.2, sample the intermediate points on the curve at the triangle resolution, and use the straight line segment as an approximation of the curve. This segment of the curve is then processed;

[0130] Step 7.3, retain the coordinate points corresponding to dmax, and divide the curve into two parts with these coordinate points as the boundary. Repeat Step 7.1 for these two parts until all dmax are < D, that is, complete the thinning of the curve;

[0131] Step 8, perform triangulation on the sampling result to obtain a triangular mesh. The specific steps of Step 8 are as follows: Call the restricted Delaunay triangulation algorithm, add points in the region according to the Delaunay triangulation criterion at the triangular mesh resolution dresolutioan to complete the Delaunay triangulation;

[0132] Step 9, sampling the grid attributes onto the triangulated network to obtain triangulated network information, wherein specifically, the search radius of the sampling step is set according to the distribution of the grid attributes, the size of the triangulated network, and the range of the attribute block, and the grid attributes are sampled onto the corresponding triangles of the triangulated network through the sampling step;

[0133] Step 10: output the geometric information, topological information and attribute information of the triangulated network information according to the requirements of numerical simulation.

[0134] Example 4

[0135] According to another aspect of the present invention, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the method for automatically generating an unstructured grid based on a two-dimensional grid model according to the present invention is implemented.

[0136] The method comprises the following steps:

[0137] Step 1, importing a two-dimensional seismic grid data model;

[0138] Step 2, analyzing the grid attributes of the two-dimensional seismic grid data model to obtain discrete attribute blocks;

[0139] Step 3, extracting the boundary of the attribute block;

[0140] Step 4, de-overlapping and merging the boundaries of the extracted attribute blocks to obtain topological separation points;

[0141] Step 5, establishing a two-dimensional cross-section topology according to the topological separation points;

[0142] Step 6, calculating the triangle resolution of the two-dimensional section topology;

[0143] Step 7, performing constraint resampling of the two-dimensional profile topology to obtain a sampling result;

[0144] Step 8, triangulating the sampling results to obtain a triangulated network;

[0145] Step 9, sampling the grid attributes onto the triangulated network to obtain triangulated network information;

[0146] Step 10: output the geometric information, topological information and attribute information of the triangulated network information according to the requirements of numerical simulation.

[0147] Example 5

[0148] In order to verify the technical effect of the unstructured grid automatic generation method based on the two-dimensional grid model according to the present invention, this embodiment selects discontinuous media containing cracks or pores, problems containing multiphase media, cross-scale problems and engineering problems containing variable media in seismic exploration for verification and processing. The test results are as follows Fig. 9 As shown. The experiment proves that the technical solution of the present invention ensures the grid quality of geophysical numerical simulation under the constraint conditions, including resolution and grid unit shape, and controls the number of grids at the same time, ensuring the robustness of the calculation model and the accuracy of the results under the condition of meeting certain computing resources and efficiency. It can minimize the process of interpreting artificial horizons and fault planes in the finite element grid generated by the grid model, thereby improving the efficiency of grid generation.

[0149] This example fully demonstrates that the present invention can effectively remove noise signals, especially strong energy linear noise, with significant denoising effect, and can retain effective wave detail information to the maximum extent with high fidelity.

[0150] For other detailed descriptions of this exemplary embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0151] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The terms used herein are selected to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An automatic generation method of unstructured grids based on a two-dimensional grid model. It is characterized in that The method comprises: Step 1, importing a two-dimensional seismic grid data model; Step 2, analyzing the grid attributes of the two-dimensional seismic grid data model to obtain discrete attribute blocks; Step 3, extracting the boundary of the attribute block; Step 4, de-overlapping and merging the boundaries of the extracted attribute blocks to obtain topological separation points; Step 5, establishing a two-dimensional cross-section topology according to the topological separation points; Step 6, calculating the triangle resolution of the two-dimensional section topology; Step 7, performing constraint resampling of the two-dimensional profile topology to obtain a sampling result; Step 8, triangulating the sampling results to obtain a triangulated network; Step 9, sampling the grid attributes onto the triangulated network to obtain triangulated network information; Step 10: output the geometric information, topological information and attribute information of the triangulated network information according to the requirements of numerical simulation.

2. The method for automatically generating an unstructured grid based on a two-dimensional grid model according to claim 1, It is characterized in that In the step 2, analyzing the grid attributes of the two-dimensional seismic grid data model specifically includes: performing data statistics on the distribution characteristics of the grid attributes, and merging or clustering to truncate the attribute value range according to the statistical results.

3. The method for automatically generating an unstructured grid based on a two-dimensional grid model according to claim 1, It is characterized in that The step 3 of extracting the boundaries of the attribute blocks specifically includes: extracting the boundaries of the same attributes according to the connectivity of the grid attributes analyzed in the step 2, and recording the boundaries of the attributes using the grid lines of the grid.

4. The method for automatically generating an unstructured grid based on a two-dimensional grid model according to claim 1, It is characterized in that The step 5 specifically includes: Step 5.1, obtain any node, such as point A; Step 5.2, obtain all arcs that share the node A, and calculate the angles of all straight line segments that start at A and end at other adjacent points, such as the angles of AB, AC, and AD; Step 5.3, according to the size of the angle, start from the positive X-axis and search for the line segment in the counterclockwise direction, such as AD, and continue to search for the nearest adjacent edge of AD in the counterclockwise direction, such as AC, to obtain the other node C of arc segment AC, and repeat step 5.2 with C as the center, starting from the CA line segment and going counterclockwise to obtain the nearest adjacent arc segment CF, and then use the steps in 5.2 to calculate the angle of the line segment related to point F. Step 5.4, repeat steps 5.2 and 5.3 to complete the tracking of a single polygon, such as obtaining ACFDA; Step 5.5, after obtaining the polygon ACFDA from AD to AC, continue to search for arc segments in the counterclockwise direction starting from the AC side, such as AB, and repeat steps 5.2, 5.3, and 5.4 to trace out the polygon ABFECA; Step 5.6, repeat steps 5.2, 5.3, 5.4 and 5.5 to complete the tracking of all polygons related to point A; Step 5.7, repeat Step 5.1, Step 5.2, Step 5.3, Step 5.4, Step 5.5 and Step 5.6 to complete the tracing of all polygons, thereby establishing the topological relationship.

5. The automatic unstructured grid generation method based on a two-dimensional grid model according to claim 1, wherein, the resampling of the constraint conditions for the two-dimensional profile topology in Step 7 specifically includes: Step 7.1, based on the two-dimensional profile topology established in Step 5, virtually connect a straight line between the start and end points of the curve segments therein, calculate the distances from all points on the curve to the straight line, and find the maximum distance value dmax. Compare dmax with a pre-given threshold D. If dmax < D, execute Step 7.2; otherwise, execute Step 7.3; Step 7.2, sample the intermediate points on the curve at the triangle resolution, then the straight line segment is used as an approximation of the curve, and this segment of the curve is processed; Step 7.3, retain the coordinate points corresponding to dmax, and divide the curve into two parts with these coordinate points as the boundary. Repeat Step 7.1 for these two parts until all dmax are < D, that is, complete the thinning of the curve.

6. The automatic unstructured grid generation method based on a two-dimensional grid model according to claim 1, wherein, Step 8 specifically includes: calling the constrained Delaunay triangulation algorithm, adding points in the region according to the Delaunay triangulation criterion at the triangle network resolution dresolutioan to complete the Delaunay triangulation.

7. The automatic unstructured grid generation method based on a two-dimensional grid model according to claim 1, wherein, the sampling of the grid attributes onto the triangular network in Step 9 specifically includes: setting the search radius of the sampling step according to the distribution of the grid attributes, the size of the triangular network, and the range of the attribute block, and sampling the grid attributes onto the corresponding triangles of the triangular network through the sampling step.

8. An automatic unstructured grid generation device based on a two-dimensional grid model, wherein, it includes: a model import unit for importing a two-dimensional seismic grid data model; a model analysis unit for analyzing the grid attributes of the two-dimensional seismic grid data model to obtain discrete attribute blocks; a model processing unit for extracting the boundaries of the attribute blocks, removing duplicates and merging the boundaries to obtain topological separation points; a topology generation unit for establishing a two-dimensional profile topology based on the topological separation points and calculating the triangle resolution of the two-dimensional profile topology; a constraint resampling unit for resampling the constraint conditions of the two-dimensional profile topology to obtain a sampling result and performing triangulation on the sampling result to obtain a triangular network; a grid generation unit for sampling the grid attributes onto the triangular network to obtain triangular network information and outputting the geometric information, topological information, and attribute information of the triangular network information according to the requirements of numerical simulation.

9. An electronic device, wherein, the electronic device includes: a memory storing executable instructions; A processor, wherein the processor runs the executable instructions in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, wherein the computer program implements the method according to any one of claims 1 to 7 when executed by a processor.

Images