Wall surface fracture grid model construction method for optimizing constraint Delaunay subdivision

Abstract

The invention discloses a wall fracture grid model construction method for optimizing constraint Delaunay subdivision. The method comprises the following steps: S1, extracting all fracture node coordinates and line segment number information; s2, carrying out smoothing treatment on the fracture; s3, drawing a smooth fracture curve in modeling software; a fracture curve is stretched in the direction perpendicular to the wall face and converted into multiple curved surfaces, and the curved surfaces intersect with wall face grids to obtain the real curve shape of the fracture on the wall face; s4, acquiring a tensile vector of the fracture; s5, carrying out two-way stretching on the crack surface; s6, modeling by using the vertex coordinate information of the crack grid; s7, converting the crack surface into a grid, and cutting; the grid of the needed part after cutting is combined with the wall face grid; and S8, performing finite element mesh re-division on the whole geometric model. According to the method, the fracture curve is simplified, and meanwhile, the fracture surface which is more practical and has the specified inclination, inclination angle and depth is generated, so that the precision and efficiency of numerical simulation are remarkably improved.

Description

Technical Field

[0001] The invention belongs to the technical field of soil crack simulation, and in particular relates to a method for constructing a wall crack grid model of optimized constrained Delaunay partitioning. Background Art

[0002] As archaeologists excavate the soil, cracks in the soil will pose a serious safety hazard to the protection of the site, so soil cracks are an issue that cannot be ignored. In addition, in the long-term development process of geological formation and human factors, cracks will inevitably appear in the soil after excavation.

[0003] Therefore, it is very important to study soil cracks in related projects. The commonly used method is to model soil cracks. There is an important problem in numerical simulation research-the establishment of a real crack model, that is, to construct a crack model that can truly express the shape and development direction of wall cracks. For example, Chinese patent, CN108090313B, discloses a complex rock crack model modeling and identification method. In the early stage, ProE's excellent modeling ability is used to model complex models. In the mid-term, ANSYS's powerful grid processor is used for grid division. In the later stage, Flac3D is used for simulation calculation, and the advantages are complementary, and the strengths are learned to make up for the weaknesses, so as to achieve error-free links between ProE, ANSYS and Flac3D. Although this method avoids the research and technical personnel spending a lot of time and energy to study the complex FISH language in Flac3D, and avoids the inability to model and grid the complex model in geotechnical engineering problems, it does not fully consider the development of cracks in space, and only considers the radial direction. Therefore, it is necessary to study the numerical simulation model of cracks in three-dimensional space to overcome the problems such as the error between the results of numerical simulation of regular cracks and the actual irregular cracks. Summary of the invention

[0004] The purpose of the present invention is to provide a wall crack mesh model construction method for optimizing constrained Delaunay partitioning, so as to solve the technical problems in the prior art that the characteristics such as the strike and dip of the cracks are not considered in the process of crack modeling research, and it is difficult to generate high-quality meshes.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] A method for constructing a wall crack mesh model by optimizing constrained Delaunay partitioning includes the following steps:

[0007] Step S1: Based on the known fracture line segment data, all fracture node coordinates and line segment number information are extracted and stored in a file. This file not only contains the key geometric information of the fracture, but also provides the necessary basic data for subsequent processing.

[0008] Step S2: Use the three-dimensional Douglas Peucker algorithm to smooth the data file of the fracture. Eliminate the irregularity of the curve while ensuring the rationality of the fracture, and ensure that the surface geometry of the fracture is more continuous and realistic, so as to avoid errors caused by numerical calculations.

[0009] Step S3: Using the drawing function, a smooth crack curve is drawn in the modeling software according to the data removed in step S2; then, the crack curve is stretched into a multiple surface by stretching the crack curve perpendicular to the wall surface, and the multiple surface is meshed. The stretched mesh surface intersects with the mesh wall surface, and the coordinate information of all nodes on the intersection line is obtained, so as to obtain the true curve shape of the crack on the wall surface.

[0010] Step S4: According to the number, strike and dip of each crack, the tensile vector of the corresponding crack is obtained, which provides directional data support for the subsequent accurate modeling of the crack surface.

[0011] Step S5: Based on the stretching vector obtained in step S4 and the depth information of the corresponding crack, the surface is bidirectionally stretched to form a crack surface; and the coordinate information of all the vertices of the crack surface in the surface mesh is obtained, which will play a key role in the subsequent modeling stage.

[0012] Step S6: Modeling is performed using the vertex coordinate information of the crack, and a data file is exported. The file contains the geometric information of the crack surface, which provides a basis for the next step of model cutting.

[0013] Step S7: mesh the crack surface and combine it with the wall mesh to form the entire crack geometry model to ensure the integrity and stability of the overall model.

[0014] Step S8: Re-divide the finite element mesh of the entire fracture geometry model to further optimize the model mesh structure and ensure that it can provide efficient and accurate results in numerical simulation.

[0015] Further optimization, in step 1, the number of the crack curve and the three-dimensional coordinates of each point are exported to a data file through a loop function.

[0016] Further optimization, in step S2, the process of smoothing the crack curve is as follows:

[0017] S2.1: Set the threshold D, find and connect the two vertices of the crack curve as the initial line segment;

[0018] S2.2: Calculate the maximum distance d between a point on the crack curve and a line segment max And determine whether it is less than the threshold D;

[0019] S2.2.1: When the maximum distance is greater than D, connect the point corresponding to the maximum distance on the fracture curve and the vertices of the two segments to form a new broken line segment; re-determine whether the maximum distance between the two segments in the newly formed broken line segment and the points on the fracture curve corresponding to their parts is less than the threshold D:

[0020] S2.2.1.1 If the maximum distance d max If it is greater than D, repeat step S2.2.1 until the maximum distance d between the two line segments in the newly formed broken line segment and the points on the corresponding fracture curve is max Less than or equal to D;

[0021] S2.2.2: When the maximum distance is less than or equal to D, the smoothing process of the crack ends;

[0022] S2.3: Repeat steps S2.1 and S2.2 until all fractal curve smoothing processes are completed.

[0023] Further optimization, in step S3, the process of obtaining the three-dimensional wall crack curve is as follows: the crack curve after smoothing in step 2 is stretched into a multiple curved surface in a direction perpendicular to the wall, so that it passes through the wall mesh, and then the multiple curved surface is converted into a mesh, and finally the wall mesh and the stretched mesh are intersected by software to obtain the intersection line.

[0024] Further optimization, in step S4, the process of obtaining the fracture stretch vector is as follows: call the Python script to convert the fracture number, depth, and dip angle data in Excel into a dictionary, traverse each row of data, and store the name, depth, and dip angle of each group of data in a list, calculate the vector projection and unit vector, and then obtain the strike unit vector, loop each fracture to export the stretch vector pkl file.

[0025] Further optimization, in step S5, the process of the crack mesh vertex coordinate information is as follows: according to the stretching vector obtained in step S4, the crack curve is bidirectionally stretched to form a crack surface, and the crack surface vertex information is recorded through a loop function and exported as a dat file.

[0026] Further optimization, in step S6, the process of fracture surface modeling is as follows: read the data file of step S5, organize the fracture information and generate multiple components in a loop, use the midpoints of the two nodes of the fracture edge as the edge keywords, create the crack surface after generating the components, merge the components and delete the temporary fracture curve, and export the fracture surface geometry file.

[0027] Further optimization, in step S7, the process of combining the crack and the wall is as follows: converting the crack surface into a mesh, cutting the crack mesh through software, and combining the required part of the mesh after cutting with the wall mesh by small triangular faces.

[0028] Further optimization, in step S8, the process of re-meshing is as follows: remesh the entire fracture geometry model, reset the mesh side length and mesh shape, and generate a mesh model with good uniformity, wherein the mesh size is set to 0.05m and the ridge angle is set to 20° to generate a full triangular surface mesh. The mesh quality directly affects the key properties of the numerical simulation, such as stability, accuracy, and convergence speed, so high-quality mesh is crucial in the simulation process.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] The present invention can generate fracture grids with any angle of development, and can change the depth of fracture development to more accurately describe the fractures. In addition, the present invention can ensure the generation of high-quality grids, thereby improving the efficiency and accuracy of numerical simulation, providing a valuable reference for future fracture modeling, and having a strong guiding significance for actual archaeological engineering construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A flowchart of a wall crack mesh model construction method for optimizing constrained Delaunay partitioning according to the present invention;

[0032] Figure 2 Schematic diagram of the process of smoothing a crack;

[0033] Figure 3 Schematic diagram of plane wall cracks in Example 1;

[0034] Figure 4 Schematic diagram of the intersection of the wall grid and the stretched grid in Example 1;

[0035] Figure 5 It is a schematic diagram of the three-dimensional wall crack curve in Example 1;

[0036] Figure 6 Schematic diagram of solving the stretching vector in Example 1;

[0037] Figure 7 Schematic diagram of fracture surface modeling in Example 1;

[0038] Figure 8 Schematic diagram of the small triangle face welding model for the mesh in Example 1;

[0039] Fig. 9 Schematic diagram of the overall model of the wall crack mesh that is re-meshed in Example 1;

[0040] Fig.10 Schematic diagram of the local model of the wall crack mesh that is re-meshed in Example 1;

[0041] Fig.11 11(a) is a comparison diagram of the real picture and the simulated picture in Example 1, where 11(a) is a knowledge photo of the wall. Fig.11 (b) is a model diagram generated using the method described in the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Embodiment 1:

[0044] like Figure 1 As shown, a method for constructing a wall crack mesh model by optimizing constrained Delaunay partitioning includes the following steps:

[0045] Step S1: Based on the known fracture line segment data, a loop function is called to store its node coordinates and line segment numbers in a dat file.

[0046] In this embodiment, the number and information of the wall crack curve are extracted according to the wall CAD file under study.

[0047] Step S2: Call the three-dimensional Douglas Peucker algorithm to smooth the obtained fracture dat file. In order to make the generated fracture curve more realistic, set the threshold D to 0.001 meters, find and connect the points at both ends of each fracture curve as the initial line segment; then calculate the maximum distance d between the point on the fracture curve and the initial line segment max And determine whether it is less than the threshold D: when d max >D, the maximum distance d on the connecting crack curve max The corresponding points and points on the two segments of the line segment are re-determined whether the maximum distance between the two line segments and the points on the corresponding parts of the crack curve is less than the threshold. If the condition is not met, repeat this step until it is met; when d≤D, the smoothing process of the crack ends, and the next crack curve is detected until all crack curves are traversed; then, the vertex information of the smoothed crack curve number is exported as a dat file. Figure 2 Schematic diagram of crack smoothing using the three-dimensional Douglas Peucker algorithm.

[0048] Due to limited space, only part of the data is shown here, as shown in Tables 1, 2, 3, and 4, where Table 1 shows the coordinate values ​​of each node on the original crack curve numbered L149; Table 2 shows the coordinate values ​​of the point nodes of the crack curve numbered L149 after smoothing; Table 3 shows the coordinate values ​​of each node on the original crack curve numbered L150; Table 4 shows the coordinate values ​​of the point nodes of the crack curve numbered L150 after smoothing.

[0049] Table 1 Coordinate values ​​of nodes on the original crack curve of No. L149

[0050]

[0051]

[0052] Table 2 Coordinate values ​​of nodes of the crack curve No. L149 after smoothing

[0053] X Y Z Node 1' -2.499896 -3.331423 489.486755 Node 2' -2.510710 -3.332709 489.541718 Node 3' -2.500123 -3.297690 489.556641 Node 4' -2.512675 -3.283556 489.638702 Node 5' -2.503760 -3.236762 489.696259 Node 6' -2.518384 -3.243387 489.772949

[0054] Table 3 Coordinate values ​​of nodes on the original crack curve of No. L150

[0055] X Y Z Node 1 -2.380782 -3.090216 489.395447 Node 2 -2.384935 -3.081404 489.425008 Node 3 -2.391519 -3.069206 489.464246 Node 4 -2.394399 -3.060679 489.495149 Node 5 -2.399004 -3.048214 489.539520 Node 6 -2.400852 -3.049022 489.547919 Node 7 2.405836 -3.053308 489.581146 Node 8 -2.407455 -3.054708 489.591980 Node 9 -2.413510 -3.059946 489.632528 Node 10 -2.415533 -3.061377 489.644470 Node 11 -2.426592 -3.083586 489.675964

[0056] Table 4 Coordinate values ​​of nodes of the L150 crack curve after smoothing

[0057] X Y Z Node 1' -2.380782 -3.090216 489.395447 Node 2' -2.399004 -3.048214 489.539520 Node 3' -2.415533 -3.061377 489.644470 Node 4' -2.426592 -3.083586 489.675964

[0058] Step S3: Using the drawing function, a smooth crack curve is drawn in the modeling software according to the data removed in step S2; then, the crack curve is stretched into a multiple surface by stretching the crack curve perpendicular to the wall surface, and the multiple surface is meshed. The stretched mesh surface intersects with the mesh wall surface, and the coordinate information of all nodes on the intersection line is obtained, so as to obtain the true curve shape of the crack on the wall surface. Figure 3 Schematic diagram of plane wall cracks in Example 1; Figure 4 This is a schematic diagram of the intersection of the wall mesh and the stretched mesh in Example 1. The yellow line in the figure represents the curve obtained by the intersection of the mesh surface and the mesh wall; Figure 5 Schematic diagram of the three-dimensional wall crack curve obtained by intersection in Example 1.

[0059] Step 4: The Python script converts the fracture number, depth and dip angle data in Excel into a dictionary, iterates through each row of data, and stores the number, depth and dip angle of each set of data in a list, calculates the vector projection and unit vector, and then obtains the strike unit vector, looping through each fracture to export the stretch vector pkl file.

[0060] Combination Figure 6 , the calculation process of the stretching vector is as follows:

[0061]

[0062]

[0063]

[0064]

[0065]

[0066] It turns out that: For cracks The stretch vector

[0067] Among them: BD⊥AC, An arc with point A as the center and cosθ as the radius; are all unit vectors;

[0068] θ is the angle between the crack and the strike vector, and cosθ is the projection of the crack line segment on the strike vector.

[0069] Due to space limitations, only part of the data is shown here, such as the data of the two crack curves numbered L149 and L150 as shown in Tables 5, 6, and 7.

[0070] Table 5 Data of two crack curves numbered L149 and L150 in the Excel data table

[0071] name depth view angle strike angle dip direction L149 155 70 290 200 L150 72 85 265 175

[0072] Table 6 Dictionary data of two crack curves numbered L149 and L150

[0073] {'name':'L149','depth':155,'view angle':70,'strike angle':290,'dip direction':200} {'name':'L150','depth':72,'view angle':85,'strike angle':265,'dip direction':175}

[0074] Table 7 PKL files of two crack curves numbered L149 and L150

[0075]

[0076]

[0077] Step 5: According to the stretching vector obtained in step S4, the crack curve is bidirectionally stretched to form a crack surface, and the coordinate information of all the crack vertices in the surface mesh is obtained through a loop function, and a dat file is exported. Tables 8 and 9 show the coordinate data of the vertices of the two crack surfaces numbered L149 and L150.

[0078] Table 8 Data of coordinates of vertices of the crack surface numbered L149

[0079]

[0080] Table 9 Data of coordinates of vertices of the L150 crack surface

[0081]

[0082] Step 6: Read the dat file of step S5, sort out the crack information and generate multiple parts in a loop, use the midpoints of the two nodes on the edge where each crack is located as the keyword of the edge, create the crack surface after generating the parts, merge the parts and delete the temporary crack curve, and export the crack surface as an igs file. Figure 7 Schematic diagram of importing the igs file into the software for fracture surface modeling in Example 1.

[0083] Step 7: Convert the fracture surface generated in step 6 into a fracture mesh, cut the fracture mesh through software, and connect the required mesh and the wall mesh with small triangles. Figure 8 Schematic diagram of the model for connecting the crack and wall mesh with small triangular surfaces in Example 1.

[0084] Step 8: The quality of the mesh directly affects the key properties of the numerical simulation, such as stability, accuracy, and convergence speed. Therefore, the mesh size is set to 0.05m and the ridge angle is set to 20° to generate a full triangular surface mesh and mesh the model. Fig. 9 Schematic diagram of the overall model of the wall crack mesh that is re-meshed in Example 1; Fig.10 Schematic diagram of the local model of the wall crack mesh that is re-meshed in Example 1.

[0085] Fig.11 11(a) is a comparison diagram of the real picture and the simulated picture in Example 1, where 11(a) is a knowledge photo of the wall. Fig.11 (b) is a model diagram generated by the method of the present invention. It can be seen from this figure that the present invention can generate a fracture grid developed at any angle, and the simulation accuracy is very high, which provides a valuable reference for future fracture modeling and has a strong guiding significance for actual archaeological engineering construction.

[0086] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A method for constructing a wall crack mesh model using optimized constrained Delaunay meshing, characterized in that: The following steps are involved: Step S1: Extract all crack node coordinates and line segment number information based on known crack line segment data and store them in a file; Step S2: Using the 3D Douglas Peucker algorithm to smooth the cracks and remove a large number of redundant graphic data points in the file; Step S3: Using a drawing function, a smooth crack curve is drawn in the modeling software based on the data removed in step S2. The crack curve is then stretched perpendicular to the wall surface to form a polyhedron, which is then meshed. The stretched meshed surface intersects with the meshed wall surface, and the coordinate information of all nodes on the intersection line is obtained, thereby obtaining the true curve shape of the crack on the wall surface. Step S4: according to the number, strike and dip of each crack, obtain the stretching vector of the corresponding crack; Step S5: Based on the stretching vector obtained in step S4 and the depth information of the corresponding crack, the surface is bidirectionally stretched to form a crack surface; and the coordinate information of the vertices of all crack surfaces in the surface mesh is obtained; Step S6: Modeling is performed using vertex coordinate information of the crack surface, and a data file is exported; Step S7: meshing the crack surface and combining it with the wall mesh to form the entire crack geometry model; Step S8: Re-divide the finite element mesh of the entire fracture geometry model.

2. The wall crack mesh model construction method of optimized constrained Delaunay decomposition according to claim 1 is characterized in that: In step S1, the process of extracting crack information is as follows: traverse each crack curve, number all cracks, and obtain the coordinate information of each node of the corresponding crack curve according to the number.

3. The wall crack mesh model construction method of optimized constrained Delaunay decomposition according to claim 2 is characterized in that: In step S6, the process of fracture surface modeling is as follows: Read the data file generated in step S5 and generate corresponding components based on each crack information; create a crack surface for each component using the midpoints of the two nodes on the edge where the crack is located as the key of the edge; finally, merge these components and delete the temporary crack curve to export the crack surface geometry file.

4. The method for constructing a wall crack mesh model using optimized constrained Delaunay decomposition according to claim 3, characterized in that: In step S7, the process of combining the crack and the wall is as follows: meshing the crack surface and cutting the crack mesh; connecting the mesh of the required part after cutting with the wall mesh with small triangular faces to ensure the integrity and stability of the overall model.

5. The method for constructing a wall crack mesh model using optimized constrained Delaunay decomposition according to claim 4, characterized in that: In step S8, the process of re-meshing is as follows: the entire crack geometric model is remeshed, the mesh side length is reset, and a mesh model with good uniformity is generated; wherein, the mesh size is set to 0.05m and the ridge angle is set to 20° to generate a full triangular surface mesh.

Images