Method for generating quasi-three-dimensional grid based on casting body sheet image

Abstract

The invention discloses a method for generating a quasi-three-dimensional grid based on a casting body sheet image, and relates to the technical field of oil and gas development. The method comprises the following steps: preprocessing a core slice plane image to generate a pore scale binary image; performing boundary identification on the pore scale binary image, and constructing a two-dimensional porous medium pore scale geometric profile curve in combination with physical scale reduction; considering the pore connectivity, and constructing a two-dimensional porous medium pore scale geometric model; adopting a normal isomorphic embedding method to carry out quasi-three-dimensional reconstruction, and constructing a quasi-three-dimensional porous medium pore scale geometric model; discretization processing is conducted on the quasi-three-dimensional porous medium pore scale geometric model, and a quasi-three-dimensional pore scale grid structure is generated in combination with sweeping operation. According to the method, the quasi-three-dimensional porous medium pore mesh model with a real normal scale is constructed, so that geometric expression and numerical reduction with higher precision on a micro pore structure are realized, and the simulation precision of a pore scale seepage behavior is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas development, and particularly relates to a method for generating a quasi-three-dimensional grid based on a cast slice image. BACKGROUND

[0002] The study of fluid percolation behavior in complex porous media has important theoretical significance and engineering value in the fields of oil exploration and production, groundwater migration, and geological storage. For a long time, two-dimensional pore structure models have been widely used in microscale percolation simulation and mechanism research due to their simplicity in construction and high computational efficiency. However, the simplification of two-dimensional models in geometric dimensions makes it difficult to accurately reproduce the pore-throat connection structure and spatial connectivity in real porous media, leading to significant deviations in the simulation of key parameters such as capillary pressure, relative permeability, and irreducible water retention compared to microscale experimental results.

[0003] In recent years, with the development of visualization methods such as microfluidic chip observation and core slice displacement experiments, the limitations of two-dimensional models in pore network topology expression, dynamic evolution of multiphase interfaces, and identification of local flow paths have become more prominent. For example, in heterogeneous shale reservoirs, two-dimensional models are difficult to describe the synergistic mechanism between micro-nano scale adsorption-desorption processes and main flow channels; in the context of CO2 geological storage, two-dimensional simplified structures are difficult to express the retention characteristics of supercritical phases under the action of pore walls, which severely restricts the physical authenticity and predictive ability of two-dimensional models. Although some studies attempt to improve simulation accuracy through three-dimensional digital core reconstruction, such methods rely on high-resolution CT imaging or FIB-SEM scanning data, which have the bottlenecks of difficult data acquisition, time-consuming modeling and grid division, and high computational resource consumption, and are usually incompatible with existing two-dimensional image data or historical research results, making it difficult to promote at the engineering level. SUMMARY

[0004] In view of this, the present application proposes a method for generating a quasi-three-dimensional grid based on a cast slice image, which can be used for studying the percolation law characteristics of reservoir fluids at the pore scale.

[0005] The technical solution adopted by the present application to solve the above problems is a method for generating a quasi-three-dimensional grid based on a cast slice image, comprising the following steps:

[0006] S1. Selecting a core slice planar image with a connected pore structure and performing image processing to generate a pore-scale binary image with pore-matrix partitioning;

[0007] S2. Performing boundary recognition on the pore-scale binary image, combining physical scale reduction and B-spline curves to construct a two-dimensional porous medium pore-scale geometric profile curve;

[0008] S3, based on the two-dimensional porous medium pore scale geometry profile curve, considering the pore connectivity, a two-dimensional porous medium pore scale geometry model is constructed;

[0009] S4, based on the two-dimensional porous medium pore scale geometry model, a quasi-three-dimensional porous medium pore scale geometry model is constructed by using the normal isomorphism embedding method for quasi-three-dimensional reconstruction;

[0010] S5, the quasi-three-dimensional porous medium pore scale geometry model is discretized and combined with the sweeping operation to generate a quasi-three-dimensional pore scale grid structure.

[0011] The technical effects of the present application are:

[0012] 1, the present application constructs a quasi-three-dimensional porous medium pore grid model with real normal scale, realizes higher precision geometric expression and numerical reduction of micro-pore structure, and significantly improves the simulation precision of pore scale percolation behavior. Compared with the traditional mechanism model based on two-dimensional geometric hypothesis, the quasi-three-dimensional structure provided by the present application can effectively reduce the error between the simulation results and the micro experimental observation results. Based on the quasi-three-dimensional grid model generated by the method, the flow numerical simulation can be observed, and the more real bound water distribution and the characteristics of the stagnant zone can be observed, and the fluid migration process is closer to the visualization experimental results of microfluidic chip, high resolution micro displacement experiment, etc. The method significantly improves the fitting degree of model prediction and physical experiment, so as to more accurately describe the micro percolation characteristics, has certain engineering applicability and research value.

[0013] 2, the quasi-three-dimensional porous medium pore scale grid model generated by the present application breaks through the limitation of OpenFOAM numerical simulation platform on three-dimensional grid structure, solves the problem that the traditional two-dimensional micro porous medium model cannot be directly used for numerical solution in the software. The constructed quasi-three-dimensional grid model not only retains the pore connectivity and geometric characteristics of the original two-dimensional structure, but also generates a data structure compatible with three-dimensional numerical format through normal extension, which can be seamlessly integrated into the OpenFOAM simulation framework. The scheme effectively realizes the equivalent three-dimensional simulation of two-dimensional microstructure, avoids the simplification processing of complex real pore structure, and significantly expands the flexibility, platform adaptability and applicability of micro grid construction method.

[0014] 3, The pore-scale quasi-3D grid construction method provided by the application is not only suitable for fine prediction of distribution of multi-phase fluids such as remaining oil, bound water and residual gas in a porous medium, but also can be extended to inversion calculation and evaluation analysis of key physical property parameters (such as relative permeability curves, capillary pressure curves and the like) of the porous medium. Meanwhile, the method has good adaptability, can be integrated into a multi-scale modeling process, and can assist in carrying out engineering researches such as reservoir productivity analysis, injection-production optimization design and numerical reproduction of core experiments, and has wide engineering application prospect and popularization value. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings needed to be used in the embodiments will be briefly introduced below, and it should be understood that the following drawings only show some embodiments of the application, and therefore should not be regarded as a limitation to the scope, and other related drawings can also be obtained by those skilled in the art without creative labor on the basis of the drawings.

[0016] Figure 1 is a pore-scale binary image in the application.

[0017] Figure 2 is a two-dimensional porous medium pore-scale geometric profile curve in the application.

[0018] Figure 3 is a two-dimensional porous medium pore-scale geometric model in the application.

[0019] Figure 4 is a quasi-3D porous medium pore-scale geometric model in the application.

[0020] Figure 5 is a quasi-3D porous medium pore-scale grid model in the application.

[0021] Figure 6 is a defined grid boundary path in the application.

[0022] Figure 7 is a simulation result of flow simulation of the quasi-3D grid constructed by using the method of the application.

[0023] Figure 8 is a micro-visualization displacement experiment result graph.

[0024] Figure 9 is a simulation result of flow simulation of a two-dimensional grid constructed based on a two-dimensional porous medium geometric model. DETAILED DESCRIPTION

[0025] The application will be further described in detail below in combination with embodiments and drawings.

[0026] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of the present application. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application.

[0027] A method for generating a pseudo-three-dimensional grid based on a cast slice image, comprising the following steps:

[0028] S1, selecting a core slice planar image with a connected pore structure and image processing the same to generate a pore-scale binary image with a pore-matrix partition;

[0029] Specifically, this step mainly pre-processes an existing core slice planar image to facilitate the subsequent work processing flow. This step mainly includes the following sub-steps:

[0030] S11, identifying the core slice planar image with a connected pore structure by artificial recognition and obtaining a gray scale image by gray scale processing; generally speaking, a core slice with a connected pore structure can be better used for micro-flow simulation and can be identified by the naked eye. The gray scale processing operation is a conventional operation in the art, such as the gray scale processing by the weighted average method.

[0031] S12, referring to the pre-processed image, identifying the continuous pore channel in the gray scale image; although the pore channel can be identified according to the gray scale in the gray scale image, the identification difficulty is relatively high and errors are prone to occur, and since the pre-processed image can identify the pore connection structure, in this step, the continuous pore channel in the gray scale image is identified by referring to the pre-processed image, and the non-pore misidentification area caused by staining or impurities is removed.

[0032] In particular, in this step, a pigment recognition algorithm can be used to remove the non-pore misidentification area.

[0033] S13, based on a Gaussian weighted adaptive threshold method, performing binary processing on the gray scale image processed by S12 to obtain a pore-scale binary image.

[0034] In this embodiment, the existing core slice planar image is processed by this step to obtain a binary image as shown in Figure 1 .

[0035] S2. Perform boundary recognition on the pore-scale binary image, and construct the two-dimensional porous medium pore-scale geometric contour curve by combining physical scale restoration and B-spline curve;

[0036] This step is primarily used to obtain continuously accessible pore boundary curves for constructing the two-dimensional pore-scale geometric model in S3. This step includes the following sub-steps:

[0037] S21. Identify the boundary pixels of connected regions in the pore-scale binary image based on the edge detection algorithm, and extract the boundary point set of the pore phase and matrix phase using the Canny operator. This step is mainly used for boundary recognition and contour extraction. By using the edge detection algorithm, the gray-level gradient of the image can be calculated, thereby identifying the boundary pixels of connected regions in the binary image. Both the edge detection algorithm and the Canny operator are common techniques in this field, so their specific operations will not be elaborated here.

[0038] S22. Combine the original core thin section planar image and convert the image coordinates into physical coordinates. In this step, it is necessary to combine the scale of the original core thin section image to convert the image coordinates into physical coordinates. In the specific operation process, two-dimensional coordinates are added to the image, and then the image is enlarged proportionally according to the scale of the original core thin section image to obtain the physical coordinates of the image.

[0039] S23. The boundary point set is parametrically fitted using B-spline curves, and combined with the physical coordinates in S22, a continuous and differentiable two-dimensional porous medium pore-scale geometric profile curve is obtained.

[0040] In this embodiment, for Figure 1 After this step, the result is as follows: Figure 2 The geometric profile curve of the pore size of the two-dimensional porous medium is shown.

[0041] S3. Based on the geometric profile curve of the pore scale of the two-dimensional porous medium, and considering pore connectivity, construct a geometric model of the pore scale of the two-dimensional porous medium. This step includes the following sub-steps:

[0042] S31. Construct a rectangle in the x and y directions that matches the geometric contour curve of the pore scale of the two-dimensional porous medium as a base, and superimpose the geometric contour curve of the pore scale of the two-dimensional porous medium onto the rectangular base. In this step, the length of the rectangular base in the x and y directions is equal to the maximum length of the geometric contour curve of the pore scale of the two-dimensional porous medium in the x and y directions, so that the geometric contour curve of the pore scale of the two-dimensional porous medium can be completely embedded in the rectangular base and superimposed.

[0043] S32, using Boolean difference set operation, using geometric profile curve to segment the rectangular base, removing non-connected or closed regions, and only retaining the effective channel region connected with the main pore structure; in this step, the Boolean difference set operation used is as follows: Ω 2D =B pore ∩R base , in the formula, B pore represents a two-dimensional porous medium pore scale geometric profile curve, R base represents a rectangular base, Ω 2D represents the calculated region.

[0044] S33, converting the retained region into a two-dimensional solid domain to obtain a two-dimensional porous medium pore scale geometric model.

[0045] In this embodiment, the obtained two-dimensional porous medium pore scale geometric model is as shown in Figure 3 .

[0046] S4, based on the two-dimensional porous medium pore scale geometric model, using the normal isomorphism embedding method to perform quasi-three-dimensional reconstruction, and constructing a quasi-three-dimensional porous medium pore scale geometric model; this step includes the following sub-steps:

[0047] S41, by rigid transformation, aligning the geometric center of the two-dimensional porous medium pore scale geometric model to the (x, y) plane of the three-dimensional Cartesian coordinate system, and at the same time, solidifying its boundary conditions and pore phase relationship to obtain a two-dimensional geometric domain; through this operation, the unified definition of the working plane is completed.

[0048] S42, according to the spatial scale requirement, obtaining a normal extension depth as a high-dimensional reference domain, based on the high-dimensional reference domain, constructing a topological direct product space of the two-dimensional geometric domain; the high-dimensional reference domain is as follows: I z =[0, H], in the formula, H represents the normal extension depth; the topological direct product space is as follows: Γ 3D =Ω 2D ×I Z , the constructed topological direct product space is used as the framework of three-dimensional pore topology.

[0049] S43, applying a shape-preserving embedding mapping operator to map the topological direct product space to a three-dimensional Euclidean space, the shape-preserving embedding mapping operator is: Φ: (x, y) ∈ Ω 2D , z ∈ I z → (x, y, z) ∈ Π 3 , in the formula, Φ is a shape-preserving embedding mapping operator, Ω 2D is a two-dimensional geometric domain, I z represents a normal extension depth, and Π 3represents a three-dimensional Euclidean space, x, y, z represent the x-axis, y-axis and z-axis of a three-dimensional Cartesian coordinate system, respectively;

[0050] S45, based on the spatial attribute inheritance principle, defining three-dimensional pore phase domain, constructing a quasi-three-dimensional porous medium pore scale geometric model: Ω 3D = Φ(Ω 2D × I z ), which ensures that the spatial distribution of pores and matrix in three-dimensional geometry remains consistent and topologically continuous with the two-dimensional prototype.

[0051] Through the operation of this step, the obtained image is as shown in Figure 4 .

[0052] S5, discretizing the quasi-three-dimensional porous medium pore scale geometric model, and generating a quasi-three-dimensional pore scale grid structure by combining the sweeping operation. This step includes the following sub-steps:

[0053] S51, selecting the interface of the quasi-three-dimensional porous medium pore scale geometric model in the (x, y) plane as the starting surface for subdivision, and using a free triangular subdivision algorithm to perform two-dimensional discretization to generate a planar grid: N 2D = Triangulate(Ω 2D ), wherein N2D represents the planar grid, and Triangulate represents the free triangular subdivision algorithm operator.

[0054] S52, based on the sweeping operation, sweeping the planar grid along the z-axis to form a quasi-three-dimensional pore scale grid structure: N 3D = Sweep(N 2D , I z ), wherein N3D represents the quasi-three-dimensional pore scale grid structure, and Sweep represents the sweeping operation. During the sweeping process, according to the specific circumstances of the casting thin section, we can make the following settings: set the number of unit layers to 1, the unit type to prism, and the distribution mode to uniform distribution.

[0055] Through the above operation, a quasi-three-dimensional pore scale grid structure can be obtained, as shown in Figure 5 . However, in actual application, we found that the obtained grid structure has certain defects. Therefore, the present embodiment also provides S6.

[0056] S6: delineating the boundary of the quasi-three-dimensional pore scale grid structure and performing topological consistency checking to obtain its geometric defects, and repairing the geometric defects to make the quasi-three-dimensional pore scale grid structure topologically closed.

[0057] In this step, we first identify the basic topological elements of the quasi-three-dimensional pore-scale grid structure, such as vertices, edges and faces, and then set boundaries, which are as follows in this embodiment:

[0058]

[0059] After setting the boundaries, the grid model is checked for topological consistency to obtain its geometric defects, which include face breakage, holes and non-manifold structures, which are common geometric defects. After obtaining the geometric defects, they need to be repaired, such as using the broken face merging method to repair the face breakage, using the face closure method to repair the hole, and using the normal correction for the non-manifold structure. After the correction, a quasi-three-dimensional pore-scale grid structure with topological integrity is obtained. For this grid model, it can be imported into existing three-dimensional model processing software for subsequent processing, such as OpenFOAM, effectively making up for the defect that existing two-dimensional models cannot be numerically simulated on the OpenFOAM platform, and providing a solution for efficient and high-fidelity pore-scale multiphase flow simulation.

[0060] In this embodiment, the quasi-three-dimensional pore-scale grid structure with topological integrity obtained is as shown in Figure 6 .

[0061] Based on the quasi-three-dimensional porous medium pore-scale grid model constructed by the present application, gas-water two-phase seepage numerical simulation is carried out. The simulation scene is the gas-drive water process, and the analysis focus is the fluid distribution characteristics in the late flow stage.

[0062] As shown in Figure 7 , it is the simulation result figure of the quasi-three-dimensional pore-scale grid structure obtained based on the method of this embodiment. The gas saturation in the model is non-uniformly distributed, where 0 represents that the local area is completely saturated with water phase, and 1 represents that it is completely saturated with gas phase. The blue area in the figure represents high water saturation, the red area represents gas saturation, and the area between the red and blue areas is the gas-water mixed transition zone. It can be seen that a obvious trapped water retention zone is formed in the upper region of the model, which shows the unexplored area caused by capillary pressure and pore structure restriction in the real reservoir.

[0063] According to the actual micro-displacement experiment, it also forms a trapped water retention zone in the upper region of the model, which shows that when the quasi-three-dimensional pore-scale grid structure with topological integrity obtained by this embodiment is used for simulation, the result conforms to the residual water distribution characteristics observed in the micro-displacement experiment. The micro-visualization displacement experiment result is as shown in Figure 8 , at the end of displacement, there is a obvious trapped water in the model.

[0064] To verify the improvement effect of the method on the two-dimensional simulation results, a two-dimensional porous medium pore scale grid is constructed based on the same two-dimensional geometric model by generating a free triangle grid, and the gas-water two-phase flow simulation under the same working condition is carried out. The simulation results are shown in Figure 9 As shown in the figure, although the late two-phase fluid tends to be stable distribution, the bound water phenomenon is not observed in the upper region of the model, and only a small amount of residual water phase exists near the boundary, and the sweep range of the injected gas is obviously expanded. The simulation results deviate in physical mechanism, and cannot reflect the local un-swept area caused by the capillary binding effect, which is inconsistent with the actual reservoir microscopic visualization experiment results.

[0065] The comparison results show that the quasi-three-dimensional modeling method proposed in the application preserves the topological characteristics of the two-dimensional structure, introduces the normal spatial dimension to construct a structure-consistent and thickness-controllable calculation grid, so that the micro mechanisms such as limited migration of multiphase interface and capillary binding can be truly expressed in the simulation process. The gas saturation distribution obtained thereby is closer to the physical experiment observation results, significantly improves the description ability and prediction accuracy of the model on the microscopic seepage process, and fully verifies the effectiveness of the application in improving the physical reality of numerical simulation.

[0066] In the description of the application, it should be pointed out that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and cannot be understood as a limitation on the application.

[0067] The above is only a preferred specific embodiment of the application, but the protection scope of the application is not limited thereto, any changes or replacements within the technical range disclosed by the embodiments of the application can be easily thought of by those skilled in the art, which should be covered within the protection scope of the application. Therefore, the protection scope of the application should be subject to the protection scope of the claims.

Claims

1. A method for generating a pseudo-3D mesh based on images of cast thin sections, characterized in that, Includes the following steps: S1. Select a planar image of a core thin section with a connected pore structure and process it to generate a pore-scale binary image with pore-matrix partitioning; S2. Perform boundary recognition on the pore-scale binary image, and construct the two-dimensional porous medium pore-scale geometric contour curve by combining physical scale restoration and B-spline curve; S3. Based on the geometric contour curve of the pore scale of the two-dimensional porous medium, and considering the pore connectivity, construct the geometric model of the pore scale of the two-dimensional porous medium. S4. Based on the two-dimensional porous medium pore scale geometric model, the normal isomorphic embedding method is used to reconstruct it into a pseudo-three-dimensional porous medium pore scale geometric model. S5. Discretize the pore-scale geometric model of the pseudo-three-dimensional porous medium and generate a pseudo-three-dimensional pore-scale mesh structure by combining it with a sweeping operation.

2. The method for generating a pseudo-3D mesh based on a thin-film image of a cast body according to claim 1, characterized in that, S1 includes the following steps: S11. Obtain a grayscale image by manually identifying planar images of core thin sections with interconnected pore structures and performing grayscale processing. S12. Refer to the image before processing to identify the continuous pore channels in the grayscale image; S13. Based on the Gaussian weighted adaptive thresholding method, the grayscale image processed by S12 is binarized to obtain a pore-scale binary image.

3. The method for generating a pseudo-3D mesh based on a thin-film image of a cast body according to claim 2, characterized in that, S12 also includes the following step: using a pigment recognition algorithm to eliminate non-porous misidentified areas.

4. The method for generating a pseudo-3D mesh based on a thin-film image of a cast body according to claim 1, characterized in that, S2 includes the following steps: S21. Identify the boundary pixels of connected regions in a pore-scale binary image based on the edge detection algorithm, and extract the boundary point set of the pore phase and matrix phase using the Canny operator; S22. Combine the original core thin section planar image with the image coordinates to convert the image coordinates into physical coordinates; S23. The boundary point set is parametrically fitted using B-spline curves, and combined with the physical coordinates in S22, a continuous and differentiable two-dimensional porous medium pore-scale geometric profile curve is obtained.

5. The method for generating a pseudo-3D mesh based on a thin-film image of a cast body according to claim 1, characterized in that, S3 includes the following steps: S31. Construct a rectangle that matches the geometric contour curve of the pore scale of the two-dimensional porous medium in the x and y directions as a base, and superimpose the geometric contour curve of the pore scale of the two-dimensional porous medium onto the rectangular base. S32. Using Boolean difference operations, the rectangular base is divided by geometric contour curves, and non-connected or closed areas are removed, leaving only the effective channel areas connected to the main pore structure. S33. Transform the preserved region into a two-dimensional solid domain to obtain a two-dimensional porous medium pore-scale geometric model.

6. The method for generating a pseudo-3D mesh based on a thin-film image of a cast body according to claim 1, characterized in that, S4 includes the following steps: S41. By rigid transformation, the geometric centroid of the two-dimensional porous medium pore scale geometric model is aligned to the (x,y) plane of the three-dimensional Cartesian coordinate system, and its boundary conditions and pore phase relationship are solidified to obtain the two-dimensional geometric domain. S42. According to the spatial scale requirements, obtain the normal extension depth as a high-dimensional reference domain, and construct the topological direct product space of the two-dimensional geometric domain based on the high-dimensional reference domain. S43. Apply the shape-preserving embedding mapping operator to map the topological direct product space to three-dimensional Euclidean space. The shape-preserving embedding mapping operator is: Φ:(x,y)∈Ω 2D ,z∈I z →(x,y,z)∈Π 3 In the formula, Φ is the row-preserving embedding mapping operator, and Ω is... 2D For a two-dimensional geometric domain, I z Indicates the depth of normal extension, Π 3 This represents three-dimensional Euclidean space, where x, y, and z represent the x-axis, y-axis, and z-axis of the three-dimensional Cartesian coordinate system, respectively. S44. Based on the principle of spatial attribute inheritance, a three-dimensional porous phase domain is defined, and a quasi-three-dimensional porous medium pore scale geometric model is constructed.

7. The method for generating a pseudo-3D mesh based on a thin-film image of a cast body according to claim 1, characterized in that, S5 includes the following steps: S51. The interface of the pore-scale geometric model of the pseudo-three-dimensional porous medium in the (x,y) plane is selected as the starting surface for meshing. The free triangulation algorithm is used to discretize it in two dimensions to generate a planar mesh. S52. Based on the sweeping operation, the planar mesh is swept along the z-axis to form a pseudo-three-dimensional pore-scale mesh structure.

8. The method for generating a pseudo-3D mesh based on a thin-film image of a cast body according to claim 1, characterized in that, It also includes S6: delineating the boundary of the pseudo-three-dimensional porosity scale mesh structure, performing a topological consistency check on it to obtain its geometric defects, and repairing the geometric defects to make the pseudo-three-dimensional porosity scale mesh structure topologically closed.

9. The method for generating a pseudo-3D mesh based on a thin-film image of a cast body according to claim 8, characterized in that, The geometric defects include surface breakage, holes, and non-manifold structures. The methods for repairing geometric defects include surface merging, surface closure, and normal correction.

Images