CN111223186A

CN111223186A - Physical modeling method, system and equipment of three-dimensional random hole structure model

Info

Publication number: CN111223186A
Application number: CN202010042734.1A
Authority: CN
Inventors: 谢乐; 蒋崇文
Original assignee: Central South University
Current assignee: Hunan Jingjian New Materials Technology Co ltd
Priority date: 2020-01-15
Filing date: 2020-01-15
Publication date: 2020-06-02
Anticipated expiration: 2040-01-15
Also published as: CN111223186B

Abstract

The invention provides a physical modeling method, a system and equipment of a three-dimensional random pore structure model, comprising the following steps: establishing a three-dimensional porous medium region physical model, carrying out grid division on the three-dimensional porous medium region physical model by using a structured hexahedral grid, and outputting a first grid file containing node position information and an element-node corresponding relation; the three-dimensional porous medium area consists of a framework and gaps, the positions of the gaps are random, grid elements representing the framework are removed, the grid elements representing the gaps are reserved, and a second grid file containing void ratio and pore size information is reconstructed; and importing the second network file into ANSYS ICEM software, identifying and removing an isolated mesh by adopting a Flood-fill/Make _ constraint consistency function, and generating a surface mesh by utilizing a mesh checking function to ensure that the volume element at the outermost layer is covered by the curved surface mesh.

Description

Physical modeling method, system and equipment of three-dimensional random hole structure model

Technical Field

The invention relates to the technical field of physical modeling and grid division of a porous medium region random pore structure, in particular to a physical modeling method, a system and equipment of a three-dimensional random pore structure model.

Background

The study of hydrodynamic characteristics inside the porous medium region is of critical importance. In order to reflect the local effect of the porous medium region and perform three-dimensional flow field simulation, three-dimensional reconstruction of the void-scale morphological structure is required. By adopting advanced experimental techniques, such as X-ray computed tomography and magnetic resonance imaging, the geometric shape characteristics of the porous medium can be captured, and a pore structure with high spatial resolution can be obtained, but the problems of high cost and high requirement on post-processing computing resources exist.

In recent years, Discrete Element Methods (DEM) have been successful in creating packing structures with randomly packed spherical or cylindrical particle features, and introducing the resulting filler structure into a CFD preprocessor (ANSYS-ICEM) to create a mesh for CFD simulation. However, performing CFD-DEM on the entire porous media area (e.g., fixed bed reactor) requires a large number of simulated particles. This makes accurate simulation results often limited by the available computing resources. In addition, meshing complex topologies (particle-particle and wall-particle contacts) is challenging. Because computational grid cells near the point of contact are typically tilted, numerical simulations tend to diverge. Therefore, it is necessary to adopt a suitable physical modeling method to perform meshing on the three-dimensional porous medium region, and then obtain the flow characteristics and pressure drop characteristics of the fluid inside the porous medium region through CFD simulation.

Disclosure of Invention

The invention provides a physical modeling method, a system and equipment of a three-dimensional random pore structure model, and aims to solve the problem that the porosity characteristics of a porous medium region are difficult to describe in physical modeling.

In order to achieve the above object, an embodiment of the present invention provides a physical modeling method of a three-dimensional random pore structure model, including:

step 1, establishing a three-dimensional porous medium region physical model, carrying out mesh division on the three-dimensional porous medium region physical model by using a structured hexahedral mesh, and outputting a first mesh file containing node position information and an element-node corresponding relation;

step 2, the three-dimensional porous medium area consists of a framework and gaps, the positions of the gaps are random, grid elements representing the framework are removed, the grid elements representing the gaps are reserved, and a second grid file containing void ratio and pore size information is reconstructed;

and 3, importing the second network file into ANSYS ICEM software, identifying and removing an isolated grid by adopting a Flood-fill/Make _ constraint consistency function, and generating a surface grid by utilizing a grid checking function to ensure that the volume element at the outermost layer is covered by the curved surface grid.

Wherein, the step 1 specifically comprises:

and carrying out meshing division on the three-dimensional porous medium region physical model by using a structured hexahedral mesh, wherein the size of the mesh represents the size of a pore diameter, an Abaqus software is used for modeling a mesh file which can be output and contains node position information, each mesh unit consists of one element and eight nodes, the total number of the elements in the output mesh file is Nt, and the elements represent the positions of gaps.

Wherein, the step 2 specifically comprises:

the number of elements representing the gaps is equal to the product of the bed porosity and the total number of elements, wherein the positions of the gaps are random and are described by non-repeated random numbers in a preset range;

and retrieving and retaining elements representing the gap positions in the grid file and node position information of the elements, and combining the gap element file and the node file to obtain a reconstructed grid file.

The embodiment of the invention also provides a physical modeling system of the three-dimensional random pore structure model, which comprises the following steps:

the model establishing module is used for establishing a three-dimensional porous medium region physical model, performing grid division on the three-dimensional porous medium region physical model by using a structured hexahedral grid, and outputting a first grid file containing node position information and an element-node corresponding relation; the three-dimensional porous medium region consists of a framework and gaps, and the positions of the gaps are random;

the reconstruction module is used for removing the grid elements representing the framework, reserving the grid elements representing the gaps, and reconstructing a second grid file containing void ratio and aperture size information;

and the grid generating module is used for importing the second network file into ANSYS ICEM software, identifying and removing an isolated grid by adopting a Flood-file/Make _ constraint consistency function, and generating a surface grid by utilizing a grid checking function to ensure that the outermost volume element is covered by the curved surface grid.

The model establishing module is specifically used for carrying out meshing on the three-dimensional porous medium region physical model by using a structured hexahedral mesh, the size of the mesh represents the size of a pore diameter, an Abaqus software is used for modeling a mesh file which can be output and contains node position information, each mesh unit consists of one element and eight nodes, the total number of the elements in the output mesh file is Nt, and the elements represent the positions of gaps.

Wherein the reconstruction module comprises:

and the retrieval module is used for retrieving and reserving elements representing the gap positions in the grid file and node position information thereof, and obtaining the reconstructed grid file by combining the gap element file and the node file.

The embodiment of the present invention further provides a physical modeling apparatus for a three-dimensional random pore structure model, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the processor implements the steps of the physical modeling method for the three-dimensional random pore structure model according to the above-mentioned embodiment.

The scheme of the invention has the following beneficial effects:

the physical modeling method, the system and the equipment of the three-dimensional random pore structure model are composed of a framework and randomly distributed gaps, a few isolated gaps are removed from the built three-dimensional random pore structure model, most of the gaps are mutually communicated, the real structural characteristics of the interior of a three-dimensional porous medium are faithfully reflected, the problem that the porosity characteristics of a porous medium region are difficult to describe in physical modeling is solved, the porous medium region with different internal structural characteristics can be obtained by carrying out grid modeling based on the void ratio and the average pore diameter parameter, meanwhile, the isolated gaps can be removed by a grid repairing means, and further, a grid model can be provided for computational fluid mechanics simulation of the porous medium region.

Drawings

FIG. 1 is a schematic diagram of an initial finite element model according to the present invention;

FIG. 2 is a diagram of element numbers and their element-node correspondence in a reconstructed element file according to the present invention;

FIG. 3 is a schematic representation of a porous media model reconstruction of the present invention;

FIG. 4 is a schematic illustration of the effect of inlet velocity on pressure drop for different pore sizes;

FIG. 5 is a graphical representation of the effect of void fraction function on pressure drop for different pore sizes.

Detailed Description

In order to make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.

The invention provides a physical modeling method, a system and equipment of a three-dimensional random pore structure model, aiming at the problem that the porosity characteristics of a porous medium region are difficult to describe in the existing physical modeling.

The embodiment of the invention provides a physical modeling method of a three-dimensional random pore structure model, which comprises the following steps:

Wherein, the step 1 specifically comprises:

Wherein, the step 2 specifically comprises:

The physical modeling method of the three-dimensional random pore structure model of the invention constructs a grid model for a porous medium area formed by randomly filling filler particles in a fixed bed reactor with the diameter of 60mm and the height of 600mm, and specifically comprises the following steps: (1) constructing a structured grid model of the three-dimensional fixed bed reactor: a structured finite element mesh model was constructed in the Abaqus software based on the mean pore size parameter (d ═ 5mm), as shown in fig. 1A. The mesh file is output with an inp suffix where each Node and Element has their own number, 8 nodes around each Element, 23353 total nodes and 20760 total elements. The position information of the node and the correspondence between the node and the element are shown in fig. 1B and 1C. And splitting the grid file into an Element file and a Node file, and reserving the Element file and the Node file for standby.

(2) Framework and void separation: the porous medium region is composed of a skeleton and voids, and the positions of the voids are represented by the positions of the elements. Assuming that the bed porosity is 50%, the total number Np of voids is 20760 × 0.5 — 10380. Since the locations of the voids in the porous media region are random, the number of voids may be represented by a non-repeating random number sequence between 1 and 20760. The non-repetitive random number sequence is generated by Matlab software, exported as a text file and used as an index file. Then, based on the index file, the Python language writing program is used for searching and reserving elements representing the gap positions in the Element file, and other truncated elements represent the skeleton, as shown in fig. 2. And finally, combining the gap element file and the node position information file to obtain a reconstructed grid file.

(3) Mesh repair and surface mesh reconstruction: for a three-dimensional mesh that can be applied to computational fluid dynamics software identification and computation, the outermost volume element should be covered by a surface mesh, and boundary conditions need to be imposed on the surface mesh connected to the volume element. And importing the grid file obtained by reconstruction into ANSYS ICEM software, and identifying and removing isolated grids by adopting a Flood-file/Make _ constraint consistency function. Isolated voids often exist in the reconstructed model, as shown in FIG. 3A, which cause subsequent calculations to not converge and errors in the calculation process. Therefore, the isolated gaps shown in fig. 3B need to be deleted before numerical simulation. The finally generated grid model of the gap structure of the porous medium area of the fixed bed is shown in FIG. 3C, and numerical simulation calculation is only carried out in the gap.

The embodiment of the invention also provides a physical modeling system of the three-dimensional random pore structure model, which comprises the following steps: the model establishing module is used for establishing a three-dimensional porous medium region physical model, performing grid division on the three-dimensional porous medium region physical model by using a structured hexahedral grid, and outputting a first grid file containing node position information and an element-node corresponding relation; the three-dimensional porous medium region consists of a framework and gaps, and the positions of the gaps are random; the reconstruction module is used for removing the grid elements representing the framework, reserving the grid elements representing the gaps, and reconstructing a second grid file containing void ratio and aperture size information; and the grid generating module is used for importing the second network file into ANSYS ICEM software, identifying and removing an isolated grid by adopting a Flood-file/Make _ constraint consistency function, and generating a surface grid by utilizing a grid checking function to ensure that the outermost volume element is covered by the curved surface grid.

Wherein the reconstruction module comprises:

The physical modeling method of the three-dimensional random pore structure model further comprises the following specific embodiments: the pressure drop characteristics of the random packing structure were determined by numerical simulation of fluid flow behavior using commercially available software (ANSYS Fluent 16.2). Single phase flow systems were selected as the subject of the study. Air (ρ 1.225kg/m3, μ 1.7894 × 10-5pa.s) was introduced at a constant rate from the bottom of the reactor and discharged at a constant pressure from the top of the column. All other boundary conditions are set as "walls" except for the velocity inlet and pressure outlet. Slip-free boundary conditions are defined for the reactor wall and framework surfaces. The gas velocity in the porous medium region ranges from 0.001 to 0.003m/s, and the fluid flow is laminar.

The pressure drop of the porous media region is related to the flow rate, particle size and bed porosity according to the classical Ergun equation. For laminar flow, the pressure drop calculation is as follows:

in equation (1), Δ P is the pressure drop, L is the tube length, v is the fluid velocity, d_pIs the particle size of the particles,

is the shape factor and ε is the bed voidage. Here, three-dimensional random pore structures were characterized in terms of porosity and pore size, and the influence of flow rate (0.001m/s, 0.0015m/s, 0.002m/s, 0.0025m/s, 0.003m/s), bed porosity (50%, 60%, 70%, 80%) and pore size (2mm, 3mm, 4mm, 5mm) on pressure drop characteristics was examined.

The effect of inlet velocity on pressure drop for different pore sizes is shown in fig. 4: (A) d is 2 mm; (B) d is 3 mm; (C) d is 4 mm; (D) d is 5 mm.

Void fraction function under different pore sizes

The effect on pressure drop is shown in fig. 5: (A) d is 2 mm; (B) d is 3 mm; (C) d is 4 mm; (D) d is 5 mm.

The trend of pressure drop with inlet velocity and bed porosity conforms to the classical Ergun equation (R)²>0.99), indicating the accuracy of the model.

While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. a physical modeling method of a three-dimensional random pore structure model, is characterized in that, comprises:

Step 1, establishing a three-dimensional porous medium region physical model, using structured hexahedral meshes to mesh the three-dimensional porous medium region physical model, and outputting a first grid file containing node position information and the corresponding relationship between elements and nodes;

Step 2, the three-dimensional porous media region is composed of skeletons and voids, and the positions of the voids are random, remove the mesh elements representing the skeleton, retain the mesh elements representing the voids, and reconstruct the information containing the porosity and pore size information. the second grid file;

Step 3, import the second network file into the ANSYS ICEM software, use the Flood-fill/Make_consistent consistency function to identify and remove isolated meshes, and use the mesh check function to generate surface meshes to ensure the outermost layer. The volume elements of are covered by the surface mesh.

2. The physical modeling method of the three-dimensional random pore structure model according to claim 1, wherein the step 1 specifically comprises:

Use structured hexahedral mesh to mesh the physical model of the three-dimensional porous medium region, the mesh size represents the size of the aperture, and the Abaqus software can be used to model a mesh file containing node position information that can be output. Each mesh element It consists of one element and eight nodes. The total number of elements in the output mesh file is Nt, and the elements represent the positions of the voids.

3. The physical modeling method of the three-dimensional random pore structure model according to claim 2, wherein the step 2 specifically comprises:

The number of elements representing voids is equal to the product of the void ratio of the bed and the total number of elements, wherein the positions of the voids are random and are described by non-repeating random numbers within a preset range;

Retrieve and retain the elements representing void positions and their node position information in the mesh file, and obtain the reconstructed mesh file by merging the void element file and the node file.

4. A physical modeling system for a three-dimensional random pore structure model, characterized in that, comprising:

The model building module is used to establish a three-dimensional porous medium region physical model, and uses the structured hexahedral mesh to mesh the three-dimensional porous medium region physical model, and outputs a first grid containing node position information and the corresponding relationship between elements and nodes document; the three-dimensional porous media region is composed of skeletons and voids, and the positions of the voids are random;

The reconstruction module is used to remove the mesh element representing the skeleton, retain the mesh element representing the void, and reconstruct the second mesh file containing the information of the void ratio and the aperture size;

A mesh generation module is used to import the second network file into ANSYS ICEM software, use the Flood-fill/Make_consistent consistency function to identify and remove isolated meshes, and use the mesh check function to generate surface meshes, Make sure the outermost volume elements are covered by the surface mesh.

5 . The physical modeling system for a three-dimensional random pore structure model according to claim 4 , wherein the model establishment module is specifically used to mesh the three-dimensional porous media region physical model by using a structured hexahedral grid. 6 . Division, the mesh size represents the size of the aperture, the mesh file containing node position information that can be output by Abaqus software modeling, each mesh unit consists of one element and eight nodes, the total number of elements in the output mesh file The mesh is Nt, and the element represents the position of the void.

6. The physical modeling system of the three-dimensional random pore structure model according to claim 5, wherein the reconstruction module comprises:

The retrieval module is used for retrieving and retaining the elements representing the position of the gap and the position information of the node in the grid file, and obtaining the reconstructed grid file by merging the gap element file and the node file.

7. A physical modeling device for a three-dimensional random pore structure model, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the The steps of implementing the physical modeling method of the three-dimensional random pore structure model according to any one of claims 1 to 3 when the computer program is described.