A random particle generation method based on particle size distribution

Abstract

The present invention proposes a method for random generation of particles based on particle size distribution, which can meet the particle size distribution characteristics of geological structures. The method adopts a random particle generation method, avoids the commonly used spherical structure, and is very similar to the real soil structure. In combination with the actual particle size distribution, batch generation is used to process different particle sizes, randomly select pore positions to generate particle seeds, and generate particles based on seeds. By proposing a random particle generation method based on particle size distribution, the present invention can restore the porous medium structure of the soil through simulation and provide a porous medium model for groundwater infiltration and fossil energy mining research. Providing technology for pore structure modeling for simulating and calculating multi-phase and multi-component flow migration will help promote the development of research on microscopic pore flow laws.

Description

Technical Field

[0001] The invention relates to a method for constructing a three-dimensional porous medium, and in particular to a method for randomly generating particles based on particle size distribution. Background Art

[0002] The geological structure of porous media is common in the research of groundwater seepage, oil and gas extraction, hydrate extraction, etc., and the transmission characteristics at the macroscopic scale depend on the characteristics of its microstructure. Therefore, obtaining a more realistic in-situ soil structure is conducive to improving the accuracy of the research results.

[0003] Currently available methods include field mining, experimental preparation, and numerical simulation. Field mining is costly, difficult to obtain, and time-consuming; experimental preparation is time-consuming and cannot be used for process analysis; and if these two methods use CT scanning to reconstruct the three-dimensional structure, there are problems such as difficulty in interface tracking and hysteresis. Therefore, many scholars have begun to use numerical simulation to construct porous media structures. At present, scholars have proposed a variety of structural models, including random models, connecting rod models, particle chain models, capillary models, statistical models, network models, and fractal models. Wang et al. (WANG M, PAN N. Numerical analyses of effective dielectric constant of multiphase microporous media [J]. Journal of Applied Physics, 2007, 101 (11): 114102-1-114102-8.) proposed a four-parameter random generation method to generate porous media based on random cluster growth theory. This method can generate porous media of various shapes, and it is intuitively similar to the real soil structure. It is currently used by many scholars for geological structure simulation (Zhong Siwei, Quantitative Study on Reconstruction of Soil Microscopic Porous Structure by Four-Parameter Random Generation Method, Wuhan University of Technology, 2018). Although this method can achieve a structure similar to that of real soil, the particle distribution is relatively uniform. In geological exploration, the particle size distribution of soil is an important data parameter for geological analysis, which shows that this method has certain defects. Summary of the invention

[0004] In order to solve the problems existing in the prior art, the present invention provides a random particle generation method based on particle size distribution, which can generate soil structure according to soil particle size distribution to meet geological characteristics, and the cluster-type generation method can generate a porous medium structure that is more similar to real soil.

[0005] The specific technical solution of the present invention is a method for randomly generating particles based on particle size distribution, comprising the following steps:

[0006] Step 1, obtaining geological structure information and determining the simulation target; the geological structure information includes particle size distribution information and porosity of the target soil;

[0007] Step 2, determining the initial parameters of the simulation system, including the three-dimensional area size of the simulation structure, the porosity of the simulation system, the particle size distribution information of the simulation system, and the growth probability of each direction of the solid phase point of the simulation system; wherein the porosity of the simulation system is the porosity of the target soil in step 1, and the particle size distribution information of the simulation system is the particle size distribution information of the target soil; the particle size distribution information of the simulation system is the corresponding relationship between the particle size and the distribution probability;

[0008] Step 3: Determine the judgment value during the simulation based on the initial parameters of step 2:

[0009] Step 3.1, the total number of solid phase points in the simulation system = the total number of grids in the simulation area × (1-porosity);

[0010] Step 3.2, the total number of solid phase points of particles with the same particle size N = the distribution probability corresponding to each particle size × the total number of system grids;

[0011] Step 3.3, the upper limit of the solid phase point of a single particle n = πD 3 / 6, where D is the particle diameter; the upper limit of the solid phase point is calculated according to the volume of the sphere;

[0012] Step 3.4, the number of random seeds with a particle size of D = the total number of particles with a particle size of D = N / n, rounded up to ensure that the total number of solid phase points is met;

[0013] Step 4, constructing porous media in batches from large to small according to particle size;

[0014] Step 4.1, construct the first batch of particles of the same size: first generate seeds of N / n particles and number them, and the subsequent growth of particles is based on the seeds; the growth of particles satisfies the growth probability of the solid phase point in each direction in step 2, and the growing solid phase point is marked with the same number as the seed, so that the volume change of the particle can be determined. By traversing the solid phase point for growth, it is determined whether the particle reaches the upper limit n of the solid phase point of a single particle body, and if so, the particle stops growing; if the traversal process reaches the total number N of solid phase points of particles of this size, the construction of the batch of particle sizes is completed;

[0015] Step 4.2, complete the generation of subsequent batches of particles: some solid phase points already exist in the simulation area. Therefore, when generating the second batch and subsequent particle sizes, it is necessary to first calculate the minimum distance between all pore points and the solid phase point. After the generation of the previous batch of particles is completed, calculate the minimum distance l between all pore points and the solid phase point, and the seeds of the next batch are randomly generated at the pore points that satisfy l ≥ 1 / 2 the particle size of the batch of particles, and then perform the same construction steps as step 4.1 until the number of solid phase points required for the batch of particles is reached;

[0016] Step 4.3, loop step 4.2 to complete the subsequent particle generation. When the last batch of particles of the particle size is generated, it is also necessary to determine whether the total number of solid phase points in the system reaches the total number of solid phase points of the simulation system in step 3.1. If so, the generation is completed.

[0017] Furthermore, in step 1, the particle size distribution information of the target soil is as follows: the particle size range is divided into n segments, each segment contains multiple data points, the weighted average of the particle size distribution probability of the data points in each segment is used as the particle size of the segment, and the sum of the distribution probabilities of the data points in each segment is used as the distribution probability of the segment.

[0018] Furthermore, in step 1, the simulation method requires a certain processing power of the computer. Therefore, when the value of n is too large, the calculation time will be too long. Therefore, in order to balance the calculation time and simulation accuracy, the n is selected from 2 to 7.

[0019] Furthermore, in step 2, the growth probability of each direction of the solid phase point of the simulation system is determined: the directional growth probability is the probability that the central solid phase point forms solid phase growth in each direction; there are 26 growth directions of the solid phase point, including 6 face center directions, 12 edge center directions and 8 corner point directions, a total of three major directions.

[0020] Furthermore, in step 2, if the particle structure is isotropic, the growth probability of each major direction is the same.

[0021] Furthermore, in step 2, the growth probability P1 in the 6 face center directions is 0.001-0.0001, the growth probability P2 in the 12 edge center directions is P1 / 4, and the growth probability P3 in the 8 corner point directions is P1 / 16.

[0022] Furthermore, in step 2, the three-dimensional region of the simulation structure is a cube structure.

[0023] Furthermore, in step 1, the particle size distribution information of the target soil can be obtained from geological surveys, papers, experiments, etc.

[0024] Furthermore, in step 1, the particle size distribution should be divided into as few segments as possible, which can improve the generation efficiency, and especially avoid directly using the particle size distribution function.

[0025] The beneficial effects of the present invention are as follows: a porous medium generation method based on particle size distribution is constructed, which can meet the particle size distribution characteristics of geological structures. The method adopts a random particle generation method, avoids the commonly used spherical structure, and is very similar to the real soil structure. Therefore, by proposing a random particle generation method based on particle size distribution, the present invention can restore the porous medium structure of the soil through simulation, and provide a porous medium model for groundwater infiltration and fossil energy mining research. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flow chart of the calculation process of the present invention.

[0027] Figure 2 This is a geological information map of the South China Sea used as a reference in the embodiments of the present invention.

[0028] Figure 3 It is an oblique view after visualization of the embodiment of the present invention.

[0029] Figure 4 It is a visualized cross-sectional view of an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings.

[0031] The following is a further detailed description of this patent through specific implementation methods and in conjunction with the accompanying drawings. The specific implementation methods of this method are as follows Figure 1 The flowchart is shown in FIG.

[0032] Example:

[0033] This method is used to simulate the clayey silt geological structure of the hydrate reservoir in the Shenhu area of ​​the South China Sea, and the geological structure is isotropic.

[0034] The first step is geological structure information: simulate the isotropy of geological structure and the particle size distribution characteristics such as Figure 2 As shown in the figure, the distribution ratios of clay particle size (<4um), silt particle size (4-63um) and sand particle size (>63um) are 18.76%-32.29%, 66.21%-75.02%, and 0.06%-10.56%, respectively. The average particle size is 12um and the porosity is 33%. Therefore, the particle size distribution information is set as 4 (26%), 12 (70%), and 64 (4%) in this simulation, which meets both the distribution range and the average particle size.

[0035] Step 2, initial data: the simulation area is 100*100*100 grids, and the side length of a single grid is 2um. Therefore, during the simulation process, the particles are arranged from large to small according to the particle radius and in units of grids. The particle size array is aimRadiu[3]={16,3,1}, and the particle size distribution probability array is solidPercent[3]={0.04,0.70,0.26}. The porosity n=0.33, and the number of grids on the three sides of the simulation area NX=NY=NZ=100. The directional growth probabilities are face center growth probability P1=0.001, edge center growth probability is face center P2=1 / 4, and corner point growth probability is face center P3=1 / 16.

[0036] Step 3: Calculate the judgment value in the simulation process according to the initial parameters: According to the size of the simulation area and the porosity, the total number of solid phase points in the simulation system is calculated as: totalSolid = NX*NX*NZ*(1-n), that is, totalSolid = 100*100*100*(1-0.33) = 670000. Through the particle size distribution information, the particle with a particle radius of 16 is calculated, the number of solid phase points of a single particle is 17158, the total number of solid phases is rounded to 26800, and the number of generated particle seeds is equivalent to 2 particles; the number of solid phase points of a single particle with a particle radius of 3 is 114, the total number of solid phase points is 469000, and the number of particle seeds is 4115; the number of solid phase points of a single particle with a particle radius of 1 is 5, the total number of solid phase points is 174200, and the number of particle seeds is 34840. (Decimal values ​​are rounded up to ensure that the total number of solid phase points meets the total number of solid phase points requirements of the same particle size). Arranged from large to small, the total number of solid phase points of particles with the same particle size is: cycleSolid[i] = totalSolid*solidPercent[i], that is, cycleSolid[3] = {26800,469000,174200}; the upper limit of the solid phase point of a single particle is: particleSolid[i] = 3.1416*(2*aimRadiu[i]) 3 / 6, that is, particleSolid[3] = {17158,114,5}; the number of random seeds with particle size D: seedNum[i] = cycleSolid[i] / particleSolid[i], that is, seedNum[3] = {2,4115,34840}.

[0037] Step 4, iterative calculation: Because there are three particle sizes, the first batch of particle growth is performed according to the program. The particle radius is 16. After randomly generating 2 seeds, random growth begins, but the total number of solid phase points of each particle cannot exceed the target volume of the particle. After the total number of solid phases reaches 26,800, the batch generation is completed and the subsequent batch particle generation begins. Starting from the second generation, it is first necessary to determine the distance between the remaining pore position and the solid phase point, and randomly generate seeds at the position that meets the conditions. Then, the same process as the first batch is carried out, and 4115 seeds are randomly generated. When the solid phase point grows to 469,000, the total number of the batch is reached. The third generation, which is also the last batch, begins. The particle radius is 1, and 34,840 seeds are randomly generated. When the solid phase point grows to 174,200, the total number of the batch is reached, and the total number of particles reaches the total number of solid phase points required by the system, and the particle generation is completed.

[0038] Result display: The generated result is a data file, which can be opened with post-processing software to view the generated porous media structure. Figure 3 is a three-dimensional view. Figure 4 It is the cross-sectional view of the 50th grid layer in the x direction.

[0039] Results verification: This example simulates the clayey silt soil structure of the hydrate reservoir in the Shenhu area of ​​the South China Sea. The geological conditions are different from those of general hydrate reservoirs. The fine particles lead to small porosity and lower permeability. The permeability of the sandstone structure of ordinary hydrate reservoirs can reach about 1D (Li JF, Ye JL, Qin XW, et al. The first offshore natural gas hydrate production test in South China Sea [J]. China Geology, 2018, 1 (1): 5-16.), while the experimentally calculated permeability value of the clayey silt sand structure in the Shenhu Sea area is only 2-200mD (Bian H, XiaY, Lu C, et al. Pore Structure Fractal Characterization and Permeability Simulation of Natural Gas Hydrate Reservoir Based on CT Images [J]. Geofluids, 2020, 2020: 1-9.) (Li JF, Ye JL, Qin XW, et al. The first offshore natural gas hydrate production test in South China Sea [J]. China Geology, 2018, 1 (1): 5-16.).

[0040] Using the embodiment model constructed by the present invention, the permeability of the porous medium is calculated using the lattice Boltzmann method, and the result is 72.8 mD. The simulation result is within a reasonable range. Therefore, this method can be used to obtain a structural model that meets the geological characteristics.

Claims

1. A method for random generation of particles based on particle size distribution, characterized in that: The steps include: Step 1, obtaining geological structure information and determining the simulation target; the geological structure information includes particle size distribution information and porosity of the target soil; Step 2, determining the initial parameters of the simulation system, including the three-dimensional area size of the simulation structure, the porosity of the simulation system, the particle size distribution information of the simulation system, and the growth probability of each direction of the solid phase point of the simulation system; wherein the porosity of the simulation system is the porosity of the target soil in step 1, and the particle size distribution information of the simulation system is the corresponding relationship between the particle size of the target soil and the distribution probability; Step 3: Determine the judgment value during the simulation based on the initial parameters of step 2: Step 3.1, the total number of solid phase points in the simulation system = the total number of grids in the simulation area × (1-porosity); Step 3.2, the total number of solid phase points of particles with the same particle size N = the distribution probability corresponding to each particle size × the total number of system grids; Step 3.3, the upper limit of the solid phase point of a single particle n=πD 3 / 6, where D is the particle diameter; Step 3.4, the number of random seeds with a particle size of D = the total number of particles with a particle size of D = N / n, rounded up; Step 4, constructing porous media in batches from large to small according to particle size; Step 4.1, construct the first batch of particles of the same size: first generate seeds of N / n particles and number them, and the subsequent growth of particles is based on the seeds; the growth of particles meets the growth probability of the solid phase point in each direction of step 2, and the growing solid phase point is marked with the same number as the seed; grow by traversing the solid phase point to determine whether the particle reaches the upper limit n of the solid phase point of a single particle body, and if so, the particle stops growing; if the traversal process reaches the total number N of the solid phase points of the particles of the particle size, the construction of the batch of particle sizes is completed; Step 4.2, complete the generation of subsequent batches of particles: after the generation of the previous batch of particles is completed, calculate the minimum distance l between all pore points and the solid phase point, and randomly generate the seeds of the next batch at the pore point that satisfies l ≥ 1 / 2 of the particle size of the batch of particles, and then perform the same construction steps as step 4.1 until the number of solid phase points required for the batch of particles is reached; Step 4.3, loop step 4.2 to complete the subsequent particle generation. When the last batch of particles of the particle size is generated, it is also necessary to determine whether the total number of solid phase points in the system reaches the total number of solid phase points of the simulation system in step 3.

1. If so, the generation is completed.

2. The method for random generation of particles based on particle size distribution according to claim 1, characterized in that: In step 1, the particle size distribution information of the target soil is as follows: the particle size range is divided into n segments, each segment contains multiple data points, the weighted average of the particle size distribution probability of the data points in each segment is used as the particle size of the segment, and the sum of the distribution probabilities of the data points in each segment is used as the distribution probability of the segment.

3. The method for random generation of particles based on particle size distribution according to claim 2, characterized in that: In step 1, n is selected from 2 to 7.

4. The method for random generation of particles based on particle size distribution according to claim 1, characterized in that: In step 2, the growth probability of each direction of the solid phase point of the simulation system is determined: the directional growth probability is the probability of the central solid phase point forming solid phase growth in each direction; there are 26 growth directions of the solid phase point, including 6 face center directions, 12 edge center directions and 8 corner point directions, a total of three major directions.

5. The method for random generation of particles based on particle size distribution according to claim 4, characterized in that: In step 2, if the particle structure is isotropic, the growth probability of each major direction is the same.

6. The method for random generation of particles based on particle size distribution according to claim 5, characterized in that: In step 2, the growth probability P1 in the 6 face center directions is 0.001-0.0001, the growth probability P2 in the 12 edge center directions is P1 / 4, and the growth probability P3 in the 8 corner point directions is P1 / 16.

7. The method for random generation of particles based on particle size distribution according to claim 1, characterized in that: In step 2, the three-dimensional region of the simulated structure is a cube structure.

8. The method for random generation of particles based on particle size distribution according to claim 1, characterized in that: In step 2, the three-dimensional region of the simulation structure is a cube structure with a size of 100*100*100 grid.

Images