A pore topography feature-based mud penetration discrete element fluid-solid coupling optimization simulation method

Abstract

The present application provides a kind of pore topography feature-based mud penetration discrete element fluid-solid coupling optimization simulation method, the spatial geometric information of sand is obtained by CT scanning method and digital three-dimensional reconstruction technology, and the STL three-dimensional graphic file of sand formation is established;The STL file is imported into the computational fluid dynamics-discrete element coupling program to establish the pore flow field grid and the discrete element wall boundary, and the discrete element fluid-solid coupling calculation model is established, so as to realize the calculation requirement of restoring the pore topography feature of formation, to reduce the calculation amount and improve the calculation efficiency, in the discrete element module, the total number of mud particles in real-time operation is reduced by inserting the speed particle and deleting the filter loss particle in a loop, using this method, the actual formation can be simulated according to the actual topography and the actual size of the slurry particles, and a new idea is provided for the research method of the micro mechanism of the mud penetration of the shield tunnel excavation face.

Description

TECHNICAL FIELD

[0001] The present application relates to the fields of three-dimensional digital core reconstruction, porous structure seepage flow field analysis, shield tunnel excavation face stability analysis and mud penetration discrete element fluid-structure coupling numerical simulation. BACKGROUND

[0002] The slurry balance shield method is the main method for tunnel construction in water-rich strata in China, and the quality of the mud film is the key to the stability of the excavation face. Once the excavation face loses stability, it will pose a serious threat to the safety of personnel and property in the tunnel. The research methods for mud film formation are as follows: in the macroscopic aspect, the mutual relationship between the mud and the stratum is qualitatively analyzed through the mud penetration column test, but the mechanism of mud film formation is complex, and the micro-mechanism research based on this method is not in-depth; in the microscopic aspect, the CFD-DEM (computational fluid dynamics-discrete element) coupling method is usually used to simulate the slurry flow from the perspective of the interaction between particles and fluid. This method can capture the dynamic trend of mud particles and analyze the whole process of mud particle penetration and film formation. However, the commonly used method is to simulate the sand particles in the stratum as spherical particles, ignoring the flow field changes caused by the differences in sand morphology, and the calculation resource conditions severely limit the number of particles calculated.

[0003] With the advancement of technology, techniques such as scanning electron microscopy (SEM) and X-ray computed tomography (CT scanning) have begun to be used for the analysis of the pore characteristics of geotechnical bodies. Unlike SEM, which only observes the two-dimensional structural characteristics of the sample surface, CT scanning, as a non-destructive testing technique, can observe the internal characteristics of the sample. Numerous studies have shown that CT scanning technology is one of the most reasonable and accurate methods for characterizing the micro-pore structure of geotechnical bodies. SUMMARY

[0004] The present application aims to optimize the existing computational fluid dynamics-discrete element coupling model based on CT scanning technology, achieving the purpose of characterizing the pore morphology and improving the calculation speed.

[0005] TECHNICAL SCHEME

[0006] A mud penetration discrete element fluid-structure coupling optimization simulation method based on pore morphology characteristics, which obtains the spatial geometric information of sand soil through CT scanning method and digital three-dimensional reconstruction technology, establishes the STL three-dimensional graphic file of sand stratum; imports the STL three-dimensional graphic file into the computational fluid dynamics-discrete element coupling program to establish the pore flow field grid and discrete element wall boundary, and establishes the discrete element fluid-structure coupling calculation model; in the discrete element module, the total number of mud particles in real-time operation is reduced by the method of inserting particles with speed and deleting lost particles through circulation.

[0007] The method includes the following four steps:

[0008] Step 1: pretreat the sand sample, and use CT scanning technology for three-dimensional reconstruction;

[0009] Step 2: pretreat the mud, and use the instrument to measure the rheological parameters of the slurry;

[0010] Step 3: according to the actual formation and slurry conditions, use the computational fluid dynamics-discrete element coupling program to establish a discrete element fluid-solid coupling calculation model for mud penetration;

[0011] Step 4: use the computational fluid dynamics-discrete element coupling program to insert the velocity particles and delete the filtration particles in the discrete element module at the coupling time step.

[0012] The beneficial effects of the present application are:

[0013] The present application optimizes the conventional computational fluid dynamics-discrete element coupling model based on the pore network of the rock-soil body established by CT scanning, so as to realize the characterization of pore topography and improve the calculation rate. The method can simulate the actual morphology of the actual formation and the actual size of the actual slurry particles, and provides a new idea for the research method of the micro mechanism of mud penetration of the shield tunnel excavation face. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 Flowchart of mud penetration discrete element fluid-solid coupling optimization simulation method based on pore topography characteristics.

[0015] Figure 2 Schematic diagram of computational fluid dynamics-discrete element (CFD-DEM) coupling method.

[0016] Figure 3 Schematic diagram of mud suspension shear model ("Model Two").

[0017] Figure 4 Schematic diagram of circulating injection and deletion of velocity particles. DETAILED DESCRIPTION

[0018] The technical solutions provided by the present application will be further described below in combination with specific embodiments and their drawings. The advantages and features of the present application will be clearer in combination with the following description.

[0019] As Figure 1 shown in detail:

[0020] Step 1: pretreat the sand sample, and use CT scanning technology for three-dimensional reconstruction.

[0021] Step 1.1 Sieving and density measurement of sand soil, preparation of sand soil sample by layer compaction method according to the requirements of gradation and porosity, finally forming a cylindrical sample with diameter D and height H.

[0022] Step 1.2 CT scanning and image processing of the prepared sand soil sample, the image processing includes extracting the range of sand soil layer by threshold segmentation method based on watershed algorithm, and the complement of the range of sand soil layer is the pore area.

[0023] Step 2: Pretreatment of mud and measurement of rheological parameters of slurry using instrument.

[0024] Step 2.1 Preparation of mud with concentration n according to the required mass fraction of mud using bentonite.

[0025] Step 2.2 After complete hydration of bentonite, measure the particle size distribution of mud particles using laser particle size analyzer, and measure the viscosity, shear stress, shear rate and other rheological parameters of slurry using rheometer.

[0026] Step 3: Establish the discrete element fluid-structure coupling calculation model (abbreviated as "Model One") of mud penetration using computational fluid dynamics-discrete element coupling program (abbreviated as "Software One") according to the actual formation and slurry conditions, as shown in Figure 2 .

[0027] Wherein the computational fluid dynamics-discrete element coupling program ("Software One") contains two subordinate libraries, namely PISO (Pressure Implicit with Splitting of Operators) algorithm solver ("Software Two") and discrete element solver based on Newton's third law ("Software Three"), and "Software One" exchanges data calculated by "Software Two" and "Software Three" at fixed time steps.

[0028] Wherein the discrete element fluid-structure coupling calculation model ("Model One") contains a computational fluid dynamics model and a discrete element model.

[0029] Step 3.1 Generate STL three-dimensional image file of sand soil for subsequent model establishment;

[0030] 3.1.1 After completing the three-dimensional reconstruction of the sand soil sample, a cubic representative unit is extracted as an STL three-dimensional image file in ASCII encoding standard.

[0031] 3.1.2 Triangular mesh optimization is performed on the obtained STL three-dimensional image file, including deleting overlapping vertices and faces, simplifying the mesh, reducing the number of neighbors of the faces, and repairing holes to meet the file import requirements.

[0032] Step 3.2 The simulation of computational fluid dynamics is performed in the slave library PISO algorithm solver of "software one" (referred to as "software two"), a computational fluid dynamics model is established, and the physical and mechanical parameters are calibrated, and the boundary conditions of the flow field are set, including the following:

[0033] 3.2.1 The specified hexahedral mesh is generated by the command, and the generated hexahedral mesh is fitted with the topology of the STL three-dimensional image file based on the STL three-dimensional image file to generate the pore fluid mesh of the formation.

[0034] 3.2.2 Considering that the mud is a non-Newtonian fluid, the liquid phase part of the slurry is simulated as a pore fluid mesh during model simplification, and the solid phase part of the slurry is simulated as a discrete element particle. Since the particles will disturb the fluid and cause the viscosity of the fluid to rise, a mud suspension shear model ("model two") needs to be established for parameter calibration, as shown in Figure 3 The flow field mesh is periodic on both sides, and the top and bottom are set as fixed velocity boundaries. After determining the density p s , the elastic modulus E s and the density p w of the fluid unit, the viscosity m w of the fluid unit is changed to calculate the shear force on the fluid until the force result meets the measured rheological parameters (shear stress, shear rate, etc.) to determine the final viscosity of the fluid unit.

[0035] 3.2.3 In the computational fluid dynamics model of the discrete element fluid-structure interaction calculation model ("model one"), the required pressure difference boundary is set at the slurry inlet and outlet, and the remaining boundaries are set as non-slip boundaries to meet the requirements.

[0036] Step 3.3 The simulation of discrete elements is performed in the discrete element solver based on Newton's third law ("software three") of the computational fluid dynamics-discrete element coupling program ("software one"), and a discrete element model is established, which is as follows:

[0037] The STL three-dimensional image file is imported into the discrete element model in the form of a wall to restrict the movement of the particles. Since bentonite can form a stable suspension in water, it exhibits colloidal properties, so the DLVO (Derjaguin-Landau-Verwey-Overbeek) contact model based on the diffusion electron layer theory is added to the simulated solid phase, which considers that there is an electrostatic repulsive force and van der Waals attractive force before the particles come into contact, and their size is related to the distance between the particle surfaces.

[0038] The van der Waals attractive force can be expressed as:

[0039]

[0040] wherein:

[0041] D is the particle outer surface distance (unit: m);

[0042] A is the Hamaker constant, often taken as 7.8 x 10 -20 J;

[0043] R i is the radius of the i-th particle (unit: m);

[0044] The electrostatic repulsion force can be expressed as:

[0045]

[0046] wherein:

[0047] K is the inverse of the Debye length, taken as 4.136 x 10 6 m -1 ;

[0048] Z is the interaction constant (unit: J / m), which can be expressed as:

[0049]

[0050] wherein:

[0051] Y0 is the particle surface potential, the value of which is related to the zeta potential.

[0052] After the parameter selection of "Model One" is completed, the coupling step of computational fluid dynamics and discrete elements is set in "Software One" for exchanging the data of the flow field and particles during coupling.

[0053] Step 4: Use the computational fluid dynamics-discrete element coupling program ("Software One") to insert the particles with a speed and delete the lost particles in the discrete element module at the coupling time step.

[0054] A particle dropping area (referred to as "Area One") is set on the upper part of the wall established by the STL file, a particle speed detection area (referred to as "Area Two") is set between the STL file and "Area One", and a particle deletion area (referred to as "Area Three") is set on the lower part of the STL file, as shown in Figure 4 .

[0055] When the data of computational fluid dynamics and discrete elements are exchanged each time, the average speed of the particles in "Area Two" is detected, and the particles in "Area Three" are deleted.

[0056] At each time of exchanging the data of computational fluid dynamics and discrete element, the number of particles in the "region one" is detected, when the number of particles is less than 0.1% of the single number of particles, the particles are again put into the "region one", and the speed of putting the particles into the "region one" is set as the average speed of the detected particles in the "region two".

[0057] The above description is only a description of the preferred embodiments of the present application, and is not any limitation on the scope of the present application. Any modification or modification made by any person skilled in the art according to the above disclosed technical content should be regarded as an equivalent effective embodiment, and belongs to the protection scope of the technical scheme of the present application.

Claims

1. A discrete element fluid-structure interaction optimization simulation method for mud infiltration based on pore morphology characteristics, characterized in that, Spatial geometric information of sandy soil was obtained using CT scanning and digital 3D reconstruction technology, and an STL 3D graphic file of the sandy soil strata was established. The STL 3D graphic file was imported into a computational fluid dynamics-discrete element coupled program to establish a pore flow field mesh and discrete element wall boundary, and a discrete element fluid-structure interaction calculation model was established. In the discrete element module, the total number of mud particles in real-time calculation was reduced by cyclically inserting velocity-carrying particles and deleting filter-loss particles. Step 1: Pre-treat the sand sample and perform three-dimensional reconstruction using CT scanning technology; Step 2: Pre-treat the mud and measure its rheological parameters using instruments; Step 3: Based on the actual formation and slurry conditions, use the computational fluid dynamics-discrete element coupling program to establish a discrete element fluid-structure interaction (DME) model for mud permeation; the DME model consists of two models, namely a computational fluid dynamics model and a discrete element model. Step 4: Using a computational fluid dynamics-discrete element coupling procedure, insert velocity-carrying particles and remove filtered-out particles at the time step of the coupling in the discrete element module; Step 4 includes: Step 4.1 Set up a particle delivery area, i.e. "Area 1", at the top of the wall created in the STL file; set up a particle speed detection area, i.e. "Area 2", between the STL file and "Area 1"; and set up a particle deletion area, i.e. "Area 3", at the bottom of the STL file. Step 4.2 Each time computational fluid dynamics and discrete metadata are exchanged, the average velocity of particles in "Region 2" is detected, and particles in "Region 3" are deleted; Step 4.3 Each time computational fluid dynamics and discrete metadata are exchanged, the number of particles in "Region 1" is detected. When the number of particles is less than 0.1% of the single deployment quantity, particles are deployed again in "Region 1", and the speed of particle deployment in "Region 1" is set to the average speed of particles detected in "Region 2".

2. The method as described in claim 1, characterized in that, Step 1 includes: Step 1.1 The sand is sieved and its density is measured. Based on the gradation and porosity requirements, sand samples are prepared by layered compaction. Step 1.2 Perform CT scanning and image processing on the prepared sandy soil sample. The image processing includes extracting the range of the sandy soil layer using a threshold segmentation method based on the watershed algorithm. The complement of the range of the sandy soil layer is the pore region.

3. The method as described in claim 1, characterized in that, Step 2 includes: Step 2.1 Prepare the mud using bentonite according to the required mass fraction of the mud; Step 2.2 After the bentonite is fully hydrated, the particle size distribution of the mud particles is measured using a laser particle size analyzer, and the rheological parameters of the slurry are measured using a rheometer.

4. The method as described in claim 1, characterized in that, Step 3: The computational fluid dynamics-discrete element coupling program contains two dependent libraries: a PISO algorithm solver and a discrete element solver based on Newton's third law. The computational fluid dynamics-discrete element coupling program exchanges the data calculated by the PISO algorithm solver and the discrete element solver at fixed time steps. The discrete element fluid-structure interaction computational model includes two models: a computational fluid dynamics model and a discrete element model. Step 3 includes: Step 3.1 Generate an STL 3D image file of the sand; 3.1.1 After the sand sample was reconstructed in three dimensions, it was exported as an STL three-dimensional image file with ASCII encoding standard; 3.1.2 The acquired STL 3D image file is optimized by removing overlapping vertices and faces, simplifying the mesh, reducing the number of neighboring faces, and repairing holes; Step 3.2 Perform computational fluid dynamics simulation in the PISO algorithm solver to establish a computational fluid dynamics model; Step 3.3 Perform discrete element simulation using a discrete element solver based on Newton's third law to establish a discrete element model.

5. The method as described in claim 4, characterized in that, In step 3.2, a pore fluid grid is generated in the formation. Based on the rheological parameters of the slurry measured in step 2, reasonable physical and mechanical parameters of the fluid grid are set, and the boundary conditions of the flow field are set, specifically including the following: 3.2.1 Generate a specified hexahedral mesh using commands. Based on the STL 3D image file, use commands to fit the generated hexahedral mesh with the topology of the STL 3D image file to generate a pore fluid mesh for the formation. 3.2.2 Establish a shear model for the mud suspension and calibrate its parameters, after determining the particle density. Elastic modulus and the density of the fluid unit Then, by changing the viscosity of the fluid unit The shear force on the fluid is calculated until the force result matches the measured rheological parameters, at which point the final fluid element viscosity is determined. 3.2.3 Set the required pressure difference boundary at the inlet and outlet of the slurry, and set the remaining boundaries as non-slip boundaries to meet the requirements.

6. The method as described in claim 4, characterized in that, In step 3.3: Import the STL 3D image file into the discrete element model as a wall to restrict the movement of particles; In the simulated solid phase, a DLVO contact model based on the diffused electron layer theory is added, which assumes that electrostatic repulsion and van der Waals attraction exist before particles come into contact. The van der Waals gravitational force is represented as: in: It is the distance between the outer surfaces of the particles, in meters (m). This is the Hamaker constant, usually taken as 7.8 × 10⁻⁶. -20 J; It is the radius of the i-th particle, in meters (m). Electrostatic repulsion is expressed as: in: It is the reciprocal of the length of Debye, taken as 4.136 × 10. 6 m -1 ; It is the interaction constant, with units of J / m, and is expressed as: in: denoted as σp, representing the surface potential of the particle, and its value is related to the zeta potential.

Images