A mechanical screen pipe sand prevention simulation method based on CFD-DEM coupling

Abstract

The application discloses a mechanical screen pipe sand prevention simulation method based on CFD-DEM coupling, and comprises the following steps: obtaining a particle size grading curve of reservoir sand particles, and constructing a mechanical screen pipe sand prevention numerical model; using a random algorithm to generate random particle size reservoir sand particles at random positions on an inflow plane; obtaining and assigning parameters of a contact submodel between particles in the numerical model; dividing the numerical model by using a fluid grid; CFD-DEM coupling calculation: for a flow field, calculating porosity from the total volume of reservoir sand particles in the fluid grid, estimating a permeability coefficient according to the porosity, solving a seepage equation of a steady-state incompressible fluid and obtaining flow field information; for reservoir sand particles, additionally considering forces exerted on the reservoir sand particles by the fluid, calculating the motion of the reservoir sand particles according to a discrete element theory; and terminating the simulation when the screen mesh is blocked in balance. The application can simulate the micro-dynamic process of migration and blocking of reservoir sand, and has high simulation accuracy.

Description

Technical Field

[0001] This invention relates to the field of sand control in oil and gas extraction, and in particular to a mechanical screen pipe sand control simulation method based on CFD-DEM coupling. Background Technology

[0002] Sand production is a common and serious problem encountered in the exploitation of loose oil and gas reservoirs, and has become one of the main issues hindering the normal production of wells. On the one hand, sand production may damage the reservoir structure, easily causing geological disasters such as formation collapse and landslides; on the other hand, sand production may cause wellbore blockage, causing erosion and wear on production equipment, seriously affecting the safety and efficiency of oil and gas production. Sand control and management are of great significance for the safe, efficient, and sustainable development of oil and gas reservoirs, and are one of the key measures to ensure oil and gas development. To solve the sand production problem, various sand control completion technologies have been developed. Among them, mechanical screen sand control technology is a commonly used one. The sand-blocking medium in mechanical screens is the key to sand control, and there are mainly four types: screens, slotted joints, metal wool, and pre-filled particles. Clarifying the sand-blocking and blockage mechanisms of sand-blocking media can not only provide a reference for the optimization of sand control technology and parameter design in hydrate production, but also help in the exploration of new sand control completion technologies.

[0003] Currently, research on sand control and prevention primarily relies on experiments and numerical simulations based on continuum mechanics. These studies offer significant insights into understanding the dynamic evolution of sand control, sand control design methods, and microscopic sand control mechanisms. However, experiments and numerical models based on continuum mechanics focus on macroscopic sand control characteristics, making it difficult to accurately characterize the discrete characteristics of reservoirs and the interactions between different phases, and to reveal the microscopic mechanisms of sand control using screens. Although a few experimental setups are equipped with advanced observation systems such as microscopic imaging to observe the dynamic processes of reservoir sand migration and blockage, problems remain regarding the difficulty in observing sand production and control processes and monitoring their spatiotemporal evolution. This presents significant challenges for precise characterization and quantitative representation at the microscopic scale. Summary of the Invention

[0004] To address the shortcomings of existing research on sand control and management, and specifically for the screen-type sand-blocking media used in mechanical screen-tube sand control processes, this invention proposes a CFD-DEM coupled simulation method for mechanical screen-tube sand control. This method considers the randomness of reservoir sand particle inflow and the specific physical structure of the screen, and can not only simulate the macroscopic sand-blocking characteristics of mechanical screens, but also explore the microscopic mechanisms of reservoir sand migration and blockage.

[0005] The specific technical solution is as follows:

[0006] A simulation method for sand control in mechanical screen pipes based on CFD-DEM coupling includes the following steps:

[0007] S1: Obtain the particle size distribution curve of the reservoir sand particles and construct a numerical model for sand control using mechanical screens. The numerical model for sand control using mechanical screens includes, in sequence along the fluid inflow direction, a reservoir sand migration zone, a screen zone, and a sand discharge zone. The reservoir sand and fluid enter from the inflow plane of the reservoir sand migration zone. Some of the reservoir sand is blocked by the screen zone, while the other part of the reservoir sand passes through the screen zone and enters the sand discharge zone. The screen in the screen zone is composed of multiple screen particles bonded together.

[0008] S2: Using a random algorithm, reservoir sand particles of random size are generated at random positions on the inflow plane every certain number of calculation time steps, and a constant inflow velocity is set.

[0009] S3: Obtain and assign parameters to the contact sub-model between particles in the numerical model of the mechanical screen pipe sand control;

[0010] S4: Based on the particle size distribution of reservoir sand, a fluid mesh is used to mesh the numerical model of mechanical screen pipe sand control, and the boundary conditions of the fluid are set.

[0011] S5: CFD-DEM Coupled Calculation: For the flow field, the porosity of the fluid mesh is calculated from the total volume of reservoir sand particles in the fluid mesh, and the permeability coefficient of the fluid mesh is estimated based on the porosity. Thus, the seepage equation of the steady-state incompressible fluid is solved and the flow field information is obtained. For the reservoir sand particles, the force exerted by the fluid on the reservoir sand particles is also considered. The motion of the reservoir sand particles is calculated by discrete element theory, realizing the bidirectional coupled calculation of CFD-DEM.

[0012] S6: When the cumulative sand output remains basically unchanged, the screen is in a state of blockage equilibrium, and the simulation is terminated.

[0013] Furthermore, in S4, the numerical model for sand control using a mechanical screen tube is a cuboid structure. Except for the inflow plane, wall boundaries are set around the mechanical screen tube sand control numerical model. One wall of the sand outlet area is set as a zero-pressure boundary, and the inflow plane, the screen in the screen area, and the zero-pressure boundary are arranged in parallel. The remaining walls are all set as impermeable boundaries. The mechanical screen tube sand control numerical model is set to not consider the effect of gravity, and the density and dynamic viscosity of the fluid are set to be consistent with pure water.

[0014] Furthermore, the screen area includes at least one screen, and the straight-line distance from the screen closest to the inflow plane to the inflow plane is greater than 10 times the maximum particle size of the reservoir sand.

[0015] Furthermore, the screen is made of metal wire, and its structural types include: plain weave dense mesh, twill weave dense mesh, and plain weave square mesh; the metal wire is made of multiple screen particles bonded together, and the overlap between adjacent screen particles is greater than or equal to half the radius of the screen particles.

[0016] Furthermore, when the screen area has several layers of screens, the positions of different layers of screens are shifted to misalign the mesh openings of different layers of screens, thereby changing the actual fluid seepage path and obtaining different multi-layer screen combinations.

[0017] Furthermore, in S3, the numerical model for sand control using mechanical screens contains multiple contact types, including: contact between reservoir sand particles, contact between reservoir sand particles and wall boundaries, contact between screen particles and wall boundaries, contact between screen particles and reservoir sand particles, and contact between screen particles.

[0018] The contact between reservoir sand particles and the contact between reservoir sand particles and wall boundaries are both given a linear contact sub-model. The parameters of the linear contact model are calibrated by discrete element triaxial test of reservoir sand. Considering the presence of fluid, the friction coefficient obtained by calibration is reduced.

[0019] The contacts between the screen particles and the wall boundary, and between the screen particles and the reservoir sand particles, are all assigned linear contact sub-models, while the contacts between the screen particles are assigned linear parallel cementation sub-models. The parameters of these three contact sub-models are set empirically. Without considering the failure behavior of the screen under liquid-solid phase impact, the cementation strength parameter between the screen particles is set to be greater than or equal to 10. 10 The value.

[0020] Furthermore, the calibration procedure for the discrete element triaxial test is as follows:

[0021] (1) The effective modulus of the linear contact sub-model and the linear effective modulus of the parallel cemented sub-model are the main adjustment objects, supplemented by the adjustment of the friction coefficient and the cemented effective modulus of the parallel cemented sub-model, so as to match the macroscopic elastic modulus and Poisson's ratio of the sample.

[0022] (2) Adjust the effective modulus of cementation and the cementation strength parameters of the parallel cemented sub-model to match the peak strength and residual strength of the specimen;

[0023] (3) Further adjust the bonding strength parameters of the parallel bonded sub-model based on the yield failure of the specimen;

[0024] (4) Based on the experimental results, the parameters of the linear contactor model are fine-tuned to balance the errors of different macroscopic mechanical parameters and improve the accuracy of the simulation results.

[0025] Furthermore, in step S5, the size of the fluid grid is determined based on the particle size distribution of the reservoir sand. To avoid singularities when solving the seepage equation for the steady-state incompressible fluid, the porosity of the fluid grid cells is not less than 0.005. The expression for the seepage equation for the steady-state incompressible fluid is as follows:

[0026]

[0027] In the formula, Let μ represent the Hamiltonian operator, u be the fluid velocity, K be the permeability, and μ be the permeability. f denoted as the dynamic viscosity coefficient of the fluid, n as the porosity of the fluid element, and p as the fluid pressure.

[0028] Furthermore, the specific operation of S5 is as follows:

[0029] For the flow field, the average porosity and permeability coefficient are obtained based on the velocity and position information of the reservoir sand particles. The porosity of the fluid element is determined by the overlap between the fluid element and the reservoir sand particles, where the reservoir sand particles are represented by cubes with length, width, and height equal to the particle size. Considering the presence of reservoir sand particles in the fluid flow, the permeability coefficient is set as a function of porosity and estimated using the Kozeny-Carman relation. The seepage equation of the steady-state incompressible fluid is discretized using the open-source solver FiPy, dividing the fluid domain into a structured grid, with each grid node controlling the volume of a certain region. The seepage equation of the steady-state incompressible fluid to be solved is integrated over the fluid control volume of each fluid grid to obtain the discrete equation. Based on the discrete equation, the flow field information, including fluid velocity, pressure, and pressure gradient, is calculated.

[0030] For reservoir sand particles, the flow field information calculated by the open-source solver FiPy is applied to the centroid of the sand particles in the form of fluid-particle interaction forces. The velocity and position of the sand particles are then calculated and updated using discrete element theory, thereby completing the information exchange between the fluid phase and the solid particles, and ultimately achieving bidirectional coupled calculation of CFD-DEM. Reservoir sand particles exhibit both translational and rotational motion modes. During this motion, the sand particles follow Newton's laws of motion, and the interaction forces between the fluid and the sand particles are considered and reflected in the equations of motion, which are expressed as follows:

[0031]

[0032] In the formula, m i Let be the mass of the i-th reservoir sand particle. Let be the acceleration of the i-th reservoir sand particle at time t. The number of contacts acting on the i-th reservoir sand particle. f is the contact force at the k-th contact point of the i-th reservoir sand particle. i g Let f be the volume force acting on the i-th reservoir sand grain. i d f is the damping force experienced by the i-th reservoir sand grain. i fJ represents the force exerted by the fluid on the i-th reservoir sand particle; i Let be the moment of inertia of the i-th reservoir sand particle. Let r be the angular acceleration of the i-th reservoir sand particle at time t. i,k Let $\mathbf{i}$ be the radius vector at the $k$-th contact point of the $i$-th reservoir sand grain. The number of cementing contacts acting on the i-th reservoir sand particle. Let m be the cementing torque of the i-th reservoir sand particle at the m-th cementing contact. Let be the damping torque experienced by the i-th reservoir sand particle.

[0033] Furthermore, in S5, the CFD calculation time step is set to 100 times the DEM calculation time step.

[0034] The beneficial effects of this invention are:

[0035] (1) The present invention uses a particle bonding method to construct a screen structure, which avoids the difficulty of coupling discrete element model and continuous medium model, and helps to improve the accuracy of simulation.

[0036] (2) The present invention uses a random algorithm to make the position and particle size of the reservoir sand entering the reservoir sand migration zone random, which can closely match the actual process of reservoir sand entering the wellbore annulus from the perforation hole, and helps to improve the accuracy of the simulation.

[0037] (3) This invention has wide applicability to the study of sand production problems in loose sandy oil and gas reservoirs. It can not only study the macroscopic sand-blocking characteristics of screens, but also simulate the microscopic dynamic process of reservoir sand migration and blockage in a very intuitive way. Based on the simulation results, on the one hand, the existing sand control technology can be optimized and improved, and on the other hand, the development of new sand control technology can be guided. Attached Figure Description

[0038] Figure 1 This is a flowchart of the mechanical screen tube sand control simulation method based on CFD-DEM coupling of the present invention.

[0039] Figure 2 This is a schematic diagram of the numerical model of the mechanical screen tube sand control in the simulation process in an embodiment of the present invention.

[0040] Figure 3 This is a schematic diagram of the structure of a single-layer screen in an embodiment of the present invention, wherein (a) is the main view, (b) is the auxiliary view, and (c) is the top sectional view.

[0041] Figure 4 This is a flowchart of CFD-DEM coupled calculation in an embodiment of the present invention.

[0042] Figure 5This is a schematic diagram illustrating the evolution of the cumulative sand production mass and sand production rate of reservoir sand under different inflow velocities in an embodiment of the present invention, wherein (a) represents the change of cumulative sand production mass over time, and (b) represents the change of sand production rate over time.

[0043] Figure 6 This is a schematic diagram of the clogging and sand-blocking process of a single-layer screen in an embodiment of the present invention, wherein (a) is a schematic diagram when reservoir sand particles begin to flow in, (b) is a schematic diagram when the screen begins to clog, (c) is a schematic diagram when the screen clogging intensifies, and (d) is a schematic diagram when the screen clogging is balanced.

[0044] Figure 7 These are schematic diagrams illustrating the state of reservoir sand particles blocked inside screens with different numbers of layers in embodiments of the present invention, wherein (a) is a single-layer screen, (b) is a double-layer screen, (c) is a triple-layer screen, and (d) is a quadruple-layer screen.

[0045] In the figure, reservoir sand migration zone 1, screen zone 2, first screen 2-1, second screen 2-2, third screen 2-3, sand discharge zone 3, zero pressure boundary 4, fluid grid 5, screen particles 6. Detailed Implementation

[0046] The present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. The objectives and effects of the present invention will become clearer as a result. The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0047] The essence of sand control and prevention is the migration of reservoir sand under the influence of fluids, while simultaneously being blocked by sand-blocking media. The scientific issues involved are the migration and blockage of reservoir sand caused by gas-liquid flow. Therefore, the CFD-DEM (Computational Fluid Dynamics-Discrete Element Method) coupled method has a natural advantage in clarifying the mechanisms of sand control and prevention in hydrate mining.

[0048] Based on the CFD-DEM coupling method, this invention proposes a mechanical screen pipe sand control simulation method based on CFD-DEM coupling, such as... Figure 1 As shown, the method includes the following steps:

[0049] S1: Obtain the particle size distribution curve of reservoir sand from the oil and gas production blocks, and construct a numerical model for sand control using mechanical screens. The particle size distribution curve of the reservoir sand is generated according to the actual particle size distribution of the reservoir sand, and the density of the reservoir sand is set to be the same as that of silica sand. The specific structure of the numerical model for sand control using mechanical screens is as follows:

[0050] like Figure 2 As shown, the numerical model for mechanical screen sand control is a cuboid structure. The specific dimensions are set according to requirements, ensuring a sufficient number of reservoir sand particles to represent the simulation results. Along the length of the mechanical screen sand control numerical model (i.e., the fluid inflow direction), from left to right, it includes three parts: reservoir sand migration zone 1, screen zone 2, and sand discharge zone 3. The left side of the reservoir sand migration zone 1 is set as the inflow plane, and a constant inflow velocity is set at the inflow plane. Considering computational efficiency, a relatively large inflow velocity can be set; the fluid inflow direction is along the length of the mechanical screen sand control numerical model. Reservoir sand and fluid enter from the inflow plane, pass through screen zone 2, and some reservoir sand is blocked by the unidirectional fluid flow, simulating the actual sand blocking process at the bottom of the well. Another portion of reservoir sand passes through screen zone 2 and enters sand discharge zone 3, simulating the actual sand discharge process. In addition to the inflow plane, rigid wall boundaries are set around the mechanical screen sand control numerical model.

[0051] Screening zone 2 includes at least one screen, the specific configuration depending on the problem being studied. In this embodiment, three screens are used, with screening zone 2 consisting of, from left to right: a first screen 2-1, a second screen 2-2, and a third screen 2-3. The straight-line distance from the first screen 2-1 to the inflow plane should be greater than 10 times the maximum particle size of the reservoir sand. All three screens are made of metal wire, and the structural types of the metal screens include: plain weave dense mesh, twill weave dense mesh, and plain weave square mesh. There is no significant difference in the sand-blocking mechanism among the three structural types; the selection is based on the actual situation, and the structural types of the three screens can be the same or different. Figure 3 As shown, in this embodiment, all three layers of screens are made of plain weave dense mesh.

[0052] like Figure 3 As shown in (c), a single metal wire constituting the screen is made up of multiple small-diameter screen particles 6 bonded together. The spatial position of the screen particles 6 is determined by the structural shape of the metal wire within the screen. The particle size of the screen particles 6 is the same as that of the metal wire, and the density is set to the metal density. It is important to note that the overlap between these screen particles 6 is relatively large (greater than or equal to half the radius of the screen particles 6), thus better simulating the outer contour of the metal wire. After constructing the structure of a single metal wire of the screen using the above method, several metal wires are then generated according to the structural form corresponding to the selected metal screen.

[0053] Since the spacing between multiple layers of screens is difficult to measure directly in practical applications, the distance between adjacent screens can be set empirically when constructing a multi-layer screen structure. Changing the width of the screen mesh openings can produce screens of different precision. Furthermore, the relative positions of multiple layers of screens can be rearranged to misalign the mesh openings, thereby altering the actual fluid flow path and resulting in different combinations of multi-layer screens.

[0054] S2: Using a random algorithm, reservoir sand particles of random size are generated at random positions on the inflow plane of the mechanical screen sand control numerical model every few calculation time steps. A constant inflow velocity is set at the inflow plane so that the reservoir sand particles enter the reservoir sand migration zone 1 from the inflow plane at a constant velocity, thereby simulating the actual process of reservoir sand entering the wellbore annulus from the perforation hole.

[0055] S3: Obtain and assign parameters to the contact sub-model between particles in the numerical model of the mechanical screen pipe sand control system. The numerical model of the mechanical screen pipe sand control system contains various contact types, including: contact between reservoir sand particles, contact between reservoir sand particles and the wall boundary, contact between screen particles and the wall boundary, contact between screen particles and reservoir sand particles, and contact between screen particles.

[0056] The contact between reservoir sand particles and the contact between reservoir sand particles and the wall boundary are both modeled using a linear contactor model. The parameters of this linear contactor model are calibrated using discrete element triaxial tests of reservoir sand, aiming to match the results of the discrete element triaxial tests with the physical test results as closely as possible. In addition, considering the presence of fluid, the calibrated friction coefficient can be appropriately reduced.

[0057] The contacts between screen particles and the wall boundary, and between screen particles and reservoir sand particles, are both assigned to linear contact sub-models, while the contacts between screen particles are assigned to linear parallel cementation sub-models. The parameters for these three contact sub-models are set empirically. Without considering the failure behavior of the screen under liquid-solid phase impact, the cementation strength parameter between screen particles is set to a very large value (greater than or equal to 10). 10 ).

[0058] For the discrete element triaxial test, please refer to the literature "Zhou Shichen, Huo Wenxing, Zhou Bo, et al. Discrete element simulation of mechanical properties of cemented hydrate sediments under triaxial compression of flexible boundary [J]. Journal of Central South University (Natural Science Edition), 2022, 53(03):830-845". The calibration process used in this embodiment is as follows:

[0059] (1) The effective modulus of the linear contactor model and the linear effective modulus of the parallel cemented sub-model are the main adjustment objects, supplemented by the adjustment of the friction coefficient and the cemented effective modulus of the parallel cemented sub-model, so as to match the macroscopic elastic modulus and Poisson's ratio of the sample.

[0060] (2) Adjust the effective modulus of cementation and the cementation strength parameters of the parallel cemented sub-model to match the peak strength and residual strength of the specimen.

[0061] (3) Adjust the bonding strength parameters of the parallel bonded sub-model further based on the yield failure of the specimen.

[0062] (4) Based on the experimental results, the parameters of the linear contactor model are fine-tuned to balance the errors of different macroscopic mechanical parameters and improve the accuracy of the simulation results.

[0063] S4: Based on the particle size distribution of the reservoir sand, determine the size of the structured fluid grid 5, and divide the entire domain of the mechanical screen sand control numerical model into multiple uniform fluid grids 5 to achieve spatial discretization of the mechanical screen sand control numerical model. Set the boundary conditions for the fluid: one wall of the sand outlet zone 3 is set as a zero-pressure boundary 4, and the inflow plane, the screen of the screen zone 2, and the zero-pressure boundary 4 are arranged in parallel; the remaining walls are all set as impermeable boundaries. Set the mechanical screen sand control numerical model to not consider the effect of gravity, and set the density and dynamic viscosity of the fluid to be consistent with pure water.

[0064] Because the zero porosity of the elements in fluid grid 5 will cause singularities when solving the seepage equation for steady-state incompressible fluid, the porosity of the elements in fluid grid 5 cannot be less than 0.005. Based on this, the size of fluid grid 5 can be determined.

[0065] S5: Perform CFD-DEM coupled calculation: The CFD-DEM method used is a coupled algorithm based on fluid mesh averaging, the core of which is to handle the interaction of information between fluid and particles. For the flow field, the porosity of fluid mesh 5 is calculated from the total volume of reservoir sand particles in fluid mesh 5, and then the permeability coefficient of fluid mesh 5 is estimated based on the porosity. This allows for the solution of the seepage equation for the steady-state incompressible fluid and the acquisition of flow field information. For reservoir sand particles, the force exerted by the fluid on the reservoir sand particles is also considered, and the motion of the reservoir sand particles is calculated using discrete element theory, realizing bidirectional coupled calculation of CFD-DEM. In this embodiment, the CFD calculation time step is set to 100 times the DEM calculation time step. Figure 4 As shown, the specific steps for this operation are as follows:

[0066] For the flow field, the porosity and permeability coefficient of the fluid element are calculated based on information such as the velocity and position of the reservoir sand particles. The porosity of the fluid element is determined by the overlap between the fluid element and the reservoir sand particles, where the reservoir sand particles are characterized by cubes with length, width, and height equal to their diameter. Considering the presence of reservoir sand particles in the fluid flow, the permeability coefficient is set as a function of the porosity and estimated using the Kozeny-Carman relation. The seepage equation of the steady-state incompressible fluid is discretized using the open-source solver FiPy, dividing the fluid domain into a structured grid, with each grid node controlling a certain volume. The seepage equation of the steady-state incompressible fluid to be solved is integrated over the fluid control volume of each fluid grid to obtain the discrete equation. Based on the discrete equation, the flow field information such as fluid velocity, pressure, and pressure gradient is calculated. During this process, it is necessary to determine whether the fluid calculation has converged. The construction process of the seepage equation of the steady-state incompressible fluid is as follows:

[0067] Fluid flow is described by equations of mass and momentum conservation. For porous media flow at low Reynolds numbers, assuming the fluid is incompressible, and defining the fluid domain and matrix domain as flat plates with average properties such as average porosity, tortuosity, and permeability, the mass conservation equation becomes the continuity equation, and the momentum conservation equation becomes Darcy's law. Therefore, the fluid governing equations can be described by the continuity equation and Darcy's law, and the seepage equation for steady-state incompressible fluids can be further derived, i.e.

[0068]

[0069] In the formula, u is the fluid velocity, K is the permeability, and μ f Let be the dynamic viscosity coefficient of the fluid, n be the porosity of the fluid element, and p be the fluid pressure. This represents the Hamiltonian operator.

[0070] For reservoir sand particles, the fluid velocity, pressure, and pressure gradient calculated by FiPy are applied to the centroid of the sand particles in the form of fluid-particle interaction forces. The particle motion equations are then calculated using discrete element method (DEM) theory to update the velocity and position of the sand particles, thus completing the information exchange between the fluid phase and the solid particles, and achieving bidirectional coupled CFD-DEM calculation. During this process, it is necessary to determine whether 100 calculation time steps have been completed. If so, fluid calculations continue; otherwise, particle calculations continue. Reservoir sand particles have two motion modes: translation and rotation. During this motion, the sand particles follow Newton's laws of motion, and the interaction forces between the fluid and the sand particles are considered and reflected in the motion equations. The particle motion equations are expressed as follows:

[0071]

[0072] In the formula, mi Let be the mass of the i-th reservoir sand particle. Let be the acceleration of the i-th reservoir sand particle at time t. The number of contacts acting on the i-th reservoir sand particle. f is the contact force at the k-th contact point of the i-th reservoir sand particle. i g Let f be the volume force acting on the i-th reservoir sand grain. i d f is the damping force experienced by the i-th reservoir sand grain. i f J represents the force exerted by the fluid on the i-th reservoir sand particle; i Let be the moment of inertia of the i-th reservoir sand particle. Let r be the angular acceleration of the i-th reservoir sand particle at time t. i,k Let $\mathbf{i}$ be the radius vector at the $k$-th contact point of the $i$-th reservoir sand grain. The number of cementing contacts acting on the i-th reservoir sand particle. Let m be the cementing torque of the i-th reservoir sand particle at the m-th cementing contact. Let be the damping torque experienced by the i-th reservoir sand particle.

[0073] The interaction between the fluid and the reservoir sand particles is very complex. For simplicity, this invention only considers three types of forces: pressure gradient force, buoyancy, and drag force. The drag force is calculated according to the empirical formula proposed by Di Felice.

[0074] S6: When the cumulative sand output remains basically unchanged, the screen is in a state of blockage equilibrium, and the simulation is terminated.

[0075] The present invention will be specifically described below through examples.

[0076] Example 1

[0077] S1: Construct a numerical model for sand control using mechanical screens. Set reservoir sand parameters: Set the particle size range of the reservoir sand to 0.1mm~0.4mm, with a median particle size d. 50 It is approximately 0.198 mm in diameter and has a density of 2560 kg / m³. 3 Based on the particle size distribution curve of the reservoir sand, and taking into account the principle that the mechanical screen sand control numerical model can accommodate a suitable amount of reservoir sand, the main dimensions of the mechanical screen sand control numerical model are set as follows: the side length of the inflow plane is set to 19.2d. 50 The length of the numerical model for sand control using mechanical screen pipes is set to 38.0d. 50Based on the maximum particle size of the reservoir sand, single or multiple layers of screens are installed at a distance greater than 10 times the maximum particle size of the reservoir sand from the inflow plane. The particle size of the screen mesh particles 6 constituting the screen wire is set to 0.02 mm, and the density to be 7800 kg / m³. 3 The overlap of the screen particles 6 is approximately equal to the radius of the screen particles 6, and the screen mesh width is 0.25 mm. A multi-layer screen structure is constructed layer by layer, with the interlayer spacing set to 0.18 mm.

[0078] S2: Using a random algorithm, reservoir sand particles of random size are generated at random locations on the inflow plane of the numerical model every certain number of computation time steps, and a constant inflow velocity is set at the inflow plane. In this embodiment, for ease of result analysis, a baseline inflow velocity v0 = 0.1 m / s is defined. The fluid density is set to 1000 kg / m³. 3 Its dynamic viscosity was set to 0.001 Pa·s.

[0079] S3: Set the contact sub-model parameters between different types of particles in the numerical model of the mechanical screen pipe sand control. See Table 1 for the values ​​of the contact sub-model parameters.

[0080] For the contact between reservoir sand particles and the contact between reservoir sand particles and the wall boundary, the values ​​of their contact sub-model parameters need to be calibrated using triaxial test results; at the same time, considering the presence of fluid, the friction coefficient is reduced to 0.2.

[0081] For the contact between screen particles and the wall boundary, the contact between screen particles and reservoir sand particles, and the contact between screen particles, the values ​​of their contact sub-model parameters are assigned empirically.

[0082] Among the five contactor models mentioned above, there are two types: linear contactor models and linear parallel cementation models. Therefore, the parameters of these two contactor models are divided into linear part parameters and parallel cementation part parameters. For the linear contactor model, only the linear part parameters need to be viewed, while for the linear parallel cementation model, the parameters of the linear part and the parallel cementation part need to be viewed separately.

[0083] Table 1. Values ​​of contactor model parameters

[0084]

[0085] In the table, E is the effective modulus, κ is the ratio of normal to tangential stiffness, and μ is the interparticle friction coefficient. For the effective modulus of bonding, The ratio of cemented normal to tangential stiffness, The tensile strength of the bond; The cohesive force of the adhesive bond. The internal friction angle of the bond is used to determine the shear strength of the bond; a "-" indicates that this item is not present.

[0086] S4: Create a structured fluid grid 5 and set the fluid boundaries and parameters. Based on the requirement that the element porosity of fluid grid 5 cannot be less than 0.005, and considering the particle size distribution of the reservoir sand, the side length of fluid grid 5 is set to 0.28 mm. To improve computational efficiency, a relatively large fluid inflow velocity is set here.

[0087] S5: CFD-DEM Coupled Calculation. For the flow field, the porosity and permeability coefficient of the fluid grid 5 are calculated, and the seepage equation of the steady-state incompressible fluid is solved to obtain the flow field information. For the particles, the force exerted by the fluid on the reservoir sand particles is also considered, and the motion of the reservoir sand particles is calculated by discrete element theory, thereby realizing the bidirectional coupled calculation of CFD-DEM.

[0088] S6: When the cumulative sand output remains essentially unchanged, the screen is in a state of blockage equilibrium, and the simulation terminates. The simulated physical time is used as a measure of time. The simulated physical time for each case is approximately 0.20 seconds, with approximately 1.7 million calculation steps.

[0089] To describe the macroscopic sand-blocking characteristics of screens, the following physical quantities are defined: the cumulative mass of reservoir sand passing through the screen is used as a measure of the cumulative outflow mass; the outflow mass per unit time is used as a measure of the outflow rate; and the ratio of the fluid pressure difference across the screen before and after clogging is defined as the pressure difference ratio. Since the fluid inflow velocity and the inlet cross-sectional area remain constant in a single simulation, the pressure difference ratio can not only measure the magnitude of the pressure difference across the screen but also reflect changes in the screen permeability.

[0090] Figures 5-7 Partial simulation results are presented, showing the results under different fluid inflow velocities when the screen mesh width is 0.25 mm. Figure 5 The evolution of the cumulative sand production quality and sand production rate of reservoir sand with the inflow velocity is presented, which allows us to explore the influence of fluid inflow velocity on sand production characteristics.

[0091] Figure 6 The dynamic processes of clogging and sand retention in a single-layer screen are presented. This allows for the exploration of the microscopic mechanisms of sand retention and clogging in a single-layer screen.

[0092] When the mesh openings of multiple screens are aligned Figure 7 The state of reservoir sand blocked inside screens with different numbers of screen layers under the same minimum diameter is presented. This allows us to explore the microscopic mechanism by which changes in the number of screen layers affect the blocking effect of reservoir sand.

[0093] It will be understood by those skilled in the art that the above descriptions are merely preferred examples of the invention and are not intended to limit the invention. Although the invention has been described in detail with reference to the foregoing examples, those skilled in the art can still modify the technical solutions described in the foregoing examples or make equivalent substitutions for some of the technical features. All modifications and equivalent substitutions made within the spirit and principles of the invention should be included within the scope of protection of the invention.

Claims

1. A mechanical screen sand control simulation method based on CFD-DEM coupling, characterized in that, Includes the following steps: S1: Obtain the particle size distribution curve of the reservoir sand particles and construct a mechanical screen tube sand control numerical model. The mechanical screen tube sand control numerical model includes a reservoir sand migration zone, a screen zone, and a sand discharge zone along the fluid inflow direction. The reservoir sand and fluid enter from the inflow plane of the reservoir sand migration zone. Some of the reservoir sand is blocked by the screen zone, and the other part of the reservoir sand passes through the screen zone and enters the sand discharge zone. The screen in the screen zone is made of multiple screen particles cemented together. In addition to the inflow plane, a rigid wall boundary is set around the mechanical screen tube sand control numerical model. S2: Using a random algorithm, reservoir sand particles of random size are generated at random positions on the inflow plane every certain number of calculation time steps, and a constant inflow velocity is set. S3: Obtain and assign parameters to the contact sub-model between particles in the numerical model of the mechanical screen pipe sand control; S4: Based on the particle size distribution of reservoir sand, a fluid mesh is used to mesh the numerical model of mechanical screen pipe sand control, and the boundary conditions of the fluid are set. S5: CFD-DEM Coupled Calculation: For the flow field, the porosity of the fluid mesh is calculated from the total volume of reservoir sand particles in the fluid mesh, and the permeability coefficient of the fluid mesh is estimated based on the porosity. Thus, the seepage equation of the steady-state incompressible fluid is solved and the flow field information is obtained. For the reservoir sand particles, the force exerted by the fluid on the reservoir sand particles is also considered. The motion of the reservoir sand particles is calculated by discrete element theory, realizing the bidirectional coupled calculation of CFD-DEM. S6: When the cumulative sand output remains basically unchanged, the screen is in a state of blockage equilibrium, and the simulation is terminated; In S4, the numerical model for sand control using a mechanical screen tube is a cuboid structure. Except for the inflow plane, wall boundaries are set around the mechanical screen tube sand control numerical model. One wall of the sand outlet area is set as a zero-pressure boundary, and the inflow plane, the screen in the screen area, and the zero-pressure boundary are arranged in parallel. The remaining walls are all set as impermeable boundaries. The mechanical screen tube sand control numerical model is set to not consider the effect of gravity, and the density and dynamic viscosity of the fluid are set to be consistent with pure water.

2. The CFD-DEM coupled mechanical screen sand control simulation method according to claim 1, wherein, The screen area includes at least one screen, and the straight-line distance from the screen closest to the inflow plane to the inflow plane is greater than 10 times the maximum particle size of the reservoir sand.

3. The CFD-DEM coupled mechanical screen sand control simulation method according to claim 1, wherein, The screen is made of metal wire, and its structural types include: plain weave dense mesh, twill weave dense mesh, and plain weave square mesh; the metal wire is made of multiple screen particles bonded together, and the overlap between adjacent screen particles is greater than or equal to half the radius of the screen particles.

4. The mechanical screen pipe sand control simulation method based on CFD-DEM coupling according to claim 1, characterized in that, When the screen area has several layers of screens, the positions of different layers of screens are shifted to misalign the mesh openings of different layers of screens, thereby changing the actual fluid seepage path and obtaining different multi-layer screen combinations.

5. The mechanical screen tube sand control simulation method based on CFD-DEM coupling according to claim 1, characterized in that, In S3, the numerical model for sand control using mechanical screens contains multiple contact types, including: contact between reservoir sand particles, contact between reservoir sand particles and wall boundaries, contact between screen particles and wall boundaries, contact between screen particles and reservoir sand particles, and contact between screen particles. The contact between reservoir sand particles and the contact between reservoir sand particles and wall boundaries are both given a linear contact sub-model. The parameters of the linear contact model are calibrated by discrete element triaxial test of reservoir sand. Considering the presence of fluid, the friction coefficient obtained by calibration is reduced. The contact between the screen particles and the wall boundary, the contact between the screen particles and the reservoir sand particles, and the contact between the screen particles all give linear contact sub-models, and the contact between the screen particles gives a linear parallel cementation sub-model; the parameters of the sub-models of the three kinds of contacts are all set according to experience; when the failure behavior of the screen under the impact of the liquid-solid phase is not considered, the cementation strength parameter between the screen particles is set to be greater than or equal to 10 10 .

6. The mechanical screen pipe sand control simulation method based on CFD-DEM coupling according to claim 5, characterized in that, The calibration procedure for the discrete element triaxial test is as follows: (1) The effective modulus of the linear contact sub-model and the linear effective modulus of the parallel cemented sub-model are the main adjustment objects, supplemented by the adjustment of the friction coefficient and the cemented effective modulus of the parallel cemented sub-model, so as to match the macroscopic elastic modulus and Poisson's ratio of the sample. (2) Adjust the effective modulus of cementation and the cementation strength parameters of the parallel cemented sub-model to match the peak strength and residual strength of the specimen; (3) Further adjust the bonding strength parameters of the parallel bonded sub-model based on the yield failure of the specimen; (4) Based on the experimental results, the parameters of the linear contactor model are fine-tuned to balance the errors of different macroscopic mechanical parameters and improve the accuracy of the simulation results.

7. The mechanical screen pipe sand control simulation method based on CFD-DEM coupling according to claim 1, characterized in that, In step S5, the size of the fluid grid is determined based on the particle size distribution of the reservoir sand. To avoid singularities when solving the seepage equation for the steady-state incompressible fluid, the porosity of the fluid grid cells is not less than 0.

005. The expression for the seepage equation for the steady-state incompressible fluid is as follows: ； In the formula, ▽ represents the Hamiltonian operator. Let μ be the fluid velocity, K be the permeability, and μ be the velocity. f denoted as the dynamic viscosity coefficient of the fluid, n as the porosity of the fluid element, and p as the fluid pressure.

8. The mechanical screen pipe sand control simulation method based on CFD-DEM coupling according to claim 7, characterized in that, The specific operation of S5 is as follows: For the flow field, the average porosity and permeability coefficient are obtained based on the velocity and position information of the reservoir sand particles. The porosity of the fluid element is determined by the overlap between the fluid element and the reservoir sand particles, where the reservoir sand particles are characterized by cubes with length, width, and height equal to the particle size. Considering the presence of reservoir sand particles in the fluid flow, the permeability coefficient is set as a function of porosity and estimated using the Kozeny-Carman relation. The seepage equation of the steady-state incompressible fluid is discretized using the open-source solver FiPy, dividing the fluid domain into a structured grid, with each grid node controlling the volume of a certain region. The seepage equation of the steady-state incompressible fluid to be solved is integrated over the fluid control volume of each fluid grid to obtain the discrete equation. Based on discrete equations, flow field information including fluid velocity, pressure, and pressure gradient is calculated. For reservoir sand particles, the flow field information calculated by the open-source solver FiPy is applied to the centroid of the sand particles in the form of fluid-particle interaction forces. The velocity and position of the sand particles are then calculated and updated using discrete element theory, thereby completing the information exchange between the fluid phase and the solid particles, and ultimately achieving bidirectional coupled calculation of CFD-DEM. Reservoir sand particles exhibit both translational and rotational motion modes. During this motion, the sand particles follow Newton's laws of motion, and the interaction forces between the fluid and the sand particles are considered and reflected in the equations of motion, which are expressed as follows: ； In the formula, m i For the first i The mass of each reservoir sand particle For the first i The acceleration of a reservoir sand particle at time t For the action in the i The number of contacts on each reservoir sand particle For the first i The contact force of a reservoir sand particle at the k-th contact point For the first i The volume forces acting on each reservoir sand grain For the first i The damping force experienced by each reservoir sand grain, For fluid action in the first i Forces on individual reservoir sand particles; J i For the first i The moment of inertia of each reservoir sand grain For the first i The angular acceleration of a reservoir sand particle at time t For the first i The radius vector of a reservoir sand grain at the k-th contact point For the action in the i The number of cemented contacts on each reservoir sand grain For the first i The cementation moment of a reservoir sand particle at the m-th cementation contact point For the first i The damping torque experienced by each reservoir sand particle.

9. The mechanical screen tube sand control simulation method based on CFD-DEM coupling according to claim 1, characterized in that, In step S5, the CFD calculation time step is set to 100 times the DEM calculation time step.

Images