Fluid-structure interaction wave field simulation method based on distance weighted momentum exchange strategy

Abstract

The invention provides a fluid-structure interaction wave field simulation method based on a distance weighted momentum exchange strategy, and belongs to the crossing field of geophysical exploration and computational acoustics. According to the method, a distance weighting strategy is introduced into an LSM-LBM coupling algorithm, and a dynamic adaptive multi-node momentum exchange mechanism is established, so that the simulation capability of the algorithm on a complex geological structure is remarkably improved, and a new solution is provided for high-precision wave field simulation in geophysical exploration of oil and gas. Comprising the following steps: step 1), constructing a geologic model and setting physical property parameters; 2) establishing discrete forms of a solid medium lattice spring model and a fluid medium lattice Boltzmann model; step 3), designing a fluid-solid boundary momentum exchange strategy based on a distance weighting thought; step 4), carrying out iterative calculation and parameter updating on the coupled wave field; and step 5), verification and application.

Description

Technical Field

[0001] The present application proposes a fluid-structure interaction wave field simulation method based on a distance-weighted momentum exchange strategy, belonging to the cross-field of geophysical exploration and computational acoustics. Background Art

[0002] At present, with the continuous growth of global energy demand, oil and gas exploration is gradually extending to complex unconventional reservoirs. In this context, it has become particularly important to study the propagation mechanism of seismic waves in fluid-bearing reservoirs and their petrophysical responses. Traditional seismic wave field simulation technologies are based on the wave equation theory and have formed a mature method system represented by the pseudospectral method and the finite difference method after decades of development. However, in the simulation of unconventional reservoirs involving complex structures, strong discontinuities, and multi-phase medium coupling, these methods face fundamental bottlenecks. Especially for the characterization of wave field propagation characteristics in fluid-solid two-phase media, traditional methods have significant deficiencies in interface coupling processing, numerical stability, and complex geological adaptability, and breakthrough technological innovations are urgently needed.

[0003] The wave equation method based on the continuous medium hypothesis describes the wave field propagation through differential equations, and its core lies in the discretized solution of partial differential equations. Numerical methods represented by the finite difference method (FDM) perform spatial discretization through Taylor expansion and have the advantages of high computational efficiency and simple implementation in homogeneous media. However, when there are multi-phase medium interfaces, pore structures, or complex morphological structures in the geological model, the applicability of the continuous medium hypothesis is challenged. Specifically, it is manifested in: (1) The sudden change in wave impedance at the interface causes numerical oscillations in traditional discrete formats, and artificial damping or special interface conditions need to be introduced, increasing the computational complexity; (2) The fluid-solid coupling effect in porous media is difficult to accurately describe by a single wave equation, and the equivalent medium theory is usually used for approximate treatment, resulting in high-frequency signal distortion.

[0004] To break through the limitations of the continuous medium theory, researchers have proposed simulation methods based on mesoscopic discrete models. Among them, the lattice spring model (LSM) and the lattice Boltzmann model (LBM) provide new solutions from the perspectives of solid elasticity mechanics and fluid dynamics, respectively. The LSM discretizes the solid medium into a mass-spring network and simulates the propagation of elastic waves through the mechanical interaction between nodes. The LBM solves the fluid dynamics problem based on the particle collision-migration process of the discrete velocity model, and its lattice discretization characteristics are naturally suitable for multi-phase flow simulation. Scholars such as O'Brien first proposed the LSM-LBM coupling method in 2003, realizing fluid-solid interaction through a momentum exchange strategy and providing a new tool for the study of volcanic earthquake source mechanisms. Subsequent research has continuously expanded on this basis. For example, Buxton et al. improved the boundary conditions for the coupling problem of fluid-filled and elastic shells, and Adler et al. applied this method to the wave field simulation of multi-phase porous media.

[0005] However, there are still two major technical bottlenecks in the existing LSM-LBM wave field coupling method: (1) The fluid-solid boundary adopts a "one-to-one" momentum exchange mode, that is, adjacent nodes directly transfer velocity and force information. This simplified treatment is prone to the accumulation of momentum transfer errors in complex interfaces (such as rough surfaces and porous media), especially non-physical oscillations in non-orthogonal grid regions; (2) Traditional coupling strategies do not fully consider the influence of the spatial distribution of nodes on the interaction strength. When there are micro-scale structures at the interface, the local stress concentration phenomenon significantly reduces the calculation stability. The accurate momentum transfer at the fluid-solid boundary is the core of the coupling algorithm accuracy. To address the above problems, the present invention proposes a momentum exchange strategy based on the distance weighting idea (DW). This strategy breaks through the traditional point-to-point transfer mode and introduces the idea of spatial weighting to achieve "many-to-one" dynamic data interaction: for boundary nodes, their momentum exchange values are determined by the weighted average of multiple surrounding nodes, and the weight coefficient is negatively correlated with the node spacing. This improvement has twofold advantages: (1) Physical rationality: It follows the natural law of field quantity attenuation with distance and suppresses the interference of local outliers; (2) Numerical stability: It reduces the calculation oscillations caused by grid distortion through spatial smoothing.

[0006] In view of the problems of momentum transfer error in the fluid-solid coupling wave field simulation method of the prior art, non-physical oscillations caused by grid distortion, and insufficient adaptability to multi-scale media, this patent application is specifically proposed. Summary of the Invention

[0007] The fluid-solid coupling wave field simulation method and device based on the distance-weighted momentum exchange strategy of the present invention, in view of the deficiencies of the fluid-solid coupling wave field simulation method of the prior art in terms of adaptability to complex media, calculation accuracy, and stability, innovatively proposes to introduce the distance-weighting strategy into the LSM-LBM coupling algorithm, and establish a dynamic adaptive multi-node momentum exchange mechanism, in order to significantly improve the simulation ability of the algorithm for complex geological structures, thereby providing a new solution for high-precision wave field simulation in oil and gas geophysical exploration.

[0008] To achieve the above design objectives, the fluid-solid coupling wave field simulation method based on the distance-weighted momentum exchange strategy includes the following steps:

[0009] Step 1), Geological model construction and physical property parameter setting;

[0010] Step 2), Establish the discrete forms of the lattice spring model of the solid medium and the lattice Boltzmann model of the fluid medium;

[0011] Step 3), Design a fluid-solid boundary momentum exchange strategy based on the distance weighting idea;

[0012] Adopt a grid consistency strategy so that the discrete nodes of the LSM and LBM algorithms are strictly aligned in space;

[0013] Step 4), coupled wave field iterative calculation and parameter update;

[0014] Realize the momentum exchange at the fluid-solid boundary by alternately solving the dynamic equations of the LSM and LBM models and combining with the DM strategy, and achieve the dynamic evolution of the wave field and parameter synchronization;

[0015] Step 5), verification and application;

[0016] Construct a multi-dimensional verification system and a multi-scenario test framework to systematically evaluate the performance of the algorithm in terms of calculation accuracy, numerical stability, and engineering practical value.

[0017] Furthermore, in step 1), according to the geological characteristics and exploration requirements of the research area, construct a geological model containing fluid and solid media, clarify the distribution range and boundaries of the fluid and solid media, and perform grid discretization on them.

[0018] Furthermore, in step 2), adopt a discrete spring model of D2Q8 rectangular grid (8 springs);

[0019] The position coordinates of the eight nodes of the spring model of D2Q8-LSM are defined as follows:

[0020]

[0021] Among them, Δx is the length of the spring in the horizontal direction, and Δz is the length of the spring in the vertical direction;

[0022] For the D2Q8-LSM model, it has a total of three types of springs. According to the spring length, they are divided into: springs 1 and 3 in the horizontal direction, with a length of Δx; springs 2 and 4 in the vertical direction, with a length of Δz; springs 5-8 in the diagonal direction, with a length of

[0023] When the LSM performs solid wave field simulation, the spring coefficients corresponding to the above three types of springs are:

[0024]

[0025] Among them, k x is the elastic coefficient resisting horizontal tensile / compressive deformation, k z is the elastic coefficient resisting vertical tensile / compressive deformation, k diag is the elastic coefficient resisting shear deformation, σ x and σ z are Poisson's ratios, E x and E z are Young's moduli, Gxz is the shear modulus;

[0026] Add a coupled absorption boundary condition to the outermost layer of the LBM. The form of the boundary condition is as follows:

[0027]

[0028] where τ * is the modified relaxation parameter, η is the attenuation function, L is the perturbation component, and W is an auxiliary quantity defined as the time integral of L.

[0029] Furthermore, step 3) includes the following steps

[0030] Adopt a momentum exchange mode with the idea of fusion distance weighting. Perform "many-to-one" data exchange between LSM nodes and LBM nodes, and perform distance weighting on the numerical values of multiple points and assign them to the target point;

[0031] The velocity transmitted from the LSM solid node to the adjacent LBM node on the fluid-solid boundary is defined as:

[0032]

[0033] where k represents the number of LSM solid points used, and m represents the power exponent of the distance.

[0034] The collision force transmitted from the LBM fluid node to the adjacent LSM node on the fluid-solid boundary is defined as:

[0035]

[0036] where j represents the number of LBM fluid points used, and n represents the power exponent of the distance.

[0037] Furthermore, step 3) includes the following implementation strategy process

[0038] 3.1), Interface curvature calculation;

[0039] At the fluid-solid interface node, fit the local curve using the coordinates of adjacent nodes and calculate the curvature C;

[0040] For the two-dimensional interface node Q i (x i ,y i ), take the left and right adjacent nodes Q i-1 and Q i+1 , and through the three-point curvature formula:

[0041]

[0042] 3.2), Parameter adjustment;

[0043] In the high-curvature region where C > C th , increase k and j to cover more neighboring nodes to smooth the momentum transfer;

[0044] In the low-curvature region where C ≤ C th , keep k = j;

[0045] In the high-curvature region, use m = 3.2 - 4.2 and n = 2.2 - 3.2 to strengthen the distance attenuation effect and suppress local distortion; in the low-curvature region, use m = 2.1 - 3.1 and n = 1.1 - 2.1 to conform to the natural attenuation law of the physical field.

[0046] Furthermore, in step 4), the following steps are completed within each time step,

[0047] 4.1), LSM solid wave field;

[0048] Calculate the stress propagation and node motion in the solid medium, including but not limited to spring force calculation and update of the motion equation, and output the displacement field, velocity field, and stress field of the solid nodes;

[0049] 4.2), LBM fluid wave field;

[0050] Simulate the acoustic wave propagation and interface reflection / transmission effects in the fluid, including but not limited to collision steps, migration steps, and macroscopic quantity calculation, and output the density field, velocity field, and pressure field of the fluid nodes;

[0051] 4.3), "many-to-one" momentum exchange strategy at the fluid-solid boundary;

[0052] Achieve high-precision and high-stability momentum transfer through DW, including but not limited to distributing the collision force of the LBM boundary nodes to adjacent LSM solid nodes through DW and distributing the velocity of the LSM boundary nodes to adjacent LBM solid nodes through DW;

[0053] 4.4), Time step synchronization and parameter update;

[0054] LSM and LBM use the same time step to ensure data synchronization;

[0055] Update the velocity and force of the boundary nodes according to the DW distribution result;

[0056] Iteratively update the states of all nodes until the total simulation duration is reached.

[0057] Furthermore, in step 5), first, in a homogeneous medium with clear fluid-structure interaction characteristics, the finite difference method (FDM) with a small time step is used to establish a benchmark solution, and the numerical accuracy of the LSM-LBM and DW-LSM-LBM coupling algorithms is compared and analyzed. Then, for porous media with complex interface morphology, a comparative study on the numerical stability of the two algorithms in long-term simulations is carried out. Finally, the improved DW-LSM-LBM algorithm is extended and applied to complex geological media scenarios with multi-phase interfaces and non-uniform structures to verify its engineering applicability.

[0058] In summary, the main advantages of this application are as follows:

[0059] 1. Through the distance-weighted momentum exchange strategy, it can effectively solve the problem that the "one-to-one" momentum exchange in the prior art is prone to momentum transfer errors in complex interfaces (such as rough surfaces and porous media), and truly realizes the "many-to-one" dynamic data interaction, thus significantly enhancing the geometric adaptability of the fluid-structure boundary momentum exchange.

[0060] 2. For non-uniform structures such as porous media and fractures, this application suppresses local stress concentration through spatial weighted smoothing, which can effectively reduce non-physical oscillations caused by grid distortion and improve the stability of the algorithm in long-term simulations.

[0061] 3. This application is applicable to oil and gas reservoir prediction, unconventional resource exploration, and rock physics analysis, providing a new tool for high-precision wave field simulation, and particularly having potential technical advantages in the exploration of complex geological media containing fluids. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is a flow chart of the fluid-structure coupling wave field simulation method based on the distance-weighted momentum exchange strategy;

[0063] Figure 2 is a schematic diagram of the grid division structure of the fluid-solid coupling medium; the dashed line in the figure is a schematic diagram of the actual fluid-solid interface;

[0064] Figure 3 is a comparison diagram of the prior art momentum "one-to-one" exchange mechanism and the DW momentum "many-to-one" exchange mechanism proposed in this application; in the figure, the square markers represent fluid calculation nodes, and the circular markers represent solid calculation nodes; the solid line represents the LBM collision force transfer process, and the dashed line describes the LSM velocity field transfer path.

[0065] Figure 4 is a schematic diagram of the characteristics of the fluid-solid layered coupling medium structure; among them, Figure 4 (a) Horizontal layered structure diagram, with an interface dip angle of 0°; Figure 4 (b) Inclined layered structure diagram, with an interface dip angle of about 14°;

[0066] Figure 5 Under the horizontal layered structure shown in Figure 4 (a), a schematic diagram of the comparison of wave field slices extracted at 450 ms obtained by three algorithms; among them, Figure 5 (a) is the wave field slice diagram obtained by adopting the FDM algorithm; Figure 5 (b) is the wave field slice diagram obtained by the existing LSM-LBM coupling algorithm; Figure 5 (c) is the wave field slice diagram obtained by adopting the DW-LSM-LBM coupling algorithm described in the present application;

[0067] Figure 6 Based on the three algorithms shown in Figure 5 , a schematic diagram of the comparison of wave profiles calculated at different horizontal distance positions at 450 ms; among them, Figure 6 (a) is the wave profile at 550 meters; Figure 6 (b) is the wave profile at 650 meters;

[0068] Figure 7 Under the Figure 4 tilted layered structure shown in (b), a schematic diagram of the comparison of wave field slices extracted at 450 ms obtained by three algorithms; among them, Figure 7 (a) is the wave field slice diagram obtained by adopting the FDM algorithm; Figure 7 (b) is the wave field slice diagram obtained by the existing LSM-LBM coupling algorithm; Figure 7 (c) is the wave field slice diagram obtained by adopting the DW-LSM-LBM coupling algorithm described in the present application;

[0069] Figure 8 Schematic diagram of a porous medium model;

[0070] Figure 9 Schematic diagram of the comparison of wave field snapshots extracted at 1000 ms obtained by two algorithms; among them, Figure 9 (a) is the wave field diagram obtained by the FDM algorithm; Figure 9 (b) is the wave field diagram obtained by adopting the DW-LSM-LBM coupling algorithm described in the present application;

[0071] Figure 10 Schematic diagram of the comparison of wave field snapshots extracted at different times by the existing LSM-LBM algorithm; among them, Figure 10 (a) is the wave field snapshot extracted at 460 ms; Figure 10 (b) is the wave field snapshot extracted at 530 ms;

[0072] Figure 11It is a comparison diagram of wave profiles calculated at different positions at 1000 ms through the FDM algorithm and the DW-LSM-LBM coupling algorithm described in this application; among them, Figure 11 (a) is the wave profile at a depth of 550 meters; Figure 11 (b) is the wave profile at a horizontal distance of 800 meters. Specific implementation mode

[0073] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0074] Example 1, the main limitations of the existing LSM-LBM coupling algorithm are as follows:

[0075] Accumulation of interfacial momentum transfer error: Adopting the "one-to-one" momentum exchange mode is prone to cause momentum transfer error in complex interfaces (such as porous media), especially non-physical oscillations in non-orthogonal grid regions;

[0076] Insufficient numerical stability: The existing strategy does not consider the influence of the spatial distribution of nodes, and micro-scale structures (such as cracks and pores) are prone to cause local stress concentration, thereby reducing the calculation stability;

[0077] Poor adaptability to complex geology: The equivalent medium theory approximately processes the fluid-solid coupling effect, resulting in high-frequency signal distortion and making it difficult to accurately depict the wave field response at the multi-phase medium interface.

[0078] The existing distance weighting strategy still has deficiencies in the following aspects:

[0079] Weighting parameter optimization: The number of nodes (k,j) and power exponents (m,n) in DW momentum exchange need to be selected according to actual geological structure tests. The parameter sensitivity is high. If the parameter configuration is improper, it will affect the simulation accuracy and efficiency;

[0080] Calculation efficiency balance: The multi-node weighted calculation increases the calculation amount compared with the above "one-to-one" mode. It is necessary to optimize the algorithm design to ensure that the dynamic adaptive mechanism does not significantly increase the calculation complexity.

[0081] Before this application, the existing technology still has the following complexity problems for multi-physical field coupling modeling:

[0082] Coupling accuracy of boundary conditions: It is necessary to precisely control the two-way energy transfer between the LSM solid vibration velocity and the LBM fluid dynamic pressure field to ensure the physical authenticity of the reflected / transmitted waves;

[0083] Grid consistency requirements: Although the grid alignment strategy is adopted, in extreme geometric forms (such as steep dip interfaces), grid meshing and node alignment may face the contradiction between accuracy and efficiency.

[0084] In response to this, the present application proposes a fluid-solid coupling wave field simulation method based on a distance-weighted momentum exchange strategy, as Figure 1 shown, which includes the following implementation steps:

[0085] Step 1), Geological model construction and physical property parameter setting:

[0086] According to the geological characteristics and exploration requirements of the research area, construct a geological model containing fluid and solid media, clarify the distribution range and boundaries of the fluid and solid media and discretize them into grids;

[0087] Set the physical parameters of various media. For example, for fluid media, mainly set its density, longitudinal wave velocity, viscosity, etc.; for solid media, set its density, Poisson's ratio, shear modulus, Young's modulus, longitudinal wave velocity, and transverse wave velocity, etc.; to ensure that the setting of physical parameters conforms to the actual geological situation and provide accurate basic data for subsequent simulations;

[0088] Step 2), Establish the discrete forms of the lattice spring model for solid media and the lattice Boltzmann model for fluid media;

[0089] The lattice spring model (LSM) is a numerical algorithm based on mesoscopic dynamics. The core theory of its simulation of the solid seismic wave field is to discretize the continuous medium into a regular grid (such as a rectangle or square) composed of mass points (nodes) and springs (elastic elements connecting the nodes). Each mass point has a mass, and the spring represents the elasticity of the material. The wave propagation is described by the elastic force of the spring and the motion equation of the mass point. To cope with the complexity of underground media and improve the flexibility of grid division, researchers have developed various LSM variants.

[0090] In the present application, a discrete spring model with a D2Q8 rectangular grid (8 springs) is adopted to better handle these challenges. The D2Q8-LSM model is a two-dimensional model, and the nodes are connected to each other by eight linear springs. The position coordinates of the eight nodes of the spring model are defined as follows:

[0091]

[0092] where, Δx is the length of the spring in the horizontal direction, and Δz is the length of the spring in the vertical direction;

[0093] For the D2Q8-LSM model, it has a total of three types of springs, which are classified in sequence according to the spring length: spring 1 and 3 (horizontal direction, length is Δx), spring 2 and 4 (vertical direction, length is Δz), spring 5 - 8 (diagonal direction, length is );

[0094] When the LSM performs solid wave field simulation, the spring coefficients corresponding to the above three springs are:

[0095]

[0096] Among them, k x is the elastic coefficient resisting horizontal tensile / compressive deformation, and k z is the elastic coefficient resisting vertical tensile / compressive deformation, and k diag is the elastic coefficient resisting shear deformation, and σ x and σ z are Poisson's ratios, and E x and E z are Young's moduli, and G xz is the shear modulus;

[0097] In the lattice spring model, the linear spring force and the angular spring force are two types of key mechanical components, which are respectively used to simulate the axial deformation and shear / bending deformation of materials. The linear spring connects adjacent nodes and simulates the axial elastic response (tension or compression) of materials, and dominates the propagation of longitudinal waves (P-waves). The force-displacement relationship follows Hooke's law.

[0098] The expression of the linear spring force F lin is as follows:

[0099]

[0100] Among them, k lin is the linear spring coefficient, r ij is the relative position vector of nodes i and j after deformation, is the original length of the spring;

[0101] The angular spring connects non-adjacent nodes (such as diagonal nodes) or is introduced through the rotational degree of freedom, simulates the shear stiffness and bending stiffness of materials, and dominates the propagation of transverse waves (S-waves); the expression of the angular spring force F ang is:

[0102]

[0103] Among them, k ang is the angular spring coefficient, is the original length of the diagonal spring. Therefore, the total spring force F all = F lin + F ang ;

[0104] The LSM model updates the displacement u, velocity v s and acceleration a by combining the following velocity Verlet algorithm:

[0105]

[0106] where \(m = \rho\Delta x\Delta z\) is the mass of the particle, \(\gamma\) is the viscoelastic coefficient controlling attenuation, and \(\Delta t\) is the time step, which needs to satisfy the Courant - Friedrichs - Lewy (CFL) condition:

[0107]

[0108] where \(l\) min is the minimum spring length, and \(V\) p is the longitudinal wave propagation speed;

[0109] The lattice Boltzmann model (LBM) is another numerical method based on mesoscopic particle dynamics. It describes the macroscopic motion of fluids, including wave phenomena (such as surface waves, longitudinal waves, etc.), by simulating the evolution of the particle distribution function in the discrete velocity space.

[0110] The core of LBM is to discretize the continuous particle velocity space into a finite number of directions. Taking the two - dimensional model D2Q9 as an example (9 discrete velocity directions), the discrete velocity directions \(e\) i are

[0111]

[0112] The weight coefficients \(w\) i and the lattice sound speed \(c\) s are

[0113]

[0114] where \(\Delta x\) and \(\Delta t\) are the spatial and time sampling steps of LBM respectively.

[0115] The LBM equation does not track individual molecules, but uses the core quantity \(f\) i (\mathbf{x},t)\) to statistically describe the collective behavior of the particle swarm. Specifically, \(f\) i (\mathbf{x},t)\) is the particle distribution function, which represents the number density of particles moving along the discrete velocity direction \(\mathbf{e}\) i at position \(\mathbf{x}\) and time \(t\).

[0116] The evolution of the particle distribution function is mainly divided into the following two steps: collision and migration:

[0117] The collision process (from relaxation to equilibrium state) is:

[0118]

[0119] where \(\tau\) is the relaxation time, which is related to the kinematic viscosity of the fluid:

[0120] The migration process (along the discrete velocity direction) is:

[0121] \(f\) i(x + e i Δt, t + Δt) = f i post (x, t) (10)

[0122] When simulating wave phenomena, the reflection and transmission effects of waves need to be considered at the interface of different media.

[0123] The corrected migration process is as follows:

[0124]

[0125] Among them, R and T are the reflection coefficient and transmission coefficient respectively, and both are related to the wave impedance of the media on both sides of the interface.

[0126] The equilibrium distribution function f i eq is the core component of the lattice Boltzmann method (LBM). It reduces the macroscopic hydrodynamic equations (such as the Navier - Stokes equation or wave equation) through the statistical characteristics of the microscopic particle distribution. Its design needs to satisfy the conservation of mass and momentum and reflect the constitutive relationship of the fluid.

[0127] The equilibrium distribution function f i eq The expression of it is:

[0128]

[0129] Among them, the macroscopic density ρ, velocity v f and pressure p of the fluid particles can be calculated from the mesoscopic particle distribution function f i as follows:

[0130]

[0131] To solve the problem of truncated boundary reflection in fluid wave field simulation, this application proposes to add a coupled absorption boundary condition to the outermost layer of LBM. This boundary condition can effectively reduce unnecessary reflections and improve the simulation accuracy. Its basic form is as follows:

[0132]

[0133] Among them, τ * is the modified relaxation parameter, η is the attenuation function, L is the perturbation component, and W is an auxiliary quantity defined as the time integral of L;

[0134] Step 3): Design a fluid - solid boundary momentum exchange strategy based on the idea of distance weighting;

[0135] The core problem of the LSM - LBM coupled system lies in the precise control of the seismic wave energy transfer mechanism at the fluid - solid interface.

[0136] This application adopts a grid consistency strategy to improve the calculation efficiency, and the discrete nodes of the LSM and LBM algorithms are strictly aligned in space; as Figure 2 shown, the interface transition region is set as the overlapping region of the LSM solid unit and the LBM fluid unit, and two-way dynamic coupling is established to achieve energy interaction: on the one hand, the vibration velocity v of the LSM solid s is input as a boundary condition into the adjacent LBM fluid unit; on the other hand, the dynamic pressure field F generated by the LBM fluid unit during the collision process coll , through the distribution function reconstruction, reacts on the LSM solid boundary to form a complete energy closed-loop transmission path. This two-way coupling mechanism not only maintains the physical authenticity of wave propagation but also significantly reduces the numerical error of cross-media simulation.

[0137] Generally, the momentum exchange at the fluid-solid boundary is based on the "one-to-one" mode, that is, the value of one point is directly assigned to another point. Although this mode is simple and easy to implement, for complex media, the existence of certain extreme values on the boundary will obviously cause concerns about the simulation accuracy and robustness.

[0138] Therefore, this application adopts a momentum exchange mode that integrates the distance weighting idea (DW), as Figure 3 shown, to achieve the "many-to-one" data exchange between LSM nodes and LBM nodes, that is, the numerical values of multiple points are distance-weighted and assigned to the target point, so as to achieve data smoothing and reduce errors. The DW momentum exchange can ensure that nodes farther away contribute less to the data value, which conforms to the actual physical laws.

[0139] Based on the DW "many-to-one" mode, the velocity v transmitted from the LSM solid node to the adjacent LBM node on the fluid-solid boundary * s is defined as:

[0140]

[0141] where k represents the number of LSM solids used, and m represents the power exponent of the distance.

[0142] Similarly, the collision force transmitted from the LBM fluid node to the adjacent LSM node on the fluid-solid boundary is defined as

[0143]

[0144] where j represents the number of LBM fluids used, and n represents the power exponent of the distance;

[0145] while the F in the prior art "one-to-one" mode coll is obtained by the collision effect of LBM:

[0146]

[0147] For the traditional "one-to-one" momentum exchange mode, k = 1, j = 1, m = 0, n = 0.

[0148] The DW-based "many-to-one" momentum exchange mode adopted in this application is an adaptive parameter adjustment mechanism based on the interface morphology. By real-time monitoring of the geometric features, numerical stability, or error feedback of the fluid-solid interface, the node numbers k, j and the power exponents m, n are dynamically adjusted to optimize the accuracy of momentum transfer and the computational efficiency.

[0149] The following is the process of the specific implementation strategy:

[0150] 3.1), Interface curvature calculation;

[0151] At the fluid-solid interface nodes, the local curve is fitted using the coordinates of adjacent nodes to calculate the curvature C. For example, for the two-dimensional interface node Q i (x i , y i ), take the left and right adjacent nodes Q i-1 and Q i+1 , through the three-point curvature formula:

[0152]

[0153] It can be seen that the higher the curvature, the more complex the interface (such as porous and fracture regions).

[0154] 3.2), Parameter adjustment;

[0155] In the high-curvature region C > C th (for example, C th = 0.1 m -1 ), increase k and j (such as from k = 3 to k = 6, j = 3 to j = 5) to cover more adjacent nodes to smooth the momentum transfer;

[0156] In the low-curvature region C ≤ C th , keep k = 3, j = 3 to reduce the computational overhead.

[0157] In the high-curvature region, use m = 3.2 - 4.2, n = 2.2 - 3.2 to strengthen the distance decay effect and suppress local distortion; in the low-curvature region, use m = 2.1 - 3.1, n = 1.1 - 2.1 to conform to the natural decay law of the physical field.

[0158] Step 4), Coupled wave field iterative calculation and parameter update;

[0159] By alternately solving the kinetic equations of the LSM (solid) and LBM (fluid) models and combining the DM strategy to achieve momentum exchange at the fluid-solid boundary, the dynamic evolution of the wave field and parameter synchronization are realized;

[0160] The following steps need to be completed in each time step:

[0161] 4.1), LSM solid wave field;

[0162] Calculate the stress propagation and node motion in the solid medium, including but not limited to spring force calculation and update of the motion equation, and output the displacement field, velocity field and stress field of the solid nodes;

[0163] 4.2), LBM fluid wave field;

[0164] Simulate the acoustic wave propagation and interface reflection / transmission effects in the fluid, including but not limited to collision steps, migration steps and macroscopic quantity calculation, and output the density field, velocity field and pressure field of the fluid nodes;

[0165] 4.3), "many-to-one" momentum exchange strategy at the fluid-solid boundary;

[0166] Achieve high-precision and high-stability momentum transfer through DW, including but not limited to distributing the collision force of the LBM boundary nodes to adjacent LSM solid nodes through DW and distributing the velocity of the LSM boundary nodes to adjacent LBM solid nodes through DW;

[0167] 4.4), Time step synchronization and parameter update;

[0168] LSM and LBM adopt the same time step to ensure data synchronization;

[0169] Update the velocity and force of the boundary nodes according to the DW distribution result;

[0170] Iteratively update the states of all nodes until the total simulation duration is reached;

[0171] Step 5), Verification and application;

[0172] Construct a multi-dimensional verification system and a multi-scenario test framework to systematically evaluate the performance of the algorithm in terms of calculation accuracy, numerical stability and engineering practical value;

[0173] First, in a homogeneous medium with clear fluid-solid coupling characteristics, a benchmark solution is established using the finite difference method (FDM) with a small time step, and the numerical accuracy of the LSM-LBM and DW-LSM-LBM coupling algorithms is compared and analyzed;

[0174] Then, for porous media with complex interface morphology, a comparative study on the numerical stability of the two algorithms in long-time simulation is carried out;

[0175] Finally, the improved DW-LSM-LBM algorithm is extended and applied to complex geological medium scenarios with multi-phase interfaces and non-uniform structures to verify its engineering applicability.

[0176] In terms of the layered medium example, to verify the effectiveness of the fluid-solid coupling wave field simulation method based on the distance-weighted momentum exchange strategy described in this application, it can be combined with the Figure 4 double-layer horizontal medium benchmark model shown in the figure. This model uses a unified grid system to describe the fluid and solid regions. The computational grid scale is 1500×1500, the spatial sampling interval is 1m×1m, the time sampling step is 0.5ms, and the total duration of the wave field record is 600ms.

[0177] The above model parameters are set as follows: the longitudinal wave velocity of the overlying fluid layer is 1500m / s, and the density is 1200kg / m 3 ; the longitudinal wave velocity of the underlying solid layer is 3000m / s, the shear wave velocity is 2000m / s, and the density is 2400kg / m 3 . The source is placed at the center point of the model (750m, 750m), and a Ricker wavelet with a main frequency of 30Hz is used for excitation. The number of nodes participating in the calculation at the fluid-solid boundary is k = 3, j = 3, and the power exponents are m = 2.4 and n = 1.6.

[0178] To verify the algorithm accuracy, FDM with a refined time sampling step of 0.25ms is selected as the reference solution, and the other parameters are kept consistent with the DW-LSM-LBM algorithm.

[0179] As Figure 5 shown, at the 450ms moment, the wave field slices generated by FDM, LSM-LBM, and the DW-LSM-LBM algorithm described in this application are compared. The three groups of results show a high degree of consistency in the waveform main frequency characteristics and polarity distribution, but there are slight differences in the energy distribution of the internal reflected waves at the fluid-solid interface (the area marked by the red arrow). Quantitative analysis shows that the wave field difference between DW-LSM-LBM and FDM is the smallest, showing better in-phase axis continuity and better energy balance, while the deviation of the traditional LSM-LBM from the reference solution is relatively significant. In the figure, the shear wave (S), longitudinal wave (P), and their derived transmitted waves, reflected waves, and converted waves are also marked, which enhances the recognizability of the complex wave system. To further quantify the algorithm differences, the wave profiles at the 550m and 650m survey line positions are extracted from Figure 5 , as Figure 6 shown. The curve fitting degree of the direct wave component (the edge region of the waveform) is relatively high (especially see Figure 6 b), while there are amplitude and phase differences in the interface reflected wave component (the central region of the waveform). It should be noted that the calculation result of DW-LSM-LBM is closer to the reference solution ( Figure 6(a), the difference from the reference solution stems from the different treatment strategies of the fluid-structure interface coupling mechanism by the two algorithms, which is in line with the theoretical expectation. This comparison verifies the effectiveness of the DW-LSM-LBM coupling algorithm described in this application in the modeling of double-layer horizontal fluid-structure media and confirms that its simulation accuracy has been significantly improved compared with traditional methods.

[0180] To further study the applicability of the fluid-structure coupled wave field simulation method based on the distance-weighted momentum exchange strategy described in this application, a tilted layered geological model as shown in Figure 4 (b) was constructed. Except for the adjustment of the fluid-solid interface morphology, the other parameters are the same as those in Figure 4 (a).

[0181] As shown in Figure 7 , snapshots of the wave field at 450 ms calculated by FDM and two LSM-LBM numerical methods are presented. Comparative analysis shows that the three methods are highly consistent in terms of waveform phase and amplitude characteristics. It is worth noting that a detailed observation of the reflected waves at the fluid-solid interface shows that Figure 7 (a) and Figure 7 (c) have better in-phase axis continuity. These results indicate that while maintaining equivalent computational efficiency in tilted layered structures, the DW-LSM-LBM coupling method exhibits better numerical accuracy compared to the traditional LSM-LBM method.

[0182] In the case of porous media, to verify the effectiveness of the fluid-structure coupled wave field simulation method based on the distance-weighted momentum exchange strategy described in this application, the DW-LSM-LBM coupling algorithm can be extended and applied to the three-dimensional porous media model as shown in Figure 8 . The model is constructed with a computational domain of 2000 m × 2000 m, and the spatial sampling interval is 1 m × 1 m. The background medium is a layered solid matrix (interface burial depth 1300 m), and three fluid-filled cylindrical holes (radius of each is 50 m) are embedded inside. The physical property parameters of the fluid and solid are shown in Table 1 below.

[0183] Table 1. Physical property parameters of porous media

[0184]

[0185] The seismic source uses a Ricker wavelet with a main frequency of 40 Hz, which is located at the coordinate point (1000 m, 850 m) in the solid matrix area. The number of nodes k = 6 and j = 5 participating in the calculation at the fluid-solid boundary, and the power exponents m = 3.5 and n = 2.7.

[0186] For comparative verification, the FDM solver uses a smaller time step to ensure the accuracy of the reference solution.

[0187] As shown in Figure 9As shown, it presents the wave field snapshots calculated by the FDM algorithm and the DW-LSM-LBM coupling algorithm at the same time step. Comparative analysis shows that the two methods show a high degree of consistency in the phase and amplitude characteristics of the main wave components such as shear waves, longitudinal waves and reflected waves, but there are differences in the wavefront morphology and energy distribution of some transmitted / reflected waves at the fluid-solid interface (as indicated by the blue markings).

[0188] As Figure 10 shown, it presents the wave field snapshots of the LSM-LBM algorithm at 460 ms and 530 ms. It can be seen that abnormal fluctuations have occurred in the LSM-LBM algorithm in the complex geological model (especially the fluid-solid coupling interface), which verifies the improvement of the DW-LSM-LBM algorithm described in this application in terms of calculation stability.

[0189] As Figure 11 shown, it presents the wave profiles extracted along Figure 9 the depth and horizontal directions. Figure 11 (a) The blue marked area shows the wave characteristics passing through the fluid region, and significant differences occur between the two algorithms here, which is due to the different influencing mechanisms of different fluid-solid coupling treatment methods. Figure 11 (b) The profile shown mainly passes through the solid region, and the calculation results of the two algorithms show good agreement. The results of the numerical example show that the DW-LSM-LBM coupling algorithm is still applicable in the porous medium model. Even when different fluid-solid coupling strategies are adopted, it can accurately and stably depict the characteristics of the seismic wave field, further verifying the stability and universality of this method in depicting the wave field in complex media.

[0190] As mentioned above, the fluid-solid coupling wave field simulation method based on the distance-weighted momentum exchange strategy described in this application can significantly improve the geometric adaptability of the momentum exchange at the fluid-solid boundary, accurately depict the wave field responses of inclined, curved and porous interfaces; suppress the abnormal stress concentration caused by local grid distortion, and enhance the calculation stability of the algorithm in heterogeneous media such as fractures and pores; establish a dynamic adaptive multi-node coupling mechanism, reduce the computational complexity of complex model simulation, and meet the engineering requirements of large-scale geological models.

[0191] It shows obvious synergistic advantages in many aspects such as multi-physical field coupling modeling, multi-scenario applicability, calculation accuracy control, and calculation stability. Its innovative fluid-solid coupling solution strategy provides a high-confidence numerical experiment platform for the simulation of wave field evolution with complex interfaces, and has important application value in engineering fields such as seabed resource exploration and reservoir fracture detection.

[0192] As mentioned above, combined with the solution content given in the accompanying drawings and descriptions, similar technical solutions can be derived, which still fall within the scope of the claims of the technical solutions of the present invention.

Claims

1. A fluid-structure interaction wave field simulation method based on a distance-weighted momentum exchange strategy, characterized in that: It includes the following implementation steps: Step 1), Geological model construction and physical property parameter setting; Step 2), Establish the discrete forms of the lattice spring model for solid media and the lattice Boltzmann model for fluid media; Step 3), Design a fluid-solid boundary momentum exchange strategy based on the idea of distance weighting; Adopt a grid consistency strategy, and the discrete nodes of the two algorithms of LSM and LBM are strictly aligned in space; Step 4), Coupled wave field iterative calculation and parameter update; Realize the momentum exchange at the fluid-solid boundary by alternately solving the dynamic equations of the LSM and LBM models and combining the DM strategy, and realize the dynamic evolution of the wave field and parameter synchronization; Step 5), Verification and application; Construct a multi-dimensional verification system and a multi-scenario test framework to systematically evaluate the performance of the algorithm in terms of calculation accuracy, numerical stability and engineering practical value.

2. The fluid-structure interaction wave field simulation method based on the distance-weighted momentum exchange strategy according to claim 1, characterized in that: In the said step 1), according to the geological characteristics and exploration requirements of the research area, construct a geological model including fluid and solid media, clarify the distribution range and boundary of the fluid and solid media, and discretize them into grids.

3. The fluid-structure interaction wave field simulation method based on the distance-weighted momentum exchange strategy according to claim 1, characterized in that: In the said step 2), adopt a discrete spring model of D2Q8 rectangular grid (8 springs); The position coordinates of the eight nodes of the spring model of D2Q8-LSM are defined as follows: Among them, Δx is the length of the spring in the horizontal direction, and Δz is the length of the spring in the vertical direction; For the D2Q8-LSM model, there are three types of springs in total. According to the spring length, they are divided into: springs 1 and 3 in the horizontal direction, with a length of Δx; springs 2 and 4 in the vertical direction, with a length of Δz; springs 5-8 in the diagonal direction, with a length of When the LSM is used to simulate the solid wave field, the spring coefficients corresponding to the above three springs are: Among them, k x is the elastic coefficient for resisting horizontal tensile / compressive deformation, k z is the elastic coefficient for resisting vertical tensile / compressive deformation, k diag is the elastic coefficient for resisting shear deformation, σ x and σ z are Poisson's ratios, E x and E z are Young's moduli, G xz is the shear modulus; Add a coupled absorption boundary condition to the outermost layer of the LBM, and the form of the boundary condition is as follows: where τ * is the modified relaxation parameter, η is the decay function, L is the perturbation component, and W is an auxiliary quantity defined as the time integral of L.

4. The fluid-structure interaction wave field simulation method based on the distance-weighted momentum exchange strategy according to claim 1, characterized in that: The said step 3) includes the following steps: Adopt a momentum exchange mode integrating the idea of distance weighting, and perform "many-to-one" data exchange between the LSM nodes and the LBM nodes, and perform distance weighting on the values of multiple points and assign them to the target point; The velocity transferred from the LSM solid nodes to the adjacent LBM nodes at the fluid-solid boundary is defined as: Among them, k represents the number of LSM solid points used, and m represents the power exponent of the distance. The collision force transmitted from the LBM fluid nodes to the adjacent LSM nodes on the fluid-solid boundary is defined as: Among them, j represents the number of LBM fluid points used, and n represents the power exponent of the distance.

5. The fluid-structure interaction wave field simulation method based on the distance-weighted momentum exchange strategy according to claim 4, characterized in that: The said step 3) includes the following implementation strategy process: 3.1), Interface curvature calculation; At the fluid-solid interface nodes, fit the local curve using the coordinates of adjacent nodes and calculate the curvature C; For the two-dimensional interface node Q i (x i , y i ), take the adjacent left and right nodes Qi-1 and Qi +1 , through the three-point curvature formula: 3.2), Parameter adjustment; In the high-curvature region C > C th , increase k and j to cover more neighboring nodes to smooth momentum transfer; In the low curvature region C ≤ C th , keep k = j; In the high-curvature region, m = 3.2 - 4.2 and n = 2.2 - 3.2 are adopted to strengthen the distance attenuation effect and suppress local distortion; in the low-curvature region, m = 2.1 - 3.1 and n = 1.1 - 2.1 conform to the natural attenuation law of the physical field.

6. The fluid-structure interaction wave field simulation method based on the distance-weighted momentum exchange strategy according to claim 1, characterized in that: In the said step 4), the following steps are completed within each time step: 4.1), LSM solid wave field; Calculate the stress propagation and node motion in the solid medium, including but not limited to spring force calculation and update of the motion equation, and output the displacement field, velocity field and stress field of the solid nodes; 4.2), LBM fluid wave field; Simulate the acoustic wave propagation and interface reflection / transmission effects in the fluid, including but not limited to collision steps, migration steps and macroscopic quantity calculation, and output the density field, velocity field and pressure field of the fluid nodes; 4.3), Fluid-solid boundary "many-to-one" momentum exchange strategy; Achieve high-precision and high-stability momentum transfer through DW, including but not limited to distributing the collision force of LBM boundary nodes to adjacent LSM solid nodes through DW and distributing the velocity of LSM boundary nodes to adjacent LBM solid nodes through DW; 4.4), Time step synchronization and parameter update; LSM and LBM use the same time step to ensure data synchronization; Update the velocity and force of boundary nodes according to the DW distribution result; Iteratively update the states of all nodes until the total simulation duration is reached.

7. The fluid-structure interaction wave field simulation method based on the distance-weighted momentum exchange strategy according to claim 1, wherein: For step 5) described above, first, in a homogeneous medium with clear fluid-structure interaction characteristics, establish a benchmark solution using the finite difference method (FDM) with a small time step, and compare and analyze the numerical accuracy of the LSM-LBM and DW-LSM-LBM coupling algorithms; then, for porous media with complex interface morphologies, conduct a comparative study on the numerical stability of the two algorithms in long-term simulations; finally, apply the improved DW-LSM-LBM algorithm to complex geological media scenarios with multi-phase interfaces and non-uniform structures to verify its engineering applicability.

Images