Method, platform, cloud equipment and computer device for numerical simulation of seismic wave propagation

By using a combination of viscous boundary and free field boundary in SPH numerical simulation, the open waves are absorbed and the seismic load is effectively input, which solves the problem of incomplete absorption of external travel waves and inconsistent calculation of side boundary attributes in traditional SPH methods, and improves the seismic wave propagation simulation accuracy and reliability.

CN118536302BActive Publication Date: 2024-11-08SOUTHWEST JIAOTONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410673301.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-28
Publication Date
2024-11-08
Estimated Expiration
2044-05-28

AI Technical Summary

Technical Problem

In the seismic wave propagation simulation of traditional SPH methods, the calculation of the velocity of the virtual particles does not meet the complete slip-free boundary conditions, resulting in the inability to fully absorb the outside waves, affecting the numerical simulation accuracy, and the calculation of the physical properties of the side boundary particles does not meet the actual situation.

Method used

By constructing the SPH numerical test model, using a combination of viscous boundaries and free field boundaries, the equivalent force of the truncated boundaries when simulating seismic wave propagation is converted into the load of the bottom and lateral boundary particles of the model, and the damper is used to absorb the lay wave. And directly apply load through the momentum equation of SPH to achieve effective input and absorption of seismic waves.

Benefits of technology

The accuracy and reliability of numerical simulation of seismic wave propagation is improved, making the propagation of seismic waves in the SPH model more in line with the actual situation and improving the accuracy of the simulation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118536302B_ABST
    Figure CN118536302B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of numerical simulation of earthquake motion, and specifically to a method, platform, cloud equipment and computer device for numerical simulation of seismic wave propagation. The scheme includes constructing an SPH numerical test model, drawing corresponding geometric block graphics based on the size of the study area, arranging particles with an equivalent diameter of D according to a regular interval distance D until the entire study area is covered, the particles in the study area are real particles, the outermost particles of the truncation boundary are boundary particles, the boundary particles are 0.5D away from the truncation boundary, and continue to expand at least three layers of particles outside the study area along the outer normal direction of the truncation boundary, which are fixed virtual particles, and finally allocating physical parameters to all particles according to material properties, selecting the corresponding kernel function W, setting the basic parameters of the SPH numerical model, and taking D as the smooth length of the particles to construct a complete SPH numerical model; calculating the equivalent force of the boundary particles based on the SPH numerical model; and inputting external loads. The present invention is suitable for numerical simulation experiments of seismic wave propagation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of numerical simulation of earthquake motion, and in particular to a method, platform, cloud equipment and computer device for numerical simulation of earthquake wave propagation. Background Art

[0002] In the field of earthquake engineering, numerical simulation technology is highly favored as a powerful tool for understanding the large deformation and motion characteristics of rock and soil, and is crucial for deepening the understanding of soil damage and catastrophic evolution processes induced by complex earthquakes. SPH (Smoothed Particle Hydrodynamics), as a meshless numerical analysis method, uses the Naiver-Stokes (NS) equations to manage the motion of system particles. It is suitable for describing the motion of fluid-like materials, such as Newtonian or non-Newtonian fluids and some special soils. In recent years, the SPH method has been redeveloped by many scholars and expanded to seismic research fields, such as numerical simulation of soil shaking table tests and numerical simulation of soil slope landslides under earthquake action.

[0003] The traditional SPH method simulates seismic wave propagation by allocating virtual seismic wave velocities to virtual particles at the bottom boundary, establishing the velocity drop of particles inside and outside the boundary, and then realizing the input of seismic loads through the mass control equation. In this process, the velocity calculation of virtual particles is based on the seismic wave velocity at the bottom boundary of the model. When the seismic wave is input into the traditional SPH model, the velocity of the particles at the bottom boundary is actually "not equal to 0", which no longer satisfies the "no slip" boundary condition of the SPH model in a complete sense. Therefore, when the seismic wave reflected by the surface returns to the incident boundary again, the wave velocity cannot be completely "decelerated" to "0", that is, the outgoing wave cannot be completely absorbed, resulting in part of the outgoing wave being reflected again at the bottom boundary and re-entering the model, thereby affecting the accuracy of the numerical simulation. In addition, when the traditional SPH model simulates seismic wave propagation, the calculation of the physical properties (velocity, displacement, etc.) of the side boundary particles is usually based on the "slip" boundary condition. When the outgoing wave in the model approaches the truncation boundary, due to the reaction force of the virtual particles at the boundary, the normal velocity tends to "0", while the tangential velocity remains unchanged, and its propagation process does not conform to the actual situation. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method, platform, cloud equipment and computer device for numerical simulation of seismic wave propagation, thereby improving the accuracy of simulation calculations.

[0005] The present invention adopts the following technical solutions to achieve the above-mentioned purpose. In a first aspect, the present invention provides a method for numerical simulation of seismic wave propagation, comprising:

[0006] S1. Construct SPH numerical test model;

[0007] Draw the corresponding geometric block graphics based on the size of the research area, and arrange the particles with an equivalent diameter of D according to the interval distance D until they cover the entire research area. The particles in the research area are real particles, and the outermost particles of the truncation boundary are boundary particles, which are 0.5D away from the truncation boundary. Then, along the outward normal direction of the truncation boundary, continue to expand at least three layers of particles to the outside of the research area, which are fixed virtual particles. Finally, assign physical parameters to all particles according to material properties. Virtual particles have the same physical properties as real particles. Select the corresponding kernel function W, set the basic parameters of SPH, and use D as the smooth length of the particles to construct a complete SPH numerical test model.

[0008] S2. According to the boundary type and boundary conditions of the SPH numerical model, the equivalent force on the truncation boundary when simulating the propagation of seismic waves is converted into two concentrated forces: the boundary particle load at the bottom of the model and the lateral boundary particle load;

[0009] S3, the external load input is directly applied to the boundary particles of the numerical model in the form of concentrated load through the momentum equation of the SPH method.

[0010] The momentum equation is:

[0011] Among them, ρ represents the density of the particle; v represents the velocity of the particle; p is the pressure on the particle; Π ij is the artificial viscosity term;

[0012] For the bottom boundary and lateral boundary, the boundary particle momentum calculation equations when the seismic wave is input are:

[0013] Bottom border: F b It represents the earthquake load at the bottom of the model, that is, the force on any bottom boundary particle;

[0014] Side border: F l It represents the load generated by the lateral free field motion of the model, that is, the force acting on any lateral boundary particles.

[0015] Furthermore, the equivalent force on the truncation boundary during the simulation of seismic wave propagation is converted into the model bottom boundary load, specifically including:

[0016] The bottom adopts a viscous boundary, and the equivalent load acting on the boundary particles is applied by arranging on the bottom boundary particles, and the seismic reflection waves from the outside of the study area are absorbed. For any viscous boundary particles, the normal force and tangential force provided by the damper are expressed as:

[0017] σ=-ρc p v n

[0018] τ=-ρc s v s

[0019] where σ and τ represent the normal and tangential forces generated by the damper, respectively; ρ represents the density of the bulk medium; v n and v s They represent the moving speed of the action point on the boundary block in the normal and tangential directions respectively; c p and c s are the propagation speeds of seismic P waves and S waves in the medium, respectively, both of which are related to the physical and mechanical properties of the medium and can be expressed as:

[0020]

[0021]

[0022] Where μ represents the Poisson's ratio of the material; E is Young's modulus;

[0023] For any bottom boundary particle b(x,y), the force F b It is expressed as:

[0024] F b =F b1 +F b2 , F b1 represents the damping force of the viscous boundary on the boundary particles, F b2 It represents the equivalent load of earthquake motion acting on the boundary particles;

[0025] For the bottom boundary particles, the damping force F b1 for:

[0026]

[0027] Among them, F b1n and F b1s are the normal and tangential damping forces acting on the bottom boundary particles respectively; α is the direction angle of the model cutoff boundary; v bx and v by Respectively represent the moving speed of the particle in the x direction and y direction;

[0028] The equivalent seismic load F that the bottom boundary particles are ultimately subjected to is b2 It is expressed as:

[0029]

[0030] Among them, F b2n and F b2s They represent the normal stress and shear stress on the particles respectively; v sx and v syThey represent the seismic wave velocities in the x and y directions acting on the boundary particles respectively.

[0031] Furthermore, the equivalent force on the truncation boundary during the simulation of seismic wave propagation is converted into the model lateral boundary load, specifically including:

[0032] For any side boundary particle l(x,y), the total force F l It mainly includes the damping force F on the boundary particles l1 and the free field motion generates a load F on the particle l2 , expressed as: F l =F l1 +F l2 ;

[0033]

[0034] Among them, F l1n and F l1s are the normal and tangential damping forces acting on the lateral boundary particles, respectively; v lx and v ly They represent the moving speed of the particle in the x and y directions respectively; v fx and v fy They represent the moving speed of the boundary particle in the x direction and the y direction in the free field, respectively, and are obtained by the seismic wave delay input method; F l2n and F l2s They are normal and shear loads on the lateral boundary particles caused by free field motion.

[0035] Furthermore, the method further includes S4, virtual particle attribute calculation; specifically, when the jth virtual particle falls within the support domain of the ith real particle, the speed v of the virtual particle j Expressed as: v j =(1-β)v i , When the virtual particle is outside the real particle support domain, there is no need to calculate the velocity; when the virtual particle falls within the real particle support domain, the stress levels of the real and virtual particles are considered to be equal, that is, σ j =σ i ; otherwise, no calculation is required.

[0036] In a second aspect, the present invention provides a seismic wave propagation numerical simulation platform for implementing the above-mentioned seismic wave propagation numerical simulation method, the platform comprising:

[0037] The model building module is used to build the SPH numerical test model. The corresponding geometric block graphics are drawn based on the size of the research area. Particles with an equivalent diameter of D are arranged according to the interval distance D until they cover the entire research area. The particles in the research area are real particles, and the outermost particles of the truncation boundary are boundary particles. The boundary particles are 0.5D away from the truncation boundary. Then, along the outward normal direction of the truncation boundary, at least three layers of particles are continued to expand to the outside of the research area, which are fixed virtual particles. Finally, physical parameters are assigned to all particles according to material properties. Virtual particles have the same physical properties as real particles. The corresponding kernel function W is selected, the basic parameters of the SPH model are set, and D is the smooth length of the particles. In this way, a complete SPH numerical test model is constructed;

[0038] The equivalent force conversion module is used to convert the equivalent force on the truncation boundary when simulating seismic wave propagation into two concentrated forces as loads on the bottom boundary particles and the lateral boundary particles of the model according to the boundary type and boundary conditions of the SPH model;

[0039] The external load input module is used to apply external load input in the form of concentrated load directly to the model boundary particles through the momentum equation of SPH. The momentum equation is:

[0040] Among them, ρ represents the density of the particle; v represents the velocity of the particle; p is the pressure on the particle; Π ij is the artificial viscosity term;

[0041] For the bottom boundary and lateral boundary, the boundary particle momentum calculation equations when the seismic wave is input are:

[0042] Bottom border: F b It represents the earthquake load at the bottom of the model, that is, the force on any bottom boundary particle;

[0043] Side border: F l It represents the load generated by the lateral free field motion of the model, that is, the force acting on any lateral boundary particles.

[0044] Furthermore, the equivalent force conversion module is specifically used to apply equivalent loads on boundary particles by arranging them on the bottom boundary particles, and to absorb seismic reflection waves from outside the study area. For any viscous boundary particles, the normal force and tangential force provided by the damper are expressed as:

[0045] σ=-ρc p v n

[0046] τ=-ρc s v s

[0047] where σ and τ represent the normal and tangential forces generated by the damper, respectively; ρ represents the density of the bulk medium; v n and v s They represent the moving speed of the action point on the boundary block in the normal and tangential directions respectively; c p and c s are the propagation speeds of seismic P waves and S waves in the medium, respectively, both of which are related to the physical and mechanical properties of the medium and can be expressed as:

[0048]

[0049] Where μ represents the Poisson's ratio of the material; E is Young's modulus;

[0050] For any bottom boundary particle b(x,y), the force F b It is expressed as:

[0051] F b =F b1 +F b2 , F b1 represents the damping force of the viscous boundary on the boundary particles, F b2 It represents the equivalent load of earthquake motion acting on the boundary particles;

[0052] For the bottom boundary particles, the damping force F b1 for:

[0053]

[0054] Among them, F b1n and F b1s are the normal and tangential damping forces acting on the bottom boundary particles respectively; α is the direction angle of the model cutoff boundary; v bx and v by Respectively represent the moving speed of the particle in the x direction and y direction;

[0055] The equivalent seismic load F that the bottom boundary particles are ultimately subjected to is b2 It is expressed as:

[0056]

[0057] Among them, F b2n and F b2s They represent the normal stress and shear stress on the particles respectively; v sx and v sy They represent the seismic wave velocities in the x and y directions acting on the boundary particles respectively.

[0058] Furthermore, the equivalent force conversion module is specifically used to convert the total force F to any side boundary particle l(x,y) l It mainly includes the damping force F on the boundary particles l1 and the free field motion generates a load F on the particle l2 , expressed as: F l =F l1 +F l2 ;

[0059]

[0060] Among them, F l1n and F l1s are the normal and tangential damping forces acting on the lateral boundary particles, respectively; v lx and v ly They represent the moving speed of the particle in the x and y directions respectively; v fx and v fy They represent the moving speed of the boundary particle in the x direction and the y direction in the free field, respectively, and are obtained by the seismic wave delay input method; F l2n and F l2s They are normal and shear loads on the lateral boundary particles caused by free field motion.

[0061] Furthermore, the platform also includes a virtual particle property calculation module for calculating the speed and stress of virtual particles, including the speed v of virtual particles when they fall into the support domain of real particles. j Expressed as: v j =(1-β)v i , When the virtual particle is outside the support domain of the real particle, there is no need to calculate the velocity; when the virtual particle falls within the support domain of the real particle, the stress levels of the real and virtual particles are considered to be equal, that is, σ j =σ i ; otherwise, no calculation is required.

[0062] In a third aspect, the present invention provides a cloud device, which includes a server, and the seismic wave propagation numerical simulation platform as described above is installed on the server.

[0063] In a fourth aspect, the present invention provides a computer device, comprising a memory, wherein the memory stores program instructions, and when the program instructions are run, the method for numerical simulation of seismic wave propagation as described above is executed.

[0064] The beneficial effects of the present invention are:

[0065] The present invention realizes the input of seismic waves outside the model in the form of loads; it also realizes the absorption function of traveling waves outside the model truncation boundary and the coupling with the free field motion by combining with artificial absorption boundaries and free field boundaries, so that the propagation of seismic waves in the SPH numerical model is more in line with the actual situation, thereby improving the accuracy and reliability of numerical simulation research on seismic wave propagation using the SPH method. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a flow chart of a numerical simulation method of seismic wave propagation based on the SPH model provided by an embodiment of the present invention;

[0067] Figure 2 Schematic diagram of SPH numerical model modeling provided by an embodiment of the present invention;

[0068] Figure 3 It is a schematic diagram of numerical simulation of seismic wave propagation using an SPH model provided by an embodiment of the present invention;

[0069] Figure 4 is a schematic diagram of the force on viscous boundary particles provided by an embodiment of the present invention;

[0070] Figure 5 It is a schematic diagram of the forces on the bottom boundary particles of the SPH model provided by an embodiment of the present invention;

[0071] Figure 6 It is a schematic diagram of the lateral boundary particle force of the SPH model provided by an embodiment of the present invention;

[0072] Figure 7 It is a schematic diagram of virtual particle property calculation provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0073] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0074] The SPH method is a Lagrangian meshless particle method for solving fluid dynamics equations. Its core idea includes two key steps: particle approximation and kernel function approximation. A set of Naiver-Stokes partial differential equations containing field variables such as density, velocity and energy is established to simulate large deformation problems and impact problems.

[0075] Particle approximation is to discretize the continuous problem domain into a series of particles, which are given a certain initial mass, density, etc. The essence of particle approximation is to convert the continuous integral form of the field function and its derivative into the form of weighted summation of discrete particles in the support domain. The kernel interpolation approximation of particle physics variables can be expressed as:

[0076]

[0077] W ij =W(x i -x j ,h)=W(x i -x j |,h)

[0078]

[0079] Where W represents the kernel function; x is the coordinate vector of the interpolated particle, and x′ is the coordinate vector of the particles around it; Ω represents the support domain, which is determined by the smoothing length h of the kernel function W. The choice of kernel function W is based on a specific usage scenario, but it must meet the compact support and normalization conditions. Common kernel function types include Gaussian kernel, cubic spline kernel, and QuanticWendland kernel. For soil, the mass conservation and momentum conservation equations based on continuity conditions can be expressed as:

[0080] Mass equation:

[0081] Momentum equation:

[0082] Among them, ρ represents the density of the particle; v represents the velocity of the particle; p is the pressure on the particle; Π ij It is an artificial viscosity term; it is used to improve the stability of numerical calculations and suppress undesirable oscillations. When constructing an SPH numerical test model, it is usually necessary to arrange virtual particles on the outside of the model to generate a strong repulsive force on the real particles in the model, so as to prevent the boundary instability caused by the non-physical penetration of real particles near the boundary in the simulation. The fixed virtual particle method is a commonly used boundary processing method in SPH simulation. Its basic principle is to imitate the spatial distribution position of the real particles at the boundary of the model and continue to expand three layers of virtual particles outward. Figure 2 As shown in the figure, the SPH numerical model uses the midline between virtual particles and real particles as the boundary of the model. Fixed virtual particles have the same initial physical parameters as real particles (such as density, mass, and velocity). They participate in the calculation during the numerical simulation but do not undergo displacement. Instead, they are assigned virtual velocities and virtual stresses to balance the unstable forces at the boundary. The velocity of virtual particles is usually calculated based on the no-slip boundary conditions of the model (assuming that the velocity at the truncation boundary of the model is 0) and the motion state of the real particles in its vicinity (density, velocity, and displacement, etc.). Among them, the method of allocating the velocity of virtual particles using the ratio of the vertical distance from the truncation boundary to the real particles and virtual particles on both sides is one of the most commonly used methods. Figure 3 As an example, in the traditional SPH seismic wave propagation simulation, the virtual motion velocity v of the bottom virtual particle j is j The actual motion speed v of the real particle i in its support domain iand the boundary velocity v wall The following relations are satisfied:

[0083] v ij =v i -v j =β(v i -v wall )

[0084] v j =(1-β)v i +βv wall

[0085] in, d i d j are the distances from the ith real particle and the jth virtual particle to the cutoff boundary. When the model does not consider the seismic load, the boundary velocity v wall = 0. In the simulation of seismic wave propagation, the boundary wall velocity v wall Equal to the seismic wave velocity v seismic , that is, v wall =v seismic Since the model boundary is not a real physical wall, it is necessary to use v j The seismic waves that should act on the boundary are converted to virtual particles outside the model, thereby generating a velocity difference v between virtual and real particles on both sides of the truncation boundary. ij ≠0, and then use the quality control equation of SPH to realize the seismic wave input.

[0086] However, the method of simulating the propagation of seismic waves in the model by constructing the particle velocity difference between the inside and outside of the model only completes the input of the outward-travelling wave, and has not yet considered the absorption of the outward-travelling wave. In other words, when the input seismic wave is reflected back to the incident boundary by the free surface, since there is no stress wave absorption condition at the truncated boundary, the outward-travelling wave becomes an inward-travelling wave after being reflected by the boundary and enters the model for the second time, thus affecting the accuracy of the numerical simulation.

[0087] Based on this, the present invention defines the outermost real particles at the truncation boundary of the SPH model as the boundary particles, such as Figure 3As shown, the seismic waves outside the model are applied to the bottom boundary particles of the model in the form of loads to realize the input of earthquake motion; at the same time, an artificial absorption boundary is set outside the bottom boundary particles to absorb the outward traveling waves propagating from the inside of the model to the boundary, simulating the diffusion of seismic waves to the infinite field. A free field boundary is set outside the lateral boundary particles of the model, on the one hand to absorb the outward traveling waves of the earthquake, and on the other hand to realize the effective coupling of the free field motion with the model boundary, thereby improving the simulation accuracy of the seismic wave propagation of the model. Considering that the SPH model of the present invention is mainly used to simulate the seismic dynamic response of solid media, the properties (density, velocity and displacement, etc.) of real particles and virtual particles near the model boundary are calculated according to the "no slip" boundary condition.

[0088] Specifically, the present invention provides a numerical simulation method for seismic wave propagation based on the SPH model, such as Figure 1 As shown, specifically including:

[0089] S1. Construct SPH model;

[0090] like Figure 3 As shown, the SPH numerical model is constructed. First, the corresponding geometric block graphics are drawn based on the size of the study area. The particles with an equivalent diameter of D are arranged according to the spacing D rule until they cover the entire study area. The particles in the study area are real particles, and the outermost particles of the truncation boundary are boundary particles. The boundary particles are 0.5D away from the truncation boundary. Then, at least three layers of particles are continued to expand to the outside of the study area along the outward normal direction of the truncation boundary. They are fixed virtual particles. Finally, physical parameters are assigned to all particles according to the material properties. Virtual particles have the same physical properties as real particles. The corresponding kernel function W is selected, the SPH basic parameters are set, and D is the smooth length of the particles. In this way, a complete SPH numerical test model is constructed.

[0091] S2. Calculation of equivalent force of boundary particles

[0092] In the present invention, the boundary particles of the SPH model not only consider the seismic wave load input by the bottom boundary, but also consider the force of the bottom absorbing boundary and the lateral free field boundary. In addition, the repulsive force generated by the virtual particles outside the model cannot be ignored. The basic principle of the absorbing boundary is to eliminate the external traveling wave stress by setting a series of basic mechanical elements on the model boundary to prevent the stress wave reflection on the truncated boundary. The free field boundary also simulates the radiation of the external traveling wave from the finite domain to the infinite domain with the help of mechanical elements, and at the same time, realizes the effective coupling of the free field motion and the numerical model at the same time. The equivalent stress of the boundary particles will be calculated according to the boundary type below.

[0093] (1) Bottom border

[0094] In the present invention, the bottom adopts a viscous boundary, that is, an absorbing boundary is arranged on the bottom boundary particles to absorb the seismic reflection waves from outside the study area. Figure 4 As shown in Figure 2, for any viscous boundary particle, the normal force and tangential force provided by the damper can be expressed as:

[0095] σ=-ρc p v n

[0096] τ=-ρc s v s

[0097] where σ and τ represent the normal and tangential forces generated by the damper, respectively; ρ represents the density of the bulk medium; v n and v s They represent the moving speed of the action point on the boundary block in the normal and tangential directions respectively; c p and c s are the propagation speeds of seismic P waves and S waves in the medium, respectively, both of which are related to the physical and mechanical properties of the medium and can be expressed as:

[0098]

[0099] Where μ represents the Poisson's ratio of the material; E is Young's modulus;

[0100] The bottom boundary mainly considers the damping force F of the viscous boundary on the boundary particles. b1 and the equivalent load F acting on the boundary particles due to the earthquake b2 ;like Figure 5 As shown, for any bottom boundary particle b(x,y), the force F b It is expressed as:

[0101] F b =F b1 +F b2 , F b1 represents the damping force of the viscous boundary on the boundary particles, F b2 It represents the equivalent load of earthquake motion acting on the boundary particles;

[0102] For the bottom boundary particles, the damping force F b1 for:

[0103]

[0104] Among them, F b1n and F b1s are the normal and tangential damping forces acting on the bottom boundary particles respectively; α is the direction angle of the model cutoff boundary; v bx and v byRespectively represent the moving speed of the particle in the x direction and y direction;

[0105] The equivalent seismic load F that the bottom boundary particles are ultimately subjected to is b2 It is expressed as:

[0106]

[0107] Among them, F b2n and F b2s They represent the normal stress and shear stress on the particles respectively; v sx and v sy They represent the seismic wave velocities in the x and y directions acting on the boundary particles respectively.

[0108] (2) Lateral boundary

[0109] The local vibration forces the free field to move. The free field boundary is mainly used to simulate the radiation of the traveling wave outside the study area to the infinite domain, so that the seismic reflection wave generated outward in the study area can be properly absorbed. This process is also achieved with the help of damping elements in the present invention. At the same time, the free field boundary will also transfer the unbalanced force generated by the free field movement to the boundary particles. Figure 6 As shown, for any side boundary particle l(x,y), the total force F l It mainly includes the damping force F on the boundary particles l1 and the free field motion generates a load F on the particle l2 , expressed as: F l =F l1 +F l2 .

[0110] The calculation of the damping force on the free field boundary needs to take into account the relative motion between the study area and the free field. l1 and the free field motion generates a load F on the particle l2 Can be expressed as:

[0111]

[0112] Among them, F l1n and F l1s are the normal and tangential damping forces acting on the lateral boundary particles, respectively; v lx and v ly They represent the moving speed of the particle in the x and y directions respectively; v fx and v fy They represent the moving speed of the boundary particle in the x direction and the y direction in the free field, respectively, and are obtained by the seismic wave delay input method; F l2n and F l2s They are normal and shear loads on the lateral boundary particles caused by free field motion.

[0113] S3, External load input

[0114] In S2, according to the boundary type and boundary conditions of the SPH model, the equivalent force that the truncation boundary may be subjected to during the seismic wave propagation simulation is converted into the model bottom boundary load F b and the lateral boundary load F l Two concentrated forces. In the present invention, the external load input is directly applied to the model boundary particles in the form of concentrated loads through the momentum equation of SPH. For the bottom boundary and lateral boundary, the boundary particle momentum calculation equations considering the earthquake input are:

[0115] Bottom border:

[0116] Side border:

[0117] S4. Virtual particle property calculation

[0118] In the present invention, the density of virtual particles remains unchanged during the calculation process and is equal to the initial density of the material. The speed of virtual particles is directly related to the speed of real particles near them and the distance between them, and is usually calculated based on no-slip boundary conditions.

[0119] like Figure 7 As shown in the figure, when the virtual particle falls into the support domain of the real particle, the speed v of the virtual particle j It can be expressed as:

[0120] v j =(1-β)v i ;

[0121] in, When the virtual particle is outside the support domain of the real particle, there is no need to calculate the velocity. The stress of the virtual particle is based on the assumption of local stress uniformity and is also calculated based on its position relative to the real particle. For a no-slip boundary, when the virtual particle falls within the support domain of the real particle, the stress levels of the real and virtual particles are assumed to be equal, that is, σ j =σ i ; Otherwise, no calculation is required. For the same virtual particle, when it is in different real particle support domains, its virtual velocity and stress level will be different when participating in the calculation.

[0122] By iterating S2-S4, a more accurate simulation of seismic wave propagation of the SPH model can be achieved.

[0123] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not deviate from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for numerical simulation of seismic wave propagation, characterized in that: The method includes: S1. Construct SPH numerical test model; Draw the corresponding geometric block graphics based on the size of the research area, and arrange the particles with an equivalent diameter of D according to the interval distance D until they cover the entire research area. The particles in the research area are real particles, and the outermost particles of the truncation boundary are boundary particles, which are 0.5D away from the truncation boundary. Then, along the outward normal direction of the truncation boundary, continue to expand at least three layers of particles to the outside of the research area, which are fixed virtual particles. Finally, assign physical parameters to all particles according to material properties. Virtual particles have the same physical properties as real particles. Select the corresponding kernel function W, set the basic parameters of SPH, and use D as the smooth length of the particles to construct a complete SPH numerical test model. S2. According to the boundary type and boundary conditions of the SPH numerical model, the equivalent force on the truncation boundary when simulating the propagation of seismic waves is converted into two concentrated forces: the boundary particle load at the bottom of the model and the lateral boundary particle load; S3, the external load input is directly applied to the boundary particles of the numerical model in the form of concentrated load through the momentum equation of the SPH method. The momentum equation is: Among them, ρ represents the density of the particle; v represents the velocity of the particle; p is the pressure on the particle; Π ij is the artificial viscosity term; For the bottom boundary and lateral boundary, the boundary particle momentum calculation equations when the seismic wave is input are: Bottom border: F b It represents the earthquake load at the bottom of the model, that is, the force on any bottom boundary particle; Side border: F l It represents the load generated by the lateral free field motion of the model, that is, the force acting on any lateral boundary particle; The equivalent force on the truncation boundary during the simulation of seismic wave propagation is converted into a concentrated load acting on the bottom boundary particles of the model. Specifically, it includes: The bottom of the model adopts a viscous boundary. The equivalent load acting on the boundary particles is applied by arranging them on the bottom boundary particles, and the seismic reflection waves from the outside of the study area are absorbed. For any viscous boundary particles, the normal force and tangential force provided by the damper are expressed as: σ=-ρc p v n τ=-ρc s v s where σ and τ represent the normal and tangential forces generated by the damper, respectively; ρ represents the density of the bulk medium; v n and v s They represent the moving speed of the action point on the boundary block in the normal and tangential directions respectively; c p and c s are the propagation speeds of seismic P waves and S waves in the medium, respectively, both of which are related to the physical and mechanical properties of the medium and can be expressed as: Where μ represents the Poisson's ratio of the material; E is Young's modulus; For any bottom boundary particle b(x,y), the force F b It is expressed as: F b =F b1 +F b2 , F b1 represents the damping force of the viscous boundary on the boundary particles, F b2 It represents the equivalent load of earthquake motion acting on the boundary particles; For the bottom boundary particles, the damping force F b1 for: Among them, F b1n and F b1s are the normal and tangential damping forces acting on the bottom boundary particles respectively; α is the direction angle of the model cutoff boundary; v bx and v by Respectively represent the moving speed of the particle in the x direction and y direction; The equivalent seismic load F that the bottom boundary particles are ultimately subjected to is b2 It is expressed as: Among them, F b2n and F b2s They represent the normal stress and shear stress on the particles respectively; v sx and v sy They represent the seismic wave velocities in the x and y directions acting on the boundary particles respectively.

2. The method for numerical simulation of seismic wave propagation according to claim 1, characterized in that: The equivalent force on the truncation boundary during the free field motion in the simulated earthquake process is converted into the load acting on the lateral boundary particles of the model, specifically including: For any side boundary particle l(x,y), the total force F l Including the damping force F on the boundary particles l1 and the free field motion generates a load F on the particle l2 , expressed as: F l =F l1 +F l2 ; Among them, F l1n and F l1s are the normal and tangential damping forces acting on the lateral boundary particles, respectively; v lx and v ly They represent the moving speed of the particle in the x and y directions respectively; v fx and v fy They represent the moving speed of the boundary particle in the x direction and the y direction in the free field, respectively, and are obtained by the seismic wave delay input method; F l2n and F l2s They are normal and shear loads on the lateral boundary particles caused by free field motion.

3. The method for numerical simulation of seismic wave propagation according to claim 1, characterized in that: The method also includes S4, virtual particle property calculation; specifically, when the jth virtual particle falls within the support domain of the ith real particle, the speed v of the virtual particle j Expressed as: v j =(1-β)v i , When the virtual particle is outside the support domain of the real particle, there is no need to calculate the velocity; when the virtual particle falls within the support domain of the real particle, the stress levels of the real and virtual particles are considered to be equal, that is, σ j =σ i ; otherwise, no calculation is required.

4. A seismic wave propagation numerical simulation platform, used to implement the seismic wave propagation numerical simulation method according to any one of claims 1 to 3, characterized in that: The platform includes: The model building module is used to build the SPH numerical test model. The corresponding geometric block graphics are drawn based on the size of the research area. Particles with an equivalent diameter of D are arranged according to the interval distance D until they cover the entire research area. The particles in the research area are real particles, and the outermost particles of the truncation boundary are boundary particles. The boundary particles are 0.5D away from the truncation boundary. Then, at least three layers of particles are continued to expand outside the research area along the outward normal direction of the truncation boundary. They are fixed virtual particles. Finally, physical parameters are assigned to all particles according to material properties. Virtual particles have the same physical properties as real particles. The corresponding kernel function W is selected, the basic parameters of the SPH model are set, and D is the smooth length of the particles to build a complete SPH numerical test model. The equivalent force conversion module is used to convert the equivalent force on the truncation boundary when simulating seismic wave propagation into two concentrated forces as loads on the bottom boundary particles and the lateral boundary particles of the model according to the boundary type and boundary conditions of the SPH model; The external load input module is used to apply external load input in the form of concentrated load directly to the model boundary particles through the momentum equation of SPH. The momentum equation is: Among them, ρ represents the density of the particle; v represents the velocity of the particle; p is the pressure on the particle; Π ij is the artificial viscosity term; For the bottom boundary and lateral boundary, the boundary particle momentum calculation equations when the seismic wave is input are: Bottom border: F b It represents the earthquake load at the bottom of the model, that is, the force on any bottom boundary particle; Side border: F l It represents the load generated by the lateral free field motion of the model, that is, the force acting on any lateral boundary particles.

5. The seismic wave propagation numerical simulation platform according to claim 4, characterized in that: The equivalent force conversion module is specifically used to absorb the seismic reflection waves from the outside of the study area by arranging them on the bottom boundary particles. For any viscous boundary particles, the normal force and tangential force provided by the damper are expressed as: σ=-ρc p v n τ=-ρc s v s where σ and τ represent the normal and tangential forces generated by the damper, respectively; ρ represents the density of the bulk medium; v n and v s They represent the moving speed of the action point on the boundary block in the normal and tangential directions respectively; c p and c s are the propagation speeds of seismic P waves and S waves in the medium, respectively, both of which are related to the physical and mechanical properties of the medium and can be expressed as: Where μ represents the Poisson's ratio of the material; E is Young's modulus; For any bottom boundary particle b(x,y), the force F b It is expressed as: F b =F b1 +F b2 , F b1 represents the damping force of the viscous boundary on the boundary particles, F b2 It represents the equivalent load of earthquake motion acting on the boundary particles; For the bottom boundary particles, the damping force F b1 for: Among them, F b1n and F b1s are the normal and tangential damping forces acting on the bottom boundary particles respectively; α is the direction angle of the model cutoff boundary; v bx and v by Respectively represent the moving speed of the particle in the x direction and y direction; The equivalent seismic load F that the bottom boundary particles are ultimately subjected to is b2 It is expressed as: Among them, F b2n and F b2s They represent the normal stress and shear stress on the particles respectively; v sx and v sy They represent the seismic wave velocities in the x and y directions acting on the boundary particles respectively.

6. The seismic wave propagation numerical simulation platform according to claim 4, characterized in that: The equivalent force conversion module is specifically used to convert the total force F to any side boundary particle l(x,y) l It mainly includes the damping force F on the boundary particles l1 and the free field motion generates a load F on the particle l2 , expressed as: F l =F l1 +F l2 ; Among them, F l1n and F l1s are the normal and tangential damping forces acting on the lateral boundary particles, respectively; v lx and v ly They represent the moving speed of the particle in the x and y directions respectively; v fx and v fy They represent the moving speed of the boundary particle in the x direction and the y direction in the free field, respectively, and are obtained by the seismic wave delay input method; F l2n and F l2s They are normal and shear loads on the lateral boundary particles caused by free field motion.

7. The seismic wave propagation numerical simulation platform according to claim 4, characterized in that: The platform also includes a virtual particle property calculation module for calculating the speed and stress of virtual particles, including the speed v of virtual particles when they fall into the support domain of real particles. j Expressed as: v j =(1-β)v i , When the virtual particle is outside the support domain of the real particle, there is no need to calculate the velocity; when the virtual particle falls within the support domain of the real particle, the stress levels of the real and virtual particles are considered to be equal, that is, σ j =σ i ; otherwise, no calculation is required.

8. A cloud device, comprising a server, characterized in that: The seismic wave propagation numerical simulation platform as described in any one of claims 4 to 7 is installed on the server.

9. A computer device comprising a memory, wherein the memory stores program instructions, characterized in that: When the program instructions are executed, the numerical simulation method for seismic wave propagation as described in any one of claims 1 to 3 is executed.