Slope instability whole process calculation method, system and equipment based on particle method

Abstract

The invention discloses a slope instability whole process calculation method, system and equipment based on a particle method. The method is based on a smoothed particle hydrodynamics (SPH) particle calculation method, combines a Riemannian large deformation theory, and realizes landslide simulation under a regional scale through a C + + / CUDA writing program. According to the method, the whole process of generation, connection, starting and accumulation of the sliding surface can be truly represented, the influence mechanism of rock and earth mass properties such as an internal friction angle, cohesion, particle size, elasticity modulus and Poisson's ratio on the landslide process can be analyzed, meanwhile, through a parallel heterogeneous CPU / GPU programming architecture, the calculation efficiency is improved by nearly a hundreds of times compared with the traditional CPU programming efficiency, and the calculation cost is greatly reduced. The watershed scale landslide flow velocity and accumulation range prediction is realized, the adaptability is very high, the defect that discrete element and finite element methods are limited by simulation scale and calculation efficiency is overcome, and a new thought is provided for large-scale landslide disaster prediction.

Description

Technical Field

[0001] The invention belongs to the technical field of landslide prediction, and in particular relates to a method, system and equipment for calculating the whole process of slope instability based on a particle method. Background Art

[0002] Landslides are common in all parts of my country. Once a landslide occurs, it will have a huge impact and burial effect, which will pose a serious threat to the lives and property of residents. Numerical simulation is a commonly used engineering safety analysis method, and the traditional finite element method is often used to simulate the startup stage of sliding surface generation, but it is difficult to simulate the large deformation stage due to the limitation of mesh distortion, and it is impossible to predict the operation-accumulation stage of landslides that risk control personnel are most concerned about; the discrete element method can make up for the shortcomings of the finite element method to achieve large deformation simulation, but it is constrained by the fact that its core contact calculation algorithm leads to low calculation efficiency. At present, it is more used in laboratory scale mechanism research, and it is difficult to carry out prediction analysis for large-scale slopes.

[0003] Therefore, realizing the simulation of rock and soil masses from small deformation to large deformation and solving the calculation problems caused by the real terrain are the current problems that need to be solved urgently for on-site landslide prediction and analysis. Summary of the invention

[0004] In order to solve the problem that the existing technology is difficult to simulate the deformation of rock and soil from small to large deformation and is limited by the calculation explosion caused by the real terrain, the present invention provides a method, system and equipment for calculating the whole process of slope instability based on the particle method. Through multi-threaded parallel calculation, based on the consideration of the constitutive model of rock and soil, efficient numerical calculation of the whole process of simulating the initiation, migration and accumulation of landslide is realized.

[0005] According to one aspect of the present invention, a method for calculating the whole process of slope instability based on a particle method is provided, comprising:

[0006] Obtain the STL real terrain data of the slope site and import it into the smooth particle fluid dynamics calculation framework;

[0007] In the smooth particle fluid dynamics calculation framework, particles are generated, the calculation domain is divided and the particles are numbered, and the rock and soil parameter attributes are assigned to each particle.

[0008] In the way that each particle corresponds to one thread, all particles are calculated in parallel with multiple threads, where each particle performs the following cyclic calculations: calculate the stress rate and adaptive time step of each rock and soil particle, update the rock and soil particle stress and store it; when the particle stress does not meet the ideal elastic-plastic criterion, calculate the large deformation of the rock and soil, update the particle velocity and density and store it; perform time integration according to the Verlet algorithm, update the particle position and store it; when the set single-step output time is reached, output the stored particle stress, particle velocity, particle density and particle position;

[0009] Repeat the loop calculation process for each particle until the current physical simulation time reaches the preset simulation time.

[0010] As a further technical solution, the STL real terrain data of the slope site is obtained and imported into the smooth particle hydrodynamics calculation framework, including:

[0011] Obtain site point cloud data of the slope;

[0012] An initial slope terrain in STL format is generated according to the site point cloud data and is imported into a smooth particle fluid dynamics calculation framework.

[0013] As a further technical solution, under the smooth particle fluid dynamics calculation framework, particles are generated and the calculation domain is divided and the particles are numbered. At the same time, the rock and soil parameter attributes are assigned to each particle, including:

[0014] Generate SPH particles according to the set initial particle spacing and search radius, and obtain the terrain represented by SPH particles with fixed spacing;

[0015] Divide the entire computational domain into grids and number the particles and grids respectively;

[0016] Determine the site's rock and soil parameter attributes and assign them to the SPH particles, and determine the SPH particle artificial sound velocity.

[0017] As a further technical solution, the determined SPH particle artificial sound speed is greater than 10 times the maximum particle speed, or meets Where E represents the elastic modulus and ρ represents the particle density.

[0018] As a further technical solution, the stress rate and adaptive time step of each rock mass particle are calculated, and the stress of the rock mass particle is updated and stored, including:

[0019] Conduct adjacent particle search, calculate the particles in the grid whose distance from the calculated particle is less than the set value, and determine the particle pairs in the search domain;

[0020] According to the ideal elastic-plastic constitutive model and combined with Riemann large deformation theory, the stress rate of rock and soil particles is solved;

[0021] Calculate the time step;

[0022] According to the solved stress rate and time step, the stress information of the rock and soil particles is updated and stored.

[0023] As a further technical solution, when the particle stress does not meet the ideal elastic-plastic criterion, the large deformation of the rock and soil body is calculated, and the particle velocity and density are updated and stored, including:

[0024] Solve the artificial viscosity and artificial stress of each rock mass particle;

[0025] According to the artificial viscosity, artificial stress and updated stress information, the mass conservation and momentum conservation equations under SPH discretization are solved to obtain the density change rate and the particle change rate;

[0026] Update particle velocity and density information and store them.

[0027] As a further technical solution, time integration is performed according to the Verlet algorithm, and the particle position is updated and stored, including:

[0028] After waiting for all threads to complete calculations, the Verlet algorithm is used to perform time integration, update the particle position information, and store it.

[0029] According to one aspect of the present invention, a slope instability whole process calculation system based on particle method is provided, comprising a first processor and a second processor, wherein:

[0030] The first processor is used to obtain the STL real terrain data of the slope site and import it into the smooth particle fluid dynamics calculation framework; under the smooth particle fluid dynamics calculation framework, generate particles and divide the calculation domain and number the particles, and assign rock and soil parameter attributes to each particle;

[0031] The second processor is used to receive the information transmitted by the first processor, and perform multi-threaded parallel calculation on all particles in a manner that each particle corresponds to one thread, wherein each particle performs the following cyclic calculation: calculate the stress rate and adaptive time step of each rock and soil particle, update the rock and soil particle stress and store it; when the particle stress does not meet the ideal elastic-plastic criterion, calculate the large deformation of the rock and soil, update the particle velocity and density and store it; perform time integration according to the Verlet algorithm, update the particle position and store it; when the set single-step output time is reached, output the stored particle stress, particle velocity, particle density and particle position;

[0032] The first processor is further used to receive data output by the second processor and determine whether the current numerical simulation actually reaches a preset simulation time. If the preset simulation time is reached, the sequential calculation process of the second processor is stopped.

[0033] According to one aspect of the present invention, there is provided a device for calculating the whole process of slope instability based on the particle method, comprising a processor and a memory for storing a computer program that can be run on the processor, wherein the processor is used to execute the steps of the method for calculating the whole process of slope instability based on the particle method when running the computer program.

[0034] According to one aspect of the present invention, a computer storage medium is provided, wherein a computer program is stored in the computer storage medium, and when the computer program is executed by a processor, the steps of the method for calculating the whole process of slope instability based on the particle method are implemented.

[0035] Compared with the prior art, the present invention provides an efficient numerical calculation method that can consider the constitutive model of rock and soil and simulate the whole process of landslide initiation, migration and accumulation, and its beneficial effects are:

[0036] (1) Based on the Riemann large deformation theory, the present invention breaks through the limitation that the traditional method cannot consider both small deformation and large deformation when calculating landslides, and establishes a method that can predict the entire destruction process of slope rock and soil from small deformation of the soil to large deformation of landslide.

[0037] (2) The method of the present invention is based on the ideal elastic-plastic rock and soil constitutive model, completes the simulation of the plastic zone generation and penetration process of the soil-rock mixture, and can accurately predict the location of the plastic zone generation and the penetration time of the rock and soil.

[0038] (3) The method of the present invention introduces the ideal elastic-plastic criterion into the smooth fluid particle dynamics, completes the sliding and accumulation process of the soil-rock mixture, and can predict the sliding speed and accumulation range of the landslide body.

[0039] (4) The method of the present invention takes into account the influence of the initial conditions of the rock and soil mass, and parameter sensitivity analysis can be carried out on the site rock and soil mass parameters (internal friction angle, cohesion, density, etc.) as needed.

[0040] (5) The method of the present invention adopts GPU parallel computing programming method, which avoids the time-consuming calculation of traditional methods and can be used for rapid assessment of disaster sites.

[0041] (6) The method of the present invention adopts C++ / CUDA programming, which is easy to encapsulate and can be imported into large-scale commercial software such as Multiphase ABAQUS after simple processing for more complex working condition calculations. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, a brief introduction is given below to the drawings used in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 A schematic flow chart of a method for calculating the entire process of slope instability based on a particle method is provided in an embodiment of the present invention.

[0044] Figure 2 A schematic diagram of the entire process of slope instability provided by an embodiment of the present invention.

[0045] Figure 3 A particle-based method-based method for calculating the entire process of slope instability provided in an embodiment of the present invention is used for a schematic diagram of a landslide result image in the southwest.

[0046] Figure 4 A schematic diagram of a method for calculating the entire process of slope instability based on a particle method and implemented by CPU+GPU provided in an embodiment of the present invention.

[0047] Figure 5 A schematic diagram of a whole-process slope instability calculation system based on a particle method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] It should be noted that:

[0049] The smoothed particle hydrodynamics (SPH) method calculates particles with both position information and state information (stress, strain, velocity, etc.). It has the advantages of easy parallel computing and the ability to embed multiple constitutive models, providing a direction for solving large-scale calculations of rock and soil landslides under real terrain. In addition, the resulting landslide simulation method should be able to be embedded in complex water-soil and water-soil-solid numerical calculations, thus having greater engineering practical value.

[0050] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention are arbitrarily combined with each other to form a new technical solution. This combination is not restricted by the sequence of steps and / or the structural composition mode, but must be based on the ability of ordinary technicians in this field to achieve. When the combination of technical solutions is contradictory or cannot be achieved, it should be considered that this combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0051] The present invention provides a method for calculating the whole process of slope instability based on particle method, such as Figure 1 As shown, first, the initial slope terrain in STL format is obtained, and the initial slope terrain is converted into terrain represented by SPH particles; then, the entire calculation domain is gridded and numbered, and soil parameters are assigned to the SPH particles; then, each SPH particle is cyclically calculated in a multi-threaded parallel manner, and each cycle process calculates the particle stress, velocity, density and position information respectively, and updates and stores the information; finally, the actual physical simulation time is monitored, and when the actual physical simulation time reaches the preset simulation time, the cyclic calculation process is stopped.

[0052] The embodiment of the present invention adopts the method of storage in CPU and calculation in GPU, and performs multi-threaded parallel calculation in GPU for each particle in the terrain, avoiding the problem of long calculation time of traditional methods, and can be used for rapid assessment of disaster sites. The calculation process for each particle: First, the stress of rock and soil particles is solved by using the ideal elastic-plastic constitutive method combined with the Riemann large deformation theory, and a method is established to predict the whole process of slope rock and soil destruction from small deformation of soil to large deformation of landslide; then, the ideal elastic-plastic criterion (i.e., Drucker-Parager criterion) is introduced into the smooth fluid particle dynamics to complete the sliding and accumulation process of soil-rock mixture, and the sliding speed and accumulation range of landslide can be predicted; then, based on the ideal elastic-plastic rock and soil constitutive model, the plastic zone generation and penetration process simulation of the soil-rock mixture is completed, and the location and penetration time of the plastic zone of the rock and soil can be accurately predicted.

[0053] The embodiment of the present invention provides an efficient numerical calculation method that can consider the constitutive model of rock and soil mass and simulate the whole process of landslide initiation, migration and accumulation, which specifically includes the following steps:

[0054] Step 1. Obtain site cloud data through remote sensing images, drone photogrammetry, laser scanners, etc., generate triangular STL file meshes, and establish real slope terrain.

[0055] Step 2. Import the real terrain data of STL (Standard Tessellation Language) into the particle method calculation framework, select the initial particle spacing Dp (SPH particle radius), determine the search radius (h = 1.4Dp), generate SPH particles in the real terrain, and use SPH particles with fixed spacing to represent the terrain. At the same time, divide the entire calculation domain into grids with a side length of 2h, and number the particles and grids respectively.

[0056] Step 3. According to the site conditions, determine the internal friction angle (Φ), cohesion (c), elastic modulus (E), Poisson's ratio (e), density (ρ) and other rock and soil parameter attributes and assign them to the SPH particles; determine the SPH particle artificial sound speed (Cs), write it into the .exe file compiled by Visual Studio by reading the XML file, and store all the information in the CPU (central processing unit) memory in Visual Studio through the C++ language.

[0057] It should be noted that when assigning the parameters of rock and soil particles, the goal of fast assignment and change of assignment is achieved by writing XML reading files in Visual studio, and the input and calculation modules are separated to achieve the purpose of easy operation.

[0058] Furthermore, the artificial sound velocity has a regulating effect on the stability and efficiency of the calculation. Once the real value of the rock and soil (600-700m / s) is taken, the calculation step size is too small, which affects the calculation efficiency; at the same time, the artificial sound velocity cannot be too small, which is easy to generate calculation noise. Therefore, it is generally believed that Cs>10Vmax, that is, the artificial sound velocity only needs to be greater than 10 times the maximum particle velocity, or meet That is, high calculation accuracy can be achieved while ensuring calculation efficiency.

[0059] Step 4. In Visual Studio, use CUDA language to call the empty GPU (graphics processing unit) memory and copy the information in the CPU to the GPU through pointer transfer. Carry out the calculation part in the GPU, put each particle in a separate thread, and each thread performs the same calculation steps.

[0060] In the embodiment of the present invention, the information in the CPU is copied to the GPU by pointer transmission using the CUDA language through Visual Studio, so as to prepare for rapid calculation in the GPU and complete the core link of the calculation framework of the slope instability whole process simulation method, which is storage in the CPU and calculation in the GPU. By using this framework, the calculation efficiency is improved by hundreds of times.

[0061] Step 5. Calculate the stress (σ) of the rock mass particles. Taking a certain computational particle as an example, search for nine grids (in two-dimensional case) or twenty-seven grids (in three-dimensional case) adjacent to the center of the particle, calculate the particles in the grid whose distance from the computational particle is less than 2h, and call the Kernel particle sorting calculation method in CUDA to quickly determine the particle pairs in the search domain.

[0062] Step 6. Solve the stress rate of rock and soil particles by combining the ideal elastic-plastic constitutive model with the Riemann large deformation theory

[0063]

[0064] In the formula is the stress rate, w is the moment of inertia, is the plastic strain rate operator, G is the shear modulus, K is the bulk modulus, is the strain rate, ψ is the dilatancy angle, J2 is the second stress tensor invariant, s is the shear stress, is the deviatoric stress tensor rate, δ is the Dirac factor, and s is the deviatoric stress tensor. The stress information of the particle is saved in the agreed GPU.

[0065] Step 7. Calculate the time step. The stable variable time step is solved by the CFL condition, which can generally be determined by the following formula:

[0066]

[0067] Δt=C CFL min(Δt f ,h / C s )

[0068] Where f a is the maximum stress value, C CFL Generally, 0.05 is taken, and h is the radius of the calculation core. After solving the stress rate and time step, the stress of each rock mass particle can be updated.

[0069] Step 8. Substitute stress into the mass conservation and momentum conservation equations under SPH discretization, and use the following formula to obtain stable and accurate stress under SPH method calculation:

[0070]

[0071]

[0072] Among them, v ij is the particle velocity difference, I is the unit matrix, g is the gravitational acceleration, Π ij is artificial viscosity, S ij is the artificial stress used to stabilize the calculation equation, is the divergence, ρ i is the density of the ith particle, ρ j is the density of the jth particle, m j is the mass of the jth particle, W ij is the Winrider kernel function, σ i is the stress of the ith particle, σ j is the stress of the jth particle. Thus, the ideal elastic-plastic constitutive relation of soil is embedded into the SPH discrete framework, the calculation of relative force is completed, and the velocity and density information of particles are saved in the agreed GPU.

[0073] When calculating, for high and steep slopes, we usually use stress correction. The correction method is to let the particles reach equilibrium using their own gravity in a short time, then set their displacement to zero, return their particle positions to their initial positions, and retain the stress distribution obtained at this time as the initial stress distribution, so as to quickly obtain the true stress distribution.

[0074] When solving the main governing equations of conservation of mass and momentum and storing particle velocity (v) information, the embodiment of the present invention calls the syncthreads() function in CUDA to wait for all threads in the GPU to complete calculations. This ensures that out-of-bounds access does not occur during GPU calculations.

[0075] Specifically, solve the artificial viscosity (Π ij ), artificial stress (S ij ), when solving for artificial viscosity,

[0076]

[0077] Its coefficient α ΙΙ Usually the value is 0.5, β ΙΙ Generally, we do not consider taking 0, c ij is the average artificial sound speed of the particles, is the average density, h ij is the average nuclear radius, r ij is the interparticle distance, φ ij The calculation is as follows:

[0078]

[0079] Sij Artificial stress can be expressed as follows:

[0080]

[0081] in, R i and R j Calculated by the following formula:

[0082]

[0083] ∈ is a constant between [0,1], in this case 0.2, P i If the artificial viscosity coefficient is not 0.5 and the artificial stress coefficient is not 0.2 in the calculation, it will be difficult to achieve the goal of stabilizing the calculation speed.

[0084] Step 9. Perform time integration. Using the modified Verlet integration format, at each step, the interactions between particle pairs are calculated, and then the variables of each particle are updated using the following explicit integral,

[0085]

[0086]

[0087]

[0088] Among them, F a is the mass force, v a is the relative velocity of the particle pair, r a is the particle spacing, σ is stress, ε is strain, S is the stress change rate, M is the strain change rate, z is the current moment, and Z+1 is the next time step. When solving the rock and soil mass, it is not necessary to update the density of the particles, but it is necessary to update the stress and strain of the particles, save the new position information of the particles in the agreed GPU, and call the -syncthreads() function in CUDA to wait for the thread calculation in all GPUs to complete.

[0089] Step 10. Output the saved position, velocity, stress and other information as a binary file (VTK), and determine whether the current simulation time has reached the required simulation physical time. If not, repeat steps 5 to 10. If it has, exit the loop and end the calculation.

[0090] See also Figure 2 and Figure 3 ,in, Figure 2 A schematic diagram of the whole process of slope instability is given. Figure 3 A schematic diagram of the result image of a landslide in the southwest is given by using the particle-based method for calculating the whole process of slope instability described in an embodiment of the present invention.

[0091] Table 1 GPU parallel computing

[0092]

[0093] It should be noted that the embodiment of the present invention calls GPU calculation, which takes advantage of the fact that the GPU has a large number of computing units. By assigning particle information to each minimum computing unit thread, the position of the corresponding thread in the GPU is found when calling to carry out calculation. Specifically, by taking advantage of the characteristics of GPU multi-core and multi-computing channels, the SPH particles are placed in a separate minimum GPU computing unit (thread). When calculating the interaction force and self-stress, independent calculation can be carried out by finding the position of the corresponding thread in the GPU without the phenomenon of calculation waiting. The calling process is shown in Table 1.

[0094] As a preferred embodiment, Figure 4 As shown, a method for calculating the whole process of slope instability based on a particle method is provided in an embodiment of the present invention, comprising the following steps:

[0095] (1) Obtain STL terrain data and set up D in Visual Studio p = 10m (SPH particle radius), generate SPH particles, a total of 328439 particles, and number the particles sequentially;

[0096] (2) Determine the search radius h = 14 m, input the adaptive step size CFL = 0.05, gravity acceleration g = 9.8 N / kg, density ρ = 2000 kg / m 3 , artificial sound velocity Cs = 120m / s, soil internal friction angle Φ = 30°, cohesion 4kPa, elastic modulus E = 20MPa, Poisson's ratio e = 0.2;

[0097] (3) In Visual Studio, select the CPU (central processing unit) free memory through the C++ language and store the particle position information (x, y, z) and attribute information (initial velocity, stress, soil elastic modulus, etc.).

[0098] (4) In Visual Studio, use CUDA to call the empty GPU (graphics processing unit) memory and copy the information in the CPU to the GPU through pointer transfer.

[0099] (5) Carry out the calculation part on the GPU, place each particle in a separate thread, and each thread performs the same calculation steps.

[0100] (6) Solve the stress rate of each rock particle in the thread As well as the adaptive time step (Δt), the stress (σ) of the rock and soil particles is updated and the stress (σ) information is stored.

[0101] (7) Solve the artificial viscosity (Π ij ), artificial stress (S ij ), when solving for artificial viscosity,

[0102]

[0103] Its coefficient α ΙΙ Usually the value is 0.5, β ΙΙ Generally, we do not consider taking 0, c ij is the average artificial sound speed of the particles, is the average density, φ ij The calculation is as follows:

[0104]

[0105] S ij Artificial stress can be expressed as follows:

[0106]

[0107] in, R i and R j Calculated by the following formula:

[0108]

[0109] ∈ is a constant between [0,1], and in this case 0.2 is used. i is the hydrostatic pressure.

[0110] (8) Solve the main equations of conservation of mass and momentum, store the particle velocity (v) information, and call the CUDA -syncthreads() function to wait for all GPU threads to complete calculations.

[0111] (9) Update the particle position information through the Verlet algorithm and store it in the GPU.

[0112] (10) The particle stress information, velocity information, and position information in the GPU are stored in the CPU, and the memory in the GPU is deleted. The information in the CPU is imported and written into a binary file for storage.

[0113] (11) Check whether the current physical simulation time reaches the planned simulation time. If not, repeat steps 5 to 11. If satisfied, exit the calculation.

[0114] Based on the same inventive concept as the above method embodiment, the embodiment of the present invention also provides a slope instability whole process calculation system based on particle method, including a first processor and a second processor, such as Figure 5 As shown,

[0115] The first processor is used to obtain the STL real terrain data of the slope site and import it into the smooth particle fluid dynamics calculation framework; under the smooth particle fluid dynamics calculation framework, generate particles and divide the calculation domain and number the particles, and assign rock and soil parameter attributes to each particle;

[0116] The second processor is used to receive the information transmitted by the first processor, and perform multi-threaded parallel calculation on all particles in a manner that each particle corresponds to one thread, wherein each particle performs the following cyclic calculation: calculate the stress rate and adaptive time step of each rock and soil particle, update the rock and soil particle stress and store it; when the particle stress does not meet the ideal elastic-plastic criterion, calculate the large deformation of the rock and soil, update the particle velocity and density and store it; perform time integration according to the Verlet algorithm, update the particle position and store it; when the set single-step output time is reached, output the stored particle stress, particle velocity, particle density and particle position;

[0117] The first processor is further used to receive data output by the second processor and determine whether the current numerical simulation actually reaches a preset simulation time. If the preset simulation time is reached, the sequential calculation process of the second processor is stopped.

[0118] In the embodiment of the present invention, the first processor can be implemented by a CPU, and the second processor can be implemented by a GPU. The embodiment of the present invention adopts a storage-in-CPU-computation method to perform multi-threaded parallel calculations in the GPU for each particle in the terrain, thereby avoiding the problem of long calculation time in traditional methods, and can be used for rapid assessment of disaster sites.

[0119] An embodiment of the present invention also provides a device for calculating the entire process of slope instability based on the particle method, comprising a processor and a memory for storing a computer program that can be run on the processor, wherein the processor is used to execute the steps of the method for calculating the entire process of slope instability based on the particle method when running the computer program.

[0120] The memory in the embodiment of the present invention is used to store various types of data to support the operation of the device for calculating the whole process of slope instability based on the particle method. Examples of these data include: any computer program for operating on the device for calculating the whole process of slope instability based on the particle method.

[0121] An embodiment of the present invention further provides a computer storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of the method for calculating the whole process of slope instability based on the particle method are implemented.

[0122] In summary, the present invention discloses a method for calculating the whole process of slope instability based on the particle method. The method is based on the smooth particle hydrodynamics (SPH) particle calculation method, combined with the Riemann large deformation theory, and realizes regional scale landslide simulation by programming in C++ / CUDA. The method can truly reproduce the whole process of sliding surface generation, penetration, startup, and accumulation, and can analyze the influence mechanism of rock and soil properties such as internal friction angle, cohesion, particle size, elastic modulus, Poisson's ratio, etc. on the landslide process. At the same time, the method uses a parallel heterogeneous CPU / GPU programming architecture, and the calculation efficiency is nearly 100 times higher than that of traditional CPU programming. It realizes the prediction of watershed scale landslide flow velocity and accumulation range, has strong adaptability, overcomes the defects of discrete element, finite element and other methods that are limited by simulation scale and calculation efficiency, and provides a new idea for the prediction of large-scale landslide disasters.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A method for calculating the whole process of slope instability based on particle method, characterized in that: include: Obtain the STL real terrain data of the slope site and import it into the smooth particle fluid dynamics calculation framework; In the smooth particle fluid dynamics calculation framework, particles are generated, the calculation domain is divided and the particles are numbered, and the rock and soil parameter attributes are assigned to each particle. In the way that each particle corresponds to one thread, all particles are calculated in parallel with multiple threads, where each particle performs the following cyclic calculations: calculate the stress rate and adaptive time step of each rock and soil particle, update the rock and soil particle stress and store it; when the particle stress does not meet the ideal elastic-plastic criterion, calculate the large deformation of the rock and soil, update the particle velocity and density and store it; perform time integration according to the Verlet algorithm, update the particle position and store it; when the set single-step output time is reached, output the stored particle stress, particle velocity, particle density and particle position; Repeat the loop calculation process for each particle until the current physical simulation time reaches the preset simulation time.

2. The whole process calculation method of slope instability based on particle method according to claim 1 is characterized in that: Obtain the STL real terrain data of the slope site and import it into the smooth particle fluid dynamics calculation framework, including: Obtain site point cloud data of the slope; An initial slope terrain in STL format is generated according to the site point cloud data and is imported into a smooth particle fluid dynamics calculation framework.

3. The whole process calculation method of slope instability based on particle method according to claim 1 is characterized in that: In the smooth particle fluid dynamics calculation framework, particles are generated, the calculation domain is divided and the particles are numbered. At the same time, the rock and soil parameter attributes are assigned to each particle, including: Generate SPH particles according to the set initial particle spacing and search radius, and obtain the terrain represented by SPH particles with fixed spacing; Divide the entire computational domain into grids and number the particles and grids respectively; Determine the site's rock and soil parameter attributes and assign them to the SPH particles, and determine the SPH particle artificial sound velocity.

4. The method for calculating the whole process of slope instability based on the particle method according to claim 3 is characterized in that: The artificial sound speed of the determined SPH particles is greater than 10 times the maximum particle speed, or meets Where E represents the elastic modulus and ρ represents the particle density.

5. The method for calculating the whole process of slope instability based on particle method according to claim 1 is characterized in that: Calculate the stress rate and adaptive time step of each soil particle, update the stress of soil particles and store it, including: Conduct adjacent particle search, calculate the particles in the grid whose distance from the calculated particle is less than the set value, and determine the particle pairs in the search domain; According to the ideal elastic-plastic constitutive model and combined with Riemann large deformation theory, the stress rate of rock and soil particles is solved; Calculate the time step; According to the solved stress rate and time step, the stress information of the rock and soil particles is updated and stored.

6. A method for calculating the whole process of slope instability based on particle method according to claim 5, characterized in that: When the particle stress does not meet the ideal elastic-plastic criterion, the large deformation of the rock mass is calculated, and the particle velocity and density are updated and stored, including: Solve the artificial viscosity and artificial stress of each rock mass particle; According to the artificial viscosity, artificial stress and updated stress information, the mass conservation and momentum conservation equations under SPH discretization are solved to obtain the density change rate and the particle change rate; Update particle velocity and density information and store them.

7. A method for calculating the whole process of slope instability based on particle method according to claim 6, characterized in that: Perform time integration according to the Verlet algorithm, update the particle position and store it, including: After waiting for all threads to complete calculations, the Verlet algorithm is used to perform time integration, update the particle position information, and store it.

8. A slope instability calculation system based on particle method, characterized in that: The method comprises a first processor and a second processor, wherein: The first processor is used to obtain the STL real terrain data of the slope site and import it into the smooth particle fluid dynamics calculation framework; under the smooth particle fluid dynamics calculation framework, generate particles and divide the calculation domain and number the particles, and assign rock and soil parameter attributes to each particle; The second processor is used to receive the information transmitted by the first processor, and perform multi-threaded parallel calculation on all particles in a manner that each particle corresponds to one thread, wherein each particle performs the following cyclic calculation: calculate the stress rate and adaptive time step of each rock and soil particle, update the rock and soil particle stress and store it; when the particle stress does not meet the ideal elastic-plastic criterion, calculate the large deformation of the rock and soil, update the particle velocity and density and store it; perform time integration according to the Verlet algorithm, update the particle position and store it; when the set single-step output time is reached, output the stored particle stress, particle velocity, particle density and particle position; The first processor is further used to receive data output by the second processor and determine whether the current numerical simulation actually reaches a preset simulation time. If the preset simulation time is reached, the sequential calculation process of the second processor is stopped.

9. A device for calculating the whole process of slope instability based on particle method, characterized in that: It comprises a processor and a memory for storing a computer program that can be run on the processor, and when the processor is used to run the computer program, it executes the steps of a method for calculating the whole process of slope instability based on the particle method as described in any one of claims 1 to 7.

10. A computer storage medium storing a computer program, wherein: When the computer program is executed by a processor, the steps of a method for calculating the whole process of slope instability based on a particle method as described in any one of claims 1 to 7 are implemented.

Images