Parameterized design and optimization method of slope protection net based on real scene three-dimensional model

Abstract

The present application relates to the technical field of civil engineering and computer graphics, and particularly relates to a parameterized design and optimization method of slope protection net based on real scene three-dimensional model, comprising: collecting target slope region data, and constructing real scene three-dimensional terrain model; constructing protection net three-dimensional geometric surface model based on the real scene three-dimensional terrain model; constructing discrete element model facing dynamics analysis; constructing dynamics simulation environment, and performing dynamics simulation; and performing multi-target iterative optimization based on the dynamics simulation result. The present application provides a new method of slope protection net design and optimization integrating high-fidelity modeling, dynamic interactive design, high-precision simulation and multi-target optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of civil engineering and computer graphics, in particular to a parameterized design and optimization method for slope protection nets based on real scene three-dimensional models. BACKGROUND

[0002] High-precision real scene three-dimensional model technology can quickly and accurately restore the topography and geological structure of a slope by fusing unmanned aerial vehicle oblique photography, three-dimensional laser scanning and other means, and has gradually become a core tool for slope geological disaster prevention. However, the current application of this technology still has the following problems:

[0003] 1) Design and reality are disconnected. Traditional design is mostly carried out on two-dimensional drawings or simplified three-dimensional terrain, and the protection net model cannot accurately reflect its real spatial form in complex terrain, resulting in inaccurate engineering quantity estimation and deviation of protection effect from expectation.

[0004] 2) The design process is fragmented and inefficient. The layout, modeling, simulation analysis and scheme evaluation of the protection net are usually completed in different steps by different software, and the data import and export is cumbersome, and the information is blocked. Scheme adjustment often needs to start from scratch, the iteration cycle is long, and it is difficult to quickly compare multiple schemes.

[0005] 3) Lack of quantitative evaluation and scientific decision basis. The evaluation of the protection scheme depends on the experience of engineers, and lacks comprehensive quantitative evaluation of multiple key indicators such as safety, reliability and economy. It is difficult to make scientific trade-offs among multiple mutually restrictive targets (such as increasing the interception rate may increase the cost), and it is difficult to find the optimal solution.

[0006] 4) The simulation model is not accurate enough. The traditional rockfall simulation model simulates the rockfall source area and the flexible structure of the protection net roughly, and it is difficult to truly reflect the interaction mechanism of rockfall and complex terrain and flexible protection net.

[0007] As one of the mainstream discrete element methods, the particle discrete element method needs to use a complex three-dimensional model as the basis to generate block structure units or surface structure units in the pre-processing process. This process is relatively cumbersome, and the dynamic regulation and control of particle unit production through a parameterized and visual framework can promote the pre-processing of discrete element simulation from an experience-dependent method to an efficient, accurate and exploratory intelligent design process. SUMMARY

[0008] The purpose of the present application is to provide a new scheme for slope protection net design and optimization that integrates high-fidelity modeling, dynamic interactive design, high-precision simulation and multi-objective optimization, in view of the deficiencies in the above background technology.

[0009] In order to achieve the above object, the application provides a parameterized design and optimization method of slope protection net based on real scene three-dimensional model, comprising the following steps:

[0010] S1, collecting target slope region data, and constructing real scene three-dimensional terrain model;

[0011] S2, based on the real scene three-dimensional terrain model, planning the two-dimensional plane line position of the protection net, projecting the two-dimensional plane line position to the surface of the real scene three-dimensional terrain model to generate the bottom installation baseline of the protection net which is accurately fitted to the ground surface, extending along the slope normal direction or gravity direction of each point position on the baseline according to the preset protection net height parameter to generate the top boundary line, and connecting the bottom installation baseline and the top boundary line to form a curved surface, thereby constructing the three-dimensional geometric curved surface model of the protection net which is completely coupled with the real terrain;

[0012] S3, constructing the discrete element model for dynamic analysis, including the discrete element model of the protection net and the discrete element model of the rockfall source area, the discrete element model of the protection net is divided according to the components and each component is given corresponding physical and mechanical parameters, and the discrete element model of the rockfall source area generates rockfall particles in the corresponding space;

[0013] S4, constructing the dynamic simulation environment, configuring the gravity field and the contact model between objects, performing dynamic simulation, recording and analyzing the simulation results; the contact model includes the contact model between rockfall and slope, the contact model between rockfall and rockfall, and the contact model between rockfall and protection net;

[0014] S5, based on the dynamic simulation results, performing multi-objective iterative optimization to obtain a series of different design schemes and corresponding design evaluation indexes, selecting the design scheme to complete the parameterized design and optimization of the slope protection net.

[0015] Further, in S1, the obtained three-dimensional point cloud or image data of the target slope region is processed to generate a three-dimensional point cloud data model with reasonable density and accuracy, the three-dimensional point cloud data model is constructed into a triangular mesh model with real geographic coordinates and elevation information, and real texture mapping is performed to obtain a slope surface real scene three-dimensional terrain model which is completely consistent with the actual topography.

[0016] Further, in S2, according to the engineering design specification, the geological disaster risk assessment result and the terrain analysis result, one or more two-dimensional plane line positions of the protection net are planned and laid out in the two-dimensional plan view or three-dimensional view of the real scene three-dimensional terrain model to determine the trend of the protection net in the horizontal plane.

[0017] Further, in S2, the two-dimensional plane line is stretched in the direction perpendicular to the horizontal plane to construct one or more vertical virtual projection planes, and the intersection of the virtual projection plane and the real three-dimensional terrain model is calculated, resulting in one or more spatial three-dimensional intersection lines that accurately fit the real ground surface, to obtain the bottom installation baseline of the protective net on the real terrain. Starting from the bottom installation baseline, the top boundary line of the protective net is generated by extending a specified distance along the slope normal direction or the gravity direction of each baseline point, according to the design height of the protective net. The bottom installation baseline and the top boundary line are connected to generate a three-dimensional geometric curved surface model of the protective net.

[0018] Further, in S3, when constructing the discrete element model of the protective net, the model parameters are defined as , representing the chord length interval of the particles along the bottom installation baseline, representing the vertical interval of the particles, representing the vertical level of the particles, representing the height of the support column, representing the layout interval frequency of the support column, representing the visualization radius of the particles;

[0019] The continuous bottom installation baseline is discretized into a set of ordered reference points, and the reference points are repeatedly vertically translated to construct the transverse particle row;

[0020] Each reference point is used to generate each vertical particle column independently to construct the longitudinal particle column, and the complete protective net particle matrix is obtained by combining all the transverse particle rows and the longitudinal particle columns.

[0021] Further, when constructing the support column, based on the generated reference point set, the parameter is used as the step size to determine each pair of reference points corresponding to the adjacent two support columns, so that each pair of reference points is separated by a horizontal interval , and the vertical interval and height of the column particles are the same as those of the protective net itself.

[0022] Further, in S4, for the rockfall and slope body contact model, the slope surface terrain is set as a rigid boundary to prevent the rockfall from penetrating, and at each iteration step, it is judged whether each rockfall particle is located inside the closed grid of the slope body. When located inside, it is moved out by the following formula:

[0023] ;

[0024] wherein, represents the center point coordinates of the rockfall particle, is the coordinates of the point on the slope surface terrain closest to the center point of the rockfall particle, is the shortest vector that pushes the center point of the rock particle out of the solid; the bouncing behavior of the rock is controlled by a spring solver, which reverses the component of the rock velocity vector perpendicular to the collision surface when a collision is detected, while setting a restitution coefficient ; friction is simulated by decaying the tangential velocity component after a collision, introducing a dynamic friction coefficient , the larger the dynamic friction coefficient, the faster the tangential velocity decays, the more the rock tends to stop sliding;

[0025] For the rock-rock contact model, multiple rock- rock collisions and bounces are set up between rocks, for any two rock particles and , the distance between their center points and is calculated ; a collision is determined to occur if the following condition is satisfied:

[0026] ;

[0027] where and are the radii of the rock particles and , respectively;

[0028] For the rock-net contact model, the contact mechanics relationship between the rock particle and the discrete element model of the protective net is defined, simulating the process of energy transfer from the steel wire mesh to the pressure relief ring and the supporting structure when the impact occurs.

[0029] Further, in S4, the protective net is discretized into a flexible structure connected by a large number of particles through a contact model, and real material mechanics parameters are assigned; the particle cluster technique is used to rigidly bond multiple spherical particles, constructing a non-spherical block model that can reflect the real irregular shape, mass, center of mass, and moment of inertia of the rock; the slope surface is set as a boundary wall with real contact stiffness, friction coefficient, and restitution coefficient.

[0030] Further, in S3, three-dimensional design evaluation indicators are set: safety indicators, reliability indicators, and economic indicators;

[0031] The safety indicator is the interception rate calculated according to the number or mass of intercepted rocks; the reliability indicator is the impact energy margin obtained by comparing the maximum impact energy of the simulation with the energy absorption level of the protective net design; the economic indicator is the unit protective cost calculated according to the geometric parameters and material unit price of the protective net.

[0032] Further, in S5, the key design parameters of the protective net are taken as input variables, and the design evaluation index is taken as an output target to establish a parameterized optimization model, the input variables are adjusted, and S2 to S5 are repeatedly executed to obtain a plurality of design schemes and corresponding design evaluation indexes, and a Pareto optimal solution is screened from the plurality of design schemes.

[0033] The above scheme of the present application has the following beneficial effects:

[0034] The parameterized design and optimization method of the slope protective net based on the real three-dimensional model provided by the present application improves the precision and reality of the protective net design: the generated protective net model is completely coupled with the real terrain based on the high-precision real three-dimensional model, ensuring the accuracy of the design and avoiding the "distortion" problem of the traditional two-dimensional design.

[0035] The present application improves the fidelity of the simulation: high-precision modeling of the terrain, the protective net and the rockfall source area ensures the input quality of the dynamics simulation, thereby obtaining a simulation result closer to the real physical process, and providing a solid foundation for the reliability verification of the scheme.

[0036] The present application realizes the scientization and optimization of the decision: a quantitative evaluation system of "safety-reliability-economy" is established, and a multi-objective optimization method is introduced, which can make a trade-off decision from a series of Pareto optimal solutions, and say goodbye to simple experience design, and find the best balance point between safety and cost.

[0037] The present application realizes the dynamic interaction and efficiency improvement of the design: through parameterization and automation process, any adjustment of the design parameters can trigger the instant update of the model and the rapid feedback of the results, which can shorten the design iteration period of several days or even several weeks to several minutes or several hours, greatly improving the design efficiency.

[0038] Other beneficial effects of the present application will be described in detail in the subsequent specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 The step flow chart of the present application is shown in the figure;

[0040] Figure 2 The two-dimensional plane line position diagram of the present application is shown in the figure;

[0041] Figure 3 The virtual projection plane diagram of the present application is shown in the figure;

[0042] Figure 4 The protective net bottom installation baseline diagram of the present application is shown in the figure;

[0043] Figure 5 The projection diagram of the protective net entity interception position on the three-dimensional terrain of the present application is shown in the figure;

[0044] Figure 6 schematic diagram of a particle matrix of a protective net of the present application;

[0045] Figure 7 schematic diagram of a set of column reference points of the present application. DETAILED DESCRIPTION

[0046] The forgoing descriptions and embodiments are examples only of the present disclosure, and the present disclosure can be implemented or applied in other different embodiments without departing from the spirit of the present disclosure. It should be understood that the embodiments described above and the features thereof can be combined with each other without conflict, and all technical solutions obtained by the combination of the embodiments and the features thereof without departing from the spirit of the present disclosure shall fall within the scope of the present disclosure.

[0047] It should be noted that various aspects of the embodiments described below are within the scope of the appended claims. It should be apparent that the aspects described herein can be embodied in a wide variety of forms and that any specific structure and / or function described herein is merely illustrative. Based on the teachings herein one skilled in the art should appreciate that an aspect described herein can be implemented independently of any other aspects and that an aspect described herein can be implemented both as any number of software and / or hardware structures and as any number of processes and / or operations. For example, an aspect can be implemented as a software program running on hardware that can process information such as, for example, information stored in a computer-readable storage medium. As another example, an aspect can be implemented as a software program running on hardware that can process information such as, for example, information stored in a computer-readable storage medium.

[0048] It should also be noted that the figures provided in the following embodiments are only schematic and that the dimensions of the figures can not be to scale with one another. Further, the following embodiments can have been simplified for a better understanding and ease of disclosure. One skilled in the art will appreciate that the drawings provided herein are meant to be illustrative only and other embodiments can be implemented without departing from the scope of the present disclosure.

[0049] As Figure 1As shown, the embodiment of the present application provides a parameterized design and optimization method of slope protection net based on real three-dimensional model, taking typical rock slope as the research object, other types of slope can also use this method for parameterized design and optimization of slope protection net. The method specifically includes the following steps:

[0050] S1, collecting target slope region data, constructing high-precision real three-dimensional terrain model.

[0051] In this step, first, the high-precision three-dimensional point cloud or image data of the target slope region is obtained, and then the data is processed to generate a slope surface real three-dimensional terrain model consistent with the real topography as the digital model basement for all subsequent steps.

[0052] For example, high-precision three-dimensional point cloud data or image data can be collected by unmanned aerial vehicle equipped with high-resolution oblique photography camera on target slope region. The unmanned aerial vehicle needs to follow the preset route to collect three-dimensional (laser) point cloud data or image data of the slope from multiple angles and high overlap, ensuring no data blind area.

[0053] When processing the data, the collected data is imported into the corresponding processing software, and the collected data is processed by the processing software for air triangulation, multi-view image matching or point cloud registration, denoising and other operations, so as to generate high-density and high-precision three-dimensional point cloud data model. Then the three-dimensional point cloud data model is constructed into a triangular mesh model with real geographic coordinates and elevation information, and real texture mapping is performed, and finally the slope surface real three-dimensional terrain model completely consistent with the actual topography is obtained.

[0054] S2, constructing a three-dimensional geometric model of the protection net based on the real three-dimensional terrain model.

[0055] In this step, first, the two-dimensional plane line of the protection net is operated and preliminarily planned in the software. Then the two-dimensional plane line is projected onto the surface of the real three-dimensional terrain model to generate a spatial three-dimensional protection net bottom installation baseline accurately fitted to the ground surface. Then, based on the baseline, according to the preset protection net height parameter, the top boundary line is extended along the slope normal direction or gravity direction of each point on the baseline, and the curved surface is connected between the bottom installation baseline and the top boundary line to construct a three-dimensional geometric curved surface model of the protection net completely coupled with the real terrain.

[0056] For example, this step can be completed in the 3D modeling software Rhinoceros (hereinafter referred to as Rhino) and its parametric plugin Grasshopper. First, import the real-world 3D terrain model generated by S1 (usually in .obj or .osgb format) into Rhino. Perform elevation, slope, and aspect analysis on the model, generating visualization files (such as elevation rendering maps, contour maps, etc.). Based on engineering design specifications, geological hazard risk assessment results (such as potential rockfall source areas, trajectory predictions), and terrain analysis results, preliminarily plan and lay out the 2D planar alignment of one or more protective nets in the 2D top view or 3D view of the model, determining the orientation of the protective nets on the horizontal plane, such as... Figure 2 As shown.

[0057] The obtained two-dimensional planar alignment is then precisely projected onto the surface of the real-world three-dimensional terrain model to initially determine the bottom installation position of the protective netting. Specifically, as follows... Figure 3 As shown, the two-dimensional plane line position Along the vertical plane (i.e.) Stretch along the axis to construct one or more vertical virtual projection planes. Calculate the virtual projection plane. With real-world 3D terrain model The intersection of these lines results in one or more spatial three-dimensional intersection lines that precisely conform to the undulations of the actual ground surface. These intersection lines serve as the baseline for the bottom installation of the protective netting structure on the real terrain. ,like Figure 4 As shown.

[0058] More specifically, in this embodiment, the point coordinates read from the 3D point cloud data model are: ,in , It is the total number of points. Let it be. Coordinates are Set initial value The generated 3D point set can be represented as , where each point The coordinates are These points are all located in of On a plane. The generated two-dimensional plane lines. It is a piecewise linear curve connecting these points.

[0059] Two-dimensional plane line position When stretching along a direction perpendicular to the horizontal plane, let the stretching height be... ,Pick The stretching vector is The first stretched surface generated It can be represented as:

[0060] ;

[0061] wherein, is a parameter along the curve, is a parameter in the direction of stretching, and . The first stretched surface can be regarded as a vertical surface with as the bottom line, extending height in the positive direction of axis.

[0062] Translate the two-dimensional planar line in the direction of , set the translation distance in the direction of , and , the translation vector is , and the translated curve can be expressed as Therefore, similarly, the generated second stretched surface can be expressed as:

[0063] ;

[0064] wherein, is another vertical surface parallel to , offset by a distance of in the direction of axis.

[0065] Convert the continuous first stretched surface S1 and the second stretched surface S2 into meshes, and represent the two meshes as and , and , The terrain mesh of the real three-dimensional terrain model can be expressed as , and the intersection of and is a series of discontinuous line segments:

[0066] ;

[0067] These line segments need to be connected to form one or more continuous multi-segment lines, called the bottom installation baseline:

[0068] ;

[0069] Similarly, calculate the intersection of and , and finally get:

[0070] ;

[0071] ;

[0072] wherein, is the projection of the interception position of the protective screen entity on the three-dimensional terrain, and is achieved by and to quickly build the interception effect of the protective screen entity simulation protective screen, as shown in Figure 5 .

[0073] Finally, taking the bottom installation baseline as the starting edge, according to the design height of the protective screen, extending a specified distance along the normal direction of the slope surface or the gravity direction of each baseline point, the top boundary line of the protective screen is generated. Connecting the bottom installation baseline and the top boundary line, one or more protective screen three-dimensional geometric surface models are generated, which are fully coupled with the real terrain and have a specified height and trend, and the model truly reflects the form and position of the protective screen in space.

[0074] S3, build a discrete element model (DEM) for dynamic analysis.

[0075] In this step, the protective screen three-dimensional geometric surface model and the rockfall source area model dynamically and automatically generated on the real three-dimensional terrain model through adjustable parameters (such as slope, curvature threshold, etc.) are converted into a discrete element model composed of discrete particles. Among them, the protective screen three-dimensional geometric surface model is divided according to its components (such as support columns, steel wire meshes, etc.) to generate a discrete element model respectively, and independent physical and mechanical parameters are assigned to each component; the rockfall source area model generates one or more rockfall particles in the corresponding space through a volume filling algorithm.

[0076] It should be noted that the rockfall source area model generation can be divided into the following steps: determining the rockfall area according to geological exploration data, surface weathering information and main control structure surface position; obtaining a plurality of closed rockfall area grid entities through grid Boolean operation according to the rockfall area and structure surface position; discretizing the continuous grid entity into a three-dimensional grid, which is composed of many tiny cubic units, which are called "voxels"; taking all the selected voxel center points as the center of the sphere, and taking half of the voxel size as the radius, generating a sphere to obtain a spherical particle combination. Subsequently, the spherical particle combination is converted into a discrete element model.

[0077] More specifically, in this embodiment, when building a discrete element model of the protective screen, the model parameters are defined as , wherein represents the chord length interval (horizontal interval) of the particles along the bottom installation baseline, represents the vertical interval (vertical interval) of the particles, represents the vertical level of the particles, Indicates the height of the supporting column. Indicates the frequency of the support column installation (per (One support column is set at each reference point). This represents the visual radius of the particle.

[0078] Continuous bottom mounting baseline Discretize into a set of ordered reference points :

[0079] ;

[0080] The generation of this set of points satisfies: , that is, the starting point; for any ,point It is on the curve From Moving forward The point obtained by the chord length distance, i.e. ,in This represents the integer index of the reference point. It is understandable that... The 0th row of particles constitutes the entire protective net, that is, the bottommost row of particles.

[0081] When constructing the other granular rows of the protective net, it is done by repeatedly vertically translating the reference point row. Therefore, for each vertical level index... Generate a corresponding horizontal granular row. Obviously, line 0 The reference point set itself, .set up for Unit vector of the axis The vector of a single vertical translation .therefore, By base row each point translates upwards times The distance is obtained, as follows:

[0082] ;

[0083] All horizontal particle rows The union of these points constitutes a complete set of horizontal protection points.

[0084] For vertical particle columns, each reference point is considered as the starting point of a "pillar," and each vertical particle column is generated independently. any reference point in , take it as the first The generation of a column of particles starts with a point. Thus, the first A column of particles is also a set of points, generated from a reference point along a step size of , repeated times, as follows:

[0085] ;

[0086] The union of all longitudinal columns of particles also constitutes the complete set of points of the longitudinal fence.

[0087] Obviously, the combination of a transverse row of particles and a longitudinal column of particles results in a complete matrix of particles of the fence , as shown in Figure 6 . Let any particle be uniquely defined by its transverse index and longitudinal index , the coordinates of this point can be generated by the following functions:

[0088] ;

[0089] Thus, the complete matrix of particles of the fence can be expressed in set theory as:

[0090] ;

[0091] Thus, the fence particles are decomposed and constructed.

[0092] When constructing the support posts, based on the generated set of reference points , each pair of reference points corresponding to two adjacent support posts is determined with a step size of , so that each pair of reference points is separated by a horizontal distance of . Thus, the first pair of reference points is , the second pair of reference points is , and so on, the th pair of reference points (counting from 0) is , and the pair of reference points can also be expressed in set theory as:

[0093] ;

[0094] where is the set of natural numbers (including 0), and each element in is an ordered pair containing two particles, limiting the position and outer diameter of the post, etc.

[0095] The vertical spacing and height of the post particles are the same as those of the main steel mesh, i.e., for any post reference point​​ its corresponding vertical particle sequence generated by the following function:

[0096] ;

[0097] Therefore, for each column reference point (i.e. the particle in the ordered pair), a set of points is generated, starting from this point, extending upwards with a step size of to a height of , as shown in Figure 7 . The complete point set of all column particles is the union of all individual column particle sequences .

[0098] In addition, in order to facilitate the import of these geometric entities into a computational script (such as Python), the above mathematical set is further transformed into an ordered list or array in this embodiment. Specifically, the point set is transformed into a one-dimensional list with the following structure:

[0099] ;

[0100] Each element in this list is a three-dimensional coordinate point , and the total length of the list is . For the column particle point set, it is transformed into a two-dimensional list, with each sub-list representing an independent column, and the structure is as follows:

[0101] ;

[0102] where is the point in , i.e. the point column reference point.

[0103] For each point in , a sphere is constructed with as the center and as the radius, and the union of all spheres constitutes the visualization model of the discrete element of the protective net.

[0104] S4, a dynamic simulation environment is constructed, the gravity field and the contact model between objects are configured, dynamic simulation is performed, the whole process of rockfall starting from the rockfall source area, moving along the slope surface and impacting the protective net is simulated, the simulation results are recorded and analyzed, and key data such as rockfall motion trajectory, impact position and impact energy are extracted.

[0105] It should be noted that the contact model between objects includes the contact model between rockfall and slope body, rockfall and rockfall, and rockfall and protective net. Considering the actual situation, these contact models are set in different ways. For example, for rockfall bouncing behavior, an elastic solver (Bouncy Solver) based on explicit dynamics is used for simulation in the embodiment. The solver can maintain the kinetic energy of the system and accurately simulate the elastic collision and momentum transfer between rigid bodies. By setting a series of physical targets such as gravity, entity collision, and sphere collision, and inputting them into the elastic solver, the whole process of rockfall from instability start to movement along the slope surface to the final impact with the protective structure can be simulated dynamically with high fidelity.

[0106] For the rockfall and slope body contact model, the Points representing the rockfall and the Solid of the closed mesh of the slope body are connected to the SolidPointCollide component, and the slope surface terrain is set as a rigid boundary to prevent the rockfall from passing through. In each iteration step, the component determines whether each rockfall particle is located inside the closed mesh of the slope body. When located inside, it is moved out by the following formula:

[0107] ;

[0108] wherein, represents the center point coordinates of the rockfall particle, is the point on the slope surface terrain closest to the center point of the rockfall particle, is the shortest vector to "push" the center point of the rockfall particle out of the entity.

[0109] The bouncing behavior of the rockfall is controlled by the Bouncy Solve (elastic solver). When a collision is detected, the elastic solver reverses the component of the rockfall velocity vector perpendicular to the collision surface, while setting the restitution coefficient , which is a parameter between 0 and 1, represents a completely inelastic collision (no bounce), represents a completely elastic collision. Friction is simulated by decaying the tangential velocity component after collision, further introducing a dynamic friction coefficient , which is a parameter between 0 and 1. The larger the dynamic friction coefficient, the faster the tangential velocity decays, and the more the rockfall tends to stop sliding.

[0110] For the rockfall and rockfall contact model, the Points representing the rockfall are connected to the SphereCollide component to allow multiple rockfalls to collide and bounce off each other. For any two rockfall particles and , the component calculates their center point coordinates and the distance between the two. If the following condition is satisfied, it is determined that a collision (overlap) has occurred:

[0111] ;

[0112] wherein and are the radii of the rockfall particle and , respectively.

[0113] For the rockfall and protective net contact model, the contact mechanics relationship between the rockfall particle and the discrete element model of the protective net is defined to accurately simulate the relatively complex process of energy transmission through the steel wire mesh to the pressure relief ring and support structure when the impact occurs. Therefore, the Points representing the rockfall and the Solid representing the protective net are connected to the SolidPointCollide component to simulate the impact of the rockfall on the protective net.

[0114] When simulating, the dynamic solver is started, and from time zero, the forces, accelerations, velocities, and displacements of all particles are iteratively calculated at each infinitesimal time step until all particles stop moving or exceed the calculation boundary. The motion trajectories of the rockfall obtained from multiple simulations are superimposed to form a three-dimensional motion trajectory envelope. The envelope is compared with the discrete element model (or three-dimensional geometric surface model) of the protective net in space to evaluate whether the layout position of the protective net can effectively intercept most of the high-risk rockfall motion trajectories. When it cannot effectively intercept, return to S1 to redesign the protective net.

[0115] Then, higher precision model optimization and simulation are performed. The protective net is discretized into a flexible structure composed of a large number of particles connected by a contact model (such as a linear spring or parallel connection model), and is given real material mechanics parameters (such as the elastic modulus, yield strength of steel wire, force-displacement curve of pressure relief ring, etc.). The particle cluster technique is used to rigidly bond multiple spherical particles to construct a non-spherical block model that can reflect the irregular shape, mass, center of mass, and moment of inertia of real rockfall. The slope surface is set as a boundary wall (Wall) with real contact stiffness, friction coefficient, and recovery coefficient.

[0116] Based on further optimization of the model, complete dynamic simulation calculation is performed, and the software will iteratively solve the motion of the rockfall and the complex nonlinear interaction between the slope surface and the protective net based on Newton's second law at an infinitesimal time step, record and analyze the simulation results, and extract key data such as the rockfall motion trajectory, impact position, and impact energy.

[0117] S5, based on the results of the dynamic simulation, multiple target iterative optimization is performed to obtain a series of different design schemes and their corresponding design evaluation indicators, and the corresponding design scheme is selected to complete the parametric design and optimization of the slope protection net.

[0118] In this step, the design evaluation index of at least three dimensions is calculated: safety index, reliability index, and economy index. Among them, the safety index is the interception rate calculated according to the number or mass of intercepted falling rocks; the reliability index is the impact energy margin obtained by comparing the maximum impact energy in simulation with the energy absorption level of the protective net design; and the economy index is the unit protection cost calculated according to the geometric parameters of the protective net and the unit price of the material. Then, the key design parameters of the protective net are taken as input variables, and the design evaluation index is taken as an output target to establish a parameterized optimization model. The input variables are systematically adjusted, and steps S2 to S5 are repeatedly executed to obtain multiple design schemes and their corresponding design evaluation indexes, and the Pareto optimal solution is selected from them to provide a scientific basis for the final decision.

[0119] More specifically, the interception rate in the present embodiment is calculated as the percentage of the number (or total mass) of falling rocks successfully intercepted by the protective net to the total number (or total mass) of falling rocks from the falling rock source area. By analyzing the final resting position of the falling rocks, the number of falling rocks landing above the protective net (intercepted) and penetrating / overcoming / bypassing the protective net is counted, and the interception rate is calculated according to the formula: interception rate = (number of intercepted falling rocks / total number of falling rocks) 100% is calculated. This index directly reflects the effectiveness of the protective net layout position.

[0120] The calculation method of the impact energy margin is the difference or ratio between the design nominal energy absorption level of the protective net and the maximum impact energy it withstands in simulation. By comparing the maximum single impact energy or cumulative impact energy obtained from simulation with the rated energy absorption level (such as 3000 kJ level) of the selected protective net model provided by the national standard or manufacturer, the rationality of protective net selection and the safety of the structure are determined.

[0121] The calculation method of the unit protection cost is the cost of materials and construction invested to achieve the current protection effect. According to the design parameters of the protective net scheme (such as total length, height, column spacing, steel wire diameter, number of pressure relief rings, etc.), combined with the material unit price and construction rate, a cost estimation function is established. This index is used for horizontal comparison of the economy of different design schemes.

[0122] For the input of the parameterized optimization model, it is the key design parameters of the protective net, including installation baseline coordinates, protective net height, column spacing, material model, etc. The optimization target of the parameter optimization model is: interception rate, impact energy margin, and unit protection cost. In the optimization process, multiple scheme comparison analysis (parameterized research) is performed, a series of different design schemes and their corresponding design evaluation indexes are obtained by systematically adjusting the input variables and running multiple simulations.

[0123] In the multi-objective trade-off decision, the "interception rate", "impact energy margin" and "unit protection cost" of all schemes are plotted in a multi-dimensional coordinate system to form a design scheme space. According to the specific requirements of the project (such as risk level, budget limit), one or more Pareto optimal solutions (Pareto Optimal Solutions) are selected from the scheme space.

[0124] In specific operation, the optimization result of the design evaluation index is presented in the form of a three-dimensional scatter plot, and the three coordinate axes represent the three optimization objectives. All "non-inferior solutions" (i.e. Pareto front) can be clearly seen on the three-dimensional scatter plot. The decision maker selects according to external constraints and value judgment, for example, the lowest cost scheme can be selected from the schemes with an interception rate of more than 98%; or under the limit of a budget of 2 million, the scheme with the best comprehensive performance of interception rate and energy impact margin is selected, so as to make a final design decision that takes into account safety, reliability and economy, and efficiently, accurately and scientifically complete the design and optimization of the slope protection net.

[0125] Based on the same inventive concept, the embodiment also provides a slope protection net parameterized design and optimization system based on a real scene three-dimensional model, which comprises a model construction module, a discrete element conversion module, a dynamics simulation module and an optimization decision module. The model construction module is used to generate a slope surface real scene three-dimensional terrain model consistent with the real topography, and then generate a protection net three-dimensional geometric model. The discrete element conversion module is used to convert the protection net three-dimensional geometric model into a discrete element model for dynamics simulation. The dynamics simulation module is used to construct a dynamics simulation environment, configure a gravity field and a contact model between objects, and perform dynamics simulation on the discrete element model. The optimization decision module is used to calculate design evaluation indexes, perform multi-objective iterative optimization, and obtain a series of different design schemes and corresponding design evaluation indexes.

[0126] The system provided by the present application has the same inventive concept and beneficial effects as the above-mentioned method, and will not be described here.

[0127] The technical features of the above embodiments can be combined in any way. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not contradict, they should be considered as within the scope of the present application.

[0128] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the application. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A parametric design and optimization method for slope protection nets based on a real-world 3D model, characterized in that, Includes the following steps: S1: Collect data on the target slope area and construct a realistic 3D terrain model; S2, based on the real-world 3D terrain model, plans the 2D plane alignment of the protective net, projects the 2D plane alignment onto the surface of the real-world 3D terrain model, generates the bottom installation baseline of the protective net that accurately fits the undulations of the ground, and extends along the slope normal or gravity direction of each point on the baseline according to the preset height parameters of the protective net to generate the top boundary line, and connects the bottom installation baseline and the top boundary line to form a curved surface, constructing a 3D geometric curved surface model of the protective net that is completely coupled with the real terrain; S3, construct a discrete element model for dynamic analysis, including a protective net discrete element model and a rockfall source area discrete element model. The protective net discrete element model is divided according to the components and each component is assigned corresponding physical and mechanical parameters. The rockfall source area discrete element model generates rockfall particles in the corresponding space. S4: Construct a dynamic simulation environment, configure the gravity field and contact models between various objects, perform dynamic simulation, record and analyze the simulation results; the contact models include the contact model between the falling rock and the slope, the contact model between falling rocks, and the contact model between the falling rock and the protective net. S5, based on the dynamic simulation results, performs multi-objective iterative optimization to obtain a series of different design schemes and corresponding design evaluation indicators. Selecting one of the design schemes completes the parametric design and optimization of the slope protection net.

2. The parametric design and optimization method for slope protection nets based on a real-scene 3D model according to claim 1, characterized in that, In S1, the acquired 3D point cloud or image data of the target slope area is processed to generate a 3D point cloud data model with reasonable density and accuracy. The 3D point cloud data model is then constructed into a triangular mesh model with real geographic coordinates and elevation information, and real texture mapping is performed to obtain a real 3D terrain model of the slope surface that is completely consistent with the actual landform.

3. The parametric design and optimization method for slope protection nets based on a real-scene 3D model according to claim 1, characterized in that, Based on engineering design specifications, geological hazard risk assessment results, and terrain analysis results, in S2, one or more protective nets are planned and laid out in the two-dimensional top view or three-dimensional view of the real-scene three-dimensional terrain model to determine the direction of the protective nets on the horizontal plane.

4. The parametric design and optimization method for slope protection nets based on a real-scene 3D model according to claim 3, characterized in that, In S2, the two-dimensional planar lines are stretched along a direction perpendicular to the horizontal plane to construct one or more vertical virtual projection planes. The intersection of the virtual projection planes and the real three-dimensional terrain model is calculated, resulting in one or more spatial three-dimensional intersection lines that accurately fit the undulations of the real surface. The bottom installation baseline of the protective net on the real terrain is obtained. Taking the bottom installation baseline as the starting edge, according to the design height of the protective net, the top boundary line of the protective net is generated by extending upwards a specified distance along the slope normal direction or gravity direction of each baseline point. The bottom installation baseline and the top boundary line are connected to generate a three-dimensional geometric surface model of the protective net.

5. The parametric design and optimization method for slope protection nets based on a real-scene 3D model according to claim 1, characterized in that, When constructing a discrete element model of a protective net in S3, define the model parameters. , This indicates the chord-length spacing of the particles along the bottom mounting baseline. This indicates the spacing between particles in the vertical direction. Indicates the vertical particle hierarchy. Indicates the height of the supporting column. This indicates the frequency of the support columns' placement. Represents the visual radius of the particle; The continuous bottom mounting baseline is discretized into a set of ordered reference points, and the reference points are repeatedly translated vertically to construct the lateral particle rows. Each reference point is used to generate a vertical column of particles independently, thus constructing a vertical particle column. All horizontal particle rows and vertical particle columns are combined to obtain a complete protective net particle matrix.

6. The parametric design and optimization method for slope protection nets based on a real-scene 3D model according to claim 5, characterized in that, When constructing the supporting columns, based on the generated set of reference points, parameters are used. The step size is used to determine each pair of reference points corresponding to two adjacent support columns, so that each pair of reference points is spaced by a horizontal distance. The vertical spacing and height of the column particles are the same as those of the protective net itself.

7. The parametric design and optimization method for slope protection nets based on a real-scene 3D model according to claim 1, characterized in that, In S4, for the rockfall-slope contact model, the slope surface is set as a rigid boundary to prevent rocks from crossing. In each iteration step, each rockfall particle is analyzed. Whether it is located inside the closed grid of the slope, if it is inside, it is moved out using the following formula: ； in, This indicates the coordinates of the center point of the falling rock particles. These are the coordinates of the point on the slope surface closest to the center of the fallen rock particle. It is the shortest vector that pushes the center point of the falling rock out of the entity; the bouncing behavior of the falling rock is controlled by the elasticity solver. When a collision is detected, the elasticity solver reverses the component of the falling rock's velocity vector perpendicular to the collision surface, and simultaneously sets the coefficient of restitution. Friction is simulated by attenuating the tangential velocity component after a collision, and a coefficient of dynamic friction is introduced. The higher the coefficient of kinetic friction, the faster the tangential velocity decays, and the more likely the falling rock is to stop sliding. For the rockfall contact model, multiple rocks are allowed to collide and bounce off each other. For any two rockfall particles... and Calculate the coordinates of their center points. and Distance between A collision is considered to have occurred if the following conditions are met: ； in and These are rock fragments and The radius; For the contact model between falling rocks and protective netting, the contact mechanics relationship between the falling rock particles and the discrete element model of the protective netting is defined to simulate the process of energy being transferred through the wire mesh to the pressure relief ring and the supporting structure when an impact occurs.

8. The parametric design and optimization method for slope protection nets based on a real-scene 3D model according to claim 7, characterized in that, When performing accurate dynamic simulations in S4, the protective net is discretized into a flexible structure composed of a large number of particles connected by a contact model, and given realistic material mechanical parameters. Particle cluster technology is used to rigidly bond multiple spherical particles to construct a non-spherical block model that can reflect the irregular shape, mass, center of mass and rotational inertia of real falling rocks. The slope is set as a boundary wall with realistic contact stiffness, friction coefficient and restitution coefficient.

9. The parametric design and optimization method for slope protection nets based on a real-scene 3D model according to claim 1, characterized in that, S3 sets three dimensions of design evaluation indicators: safety indicators, reliability indicators, and economic indicators; The safety index is the interception rate calculated based on the number or mass of the intercepted rocks; the reliability index is the impact energy margin obtained by comparing the simulated maximum impact energy with the energy absorption level of the protective net design. The economic indicator is the unit protection cost calculated based on the geometric parameters of the protective net and the unit price of the materials.

10. The parametric design and optimization method for slope protection nets based on a real-scene 3D model according to claim 9, characterized in that, In S5, the key design parameters of the protective netting are used as input variables, and the design evaluation index is used as the output target. A parameterized optimization model is established, the input variables are adjusted, and S2 to S5 are repeated to obtain multiple design schemes and corresponding design evaluation indexes. The Pareto optimal solution is then selected from these schemes.

Images