Dangerous rock falling motion feature analysis method based on discrete element simulation

Abstract

The invention discloses a dangerous rock falling motion feature analysis method based on discrete element simulation, and belongs to the technical field of engineering geology. Comprising the steps that dangerous rock data are collected through multiple means and preprocessed, and a three-dimensional visual dangerous rock mass initial model is constructed; based on the three-dimensional visual dangerous rock mass initial model, constructing a discrete element model, defining contact types and parameters among joint units by adopting a Hertz-mindlin contact theory, and determining normal contact force; based on the discrete element model and the normal contact force, a discrete element simulation technology is adopted, multi-physical field factors are considered to simulate dangerous rock falling movement, a discrete element model simulation result is obtained, a dynamic updating method is adopted for joint failure judgment, and the discrete element model is verified through an indoor physical model test and field monitoring data. And designing and implementing a prevention project according to a simulation result, thereby effectively preventing dangerous rock falling disasters and ensuring project safety.

Description

Technical Field

[0001] The present invention belongs to the technical field of engineering geology, and in particular relates to a method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation. Background Art

[0002] In mountainous engineering construction and operations, rockfalls pose a serious threat to personnel safety and the stable operation of infrastructure. According to relevant data, these disasters occur frequently, causing significant losses to transportation, water conservancy, and other projects. Accurately analyzing the movement characteristics of rockfalls is crucial for developing effective prevention and control measures.

[0003] Currently, methods for analyzing the movement characteristics of dangerous rockfalls are constantly evolving. Traditionally, the risk of dangerous rockfalls is assessed through on-site investigations and empirical judgment, but this approach is highly subjective and has limited accuracy. Technological advances have enabled the application of modern survey methods, such as 3D digital photography and drone aerial photography, to obtain basic information about dangerous rock masses, providing data support for analysis.

[0004] In order to explore a more effective impact force algorithm, some scholars proposed an MDR algorithm for rockfall impact force based on the concept of dimensionality reduction, combined with the mechanical properties, geometric relationships and energy conservation principle of rockfall impact cushions. A dimensionless algorithm was proposed based on indoor model tests and the dimensionless concept. Through Yang Qixin's experimental algorithm and the expansion coefficient concept, a triangle correction algorithm, a sine correction algorithm, and a numerical simulation correction algorithm for rockfall impact force were proposed respectively ("Derivation of rockfall impact force based on dimensionality reduction concept and dimensionless solution", Wang Xing et al., Science, Technology and Engineering, Issue 34, 2023).

[0005] In terms of simulation analysis, three-dimensional rockfall analysis and two-dimensional rockfall numerical calculations are commonly used. Three-dimensional rockfall analysis comprehensively captures the movement trends of falling rocks in complex terrain, considering the impact of factors such as slope surface hardness and terrain gradient on rockfall bounce height, movement speed, and impact kinetic energy. Two-dimensional rockfall numerical calculations select representative cross-sections for detailed analysis, combining on-site terrain conditions and different regional safety protection levels to provide a basis for the selection and placement of protective nets.

[0006] Guangqi Chen et al. from Kyushu University in Japan compared and analyzed the key advantages of two-dimensional (2D) and three-dimensional (3D) discontinuous deformation analysis (DDA) numerical simulations in rockfall analysis. Specific tests compared the performance of 2D and 3D DDA rockfall analysis in predicting trajectory and dynamic behavior. They concluded that 3D DDA simulations are more suitable for rough, sloping slopes with dense vegetation, while 2D DDA simulations have advantages in simulating certain specific conditions, such as river valley slopes. Other researchers used Pix4D software to construct high-precision 3D simulation models of slopes and dangerous rock formations collected using drones and 3D laser scanners. They then used Unity3D to simulate and analyze the motion characteristics and accumulation locations of rockfall under different conditions. The results showed that the motion patterns of plateau rockfalls based on 3D terrain simulations are more realistic and accurate. The larger the rock mass and the slope gradient, the greater the rockfall velocity, acceleration, and displacement. The greater the sphericity of the rockfall shape, the greater the rockfall velocity, acceleration, and displacement. The movement speed and acceleration of falling rocks are greater and the displacement is greater than that of toppling and sliding rocks ("Analysis of the Movement Characteristics of Plateau Rockfall Based on Three-Dimensional Terrain Simulation", Ye Tangjin et al., Plateau Science Research, Issue 2, 2022).

[0007] However, traditional simulation methods have certain defects and are difficult to accurately consider complex mechanical behaviors such as the interaction between dangerous rock blocks, energy loss during rockfall collisions, and fragmentation. For example, although existing literature tries to consider practical factors when constructing models, it is still inevitable to simplify the complex geological conditions and rockfall movement process. For example, when simulating the interaction between rockfall and slope, it is difficult to accurately consider factors such as the micro-topography undulations of the slope and the heterogeneity of the rock mass, resulting in deviations between the model and the actual situation, affecting the accuracy of the simulation results. Moreover, many parameters in the model (such as the friction coefficient, restitution coefficient, etc.) are usually determined through empirical values or simple experiments, and there is a certain degree of uncertainty. Slight changes in these parameters may have a significant impact on the simulation results of the rockfall movement characteristics.

[0008] Discrete element numerical simulation technology is gaining increasing application in geotechnical engineering, offering unique advantages in simulating interparticle interactions. However, comprehensive research on its application to the analysis of rockfall motion characteristics remains inadequate and systematic. While some existing studies have attempted to incorporate discrete element methods, they remain limited in areas such as data acquisition and integration, simulation model construction, and integration with real-world projects. Summary of the Invention

[0009] The purpose of the present invention is to address some technical problems existing in the prior art. The present invention focuses on the prevention and control of dangerous rock masses in the inlet area of the flood discharge system of a water conservancy project. The overall concept is to first use a variety of advanced technical means to collect multi-source data, and then construct a three-dimensional visual initial model of the dangerous rock mass after pre-processing. Then, the dangerous rock mass is divided into joint units, various parameters are reasonably assigned and joint fissures are processed, and discrete element simulation technology is used to simulate the movement of dangerous rockfalls by considering multi-physical field factors. The results are subjected to motion feature extraction and probability analysis, and protective measures are optimized accordingly. Finally, based on the simulation results, a prevention and control project is designed and implemented, including overall protection, removal and reinforcement of dangerous rock masses, stone protection and management, etc., and the prevention and control project volume is counted, so as to effectively prevent and control dangerous rockfall disasters and ensure project safety.

[0010] Specifically, the present invention provides a method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation, comprising the following steps:

[0011] S1. Collect and pre-process dangerous rock data through multiple means to construct a 3D visual initial model of dangerous rock mass;

[0012] S2. Based on the three-dimensional visualized initial model of the dangerous rock mass, a discrete element model is constructed and the contact type and parameters between joint units are defined using the Hertz-Mindlin contact theory to determine the normal contact force;

[0013] S3. Based on the discrete element model and the normal contact force, discrete element simulation technology is used to simulate the rockfall motion of dangerous rocks by considering multiple physical field factors, and discrete element model simulation results are obtained. Joint failure is judged by a dynamic update method, and the discrete element model is verified through indoor physical model tests and field monitoring data.

[0014] S4. Design and implement prevention and control projects based on discrete element model simulation results.

[0015] Preferably, S1 includes obtaining dangerous rock mass data and its surrounding environment information by using surveying and mapping technology and satellite remote sensing; cleaning the collected data, and extracting features of the dangerous rock mass and surrounding stable rock mass in multispectral images by using image processing and machine learning technology, classifying the images by using supervised classification or unsupervised classification algorithms, comparing and correcting the classification results with the actual situation on site, and fusing them in a unified geographic coordinate system through geographic information system technology to construct a three-dimensional visual initial model of the dangerous rock mass.

[0016] Preferably, the S2 specifically includes the following steps:

[0017] S21, acquiring joint and fissure data and establishing a joint and fissure database, and dividing the dangerous rock mass into a plurality of joint units based on the three-dimensional visualized dangerous rock mass initial model according to the distribution characteristics of the joints and fissures;

[0018] S22. Based on the classification results of S1 and the physical and mechanical properties of the dangerous rock mass, assign physical characteristics to each joint unit;

[0019] S23. The Hertz-Mindlin contact theory is used to define the contact type and parameters between joint elements and determine the normal contact force.

[0020] Preferably, the normal contact force is specifically expressed as:

[0021]

[0022] in: is the equivalent elastic modulus, E1 and E2 are the elastic moduli of the two contact elements, v1 and v2 are Poisson's ratios, is the equivalent radius, R1 and R2 are the contact element radii, δ n is the normal overlap; set the friction coefficient and joint unit parameters according to the slope surface hardness and joint crack distribution.

[0023] Preferably, in S3, the joint damage judgment includes calculating the normal stress and shear stress of the joint surface in each time step according to the relative position and force of the falling rock and the joint, comparing the shear stress with the shear strength of the joint, and judging whether the joint is damaged. When the shear stress is greater than the shear strength of the joint, the joint is damaged; when the shear stress is less than the shear strength of the joint, the joint is not damaged.

[0024] Preferably, after the joint damage is determined, the additional force on the rockfall is calculated according to the mode and degree of joint damage to adjust the motion trajectory and mechanical properties of the rockfall; the discrete element method is used to simulate the motion and interaction between the broken rock mass and the rockfall, and the broken rock mass is regarded as multiple discrete units to couple with the rockfall to calculate the additional force.

[0025] Preferably, the indoor physical model test includes making a physical model that restores the actual dangerous rock mass, simulating the rockfall movement process, recording the movement trajectory, speed and joint damage of the rockfall, and comparing and analyzing the indoor physical model test results with the discrete element model simulation results. If there is a deviation between the simulation results of the discrete element model and the actual situation, the parameters of the discrete element model are adjusted by using the parameter inversion method.

[0026] As a preference, the motion characteristics of dangerous rockfalls are extracted from the simulation results, and the simulation results are statistically analyzed to obtain the probability distribution of motion trajectory, impact kinetic energy and bounce height parameters. The probability that the impact kinetic energy of rockfall in a certain area exceeds a certain threshold is calculated, and the collision and crushing of rockfall is simulated. After crushing, the parameters are redistributed to continue the simulation.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1. One of the core innovations of the present invention is the use of multi-source data acquisition technology, integrating three-dimensional digital photography, drone aerial photography, three-dimensional laser scanning, geological surveys and satellite remote sensing data. Compared with the traditional single data acquisition method, this integration of multi-source data can obtain more comprehensive and accurate dangerous rock mass information. Through unified processing using geographic information system (GIS) technology, a three-dimensional visual initial model of dangerous rock mass is constructed, which intuitively presents the spatial distribution, shape and size of the dangerous rock mass and its relationship with the surrounding environment, providing a solid data foundation for subsequent simulation analysis. Its technical effect is that it greatly improves the accuracy of the recognition of dangerous rock mass, reduces the analysis error caused by missing or inaccurate data, and lays the foundation for the accurate analysis of the movement characteristics of dangerous rock falls.

[0029] 2. The present invention has significant innovations in the construction of discrete element models. Not only does it give the joint unit precise physical and mechanical parameters, but it also uses the Hertz-Mindlin contact theory to accurately define the contact behavior between units. At the same time, it fully considers complex geological factors such as slope surface hardness and joint fissure distribution, and dynamically adjusts the friction coefficient and joint unit parameters. This enables the discrete element model to more realistically simulate the actual mechanical properties of dangerous rock masses. Through this refined modeling, the simulation results are closer to the actual motion state of dangerous rockfalls, and can accurately reflect processes such as interaction between blocks, energy transfer and loss, providing a more reliable model basis for the analysis of the motion characteristics of dangerous rockfalls, and effectively improving the accuracy and credibility of the analysis results.

[0030] 3. During the simulation calculation process of the present invention, multiple physical field factors such as gravity, friction, and collision force are comprehensively considered, and simulation is performed based on the law of conservation of energy and Newton's laws of motion. This is another innovation of the invention. This multi-physical field coupled simulation method comprehensively covers the various mechanical effects in the movement of dangerous rockfalls. Compared with traditional simulation methods that only consider some factors, it can more completely simulate the rockfall movement process. Through this simulation calculation, key motion characteristics such as motion trajectory, speed, bounce height, and impact kinetic energy can be accurately obtained, providing a powerful means for in-depth understanding of the laws of dangerous rockfall movement and providing more accurate data support for the scientific design of subsequent protective measures.

[0031] 4. The present invention innovatively introduces a probabilistic analysis method, randomly samples simulation parameters, and simulates multiple times. By statistically analyzing a large number of simulation results, the probability distribution of the motion characteristics of dangerous rockfalls is obtained. This innovation breaks through the limitations of traditional deterministic analysis and fully considers the uncertainty factors of dangerous rockfalls. Its technical effect is that it can more scientifically assess the risk of dangerous rockfalls and provide a probabilistic basis for the reliability design of protective measures. For example, calculating the probability that the kinetic energy of the rockfall impact exceeds a certain threshold can help engineers more reasonably determine the protection level of protective facilities and improve the safety and economy of protective projects.

[0032] 5. The present invention is an important innovation in that it optimizes the design of protective measures based on the motion characteristic results obtained by simulation calculations and in combination with the safety protection levels of different areas. By referring to the statistical table of the two-dimensional simulation calculation results and the cross-sectional diagram of the analysis of the motion characteristics of dangerous rockfalls, the layout position and protection level of the protective net can be accurately determined. This simulation result-driven optimization method of protective measures has changed the blindness and empiricism of traditional protective designs. The technical effect is that the designed protective measures are more targeted, which can not only effectively intercept dangerous rockfalls and ensure the safety of personnel and facilities, but also avoid the waste of resources caused by over-design, and achieve a balance between protective effect and economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 Schematic diagram of the overall process of an embodiment of the present invention;

[0034] Figure 2 This is the overall distribution map of dangerous rock masses in the hub area according to an embodiment of the present invention;

[0035] Figure 3 3D simulation calculation analysis result diagram of the upstream right bank of the dam and the inlet of the discharge system according to an embodiment of the present invention;

[0036] Figure 4 This is a diagram showing the results of a three-dimensional simulation calculation and analysis of the right abutment area of the dam according to an embodiment of the present invention;

[0037] Figure 5 Schematic diagram of the two-dimensional analysis section position of an embodiment of the present invention. DETAILED DESCRIPTION

[0038] In response to some technical problems existing in the prior art, the main purpose of this invention is to accurately analyze the movement characteristics of dangerous rockfalls, provide a scientific basis for the prevention and control of dangerous rockfall disasters, and reduce the impact of disasters on people, infrastructure and the environment. The specific purpose is as follows:

[0039] First, the present invention achieves multi-source data fusion. Data related to dangerous rock masses is collected using a variety of methods, including 3D digital photography, drone aerial photography, 3D laser scanning, geological surveys, and satellite remote sensing. This data is then integrated and processed using Geographic Information System (GIS) technology. This allows for more comprehensive and accurate information on the dangerous rock masses, including their location, shape, size, structural surface distribution, surface texture, macroscopic topography, and surrounding environment. This provides a solid data foundation for subsequent simulations, allowing for a more accurate understanding of the initial state and movement environment of dangerous rockfalls.

[0040] Secondly, the present invention achieves a precise construction of a discrete element model. The dangerous rock mass is divided into multiple discrete units, and the Hertz-Mindlin contact theory is used to define the contact behavior between the units, taking into account factors such as physical and mechanical properties, joint and fissure distribution, and slope surface hardness. This model can realistically simulate complex mechanical processes such as the interaction between dangerous rock masses, energy transfer, and loss, making the simulation results more closely resemble the actual motion state of dangerous rockfalls, and thus accurately analyzing motion characteristics such as trajectory, velocity, bounce height, and impact kinetic energy.

[0041] Third, the present invention can realize multi-physics field coupling simulation and probability analysis. In the simulation calculation, multi-physics field factors such as gravity, friction, collision force, etc. are comprehensively considered, and simulation is performed according to the law of conservation of energy and Newton's laws of motion. At the same time, a probability analysis method is introduced to randomly sample the simulation parameters and simulate them multiple times, and statistically analyze the probability distribution of motion characteristics. This method breaks through the limitations of traditional deterministic analysis, fully considers the uncertainty of the occurrence of dangerous rockfalls, and can more scientifically assess the risk level, such as calculating the probability that the kinetic energy of the rockfall impact exceeds a certain threshold.

[0042] Finally, the present invention can use simulation results to drive protection design and optimize protection measures against dangerous rockfalls. Based on the motion characteristic results obtained by simulation, combined with the safety protection levels of different areas, and referring to the statistical table of two-dimensional simulation calculation results and the cross-sectional diagram of the motion characteristic analysis of dangerous rockfalls, the layout position and protection level of protective facilities such as protective nets can be accurately determined. This avoids the blindness and empiricism of traditional protection designs, making protective measures more targeted, effectively intercepting dangerous rockfalls and ensuring the safety of personnel and facilities, while avoiding waste of resources caused by over-design, and achieving a balance between protection effect and economic benefits.

[0043] Example 1: Figure 1 As shown in FIG, the method for analyzing the characteristics of dangerous rockfall motion based on discrete element simulation includes the following steps:

[0044] S1. Multi-source data acquisition and preprocessing. First, we utilize 3D digital photography, drone aerial photography, 3D laser scanning, geological surveys, and satellite remote sensing to obtain information on the location, shape, size, structural surface distribution, surface texture, macroscopic topography, and surrounding environment of the dangerous rock mass. We perform preprocessing operations such as denoising, filtering, and format conversion on the collected data. The data is then integrated into a unified geographic coordinate system using Geographic Information System (GIS) technology to construct an initial 3D visualization model of the dangerous rock mass.

[0045] S2, discrete element model construction. First, the dangerous rock mass is divided into multiple joint units. According to the physical and mechanical properties of the dangerous rock mass, the density, elastic modulus, Poisson's ratio and other parameters of each joint unit are assigned. The Hertz-Mindlin contact theory is used to define the contact type and parameters between units and determine the normal contact force:

[0046]

[0047] in:

[0048] is the equivalent elastic modulus, E1 and E2 are the elastic moduli of the two contact elements, v1 and v2 are Poisson's ratios, is the equivalent radius, R1 and R2 are the contact element radii, δ n is the normal overlap; set the friction coefficient and joint unit parameters according to the slope surface hardness and joint crack distribution.

[0049] S3, simulation calculation and result analysis. First, gravity, friction, and collision force are considered in the simulation, and discrete element simulation technology is used to simulate the movement of dangerous rockfalls by considering multiple physical field factors; the movement trajectory, speed, and bounce height are analyzed by combining three-dimensional and two-dimensional simulation. (υ n is the vertical velocity component after colliding with the slope, g is the acceleration of gravity), impact kinetic energy (m is the mass of the falling rock, υ is the speed) and other characteristics; introduce the probability analysis method, randomly sample the simulation parameters for multiple simulations, and statistically analyze the probability distribution of motion characteristics; based on the simulation results and combined with the safety protection level, optimize the design of protection measures.

[0050] Furthermore, in the multi-source data collection and preprocessing step, when using drone aerial photography to obtain a large-scale distribution overview of dangerous rock masses, multispectral imaging technology is used to identify dangerous rock masses of different materials and distinguish dangerous rock masses from surrounding stable rock masses.

[0051] Specifically, select a drone with stable flight performance, long endurance, and the ability to carry multispectral imaging equipment. Multispectral imaging equipment must cover specific wavelengths, such as near-infrared and short-wave infrared, that can effectively distinguish between dangerous rock masses and surrounding stable rock masses. Before flight, thoroughly inspect the performance of the drone and multispectral imaging equipment, calibrate equipment parameters, and ensure measurement accuracy.

[0052] Considering the scope of the study area, its topographical characteristics, and the possible distribution of dangerous rock masses, use professional UAV flight planning software to develop a reasonable flight route. Set appropriate flight altitude, speed, and heading overlap. The flight altitude is determined by the required resolution and equipment performance. For example, if high-resolution imagery is desired to accurately identify the boundaries of dangerous rock masses, the flight altitude can be set between 100 and 200 meters. The speed should be controlled at 3 to 5 meters per second to ensure image clarity. The heading overlap should be set at 70% to 80%, and the lateral overlap at 60% to 70% to ensure that the acquired imagery fully covers the study area.

[0053] UAV flight missions are conducted in clear weather, good lighting conditions, and low winds (generally less than level 3). Multispectral imaging equipment collects data according to preset parameters, simultaneously recording position, attitude, and other information during flight. GPS and an inertial measurement unit (IMU) are used to achieve precise positioning and attitude monitoring, ensuring the spatial accuracy of the collected data.

[0054] After the acquisition is completed, the acquired multispectral image is preprocessed. First, the noise caused by equipment noise, atmospheric scattering and other factors is removed. Filtering algorithms such as Gaussian filtering and median filtering can be used. Taking Gaussian filtering as an example, the calculation process is as follows: for each pixel in the image, the weighted average of its neighboring pixels is calculated according to the Gaussian function as the new value of the pixel. Let the image be I(x,y), the image after Gaussian filtering be G(x,y), and the Gaussian function be g(x,y,σ), then:

[0055]

[0056] Among them, (x, y) is the coordinate of the current pixel, (m, n) is the coordinate of the neighboring pixel, and σ is the standard deviation of the Gaussian function, which controls the smoothness of the filter.

[0057] Radiometric calibration is performed to convert the digital quantization value (DN value) of the image into a physical radiation brightness value, so that images acquired at different times and under different conditions are comparable. The radiometric calibration formula is:

[0058] L = G × DN + B;

[0059] Where L is the radiation brightness value, G is the gain coefficient, B is the offset, and DN is the digital quantization value of the image. G and B can be obtained through device calibration.

[0060] Next, atmospheric correction is performed to eliminate the effects of atmospheric scattering and absorption on light and restore the true reflectivity of the ground. Common atmospheric correction methods include those based on radiation transfer models (such as the 6S model and the MODTRAN model) and statistical methods (such as the dark pixel method).

[0061] Based on the above, image processing and machine learning techniques are used to extract features of the dangerous rock mass and surrounding stable rock mass from multispectral images, such as spectral characteristics (reflectance differences between different bands) and textural characteristics (grayscale variation patterns in the image). For spectral characteristics, the average reflectance of different regions in each band is calculated; for textural characteristics, the gray-level co-occurrence matrix (GLCM) can be used to calculate texture parameters such as contrast, correlation, energy, and entropy. The images are then classified using supervised or unsupervised classification algorithms. Supervised classification can use the maximum likelihood classification method, which calculates the probability of each pixel belonging to each category based on the statistical characteristics of known training samples and assigns the pixel to the category with the highest probability.

[0062] Assuming there are n categories, for pixel x, the probability that it belongs to category i is:

[0063]

[0064] Where P(x|i) is the probability density function of category i, P(i) is the prior probability of category i, and P(x) is the probability density function of pixel x. The pixel's category is determined by comparing P(x|i) (i = 1, 2, ..., n). Unsupervised classification can use the K-means clustering algorithm, which, through continuous iteration, maximizes the similarity of pixels within a cluster and minimizes the similarity of pixels between clusters.

[0065] Finally, the classification results are compared and verified with the actual situation on site. This can be done through field visits, reference to geological data, and other methods. If there are any misclassifications or omissions, the causes are analyzed and the classification model is modified. For example, if some dangerous rock masses are misclassified as stable rock masses due to similar texture features, the texture feature extraction parameters can be adjusted or more training samples can be added to reclassify the model.

[0066] Furthermore, in the discrete element model construction step, the influence of joint cracks on rockfall movement is considered. When the rockfall moves to the joint position, according to the joint shear strength formula:

[0067]

[0068] Among them, c j is the joint cohesion, σ n is the normal stress on the joint surface, is the internal friction angle of the joints) to determine whether the joints are damaged, and then adjust the trajectory of the falling rock and its mechanical properties.

[0069] Before constructing the discrete element model, after acquiring joint and fissure data, it must be systematically organized and analyzed. Classification and statistics are performed based on factors such as the occurrence and development level of the joints and fissures, and a joint and fissure database is established. Error analysis and corrections are performed on the geometric parameters of the joints and fissures to ensure data accuracy and reliability. Furthermore, the dangerous rock mass is divided into different regions based on the distribution characteristics of the joints and fissures, providing a basis for the subsequent configuration of joint units in the discrete element model.

[0070] In the discrete element model, the setting of joint units should fully consider the actual situation of joints and fissures. According to the classification results of joints and fissures, different types of joint units are assigned corresponding mechanical parameters. j , internal friction angle In addition to being determined through indoor tests and empirical formulas, parameters such as shear strength can also be modified based on in-situ test data. For example, in-situ direct shear tests can be used to obtain shear strength parameters of joint surfaces, improving their accuracy.

[0071] The geometry of the joint units should reflect the actual shape of the joints and fissures as realistically as possible. For regular joints and fissures, simple geometric shapes (such as planes and straight lines) can be used for simulation; for complex joint networks, irregular geometric shapes such as polygons and polyhedrons can be used. At the same time, the interconnectedness between joint units must be considered to simulate the impact of joint connectivity on rockfall motion.

[0072] During the rockfall, the stress state of the joints is constantly changing. Therefore, the joint failure judgment should adopt a dynamic update method. In each time step, the normal stress σ of the joint surface is calculated in real time based on the relative position and force of the rockfall and the joints. n and shear stress τ. The calculated shear stress τ is compared with the joint shear strength Make a comparison to determine whether the joints are damaged.

[0073] To improve the accuracy of joint failure assessment, damage mechanics theory can be introduced. This model considers the evolution of joint damage during stress loading and establishes a joint damage model. As joint damage increases, the cohesion and internal friction angle of the joint gradually decrease. During joint failure assessment, the joint mechanical parameters are updated in real time, ensuring that the assessment results are more consistent with actual conditions.

[0074] Once joint failure is determined, the trajectory and mechanical properties of the rockfall must be adjusted promptly. Based on the type and degree of joint failure (e.g., tensile or shear), the additional forces acting on the rockfall are calculated. For example, during shear failure, the rockfall will experience a sudden change in the shear force on the joint surface, causing a sudden change in its velocity and direction. Based on Newton's second law, F = ma, the acceleration of the rockfall under the additional forces is calculated, thereby updating its velocity and position.

[0075] At the same time, the interaction between falling rocks and the broken rock mass after joint failure must be considered. The broken rock mass may act as a barrier, cause friction, and collide with the falling rocks, affecting their trajectory and energy loss. The discrete element method can be used to simulate the movement and interaction of the broken rock mass. By treating the broken rock mass as multiple joint units and performing coupled calculations with the falling rocks, this method more accurately reflects the motion characteristics of the falling rocks after joint failure.

[0076] After constructing a discrete element model that incorporates the effects of joints and fissures, the model needs to be validated. This can be done through indoor physical model testing and field monitoring data. In indoor physical model testing, a physical model similar to the actual dangerous rock mass is constructed to simulate the rockfall process, recording data such as the rockfall trajectory, velocity, and joint damage. The results of the physical model test are then compared and analyzed with the discrete element model simulation results to assess the model's accuracy and reliability.

[0077] If the model simulation results deviate from the actual situation, parameter inversion can be used to adjust the model parameters. Using optimization algorithms (such as genetic algorithms and particle swarm optimization), with actual monitoring data as the objective function, the mechanical and geometric parameters of the joint units are continuously adjusted to ensure that the model simulation results match the actual situation as closely as possible. After multiple iterations of optimization, an optimal set of model parameters is obtained, improving the model's simulation accuracy.

[0078] Furthermore, in the simulation calculation and result analysis steps, combined with the statistical table data of the two-dimensional simulation calculation results, the partial coefficient γ of the dangerous rockfall motion parameter is used. α (The random simulation statistical value is 1.1), calculate the design value of impact kinetic energy E d =γ α E k and the design value of the bounce height h d =γ α h k . Among them E k is the standard value of impact kinetic energy, h k It is the standard value of bounce height.

[0079] In the simulation calculation and result analysis steps, the layout position and protection energy level of the protection net are determined by referring to the cross-section diagram of the dangerous rockfall motion characteristics. If the maximum impact kinetic energy of the rockfall in a certain section is E kmax , then the nominal value of the passive protection net protection energy level E B Need to meet E B ≥γ E E d (When the safety level is Ⅰ, γ E =1.5).

[0080] In the simulation calculation and result analysis steps, the collision and crushing of falling rocks is simulated. When the stress on the block exceeds its tensile strength σ t Or shear strength τ f Determine the crushing when the shear strength τ f According to the Mohr-Coulomb criterion, the block mass, velocity and other parameters are redistributed after crushing to continue the simulation.

[0081] First, the present invention achieves multi-source data fusion. Data related to dangerous rock masses is collected using a variety of methods, including 3D digital photography, drone aerial photography, 3D laser scanning, geological surveys, and satellite remote sensing. This data is then integrated and processed using Geographic Information System (GIS) technology. This allows for more comprehensive and accurate information on the dangerous rock masses, including their location, shape, size, structural surface distribution, surface texture, macroscopic topography, and surrounding environment. This provides a solid data foundation for subsequent simulations, allowing for a more accurate understanding of the initial state and movement environment of dangerous rockfalls.

[0082] Secondly, the present invention achieves a precise construction of a discrete element model. The dangerous rock mass is divided into multiple joint units, and the Hertz-Mindlin contact theory is used to define the contact behavior between these units, taking into account factors such as physical and mechanical properties, joint and fissure distribution, and slope surface hardness. This model can realistically simulate complex mechanical processes such as the interaction between dangerous rock masses, energy transfer, and loss, making the simulation results more closely aligned with the actual motion state of dangerous rockfalls, and thus accurately analyzing motion characteristics such as trajectory, velocity, bounce height, and impact kinetic energy.

[0083] Third, the present invention can realize multi-physics field coupling simulation and probability analysis. In the simulation calculation, multi-physics field factors such as gravity, friction, collision force, etc. are comprehensively considered, and simulation is performed according to the law of conservation of energy and Newton's laws of motion. At the same time, a probability analysis method is introduced to randomly sample the simulation parameters and simulate them multiple times, and statistically analyze the probability distribution of motion characteristics. This method breaks through the limitations of traditional deterministic analysis, fully considers the uncertainty of the occurrence of dangerous rockfalls, and can more scientifically assess the risk level, such as calculating the probability that the kinetic energy of the rockfall impact exceeds a certain threshold.

[0084] Finally, the present invention can use simulation results to drive protection design and optimize protection measures against dangerous rockfalls. Based on the motion characteristic results obtained by simulation, combined with the safety protection levels of different areas, and referring to the statistical table of two-dimensional simulation calculation results and the cross-sectional diagram of the motion characteristic analysis of dangerous rockfalls, the layout position and protection level of protective facilities such as protective nets can be accurately determined. This avoids the blindness and empiricism of traditional protection designs, making protective measures more targeted, effectively intercepting dangerous rockfalls and ensuring the safety of personnel and facilities, while avoiding waste of resources caused by over-design, and achieving a balance between protection effect and economic benefits.

[0085] Example 2: Figure 2-Figure 5 As shown in Figure 2, in the flood discharge system inlet area of a large-scale water conservancy project, there are a large number of dangerous rock masses on the surrounding natural slopes, which pose a serious threat to the construction and operation safety of the project. Figure 2 In order to effectively prevent and control dangerous rockfall disasters and ensure the smooth progress of the project, a dangerous rockfall motion characteristic analysis method based on discrete element simulation is adopted. The specific implementation process is as follows:

[0086] like Figure 1As shown, follow the detailed steps below:

[0087] S1, multi-source data collection and preprocessing:

[0088] Dangerous rock mass data are obtained using a variety of methods including 3D digital photography technology, drone aerial photography, 3D laser scanning, geological surveys, and satellite remote sensing.

[0089] (1) Three-dimensional digital photography and drone aerial photography. Use a high-resolution three-dimensional digital camera to photograph the dangerous rock mass from multiple angles, combined with drones equipped with multispectral imaging equipment for aerial photography, to obtain information such as the surface texture, geometry, location, and distribution range of the dangerous rock mass. Multispectral imaging technology can be used to identify dangerous rock masses of different materials and distinguish them from surrounding stable rock masses.

[0090] (2) 3D laser scanning. Use a 3D laser scanner to scan the dangerous rock mass and obtain surface point cloud data. After denoising and filtering, the data quality is improved and the shape and size information of the dangerous rock mass is accurately obtained.

[0091] (3) Geological exploration. Conduct geological drilling and geophysical exploration to obtain geological information such as the internal structure, lithology, distribution of joints and fissures, and potential sliding surfaces of the dangerous rock mass. Drill multiple holes on site and collect core samples for laboratory analysis to determine the physical and mechanical parameters of the rock mass.

[0092] (4) Satellite remote sensing: Use satellite remote sensing images to obtain macro-topography and surrounding environment information, providing background data for subsequent simulation analysis.

[0093] (5) Data preprocessing and fusion. The collected multi-source data are preprocessed by denoising, filtering, format conversion, and other operations. Then, they are fused in a unified geographic coordinate system using geographic information system (GIS) technology to construct a three-dimensional visual initial model of the dangerous rock mass.

[0094] S2, discrete element model construction

[0095] (1) Joint unit division. Based on the shape, size, and internal structure of the dangerous rock mass, combined with the distribution of joints and fissures obtained from geological exploration, the dangerous rock mass is divided into multiple joint units. When dividing, the unit boundaries are aligned with the joint and fissure surfaces as much as possible. For parts with complex shapes, irregular units are used to better simulate the actual shape.

[0096] (2) Parameter assignment. Through indoor and field tests, the physical and mechanical parameters of the dangerous rock mass, such as density, elastic modulus, and Poisson's ratio, are obtained and assigned to each joint unit. The Hertz-Mindlin contact theory is used to define the contact type and parameters between units, and the friction coefficient is adjusted according to the hardness of the slope surface. At the same time, based on the results of geological exploration, the parameters such as the cohesion and internal friction angle of the joints are determined.

[0097] (3) Joint and fissure treatment. Joint units of different strengths are set in the discrete element model to simulate the influence of joints on rockfall. When the rockfall moves to the joint position, according to the joint shear strength formula:

[0098] Among them, c j is the joint cohesion, σ n is the normal stress on the joint surface, The internal friction angle of the joint is used to determine whether the joint is damaged, and then the rockfall trajectory and mechanical properties are adjusted to determine whether the joint is damaged. If the joint is damaged, the rockfall trajectory and mechanical properties are readjusted.

[0099] S3, simulation calculation and result analysis:

[0100] (1) Simulation calculation. In the discrete element simulation software, multiple physical field factors such as gravity, friction, and collision force are set, and the movement process of dangerous rockfall is simulated according to the law of conservation of energy and Newton's laws of motion. An appropriate time step is set and multiple simulation calculations are performed. In each simulation, some parameters (such as the initial position, shape, and size of the dangerous rock body) are randomly sampled to account for the uncertainty of dangerous rockfall.

[0101] (2) Motion feature extraction. The motion features of the rockfall, such as its trajectory, velocity, bounce height, and impact kinetic energy, are extracted from the simulation results. For example, the simulation found that the motion trajectory of some rockfalls is significantly affected by the slope gradient and joint fissures of the slope. When encountering joints, the rockfall trajectory changes, and its velocity and impact kinetic energy also change accordingly.

[0102] (3) Probability analysis. Statistical analysis is performed on multiple simulation results to obtain the probability distribution of parameters such as motion trajectory, impact kinetic energy, and bounce height. The probability that the impact kinetic energy of falling rocks in a certain area exceeds a certain threshold is calculated to provide a scientific basis for risk assessment. Based on the results of the three-dimensional oblique photography and on-site investigation and analysis, a large number of isolated rocks, isolated rock groups, dangerous rock bodies, dangerous rock groups and other types of dangerous rock bodies are randomly distributed on the natural slopes on both sides of the hub area. Most of them are in a basically stable state for a long time. When external interference conditions (excavation disturbance, blasting vibration, strong winds) occur, falling rocks may occur, which is very harmful to engineering construction.

[0103] like Figure 3 and Figure 4 As shown, Figure 3 The results of a 3D simulation analysis of the dam's upstream right bank and the spillway inlet are shown. Within this slope section, the bounce height of falling rocks is mostly within 16 meters, with a few exceeding 22 meters due to the steep terrain. The maximum impact energy is approximately 2500 kJ, and the maximum movement speed reaches 25 m / s. The instability of this dangerous rockfall primarily threatens the spillway inlet area. Figure 4 The results of a 3D simulation analysis of the dam's right abutment are shown. Within this slope, the bounce height of falling rocks is mostly within 20 meters, with a few exceeding 25 meters due to the steep terrain. The maximum impact energy is approximately 2500 kJ, and the maximum speed is 25 m / s. This section of dangerous rock instability primarily threatens the dam's right abutment.

[0104] From the three-dimensional rockfall simulation analysis, it can be seen that after a rockfall occurs, the bounce height, movement speed, and impact kinetic energy of the rockfall moving downward along the slope are all controlled by the change in slope hardness caused by whether the bedrock on the surface of the slope is exposed or covered by Quaternary loose deposits, while the movement trajectory is controlled by micro-topographic conditions such as the slope gradient. For example, after a rockfall occurs, the dangerous rocks developed near the source of the channel and on the side slopes of the channel will enter the channel within a certain period of time, collide with the bottom of the channel, and then move along the channel. As the ditch moves downward, some of the fallen rocks will bounce and roll over when the ditch turns, but most of them will turn with the ditch; the larger the volume of the fallen rocks, the longer their movement path and the greater their kinetic energy, but the speed of movement and the bounce height are not closely related to the size of the volume; the steeper the slope of the terrain where the fallen rocks are located, the greater their bounce height, the greater their movement speed, and the corresponding higher the impact kinetic energy; if the fallen rocks enter the ditch and the ditch is filled with loose Quaternary deposits, the movement speed will drop significantly and the probability of them staying in the ditch will increase greatly.

[0105] The diameter of the scattered rocks in the gully obtained by interpreting the high-precision three-dimensional tilt model is used as the input condition for the size of the single dangerous rock, and 2m is taken as the input condition for the size of the single dangerous rock. 3 (The 95% percentile volume size was obtained by interpretation). Considering the simultaneous instability of up to 10 dangerous rock masses, a three-dimensional rockfall simulation analysis was conducted to simulate local small-scale collapse conditions. According to the overall calculation results, the maximum bounce height of the falling rock is around 10m-20m. Near the preliminary exploration access road (a location with relatively good traffic conditions for protective construction), the maximum impact kinetic energy is about 2500kJ or more. If it falls to the bottom of the slope, the speed increases, and most of the impact energies exceed 10000kJ.

[0106] (4) Optimization of protective measures. Based on the statistical table of two-dimensional simulation calculation results and the cross-sectional diagram of the analysis of the characteristics of dangerous rockfall movement, the reasonable layout position and protection level of the protective net are determined.

[0107] Based on the results of the three-dimensional numerical simulation, a reasonable path section was selected for two-dimensional motion characteristic analysis. Combined with the preliminary layout of the protective facilities, five typical sections (R9, R10, R11, R12, and R13) were selected for two-dimensional calculation and analysis of the rockfall trajectory, impact energy, and bounce height, which served as the basis for the selection and layout of the protective net. According to the on-site investigation and the interpretation of the oblique photography data, the collapsed rocks that stayed in the gully were mostly small rocks, and some larger rocks were larger than 1m in size.3 ~2m 3 Considering the safety protection level of different areas, for the important Class I safety protection area, the rockfall volume is 2m 3 The simulation was performed 1000 times (the 95th percentile volume was obtained by interpretation). The results of the 2D rockfall motion characteristic analysis are shown in Table 1.

[0108] Table 1: Statistics of two-dimensional simulation results of dangerous rockfall

[0109]

[0110] Note: "R" in the table represents the right bank, the standard value of impact kinetic energy E k , standard value of bounce height h k are simulated values with a 95% assurance rate.

[0111] (5) Crushing simulation. During the simulation, the impact and crushing of falling rocks are simulated. When the stress on the block exceeds its tensile strength or shear strength, the block is considered to have broken. The shear strength is calculated according to the Mohr-Coulomb criterion. After the crushing, the block's mass, velocity, and other parameters are redistributed, and the simulation is continued to more realistically reflect the movement process of the dangerous rockfall.

[0112] The present invention utilizes a method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation, combined with actual engineering conditions, to complete the analysis of dangerous rockfalls, the design and implementation of prevention and control projects, and effectively ensure the safe construction and operation of water conservancy hub projects.

Claims

1. A method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation, characterized in that: The steps include: S1. Collect and pre-process dangerous rock data through multiple means to construct a 3D visual initial model of dangerous rock mass; S2. Based on the three-dimensional visualized initial model of the dangerous rock mass, a discrete element model is constructed and the contact type and parameters between joint units are defined using the Hertz-Mindlin contact theory to determine the normal contact force; S3. Based on the discrete element model and the normal contact force, discrete element simulation technology is used to simulate the rockfall motion of dangerous rocks by considering multiple physical field factors, and discrete element model simulation results are obtained. Joint failure is judged by a dynamic update method, and the discrete element model is verified through indoor physical model tests and field monitoring data. S4. Design and implement prevention and control projects based on discrete element model simulation results.

2. The method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation according to claim 1 is characterized in that: The S1 includes using surveying and mapping technology and satellite remote sensing to obtain dangerous rock mass data and its surrounding environment information; cleaning the collected data, and using image processing and machine learning technology to extract the characteristics of the dangerous rock mass and surrounding stable rock mass in multispectral images, using supervised classification or unsupervised classification algorithms to classify the images, comparing and correcting the classification results with the actual situation on site, and fusing them in a unified geographic coordinate system through geographic information system technology to construct a three-dimensional visual initial model of the dangerous rock mass.

3. The method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation according to claim 1 is characterized in that: The S2 specifically includes the following steps: S21, acquiring joint and fissure data and establishing a joint and fissure database, and dividing the dangerous rock mass into a plurality of joint units based on the three-dimensional visualized dangerous rock mass initial model according to the distribution characteristics of the joints and fissures; S22. Based on the classification results of S1 and the physical and mechanical properties of the dangerous rock mass, assign physical characteristics to each joint unit; S23. The Hertz-Mindlin contact theory is used to define the contact type and parameters between joint elements and determine the normal contact force.

4. The method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation according to claim 3 is characterized in that: The normal contact force is specifically expressed as: in: is the equivalent elastic modulus, E1 and E2 are the elastic moduli of the two contact elements, v1 and v2 are Poisson's ratios, is the equivalent radius, R1 and R2 are the contact element radii, δ n is the normal overlap; set the friction coefficient and joint unit parameters according to the slope surface hardness and joint crack distribution.

5. The method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation according to claim 1 is characterized in that: In S3, the joint damage judgment includes calculating the normal stress and shear stress of the joint surface in each time step according to the relative position and force of the falling rock and the joint, comparing the shear stress with the shear strength of the joint, and judging whether the joint is damaged. When the shear stress is greater than the shear strength of the joint, the joint is damaged; when the shear stress is less than the shear strength of the joint, the joint is not damaged.

6. The method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation according to claim 5 is characterized in that: After determining the joint failure, the additional force acting on the rockfall is calculated according to the mode and degree of joint failure to adjust the motion trajectory and mechanical properties of the rockfall. The discrete element method is used to simulate the motion and interaction between the broken rock mass and the rockfall, and the broken rock mass is regarded as multiple discrete units to couple with the rockfall to calculate the additional force.

7. The method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation according to claim 1 is characterized in that: The indoor physical model test includes making a physical model that restores the actual dangerous rock mass, simulating the rockfall movement process, recording the movement trajectory, speed and joint damage of the rockfall, and comparing and analyzing the indoor physical model test results with the discrete element model simulation results. If there is a deviation between the simulation results of the discrete element model and the actual situation, the parameters of the discrete element model are adjusted using the parameter inversion method.

8. The method for analyzing the motion characteristics of dangerous rockfalls based on discrete element simulation according to claim 7 is characterized in that: The motion characteristics of dangerous rockfalls are extracted from the simulation results, and statistical analysis is performed on the simulation results to obtain the probability distribution of motion trajectory, impact kinetic energy, and bounce height parameters. The probability that the impact kinetic energy of rockfall in a certain area exceeds a certain threshold is calculated, and the collision and fragmentation of rockfalls are simulated. After fragmentation, the parameters are redistributed to continue the simulation.

Images