A simulation test method and system for tailings dam slope instability
By constructing a quantitative model of spatial variation in tailings dams and using digital image correlation, the problem of simulating the random distribution of lenses in tailings dams was solved, enabling accurate simulation and risk prediction of slope instability processes.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NORTH CHINA UNIVERSITY OF TECHNOLOGY
- Filing Date
- 2026-04-23
- Publication Date
- 2026-06-30
Smart Images
Figure CN122306567A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of geotechnical engineering physical simulation testing technology, and in particular to a simulation test method and system for tailings dam slope instability. Background Technology
[0002] Tailings dams are large industrial structures used to store tailings during mining operations, and their stability directly affects the safety of mining production and the lives and property of people downstream. Slope instability is one of the main types of tailings dam accidents. In actual engineering practice, it has been found that lenticular weak geological structures often exist within tailings dams. These weak geological structures are areas with low strength but no clear boundary with the surrounding tailings. The presence of these lenses reduces the dam's strength and increases the wetting line height; when these lenses are near the dam slope, they become a key factor in inducing slope instability.
[0003] Currently, the main research methods for tailings dam slope stability include numerical simulation and physical simulation experiments. While numerical simulation can handle complex mechanical models, it is difficult to fully and realistically reflect the random heterogeneity of materials and boundary conditions. Physical simulation experiments, especially bottom friction tests, are widely used because they can visually reproduce the entire process of slope deformation and failure.
[0004] Existing physical simulation test methods have the following drawbacks: Traditional methods typically simplify tailings dams into homogeneous or layered homogeneous models, which cannot accurately depict the random spatial distribution of the lens body, resulting in a large deviation between the initial experimental model and the actual situation.
[0005] Due to the lack of a physical experimental model that can truly reflect the spatial distribution of the lens, the underlying mechanisms by which the lens specifically affects the failure mode, failure surface location, slippage volume, and the entire evolution process of slope instability remain unclear.
[0006] Existing experiments are mostly based on empirical lens placement, lacking scientific quantitative methods to control the degree of lens variation, making it difficult to establish a quantitative relationship between lens characteristics and instability laws. Summary of the Invention
[0007] This application proposes a simulation test method and system for tailings dam slope instability, which is used to solve the technical problem that the existing technology cannot accurately simulate the instability process of tailings dam slopes containing randomly distributed lenses.
[0008] To achieve the above objectives, this application provides the following technical solution: Firstly, this application proposes a simulation test method for tailings dam slope instability, including: Step 1: Obtain the set of statistical characteristic parameters of tailings sand in tailings dam sediment under a random field model, and construct a quantitative model of spatial variation of tailings dam containing lenses by using the stepwise decomposition method. Step 2: Based on the quantitative model of spatial variation of tailings dam, a physical test model is built on the bottom friction test rig; Step 3: Start the bottom friction test bench to apply simulated gravity load to the physical test model, and use a high-speed camera to collect image data in real time during the test process; Step 4: Process image data using digital image correlation to obtain displacement and strain fields, and determine the instability and catastrophic data of tailings dam slope under the action of the lens body based on the evolution of displacement and strain fields.
[0009] In conjunction with the first aspect, step 1 includes: Step 101: Obtain field survey data of tailings sand deposited in tailings dams and determine the probability distribution characteristics of the physical and mechanical parameters of tailings sand deposited in tailings dams. Step 102: Based on the probability distribution characteristics, determine whether the tailings sand in the tailings dam sediment follows a Gaussian distribution; Step 103: When the material follows a Gaussian distribution, determine the mechanical properties of the tailings dam material and decompose the mechanical properties of the tailings dam material into trend components and fluctuation components; Step 104: Under a stationary random field, the fluctuation distance is set as the key parameter for controlling spatial variability based on the fluctuation components; wherein, the key parameter is characterized by the autocorrelation function, which represents the correlation coefficient between the two endpoints of the fluctuation distance in the stationary random field. Step 105: Construct a three-dimensional covariance total correlation matrix based on the key parameters; Step 106: Decompose the three-dimensional covariance total correlation matrix using the Kronecker product to determine the correlation matrix in multiple directions; Step 107: Determine the triangular matrix by performing Cholleski decomposition on the correlation matrix; Step 108: Determine the random standard vector based on the triangular matrix, and determine the random distribution field of the lens body with a specified fluctuation distance through the random standard vector to generate a quantitative model of the spatial variation of the tailings dam.
[0010] In conjunction with the first aspect, the autocorrelation function is: ; in, This represents the relative distance between different points along the x-direction; This represents the relative distance between different points along the y-direction; This represents the relative distance between different points along the z-direction; This represents the distance of fluctuation along the x-direction; This represents the distance of fluctuation along the y-direction; This represents the fluctuation distance along the z-direction.
[0011] In conjunction with the first aspect, the specific steps for constructing the physical experimental model include: Step 201: Determine the geometric parameters of the target test model based on the bottom friction test rig; Step 202: Determine multiple density regions of the target experimental model based on geometric parameters; Step 203: Based on the dry density assignment results of multiple density regions in the quantitative model of spatial variation of tailings dam, determine the dry density value of each region in the quantitative model of spatial variation of tailings dam, and output the model image of the target experimental model; Step 204: Perform grayscale processing and image recognition on the model image to obtain the model size of the target physical test model, and determine the lens body region and the dry density value of the lens body region. Step 205: Determine the tail sand material in the lens area based on the dry density value, and weigh the required tail sand material; Step 206: The tailing sand material is evenly loaded onto the belt of the bottom friction test bench and compacted according to the model size of the target test model so that the target test model reaches the designed thickness. Step 207: Start the bottom friction test bench to make the belt rotate at a uniform speed and perform pre-consolidation treatment on the physical test model; Step 208: Determine the lens body outline according to the lens body area, mark the lens body area on the target test model, remove the raw material in the marked area of the lens body, roughen the junction, and then fill with the predetermined lens body material and compact it. Step 209: After compaction, the target test model is modified to generate the target physical test model.
[0012] In conjunction with the first aspect, the raw materials for removing the lens body marking area include: performing grayscale processing and image recognition on the lens body outline of the lens body marking area to determine the lens body boundary.
[0013] In conjunction with the first aspect, step 4 includes: Step 401: Create a speckle pattern on the surface of the physical test model; Step 402: Continuously acquire image sequences of the model during the experiment using a high-speed camera; Step 403: Perform correlation calculations on the image sequence, track the motion trajectory of the speckle, and obtain the full-field displacement vector of the physical test model surface; Step 404: Based on the obtained displacement field, the displacement data is smoothed using the least squares fitting method, and the deformation gradient matrix is constructed; Step 405: Calculate the strain field based on the deformation gradient matrix.
[0014] In conjunction with the first aspect, step 404 includes: fitting the displacement data using a bilinear Lagrange polynomial or a biquadratic Lagrange polynomial.
[0015] In conjunction with the first aspect, step 405 includes: calculating the strain field using the Euler-Almansi strain tensor or the Green-Lagrange strain tensor.
[0016] In conjunction with the first aspect, based on the physical test model, the instability and catastrophic parameters of the tailings dam slope under different test commands are determined; among which, the test commands include: instability and failure identification command, slippage point location command, and slippage volume calculation command.
[0017] Secondly, this application proposes a simulation test system for tailings dam slope instability, comprising: Model building module: Obtain the statistical characteristic parameter set of tailings sand in tailings dam sediment under random field model, and construct a quantitative model of spatial variation of tailings dam including lens body by using the stepwise decomposition method. Physical test module: Based on the quantitative model of spatial variation of tailings dam, a physical test model is built on the bottom friction test platform; Test data acquisition module: The bottom friction test bench is activated to apply simulated gravity load to the physical test model, and image data during the test process is acquired in real time through a high-speed camera; Analysis module: Based on digital image correlation method, image data is processed to obtain displacement field and strain field, and the instability disaster data of tailings dam slope under the action of lens body is determined according to the evolution process of displacement field and strain field.
[0018] The beneficial effects of this invention are as follows: This application introduces a quantitative evolution method for spatial variation of lenses based on random field theory into physical simulation experiments. This method can construct a tailings dam model that is more realistic and includes randomly distributed lenses. It can capture the entire process of displacement and strain field evolution of slope deformation and failure. Through lens simulation experiments, it can determine the process that leads to the transformation of slope failure mode from lateral movement to a combination of traction and lateral movement, thereby achieving risk prediction of slope instability.
[0019] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.
[0020] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0021] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.
[0022] In the attached diagram: Figure 1 This is a flowchart of a simulation test method for tailings dam slope instability in an embodiment of the present invention; Figure 2 This is a fitting curve of the physical properties of the tailings sand in the tailings dam sediment in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the construction steps of the quantitative model for spatial variation of tailings dams in this embodiment of the invention. Figure 4 This is a schematic diagram of the model generalization process in an embodiment of the present invention; Figure 5 This is a schematic diagram of the slope calculation grid for the area where the lens body is located in an embodiment of the present invention; Figure 6 This is a schematic diagram illustrating the construction of the physical experimental model in an embodiment of the present invention; Figure 7 This is a front view of the bottom friction test bench in an embodiment of the present invention; Figure 8 This is a schematic diagram of the tailings dam slope test model in an embodiment of the present invention; Figure 9 This is a flowchart illustrating the process of processing image data using the digital image correlation method in an embodiment of the present invention. Figure 10 This is a schematic diagram of displacement and deformation in different directions in an embodiment of the present invention; Figure 11 This is a contour map of the horizontal direction of the tailings dam slope in an embodiment of the present invention; Figure 12 This is a contour map of the vertical direction of the tailings dam slope in an embodiment of the present invention; Figure 13 This is a contour map of the total displacement of the tailings dam slope in an embodiment of the present invention. Detailed Implementation
[0023] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.
[0024] Example 1: As Figure 1As shown, this application proposes a simulation test method for tailings dam slope instability, which is applied to the simulation test prediction of slope instability of iron ore tailings dams. The method of this application is as follows: Step 1: Obtain the set of statistical characteristic parameters of tailings sand in tailings dam sediment under a random field model, and construct a quantitative model of spatial variation of tailings dam containing lenses by using the stepwise decomposition method. As one embodiment of this application, step 1 is used to mathematically abstract the spatial variability of tailings dam materials using random field theory, constructing a three-dimensional digital model including lenses. The feature parameter set consists of the physical and mechanical parameters of the sedimentary tailings sand obtained from on-site surveys and laboratory tests of the target tailings dam, which also serve as the modeling input parameters. Specifically, it includes: material property parameters, spatial variability parameters, computational grid parameters, and random seed parameters.
[0025] Material property parameters are used to perform multi-peak fitting on the porosity, dry density, and moisture content curves of the tailings sand region in the tailings dam sedimentation area, such as... Figure 2 As shown, the fitting results indicate that the spatial variability of the physical properties of the tailings sand deposited in the tailings dam follows a Gaussian distribution. Spatial variability parameters include fluctuation distance and autocorrelation function type. The computational grid parameters are set according to the actual size of the tailings dam and experimental requirements, specifying the model's three-dimensional dimensions and the number of grid divisions. The random seed parameter sets a random seed value to generate independent random standard vectors, ensuring the repeatability of the randomly generated results.
[0026] The spatial variability quantitative model is a three-dimensional digital model constructed using random field theory and the stepwise decomposition method. This model assigns a dry density value to each grid cell, quantitatively characterizing the spatial distribution, morphology, and mechanical properties of the lens within the tailings dam. The model is constructed using the stepwise decomposition method, a matrix dimensionality reduction algorithm employed to address the issues of excessive computational cost and memory consumption associated with traditional covariance matrix calculations in large-scale three-dimensional random field modeling. The stepwise decomposition method decomposes the total correlation matrix in three-dimensional space into the product of one-dimensional correlation matrices in three directions using the Kronecker product. This is then combined with the Choreski decomposition to generate random distribution fields satisfying specified statistical characteristics. Multiple random distribution fields constitute the tailings dam spatial variability quantitative model. A random field model refers to a mathematical model that treats the physical and mechanical parameters of the tailings dam material as random functions varying with spatial coordinates, and describes its spatial variability through statistical characteristics such as mean, variance, and autocorrelation function. Lens bodies refer to localized weak regions or interlayers within the tailings dam body that have no clear boundary with the surrounding tailings sand but whose physical and mechanical properties are significantly lower than those of the surrounding medium. They are the main modeling areas that cause tailings dam slope instability and are also the main image recognition areas in this application. Specifically, they include two types: fine-grained interlayer lenses and low-density lens bodies.
[0027] Step 2: Based on the quantitative model of spatial variation of tailings dam, a physical test model is built on the bottom friction test rig; As one embodiment of this application, the bottom friction test bench includes a frame, a motor, a driving roller, a driven roller, an annular belt, and a model-making frame. It generates frictional force through the movement of the annular belt to simulate the gravitational force acting on a slope in its natural state, and is used to study the entire process of slope deformation and failure. The test principle is based on the principle of similarity, using frictional force to simulate the gravitational force acting on the model in its natural state. During the construction of the physical test model, the weights of various tailings sand materials required are calculated based on the model dimensions and dry density assignments, and then the physical test model is constructed.
[0028] Step 3: Start the bottom friction test bench to apply simulated gravity load to the physical test model, and use a high-speed camera to collect image data in real time during the test process; As one embodiment of this application, the bottom friction test bench starts timing, and the motor drives the annular belt to move, applying simulated gravity load to the physical test model. A high-speed camera continuously acquires image sequences of the model during the test at a preset frequency and transmits them to the data processing unit in real time. The physical test model starts timing from the start of the bottom friction test bench and ends when the tailings dam slope becomes unstable. The physical test model is named from left to right according to the slope steps as the first-level sub-dam, the second-level sub-dam, and the third-level sub-dam.
[0029] Step 4: Process image data using digital image correlation to obtain displacement and strain fields, and determine the instability and catastrophic data of tailings dam slope under the action of the lens body based on the evolution of displacement and strain fields.
[0030] As one embodiment of this application, digital image correlation refers to a non-contact optical measurement method that calculates the full-field displacement and strain fields of the model surface by tracking the positional changes of the speckle pattern on the model surface before and after deformation. The displacement field refers to the spatial displacement vector distribution of each point on the physical test model surface relative to its initial position during the test, including horizontal displacement, vertical displacement, and total displacement. The strain field refers to the strain state distribution of each point on the model surface obtained based on the displacement field calculation, including normal strain, shear strain, and principal strain. Instability catastrophic data refers to the dataset obtained through experiments and analysis used to characterize the instability features of tailings dam slopes under the action of a lens, including failure morphology characteristics, displacement field characteristics, instability mode, evolution stage, slip point location, and slip volume.
[0031] Example 2, as Figure 3 As shown, step 1 includes: Step 101: Obtain field survey data of tailings sand deposited in tailings dams and determine the probability distribution characteristics of the physical and mechanical parameters of tailings sand deposited in tailings dams. As one embodiment of this application, the field survey data comprises statistical physical and mechanical parameters of the tailings dam deposition area, which are also a set of statistical characteristic parameters. The probability distribution characteristics are the physical and mechanical parameters of the tailings sand in the tailings dam deposition obtained through field surveys and laboratory tests. Specifically, these parameters include the spatial distribution patterns and fluctuation characteristics of dry density, porosity, and water content, which are manifested as the statistical distribution type and characteristic values of these parameters. Distribution type: refers to the probability distribution that the spatial values of the physical and mechanical parameters of the tailings sand follow.
[0032] Step 102: Based on the probability distribution characteristics, determine whether the tailings sand in the tailings dam sediment follows a Gaussian distribution; As one embodiment of this application, this application uses a multi-peak fitting method to determine whether the spatial variability of the physical properties of tailings sand deposited in tailings dams follows a Gaussian distribution. The probability distribution characteristics are reflected by the multi-peak fitting of the porosity curve, dry density curve, and water content curve of the tailings sand. These curves exhibit multiple peak shapes for fitting. The fitting matches the distribution curve of the measured data with the theoretical probability distribution curve. If the fitting result follows a Gaussian distribution, it indicates that the spatial random fluctuations of the physical and mechanical parameters of the tailings sand deposited in tailings dams conform to a Gaussian distribution, and modeling and experimental analysis can be performed.
[0033] Characteristic parameters include the mean, which describes the central trend of the distribution, and the variance, which describes the amplitude of fluctuations. In this application, dry density is used as a variable, and the average dry density of the tailings sand is obtained as the mean parameter. The variance of the dry density is then obtained through statistical analysis. The variance and mean parameters are used to determine whether the tailings sand deposited in the tailings dam follows a Gaussian distribution.
[0034] Step 103: When the material follows a Gaussian distribution, determine the mechanical properties of the tailings dam material and decompose the mechanical properties of the tailings dam material into trend components and fluctuation components; As one embodiment of this application, the mechanical properties of tailings dam materials are... Decomposed into trend components and fluctuation components As shown in the following formula. Where, the trend component... Take the average , The average dry density of the tailings sand, with fluctuation components. Assume it is a stationary random field.
[0035] ; Step 104: Under a stationary random field, the fluctuation distance is set as the key parameter for controlling spatial variability based on the fluctuation components; wherein, the key parameter is characterized by the autocorrelation function, which represents the correlation coefficient between the two endpoints of the fluctuation distance in the stationary random field. As an embodiment of this application, under a stationary random field, the fluctuation distances θx, θy, and θz in the three directions of x, y, and z are set as key parameters for controlling spatial variability, and the correlation coefficient between two points in space is characterized by an autocorrelation function.
[0036] Step 105: Construct a three-dimensional covariance total correlation matrix based on the key parameters; As one embodiment of this application, this application generates a one-dimensional correlation matrix in three directions based on key parameters, namely the autocorrelation function and the fluctuation distance. , , Then, a three-dimensional covariance total correlation matrix is constructed. The three-dimensional covariance total correlation matrix is also based on the grid architecture of the three-dimensional random grid in the region where the lens is located.
[0037] Step 106: Decompose the three-dimensional covariance total correlation matrix using the Kronecker product to determine the correlation matrix in multiple directions; As one embodiment of this application, the Kronecker product decomposition involves decomposing the three-dimensional total correlation matrix into a product of three one-dimensional correlation matrices in three directions using the Kronecker product. During decomposition, it can be decomposed into a 2×2×2 one-dimensional matrix in the x, y, and z directions, or a one-dimensional and two-dimensional matrix in the x, y, and z directions, as shown in the attached figure. Figure 5 As shown.
[0038] ; in, This is the Kronecker product operator. Through this decomposition, the original process (nx × ny × nz) can be reduced to... A 2D giant matrix is transformed into three smaller matrices (nx², ny², nz²). This processing significantly reduces computational complexity.
[0039] Step 107: Determine the triangular matrix by performing Cholleski decomposition on the correlation matrix; As one embodiment of this application, the correlation matrices in each direction are decomposed using Cholleski decomposition to obtain lower triangular matrices. , , These correspond to the x, y, and z directions, respectively. Lower triangular matrix. , , After applying the Choreski decomposition method to the three-dimensional decomposition matrix, it can be decomposed into: or ; but ; Where T represents the matrix transpose operation; Step 108: Determine the random standard vector based on the triangular matrix, and determine the random distribution field of the lens body with a specified fluctuation distance through the random standard vector to generate a quantitative model of the spatial variation of the tailings dam.
[0040] In this application, L is a lower triangular matrix of C. Then, a series of independent random standard vectors U are generated. The quantitative evolution method for the spatial variation of the tailings dam lens body can then be expressed as: ; As one embodiment of this application, when generating a random field distribution, a random distribution field of the lens body with a specified fluctuation distance is obtained by calculating based on a random seed, i.e., a random standard vector, using the following formula: for The lower triangular matrix is used. The random fluctuation component S is superimposed with the mean μ to obtain the final three-dimensional dry density distribution field. In the three-dimensional dry density distribution field, each grid cell is assigned a dry density value, and regions with dry densities below a set threshold are identified as low-density lenses. Simultaneously, fine-grained sandwich lens regions are identified based on a material classification threshold. Preferably, the set material classification threshold is determined by statistically analyzing the probability distribution characteristics of the three-dimensional dry density distribution field, for example, regions with dry densities below one standard deviation or a specific quantile of the average dry density are identified as low-density lenses.
[0041] Through the above steps, a quantitative model of spatial variation is constructed, such as... Figure 4 As shown, the spatial variation quantification model is a three-dimensional digital matrix that records the dry density assignment results of each tiny unit in the physical model, serving as a digital blueprint for the construction of the physical model.
[0042] Example 3, the autocorrelation function is: ; in, This represents the relative distance between different points along the x-direction; This represents the relative distance between different points along the y-direction; This represents the relative distance between different points along the z-direction; This represents the distance of fluctuation along the x-direction; This represents the distance of fluctuation along the y-direction; This represents the fluctuation distance along the z-direction.
[0043] As one embodiment of this application, the autocorrelation function is a mathematical function used to describe the correlation between two points in space in a random field. Its value decreases as the distance between the two points increases, and the rate of decrease is controlled by the fluctuation distance. This application preferably uses an exponential autocorrelation function.
[0044] Example 4, as Figure 6 As shown, the specific steps for building the physical test model before conducting the test on the bottom friction test bench include: Step 201: Determine the geometric parameters of the target test model based on the bottom friction test rig; As one embodiment of this application, during the construction of the target physical test model, the target test model uses the scaling of the tailings sand in the tailings dam sediment as the geometric parameters of the target test model. For example, the test model is designed with a scaling ratio of 1:150, the model is approximately trapezoidal, 770mm long, 210mm high, 8mm thick, with a step width of 40mm and a step height of 70mm. This allows for the integration of the weakening effect of two occurrence forms—fine-grained interlayer lenses and low-density lens zones—on the mechanical strength of the tailings sand in the tailings dam sediment, specifically targeting the tailings clay fine-grained interlayer lenses (density 1.21). ) and low-density lens (density 1.50) The research was conducted. The model design defines a dry beach area of 100mm in length, corresponding to an actual dry beach length of 15m. Based on the survey results of the dry beach length at the tailings dam site, the dry beach length during the tailings dam operation period is 180m. Therefore, the seepage effect of water is not considered in the model.
[0045] Step 202: Determine multiple density regions of the target experimental model based on geometric parameters; As an embodiment of this application, because the physical state of the sedimentary tailings sand in different regions is different, in the process of model design, the sedimentary sand is divided into multiple different density regions by combining the geometric parameters of the overall model and the different sedimentary tailings sand regions through different dry density values, and the dry density value of each region is different.
[0046] Step 203: Based on the dry density assignment results of multiple density regions in the quantitative model of spatial variation of tailings dam, determine the dry density value of each region in the quantitative model of spatial variation of tailings dam, and output the model image of the quantitative model of spatial variation of tailings dam. As one embodiment of this application, when simulating a tailings dam using a bottom friction test rig, the physical state of the sedimentary tailings sand in different regions is determined according to a spatial variability quantitative model. Then, based on the dry density assignment results for different regions, the dry density values of different sedimentary tailings sand regions under the actual test scenario are determined. The overall simulation model, divided by dry density values, constitutes a model image. The model image is a simulation image, and the bottom friction test rig is as follows: Figure 7 As shown.
[0047] Step 204: Perform grayscale processing and image recognition on the model image to obtain the model size of the target physical test model, and determine the lens body region and the dry density value of the lens body region. As one embodiment of this application, this application achieves the identification and extraction of the lens body contour in the model by performing grayscale processing on the output model image. Combined with the dry density distribution results, the extracted lens body contour is re-identified and marked, and the marked lens body is extracted again to form a quantitative model of spatial variation of tailings dam lens body for slope instability test, so as to physically model the lens body area and determine the contour and coordinate data of the lens body area.
[0048] Step 205: Determine the tail sand material in the lens area based on the dry density value, and weigh the required tail sand material; As one embodiment of this application, the physical test model dimensions are obtained based on pre-set test objectives and field survey data, and the required weights of various tailings sand materials are calculated according to their dry density values. It is understood that in this application, the base material is tailings sand, the fine-grained interlayer lens material is tailings clay, and the low-density zone lens material is tailings silt. The material mechanical parameters are shown in the table below: Step 206: The tailing sand material is evenly loaded onto the belt of the bottom friction test bench and compacted according to the model size of the target test model so that the target test model reaches the designed thickness. As one embodiment of this application, the base material is uniformly loaded onto the annular belt, preferably a rectangular model of 80cm×22cm is compacted, and the model is ensured to reach the designed thickness during the compaction process, preferably 8mm.
[0049] Step 207: Start the bottom friction test bench to make the belt rotate at a uniform speed and perform pre-consolidation treatment on the physical test model; As one embodiment of this application, the bottom friction test bench is started, and the belt is rotated at a constant speed for 2 revolutions to pre-consolidate the physical test model and eliminate the original compressive stress generated during the slope cutting and shaping process.
[0050] Step 208: Determine the lens body outline according to the lens body area, mark the lens body area on the target test model, remove the raw material in the marked area of the lens body, roughen the junction, and then fill with the predetermined lens body material and compact it. As one embodiment of this application, based on the lens body contour in a pre-generated quantitative model, the model output image is processed into grayscale to achieve accurate identification and extraction of the lens body contour, such as... Figure 4As shown in part b. The lens body region is marked on the physical model, the raw material in this region is removed, and the junction is roughened to enhance bonding strength. Then, the predetermined lens body material is filled and compacted. Based on the dry density distribution results, the extracted lens body contour is re-identified and marked to form the final experimental model, as shown in part b. Figure 4 part c and Figure 8 As shown. In the actual implementation process, the lens body area is marked on the target test model, a thin-walled isolation sheet is inserted at the boundary of the marked area to support the surrounding material, then the raw material in the marked area of the lens body is removed, the junction is slightly compacted, or a small amount of water mist is sprayed to increase the temporary apparent cohesion, then the predetermined lens body material is filled and compacted; after filling is completed, the isolation sheet is removed.
[0051] Step 209: After compaction, the target test model is modified to generate the target physical test model.
[0052] In one embodiment of this application, after filling, the physical test model needs to be refined to achieve the predetermined geometry and size, thereby generating the target physical test model. This involves refining target parameters such as the model's length, height, thickness, step width, step height, and length of the dry beach area.
[0053] Example 5, the raw materials for removing the lens body marking area include: performing grayscale processing and image recognition on the lens body outline of the lens body marking area to determine the lens body boundary.
[0054] In one embodiment of this application, when identifying the marked area of the lens body, in order to prevent the boundary of the lens body area from being indistinguishable, it is necessary to remove the raw material of the marked area of the lens body, and then identify it based on the image after removal to distinguish the obvious boundary between the lens body and other parts of the tailings sand.
[0055] Example 6, as Figure 9 As shown, step 4 includes: Step 401: Create a speckle pattern on the surface of the physical test model; In one embodiment of this application, white speckle patterns are uniformly sprayed onto the surface of a physical test model for image recognition.
[0056] Step 402: Continuously acquire image sequences of the model during the experiment using a high-speed camera; In one embodiment of this application, a MatchID high-speed camera from the slope failure observation system is mounted, preferably 80cm above the physical test model, and connected to the data acquisition system. The sampling frequency of the high-speed camera is preferably set to 3 seconds per frame, the full-screen frequency is preferably set to 30Hz, and the displacement resolution is preferably set to 0.001-0.01 pixels. Simultaneously, supplementary lighting is used to ensure image quality, and image sequences are acquired continuously during the experiment.
[0057] Step 403: Perform correlation calculations on the image sequence, track the motion trajectory of the speckle, and obtain the full-field displacement vector of the physical test model surface; In one embodiment of this application, the correlation calculation is used to obtain the displacement vector of the image sequence: correlation calculation is performed on the image sequence to track the motion trajectory of the speckle, obtaining the full-field displacement vector of the model surface. The displacement calculation of the above displacement field is based on deformation motion analysis, such as... Figure 10 As shown, the displacement vector of a point within the object before and after deformation is obtained through the displacement approximation search.
[0058] The formula for calculating the displacement vector is as follows: The horizontal deformation of the figure is as follows: ; The vertical deformation of the figure is as follows: ; The shear strain during tilting deformation is: ; and Is and The displacement component in the direction.
[0059] Step 404: Based on the obtained displacement field, the displacement data is smoothed using the least squares fitting method, and the deformation gradient matrix is constructed; In one embodiment of this application, the purpose of smoothing is to eliminate experimental noise, and the least squares fitting method is used to smooth the displacement data. Around the discrete displacement data points to be determined ( , An N×N square strain window is selected, and the displacement is analytically approximated using a bilinear Lagrange polynomial. Specifically, the displacement data is fitted using a bilinear Lagrange polynomial or a biquadratic Lagrange polynomial.
[0060] as follows: ; ; These are the undetermined coefficients in the polynomial.
[0061] These are discrete displacement data points.
[0062] Based on the fitted displacement field, the deformation gradient matrix F is constructed. For the bilinear case: ; Step 405: Calculate the strain field based on the deformation gradient matrix.
[0063] In one embodiment of this application, the strain field is calculated using the Euler-Almansi or Green-Lagrange strain tensor based on the deformation gradient matrix F, thus obtaining the strain field. Specifically, after smoothing transformation: ; The dummy strain VSG in the strain formula includes a subset SS, a step size ST, and a strain window SW, expressed as: ; Through the above calculations, we can obtain different time points such as Figure 11 The horizontal displacement field shown, such as Figure 12 The vertical displacement field shown and as Figure 13 The total displacement field is shown.
[0064] Example 7: Based on the physical test model, the instability and catastrophic parameters of the tailings dam slope under different test commands were determined; wherein, the test commands include: instability and failure identification command, slippage point location command, and slippage volume calculation command.
[0065] In one embodiment of this application, based on the obtained displacement and strain field evolution processes, and according to different test instructions—namely, instability failure identification instructions, slippage point location instructions, and slippage volume calculation instructions—the aforementioned defined instability disaster data of the tailings dam slope under the action of the lens body are determined. This includes: instability failure identification, corresponding to the instability failure identification instruction: automatically identifying the location and propagation path of cracks through displacement field contour maps. In this application, based on the characteristics of displacement field evolution, instability modes are automatically identified: traction landslides and composite landslides. The slippage point location instruction extracts the slippage point location coordinates from the displacement field contour map. Analysis shows that when the lens body is located at the toe of the slope, the slippage point is located at the toe of the slope; as the distance between the lens body and the slope bottom increases, the slippage point moves upstream. The slippage volume calculation instruction calculates the slippage volume based on the area of the sliding region in the displacement field and the model thickness. Comparing the test results with different fluctuation distances, the relationship between the lens body position and the slippage volume is obtained: the slippage volume is largest when the lens body is located at the toe of the slope, and decreases when it is located in the middle of the slope.
[0066] Example 8: This application proposes a simulation test system for tailings dam slope instability, comprising: Model building module: This module acquires the statistical characteristic parameter set of tailings sediment in the tailings dam under a random field model, and constructs a quantitative model of the spatial variation of the tailings dam, including lenses, using a stepwise decomposition method. Specifically, this module executes the algorithm in step 1 of Example 1, including probability distribution analysis, fluctuation distance setting, autocorrelation function selection, Kronecker multiplication integral solution, Chollisky decomposition, and random field generation.
[0067] Physical Experiment Module: Based on the quantitative model of spatial variation in tailings dams, a physical experiment model is constructed on a bottom friction test bench. This module includes the bottom friction test bench and model-making tools, used to construct the physical experiment model on the bottom friction test bench according to the quantitative model of spatial variation in tailings dams. The specific structure of the bottom friction test bench includes a frame, motor, driving roller, driven roller, annular belt, model-making frame, compaction mechanism, removal mechanism, and filling mechanism. The model-making tools, based on the dry density assignment results output by the model construction module, perform operations such as material weighing, compaction, lens area removal, and material filling to generate a physical model that meets the requirements.
[0068] The test data acquisition module: This module applies simulated gravity loads to the physical test model using the bottom friction test bench and acquires image data in real time via a high-speed camera. It also includes a slope failure observation system, which is used to apply simulated gravity loads to the physical test model using the bottom friction test bench and acquire image data in real time via a high-speed camera. The slope failure observation system comprises a high-speed camera, a shooting bracket, a supplementary light, and a data acquisition unit. The high-speed camera has a sampling frequency of 75–500 Hz and a displacement resolution of 0.001–0.01 pixels.
[0069] The analysis module processes image data using digital image correlation (DIC) to obtain displacement and strain fields. Based on the evolution of these fields, it determines the instability and catastrophic events of the tailings dam slope under the action of a lens. Connected to the experimental data acquisition module, this module processes image data using DIC to obtain displacement and strain fields and determines the instability and catastrophic events of the tailings dam slope under the action of a lens. Specifically, this module executes the DIC analysis algorithm described in Example 1, including speckle tracking, displacement field calculation, least-squares fitting, deformation gradient matrix construction, strain field calculation, instability mode recognition, slippage point location, and slippage volume calculation.
[0070] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.
Claims
1. A simulation test method for tailings dam slope instability, characterized in that, include: Step 1: Obtain the set of statistical characteristic parameters of tailings sand in tailings dam sediment under a random field model, and construct a quantitative model of spatial variation of tailings dam containing lenses by using the stepwise decomposition method. Step 2: Based on the quantitative model of spatial variation of tailings dam, a physical test model is built on the bottom friction test rig; Step 3: Start the bottom friction test bench to apply simulated gravity load to the physical test model, and use a high-speed camera to collect image data in real time during the test process; Step 4: Process image data using digital image correlation to obtain displacement and strain fields, and determine the instability and catastrophic data of tailings dam slope under the action of the lens body based on the evolution of displacement and strain fields.
2. The simulation test method for tailings dam slope instability as described in claim 1, characterized in that, Step 1 includes: Step 101: Obtain field survey data of tailings sand deposited in tailings dams and determine the probability distribution characteristics of the physical and mechanical parameters of tailings sand deposited in tailings dams. Step 102: Based on the probability distribution characteristics, determine whether the tailings sand in the tailings dam sediment follows a Gaussian distribution; Step 103: When the material follows a Gaussian distribution, determine the mechanical properties of the tailings dam material and decompose the mechanical properties of the tailings dam material into trend components and fluctuation components; Step 104: Under a stationary random field, the fluctuation distance is set as the key parameter for controlling spatial variability based on the fluctuation components; wherein, the key parameter is characterized by the autocorrelation function, which represents the correlation coefficient between the two endpoints of the fluctuation distance in the stationary random field. Step 105: Construct a three-dimensional covariance total correlation matrix based on the key parameters; Step 106: Decompose the three-dimensional covariance total correlation matrix using the Kronecker product to determine the correlation matrix in multiple directions; Step 107: Determine the triangular matrix by performing Cholleski decomposition on the correlation matrix; Step 108: Determine the random standard vector based on the triangular matrix, and determine the random distribution field of the lens body with a specified fluctuation distance through the random standard vector to generate a quantitative model of the spatial variation of the tailings dam.
3. The simulation test method for tailings dam slope instability as described in claim 2, characterized in that, The autocorrelation function is: ； in, This represents the relative distance between different points along the x-direction; This represents the relative distance between different points along the y-direction; This represents the relative distance between different points along the z-direction; This represents the distance of fluctuation along the x-direction; This represents the distance of fluctuation along the y-direction; This represents the fluctuation distance along the z-direction.
4. The simulation test method for tailings dam slope instability as described in claim 1, characterized in that, The specific steps for building the physical experimental model include: Step 201: Determine the geometric parameters of the target test model based on the bottom friction test rig; Step 202: Determine multiple density regions of the target experimental model based on geometric parameters; Step 203: Based on the dry density assignment results of multiple density regions in the quantitative model of spatial variation of tailings dam, determine the dry density value of each region in the quantitative model of spatial variation of tailings dam, and output the model image of the quantitative model of spatial variation of tailings dam. Step 204: Perform grayscale processing and image recognition on the model image to obtain the model size of the lens body region, and determine the lens body region and the dry density value of the lens body region; Step 05: Determine the tail sand material in the lens body area based on the dry density value, and weigh the required tail sand material. Step 206: The tailing sand material is evenly loaded onto the belt of the bottom friction test bench and compacted according to the model size of the target test model so that the target test model reaches the designed thickness. Step 207: Start the bottom friction test bench to make the belt rotate at a uniform speed and perform pre-consolidation treatment on the physical test model; Step 208: Determine the lens body outline according to the lens body area, mark the lens body area on the target test model, remove the raw material in the marked area of the lens body, roughen the junction, and then fill with the predetermined lens body material and compact it. Step 209: After compaction, the target test model is modified to generate the target physical test model.
5. The simulation test method for tailings dam slope instability as described in claim 4, characterized in that, The raw materials for removing the lens body marking area include: performing grayscale processing and image recognition on the lens body outline of the lens body marking area to determine the lens body boundary.
6. The simulation test method for tailings dam slope instability as described in claim 1, characterized in that, Step 4 includes: Step 401: Create a speckle pattern on the surface of the physical test model; Step 402: Continuously acquire image sequences of the model during the experiment using a high-speed camera; Step 403: Perform correlation calculations on the image sequence, track the motion trajectory of the speckle, and obtain the full-field displacement vector of the physical test model surface; Step 404: Based on the obtained displacement field, the displacement data is smoothed using the least squares fitting method, and the deformation gradient matrix is constructed; Step 405: Calculate the strain field based on the deformation gradient matrix.
7. The simulation test method for tailings dam slope instability as described in claim 6, characterized in that, Step 404 includes: fitting the displacement data using a bilinear Lagrange polynomial or a biquadratic Lagrange polynomial.
8. The simulation test method for tailings dam slope instability as described in claim 6, characterized in that, Step 405 includes: calculating the strain field using the Euler-Almansi strain tensor or the Green-Lagrange strain tensor.
9. The simulation test method for tailings dam slope instability as described in claim 1, characterized in that, Step 4 further includes: determining the instability and catastrophic parameters of the tailings dam slope under different test commands based on the physical test model; wherein the test commands include: instability and failure identification command, slippage point positioning command, and slippage volume calculation command.
10. A simulation test system for tailings dam slope instability, characterized in that, include: Model building module: Obtain the statistical characteristic parameter set of tailings sand in tailings dam sediment under random field model, and construct a quantitative model of spatial variation of tailings dam including lens body by using the stepwise decomposition method. Physical test module: Based on the quantitative model of spatial variation of tailings dam, a physical test model is built on the bottom friction test platform; Test data acquisition module: The bottom friction test bench is activated to apply simulated gravity load to the physical test model, and image data during the test process is acquired in real time through a high-speed camera; Analysis module: Based on digital image correlation method, image data is processed to obtain displacement field and strain field, and the instability and disaster situation of tailings dam slope under the action of lens body is determined according to the evolution process of displacement field and strain field.