Ionic type rare earth mining area landslide early warning method, device, equipment, medium and product based on discrete element simulation

Abstract

The invention discloses an ion type rare earth mining area landslide early warning method and device based on discrete element simulation, equipment, a medium and a product, and relates to the field of landslide early warning, and the method comprises the steps: obtaining key geological parameters and environmental parameters; constructing a mining area three-dimensional geologic model and a particle discrete element model; correcting the particle discrete element model and simulating a displacement field evolution diagram, a stress distribution evolution diagram and a sliding surface evolution trend of the post-disaster mine; obtaining a multi-dimensional feature; constructing a machine learning model; inputting the multi-dimensional features into a random forest model to obtain key trigger factors; training the long-short-term memory network and predicting future mine displacement and landslide probability; the future mine displacement, the landslide probability and the key trigger factors are combined to obtain combined features; judging whether the combined features exceed an early warning value or not; and if the pre-warning value is exceeded, re-operating the particle discrete element model. The prediction precision and the prediction efficiency can be improved.

Description

TECHNICAL FIELD

[0001] The application relates to the field of landslide early warning, in particular to a landslide early warning method, device, equipment, medium and product based on discrete element simulation in an ion-type rare earth mine area. BACKGROUND

[0002] Rare earth mines are important resources for modern high-tech industries, and their exploitation and utilization directly affect the strategic industrial layout and technological innovation of a country. Ion-type rare earth mines are mainly distributed in countries such as China, the United States, and Brazil, and the ore is mainly extracted from the ore deposit by leaching. Such mine areas are usually located in mountainous or hilly areas with complex terrain, large differences in soil and rock structure, and are prone to landslides. During the mining process, a large amount of leaching solution is injected into the stope, causing changes in the structure of the soil and rock layers in the stope, and the hydrogeological conditions also fluctuate dramatically, thus providing favorable conditions for the occurrence of landslides. In recent years, with the acceleration of rare earth resource exploitation, the frequency and severity of natural disasters such as landslides in mine areas have gradually increased, especially in some specific mine areas such as ion-type rare earth mine areas, due to the special geological structure and mining methods, the occurrence of landslide disasters has a certain regularity and periodicity.

[0003] The prediction and early warning of landslide disasters are of great significance to ensure the safety of mine operations, reduce personnel casualties and property losses. Traditional landslide early warning methods rely heavily on empirical analysis and simplified theoretical models, which are difficult to accurately reflect the landslide occurrence mechanism in complex geological environments. Therefore, there is an urgent need for a landslide early warning method based on advanced technology to address the challenges of current mine landslide disasters.

[0004] The prediction and early warning of landslide disasters are an important technology in geological disaster prevention. With the development of computer technology, more and more landslide early warning methods have been proposed. However, existing landslide early warning technologies still have many shortcomings, mainly in the following aspects:

[0005] Traditional landslide early warning methods mainly rely on geological surveys, historical data analysis, and physical models. Although these methods can identify the trend of landslide occurrence to some extent, they often overlook the dynamic changes in geological conditions and complex mechanical mechanisms. For example, in ion-type rare earth mine areas, due to the particularity of the ore deposit itself, traditional empirical methods are difficult to accurately reflect the specific conditions of landslide occurrence.

[0006] The geological environment of ion-type rare earth mine areas is complex and variable, including different soil types, rock structures, and groundwater conditions, which makes the landslide occurrence mechanism complex and unstable. Traditional landslide early warning methods often overlook these dynamic and local geological changes, and cannot accurately capture potential landslide risks.

[0007] Many landslide prediction models simplify the mechanical properties of geological bodies and their interaction with external factors (such as precipitation, vibration, etc.) in the process of establishment. The prediction results of these simplified models have large errors, which are difficult to meet the high-precision and high-timeliness requirements of landslide early warning.

[0008] Most of the current landslide monitoring technologies mainly rely on ground monitoring instruments (such as geological radar, displacement sensor, etc.) for data acquisition. These technologies often have problems such as poor timeliness, limited coverage, and low spatial resolution for landslide prediction in complex geological environments, especially for large-scale landslide monitoring.

[0009] The existing numerical simulation methods (such as finite element) have low simulation accuracy for discontinuous media (such as fractured rare earth weathering layer), and cannot accurately reflect the mechanical behavior between soil particles. SUMMARY

[0010] The purpose of the present application is to provide a landslide early warning method, device, equipment, medium and product based on discrete element simulation in ion-type rare earth mine area, which can improve the prediction accuracy and efficiency.

[0011] To achieve the above purpose, the present application provides the following solutions:

[0012] In a first aspect, the present application provides a landslide early warning method based on discrete element simulation in ion-type rare earth mine area, comprising:

[0013] Obtaining key geological parameters and environmental parameters;

[0014] Building a three-dimensional geological model of the mine area based on the key geological parameters and environmental parameters;

[0015] Building a particle discrete element model based on the three-dimensional geological model of the mine area;

[0016] Correcting the particle discrete element model;

[0017] Simulating the displacement field evolution diagram, stress distribution evolution diagram and sliding surface evolution trend of the post-disaster mine based on the corrected particle discrete element model;

[0018] Obtaining multi-dimensional features; the multi-dimensional features include statistical features, derived features and environmental features;

[0019] Building a machine learning model; the machine learning model includes a random forest model and a long short-term memory network;

[0020] Inputting the multi-dimensional features into the random forest model to obtain key trigger factors;

[0021] Training the long short-term memory network based on the key geological parameters and environmental parameters;

[0022] Predict future mine displacement and landslide probability based on trained long short-term memory networks;

[0023] combining the future mine displacement and landslide probability and key triggering factors to obtain a combined feature;

[0024] Determine whether the combined features exceed the warning value;

[0025] If the warning value is exceeded, return to step "correcting the particle discrete element model" to re-simulate the displacement field evolution diagram, stress distribution evolution diagram and sliding surface evolution trend of the post-disaster mine.

[0026] Optionally, the key geological parameters and environmental parameters include: geological drilling data, surface deformation monitoring data, groundwater level and pore water pressure sensor data, precipitation and surface runoff data, and historical landslide data.

[0027] Optionally, the ionic rare earth mining area landslide early warning method based on discrete element simulation further includes, between the step of "obtaining key geological parameters and environmental parameters" and the step of "constructing a three-dimensional geological model of the mining area based on the key geological parameters and environmental parameters":

[0028] The key geological parameters and environmental parameters are preprocessed.

[0029] Optionally, the three-dimensional geological model of the mining area includes the following elements:

[0030] Stratigraphic distribution and physical parameters, fault structure and joint network, hydrogeological units, slope boundaries and mining disturbance areas; the bottom layer distribution and physical parameters include: triaxial strength, friction angle and density, etc.; the hydrogeological units include: aquifers and aquicludes.

[0031] Optionally, constructing a particle discrete element model based on the three-dimensional geological model of the mining area specifically includes the following steps:

[0032] Granular modeling: The mining area rock and soil are converted into a particle network, where each particle represents a rock unit, and the shear, compression and failure behaviors between rock layers are simulated through contact forces;

[0033] Mechanical property assignment: Based on geological drilling and experimental data, parameters such as friction angle, shear strength, and cohesion are assigned to each type of particle;

[0034] Initial boundary condition setting: setting the initial gravity field, groundwater level and external force boundary of mining disturbance;

[0035] Dynamic disturbance simulation: Input external disturbance factors such as rainfall and leaching mining operations to dynamically simulate the response behavior of the geological body.

[0036] Optionally, the statistical features include displacement change rate, pore pressure change gradient, and cumulative rainfall; the derived features include displacement acceleration and stress field energy density; and the environmental features include mining activity intensity and joint surface dip angle and other geological parameters.

[0037] In a second aspect, the application provides a landslide early warning device for an ion-type rare earth mine based on discrete element simulation, comprising:

[0038] a parameter acquisition module configured to acquire key geological parameters and environmental parameters;

[0039] a three-dimensional geological model construction module configured to construct a three-dimensional geological model of the mine based on the key geological parameters and the environmental parameters;

[0040] a particle discrete element model construction module configured to construct a particle discrete element model based on the three-dimensional geological model of the mine;

[0041] a correction module configured to correct the particle discrete element model;

[0042] a simulation module configured to simulate a displacement field evolution map, a stress distribution evolution map, and a sliding surface evolution trend of the post-disaster mine based on the corrected particle discrete element model;

[0043] a multi-dimensional feature acquisition module configured to acquire multi-dimensional features; the multi-dimensional features include statistical features, derived features, and environmental features;

[0044] a machine learning model construction module configured to construct a machine learning model; the machine learning model includes a random forest model and a long short-term memory network;

[0045] a key trigger factor determination module configured to input the multi-dimensional features into the random forest model to obtain a key trigger factor;

[0046] a training module configured to train the long short-term memory network based on the key geological parameters and the environmental parameters;

[0047] a prediction module configured to predict future mine displacement and landslide probability based on the trained long short-term memory network;

[0048] a feature combination module configured to combine the future mine displacement and landslide probability and the key trigger factor to obtain combined features;

[0049] a judgment module configured to determine whether the combined features exceed a warning value;

[0050] The re-running module is configured to return to the correction module when the early warning value is exceeded, and to re-simulate a displacement field evolution diagram, a stress distribution evolution diagram and a sliding surface evolution trend of the post-disaster mine.

[0051] In a third aspect, the present application provides a computer device, comprising a memory, a processor, a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the steps of the landslide early warning method for ion-type rare earth mine area based on discrete element simulation according to any one of the above embodiments.

[0052] In a fourth aspect, the present application provides a computer readable storage medium, having a computer program stored thereon, and the computer program is executed by a processor to implement the steps of the landslide early warning method for ion-type rare earth mine area based on discrete element simulation according to any one of the above embodiments.

[0053] In a fifth aspect, the present application provides a computer program product, comprising a computer program, and the computer program is executed by a processor to implement the steps of the landslide early warning method for ion-type rare earth mine area based on discrete element simulation according to any one of the above embodiments.

[0054] According to the specific embodiments provided in the present application, the present application has the following technical effects:

[0055] The application provides a landslide early warning method, device, equipment, medium and product based on discrete element simulation of an ion-type rare earth mine area, which can simulate the deformation and damage process of granular materials such as soil and rock in detail through a discrete element method, can accurately describe the mechanical behavior and changes of the geological body in the mine area, is especially suitable for landslide analysis under a complex geological environment, can accurately predict the time, position and scale of landslide occurrence through detailed modeling of the interaction between particles, and can simulate the discontinuity and micro damage mechanism of rock mass, is suitable for the complex geological environment of the ion-type rare earth mine area, can consider the dynamic response of soil and rock and the interaction with external factors (such as precipitation, earthquake, etc.) in the mining process of the mine area, can better simulate the occurrence mechanism of landslide disasters, and has a dynamic simulation capability that cannot be matched by traditional static mechanical models, the ion-type rare earth mine area often has irregular terrain, and a traditional model is difficult to accurately describe the geological environment, the discrete element method can better adapt to the complex and irregular terrain structure and accurately simulate landslide through granular modeling, the landslide early warning system based on discrete element simulation can be combined with real-time monitoring data (such as ground displacement, precipitation, etc.), model parameters can be dynamically adjusted, and the timeliness and accuracy of the early warning system can be improved, the real-time dynamic adjustment capability is difficult to achieve by a traditional early warning system, data-driven technology is used to automatically identify key trigger factors such as rainfall threshold and groundwater level change rate, has learning and adaptation capabilities, all landslide simulation results can be visualized and output, dynamic three-dimensional demonstration of the landslide process is realized, and an intuitive basis is provided for on-site decision-making, the discrete element method can combine geological data, meteorological data, remote sensing data and other data sources to perform comprehensive analysis and prediction, and provide more comprehensive information support for early warning of landslide disasters, and if the prediction result of the long short-term memory network exceeds a threshold value, the particle discrete element model is automatically triggered, unnecessary full calculation of the particle discrete element model is reduced, and efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0057] Figure 1 A flowchart of a landslide early warning method based on discrete element simulation of an ion-type rare earth mine area provided by an embodiment of the present application is shown in the figure.

[0058] Figure 2 A block diagram of a landslide early warning method based on discrete element simulation of an ion-type rare earth mine area provided by an embodiment of the present application is shown in the figure.

[0059] Figure 3 A structure schematic diagram of a landslide early warning device based on discrete element simulation for an ion type rare earth mine area is provided for an embodiment of the present application.

[0060] Figure 4 A structure schematic diagram of a computer device is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present application will be described clearly and completely below with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0062] The above purposes, features and advantages of the present application can be more obvious and easy to understand. The present application will be further described in detail below with the drawings and specific embodiments.

[0063] In an exemplary embodiment, as shown in Figure 1 A landslide early warning method based on discrete element simulation for an ion type rare earth mine area is provided, which is executed by a computer device, specifically, can be executed by a terminal or a server alone or by a terminal and a server together, and includes the following steps 201 to 208. Wherein:

[0064] Step 101: Obtain key geological parameters and environmental parameters.

[0065] This step obtains key geological and environmental parameters through various sensors and remote sensing systems deployed in the rare earth mine area, including but not limited to: geological drilling data (rock structure, interlayer shear strength, etc.), ground deformation monitoring data (GNSS, InSAR), groundwater level and pore water pressure sensor data, precipitation and surface runoff data, and historical landslide data (landslide location, scale, trigger condition).

[0066] The above data will be standardized by a preprocessing module, including missing value completion, data cleaning and resampling, unified time and space scale, to provide high-quality input for subsequent modeling.

[0067] Step 102: Construct a three-dimensional geological model of the mine area based on the key geological parameters and environmental parameters.

[0068] A modeling software (such as 3Dmine, surpac) is used in combination with a GIS platform to conduct three-dimensional modeling of the geological structure of the mine area. The model includes the following elements:

[0069] Strata distribution and physical property parameters: including triaxial strength, friction angle, density, etc.

[0070] Fracture structure and joint network;

[0071] Hydrogeological unit: including aquifer, aquifuge;

[0072] Slope boundary and mining disturbance area;

[0073] The model serves as the geometric basis for discrete element modeling, determining boundary conditions, initial stress field, and mechanical element division.

[0074] Step 103: Construct a particle discrete element model based on the mine area three-dimensional geological model.

[0075] Using discrete element software (such as UDEC, PFC, RockyDEM), construct a particle discrete element model based on the three-dimensional geological model, the specific process includes:

[0076] Particle modeling: convert the mine area rock-soil mass into a particle network, with each particle representing a rock element, simulating shear, compression, and failure behavior between rock layers through contact forces.

[0077] Mechanical property assignment: according to geological drilling and experimental data, assign parameters such as friction angle, shear strength, and cohesion to each type of particle.

[0078] Initial boundary condition setting: set the initial gravity field, groundwater level, and external force boundary of mining disturbance.

[0079] Dynamic disturbance simulation: input external disturbance factors such as rainfall and leaching mining operations to dynamically simulate the response behavior of geological bodies.

[0080] Simulation results will include displacement field evolution map, stress distribution evolution map, and sliding surface evolution trend.

[0081] Step 104: Correct the particle discrete element model.

[0082] To improve prediction accuracy, the model needs to be fused and corrected with real-time observation data: based on actual monitoring of ground displacement, pore water pressure, and time of crack appearance on the slope surface, use parameter inversion method to adjust model input.

[0083] Specifically, compare the observed data (pore water pressure, displacement, time of crack appearance on the slope surface) with the numerical model calculation results, adjust the discrete element model or calculation parameters, and correct the numerical model.

[0084] Step 105: Based on the corrected particle discrete element model, simulate the displacement field evolution map, stress distribution evolution map, and sliding surface evolution trend of the post-disaster mine.

[0085] Step 106: Obtain multi-dimensional features; the multi-dimensional features include statistical features, derived features, and environmental features.

[0086] Step 107: Build a machine learning model; the machine learning model includes a random forest model and a long short-term memory network.

[0087] Step 108: Input the multi-dimensional features into the random forest model to obtain key trigger factors.

[0088] Step 109: Train the long short-term memory network based on the key geological parameters and environmental parameters.

[0089] Step 110: Predict future mine displacement and landslide probability based on the trained long short-term memory network.

[0090] Specifically, first, data preparation and feature construction are performed:

[0091] 1. Input data:

[0092] Time series data: real-time sensor data such as surface displacement, pore water pressure, rainfall, and groundwater level.

[0093] Historical data: records of past landslide events and their triggering conditions (such as rainfall threshold, displacement rate).

[0094] 2. Feature construction:

[0095] Statistical features: displacement rate (mm / h), pore pressure change gradient (kPa / h), cumulative rainfall (24h / 72h).

[0096] Derived features: displacement acceleration, stress field energy density (output by particle discrete element model DEM simulation).

[0097] Environmental features: mining activity intensity, joint surface inclination, and other geological parameters.

[0098] Second, machine learning model construction is performed:

[0099] 1. Model selection:

[0100] Random forest (RandomForest)

[0101] Purpose: Identify the importance ranking of key trigger factors.

[0102] Input: Multi-dimensional features (displacement, pore pressure, rainfall, etc.).

[0103] Output: Feature importance score (such as rainfall cumulative weight accounting for 70%).

[0104] LSTM (Long Short-Term Memory Network)

[0105] Purpose: Learn the precursor evolution law of time series data (such as displacement acceleration trend).

[0106] Input: Historical displacement sequence, rainfall time series (sliding window sampling).

[0107] Output: Displacement prediction value and landslide probability in the next 1-3 days.

[0108] 2. Training and verification:

[0109] Use historical landslide event data as labels (landslide = 1, no landslide = 0).

[0110] Cross-validation to optimize model parameters (such as LSTM time step, random forest tree depth).

[0111] Step 111: Combine the future mine displacement and landslide probability and key trigger factors to get the combined features.

[0112] Step 112: Determine if the combined features exceed the warning value.

[0113] Step 113: If it exceeds the warning value, return to the step "correct the particle discrete element model", re-simulate the post-disaster mine displacement field evolution map, stress distribution evolution map and sliding surface evolution trend.

[0114] The above steps 111-113 are the process of linkage with DEM simulation system, as follows:

[0115] 1. Dynamic feedback mechanism:

[0116] When LSTM predicts displacement abnormal acceleration or random forest identifies key factor combination (such as high rainfall + pore pressure sharp rise):

[0117] Trigger DEM model to immediately recalibrate parameters.

[0118] Start a new round of DEM simulation and update the stability assessment results.

[0119] 2. Real-time data-driven update:

[0120] Every 6 hours (or custom period) input the latest monitoring data into the LSTM model to generate short-term prediction.

[0121] If the prediction result exceeds the warning threshold, automatically trigger the DEM model to run again (without manual intervention).

[0122] 3. Output and decision support

[0123] Visualization of precursor features: Mark abnormal time points (such as sudden changes in displacement rate) detected by LSTM on the early warning interface.

[0124] Displays the main triggers for random forest decisions (e.g., “the current rainfall has reached 80% of the critical value”).

[0125] Fast update effect: Traditional DEM simulation takes several hours, while machine learning models can output prediction results in minutes.

[0126] By pre-screening high-risk periods through LSTM, unnecessary full DEM calculations can be reduced, improving efficiency.

[0127] Time series dynamic simulation: Based on the latest input data, the DEM simulation is periodically rerun to obtain the current stability assessment results.

[0128] 4. Landslide risk warning and result release

[0129] Finally, the early warning module compares the DEM simulation results with the stability evaluation index to form a landslide risk level assessment and presents it to the user through a graphical interface:

[0130] Landslide risk classification map (red / orange / yellow / blue);

[0131] Landslide triggering probability curve;

[0132] Warning time window (predicting the most likely sliding start time);

[0133] Sliding body volume and motion path prediction;

[0134] Alarm release mechanism: Push landslide warning information to mine managers and staff through SMS, APP, web pages, etc.

[0135] Based on the same inventive concept, embodiments of the present application also provide an ionic rare earth mining area landslide warning device based on discrete element simulation for implementing the aforementioned ionic rare earth mining area landslide warning method based on discrete element simulation. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the ionic rare earth mining area landslide warning device based on discrete element simulation provided below can be found in the above-mentioned limitations of the ionic rare earth mining area landslide warning method based on discrete element simulation, and will not be repeated here.

[0136] In an exemplary embodiment, Figure 3 As shown, an ionic rare earth mining area landslide early warning device based on discrete element simulation is provided, comprising:

[0137] Parameter acquisition module, used to obtain key geological parameters and environmental parameters;

[0138] a three-dimensional geological model construction module configured to construct a three-dimensional geological model of the mining area based on the key geological parameters and the environmental parameters;

[0139] a particle discrete element model construction module configured to construct a particle discrete element model based on the three-dimensional geological model of the mining area;

[0140] a correction module configured to correct the particle discrete element model;

[0141] a simulation module configured to simulate a displacement field evolution map, a stress distribution evolution map and a sliding surface evolution trend of the post-disaster mine based on the corrected particle discrete element model;

[0142] a multi-dimensional feature acquisition module configured to acquire multi-dimensional features; the multi-dimensional features include statistical features, derived features and environmental features;

[0143] a machine learning model construction module configured to construct a machine learning model; the machine learning model includes a random forest model and a long short-term memory network;

[0144] a key trigger factor determination module configured to input the multi-dimensional features into the random forest model to obtain a key trigger factor;

[0145] a training module configured to train the long short-term memory network based on the key geological parameters and the environmental parameters;

[0146] a prediction module configured to predict a future mine displacement and landslide probability based on the trained long short-term memory network;

[0147] a feature combination module configured to combine the future mine displacement and landslide probability and the key trigger factor to obtain combined features;

[0148] a judgment module configured to judge whether the combined features exceed a warning value;

[0149] a re-run module configured to return to the correction module to re-simulate the displacement field evolution map, the stress distribution evolution map and the sliding surface evolution trend of the post-disaster mine when the combined features exceed the warning value.

[0150] The technical solutions of the present application are further described in detail below with reference to a specific embodiment:

[0151] Deployment of a landslide warning system for an ion-type rare earth mining area in Jiangxi

[0152] I. Project background

[0153] The mining area is located in the hilly region of southern Jiangxi Province, which is a typical ion-type rare earth mine with more than 10 years of mining history. The mining area is located in a region with frequent rainfall, surrounded by residual red soil, with loose strata and poor stability. In the past, there have been many landslide incidents, posing a great threat to safety and environmental protection.

[0154] II. System deployment scheme

[0155] This embodiment is divided into five steps to complete the construction and operation of the landslide warning system according to the foregoing technical solution:

[0156] Step 1: Data collection and sensor deployment

[0157] Various sensor devices are deployed on the slope, platform and low-lying areas of the mining area:

[0158] Pore water pressure gauge x 12 (deployed at different depths in deep and shallow layers);

[0159] Displacement monitoring points (GNSS) x 8;

[0160] Rain gauge x 4 sets;

[0161] The following historical geological and production data are collected and sorted out:

[0162] Geological profile x 5;

[0163] Landslide records x 8 (including displacement precursors and rainfall);

[0164] Three-dimensional terrain and DEM data.

[0165] Step 2: Three-dimensional geological modeling

[0166] The LeapfrogGeo and ArcGIS are used to establish a three-dimensional geological model of the mining area:

[0167] The strata are divided into 4 layers (residual soil layer, sand shale, weathered zone, bedrock);

[0168] There are 2 main joints of fault structure, named F1 and F2, with angles of 45° and 65°;

[0169] The simulation area is about 2 square kilometers in area and 100 meters in depth;

[0170] The aquifer is endowed with a permeability coefficient and a water level fluctuation range;

[0171] The model is exported in STL format for discrete element simulation.

[0172] Step 3: Discrete element simulation modeling

[0173] A particle model is established using PFC3D software:

[0174] Particle number: about 1.2 million, particle size range: 0.1-0.5 meters;

[0175] Material parameters are set according to experimental data as follows:

[0176] Friction angle: 29°-36°;

[0177] Cohesion: 10-30 kPa;

[0178] Poisson's ratio: 0.25;

[0179] The boundary conditions are set as fixed base and free slope surface. The loading history is simulated by rainfall field, and the simulation lasts for 48 hours of rainstorm event.

[0180] Step 4: Model calibration and real-time update

[0181] Inversion is performed through real-time sensor feedback data (ground displacement, pore pressure change):

[0182] Particle swarm optimization algorithm (PSO) is used to adjust the inter-particle cohesion and friction coefficient;

[0183] Compared with the actual landslide event displacement, the prediction error is controlled within ± 15%;

[0184] The model is updated every 6 hours, and the stability is re-evaluated;

[0185] In addition, an LSTM model is constructed to predict the trend of ground displacement and rainfall sequence, enhancing the ability to identify the possibility of short-term landslide.

[0186] Step 5: Landslide warning release and display

[0187] The warning system sets a three-level response mechanism:

[0188] Risk level Displacement rate threshold Abnormal pore pressure threshold Rainfall threshold Early warning measures Blue early warning ≥ 2 mm / day ≥ 5 kPa ≥ 50 mm Routine monitoring Yellow early warning ≥ 5 mm / day ≥ 10 kPa ≥ 80 mm On-site patrol Red early warning ≥ 10 mm / day ≥ 15 kPa ≥ 120 mm Evacuation plan start

[0189] The three-dimensional landslide simulation interface is deployed on the mine management center server, and the model running results, warning level, and sliding surface visualization path can be accessed through the Web. The system successfully predicted a small-scale landslide after a red warning was initiated, avoiding personnel and equipment losses.

[0190] In an exemplary embodiment, a computer device, which can be a server or a terminal, is provided, and an internal structure diagram thereof can be as shown in Figure 4As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with the terminal outside through the network connection. The computer program is executed by the processor to realize a discrete element simulation-based ion-type rare earth mine area landslide early warning method.

[0191] Those skilled in the art can understand that, Figure 4 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement. In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps in each of the method embodiments described above.

[0192] In one exemplary embodiment, a computer readable storage medium is provided, storing a computer program, which is executed by a processor to implement the steps in each of the method embodiments described above.

[0193] In one exemplary embodiment, a computer program product is provided, including a computer program, which is executed by a processor to implement the steps in each of the method embodiments described above.

[0194] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0195] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0196] The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a blockchain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0197] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described, but as long as the combinations of the technical features do not exist, they should be considered as the scope of the present application.

[0198] The principles and implementation modes of the present application are described by applying specific examples herein. The above description of the embodiments is only used to help understand the method and its core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range can be changed. In conclusion, the content of the present application should not be understood as a limitation.

Claims

1. A method for landslide warning of an ion-type rare earth mine area based on discrete element simulation, characterized in that, The ion-type rare earth mine landslide early warning method based on discrete element simulation comprises: obtaining key geological parameters and environmental parameters; constructing a mine three-dimensional geological model based on the key geological parameters and environmental parameters; constructing a particle discrete element model based on the mine three-dimensional geological model; correcting the particle discrete element model; simulating the displacement field evolution diagram, stress distribution evolution diagram and sliding surface evolution trend of the post-disaster mine based on the corrected particle discrete element model; obtaining multi-dimensional features; the multi-dimensional features include statistical features, derived features and environmental features; constructing a machine learning model; the machine learning model includes a random forest model and a long short-term memory network; inputting the multi-dimensional features into the random forest model to obtain key trigger factors; training the long short-term memory network based on the key geological parameters and environmental parameters; predicting future mine displacement and landslide probability based on the trained long short-term memory network; combining the future mine displacement and landslide probability and the key trigger factors to obtain combined features; determining whether the combined features exceed the early warning value; if the early warning value is exceeded, return to the step of "correcting the particle discrete element model" to re-simulate the displacement field evolution diagram, stress distribution evolution diagram and sliding surface evolution trend of the post-disaster mine.

2. The method of claim 1, wherein the method is characterized by: The key geological parameters and environmental parameters include geological drilling data, surface deformation monitoring data, groundwater level and pore water pressure sensor data, precipitation and surface runoff data, and historical landslide data.

3. The method of claim 1, wherein the method is characterized by: The ion-type rare earth mine landslide early warning method based on discrete element simulation further comprises, between the steps of "obtaining key geological parameters and environmental parameters" and "constructing a mine three-dimensional geological model based on the key geological parameters and environmental parameters": preprocessing the key geological parameters and environmental parameters.

4. The method of claim 1, wherein the method is characterized by: The mine three-dimensional geological model includes the following elements: stratum distribution and physical parameters, fault structure and joint network, hydrogeological unit, and slope boundary and mining disturbance area; The stratum distribution and physical parameters include triaxial strength, friction angle and density, etc.; the hydrogeological unit includes aquifer and aquiclude.

5. The discrete element simulation-based landslide warning method for ion-type rare earth mine areas according to claim 1, characterized in that, Constructing a particle discrete element model based on the mine three-dimensional geological model specifically comprises the following steps: granular modeling: converting the mine rock-soil body into a particle network, with each particle representing a rock unit, and simulating the shear, compression and failure behavior between rock layers through contact force; mechanical property assignment: assigning friction angle, shear strength, cohesion and other parameters to each type of particle according to geological drilling and experimental data; initial boundary condition setting: setting the initial gravity field, groundwater level and external disturbance of mining; dynamic disturbance simulation: inputting rainfall, leaching mining external disturbance factors to dynamically simulate the response behavior of the geological body.

6. The discrete element simulation-based landslide warning method for ion-type rare earth mine areas according to claim 1, characterized in that, The statistical features include displacement change rate, pore pressure change gradient and cumulative rainfall; the derived features include displacement acceleration and stress field energy density; and the environmental features include mining activity intensity and joint surface inclination and other geological parameters.

7. A landslide early warning device for ion-type rare earth mine area based on discrete element simulation, characterized in that, The ion-type rare earth mine landslide early warning device based on discrete element simulation comprises: The parameter acquisition module is configured to acquire key geological parameters and environmental parameters. The three-dimensional geological model construction module is configured to construct a three-dimensional geological model of a mining area based on the key geological parameters and the environmental parameters. The particle discrete element model construction module is configured to construct a particle discrete element model based on the three-dimensional geological model of the mining area. The correction module is configured to correct the particle discrete element model. The simulation module is configured to simulate a displacement field evolution diagram, a stress distribution evolution diagram, and a sliding surface evolution trend of a post-disaster mine based on the corrected particle discrete element model. The multi-dimensional feature acquisition module is configured to acquire multi-dimensional features, including statistical features, derived features, and environmental features. The machine learning model construction module is configured to construct a machine learning model, including a random forest model and a long short-term memory network. The key trigger factor determination module is configured to input the multi-dimensional features into the random forest model to obtain a key trigger factor. The training module is configured to train the long short-term memory network based on the key geological parameters and the environmental parameters. The prediction module is configured to predict future mine displacement and landslide probability based on the trained long short-term memory network. The feature combination module is configured to combine the future mine displacement and landslide probability and the key trigger factor to obtain combined features. The judgment module is configured to determine whether the combined features exceed a warning value. The re-run module is configured to return to the correction module to re-simulate the displacement field evolution diagram, the stress distribution evolution diagram, and the sliding surface evolution trend of the post-disaster mine when the warning value is exceeded.

8. A computer device comprising: The memory, the processor, and the computer program stored in the memory and executable on the processor are characterized in that the processor executes the computer program to implement the ion-type rare earth mining area landslide warning method based on discrete element simulation according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the ion-type rare earth mining area landslide warning method based on discrete element simulation according to any one of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the ion-type rare earth mining area landslide warning method based on discrete element simulation according to any one of claims 1-6.

Images