Dangerous rock monitoring system and risk assessment method thereof

Abstract

The present application relates to the technical field of dangerous rock monitoring, and particularly relates to a dangerous rock monitoring system and a risk assessment method thereof.The present application introduces a graph convolutional neural network to accurately model environmental factors such as wind speed and vegetation, overcomes the limitations of traditional wind force calculation models, corrects the influence of wind force tension, and thus more accurately assesses the stability of the rock mass;By introducing a dynamic adjustment mechanism of the environmental change coefficient and the vibration propagation coefficient, the model can adapt to changes in environmental factors such as temperature, humidity and precipitation in real time, and improve the adaptability and accuracy of the model;In addition, considering the spatial interaction and dynamic effect between rock mass units, an instability propagation equation is established to capture the propagation process of rock mass instability, so as to comprehensively consider the physical properties, spatial interaction and environmental changes of the rock mass, thereby providing more accurate and real-time instability prediction, and providing effective risk warning and emergency decision assistance for mining, slope construction and other engineering.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of dangerous rock monitoring, in particular to a dangerous rock monitoring system and a risk assessment method thereof. BACKGROUND

[0002] The monitoring method of rock mass instability relies on physical monitoring means, such as crack observation, settlement measurement, etc. These methods can provide a certain degree of rock mass state reflection, but generally cannot fully capture the dynamic characteristics in the process of rock mass instability, especially the interaction between rock mass units and the comprehensive influence of external environmental factors on rock mass stability. Moreover, the existing monitoring method often does not reflect the long-term cumulative effect of external environment (such as precipitation, temperature change, etc.) on the stability of rock mass units in real time, resulting in low prediction accuracy and timeliness.

[0003] Secondly, the current rock mass instability evaluation model ignores the comprehensive effect of multiple factors such as terrain, vegetation growth, wind speed, etc. For example, the conventional wind force influence model does not fully consider the acceleration or retardation effect of local terrain or vegetation on wind speed, resulting in inaccurate wind force calculation and affecting the accuracy of rock mass stability prediction.

[0004] In addition, the instability of rock mass is often a complex dynamic process, affected by the historical behavior of rock mass itself, external environmental changes, mutual influence of adjacent rock mass units, etc. Therefore, it is difficult to fully capture the spatial and temporal interaction and nonlinear relationship, resulting in the inability to improve the accuracy of rock mass monitoring and prediction of instability risk. SUMMARY

[0005] In view of the above-mentioned shortcomings of the prior art, the present application provides a dangerous rock monitoring system and a risk assessment method thereof, which can effectively solve the problem that the actual environmental characteristics of rock mass are not fully considered in the prior art, thereby improving the accuracy of rock mass instability risk monitoring.

[0006] To achieve the above purpose, the present application realizes the following technical solutions:

[0007] The present application provides a dangerous rock monitoring system, which at least comprises:

[0008] A regional positioning module acquires a dangerous rock region and establishes a three-dimensional geometric model to construct a rock mass unit set;

[0009] A dangerous rock body judgment module filters dangerous rock mass units based on joint surfaces or terrain;

[0010] A dangerous rock index analysis module calculates environmental indexes according to the dangerous rock mass units and determines the wind load in the environmental indexes:

[0011] The graph convolutional neural network is introduced to predict the local face wind speed gain, the node feature vector and the edge feature are established, the multi-layer graph convolution is performed to output the graph level feature vector, the wind amplification factor is calculated according to the graph level feature vector, and the wind load is established based on the wind amplification factor and the environmental data;

[0012] The rock mass instability prediction module inputs the characteristic indicators in the environmental indicators to establish exogenous variables, and preliminarily predicts the instability probability of the rock mass unit according to the time series model;

[0013] The adjacency matrix is established in response to the spatial relationship between the rock-soil units, the instability propagation equation is established according to the adjacency matrix to simulate the propagation process of instability, and the instability probability of the rock mass unit is re-predicted.

[0014] Further, the adjacency matrix simulates the mutual influence and propagation effect between the rock mass units according to the physical properties, spatial distance, vibration propagation coefficient, geological structure and external environment of the rock mass units.

[0015] Further, when the vibration propagation coefficient is determined:

[0016] Considering that the vibration propagation of the rock mass unit is affected by the environment and changes the physical properties of the rock mass, an environmental change coefficient is introduced to dynamically construct the vibration propagation coefficient.

[0017] Further, when the vibration propagation coefficient is determined:

[0018] Considering that the vibration propagation of the rock mass unit is affected by the environment and changes the physical properties of the rock mass, an environmental change coefficient is introduced to dynamically construct the vibration propagation coefficient.

[0019] Further, when the vibration propagation coefficient is determined:

[0020] Considering that the cracks of the rock mass shrink under arid environment, the vibration propagation intensity decreases, and the rock mass may become more fragile under humid environment, the vibration propagation intensity increases;

[0021] An adaptive mechanism is introduced, and the vibration propagation coefficient is defined by comprehensively considering the physical properties and the mutation effect of the rock mass.

[0022] Further, the method for judging the dangerous rock mass unit through the terrain is:

[0023] The risk assessment method is applied to the dangerous rock monitoring system, and includes the following method steps:

[0024] The dangerous rock region is collected and a three-dimensional geometric model is established, and a rock mass unit set is constructed;

[0025] The dangerous rock mass judgment module filters the dangerous rock mass unit based on the joint surface or the terrain;

[0026] The dangerous rock mass index analysis module calculates environmental indicators based on the dangerous rock mass units.

[0027] Exogenous variables are established, and the instability probability of rock mass units is preliminarily predicted based on time series models;

[0028] An adjacency matrix is ​​established in response to the spatial relationship between rock and soil elements. An instability propagation equation is established based on the adjacency matrix to simulate the instability propagation process, and the instability probability of rock mass elements is re-predicted to achieve risk assessment of rock mass elements.

[0029] The technical solution provided by this invention has the following advantages compared with the known prior art:

[0030] By introducing graph convolutional neural networks to accurately model environmental factors such as wind speed and vegetation, the limitations of traditional wind power calculation models are overcome, the influence of wind tension is corrected, and the stability of rock masses is assessed more accurately.

[0031] By introducing a dynamic adjustment mechanism for the environmental change coefficient and vibration propagation coefficient, the model can adapt to changes in environmental factors such as temperature, humidity, and precipitation in real time, thereby improving its adaptability and accuracy. In addition, considering the spatial interaction and dynamic effects between rock mass units, an instability propagation equation is established to capture the propagation process of rock mass instability. This comprehensively considers the physical properties, spatial interaction, and environmental changes of the rock mass, thus providing more accurate and real-time instability predictions and offering effective risk warnings and emergency decision-making support for mining, slope construction, and other engineering projects. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0033] Figure 1 This is an overall module block diagram of the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0035] The present invention will be further described below with reference to embodiments.

[0036] Example 1 (see) Figure 1 A rockfall monitoring system, which includes at least:

[0037] The regional positioning module takes high-resolution satellite orthophotos of the monitoring area as input, defines dangerous rock areas (which are rock mass locations with signs of instability, such as cracks, cliffs, steep slopes, etc.), obtains the geometric information of dangerous rock areas, establishes the corresponding three-dimensional geometric model M0, and realizes the spatial reference positioning of dangerous rock areas.

[0038] Random sampling of surface point clouds is performed on a 3D geometric model, and plane parameters (n) satisfying the majority of points are fitted. j ,d j ), nj represents the surface normal vector, d j By representing the plane constant, the set of joint (fault) surfaces {S} can be identified. j}, S j This represents the j-th joint surface, i.e., a single plane, where each S... j Corresponding to a plane equation:

[0039] S j :n j ·X+d j =0, where X represents the coordinate vector of any point in three-dimensional space, and d j S represents j The offset is a constant term, for plane S. j Includes all that satisfy S j :n j ·X+d j A three-dimensional coordinate vector equal to 0;

[0040] Take two pairs of planes S j S j′ Intersection line, S j′ Let the j′-th plane belong to {S} j}, combined with the elevation model (DEM), topological segmentation is performed on the three-dimensional geometric model, dividing the three-dimensional geometric model into several relatively independent sets of rock mass units. Each rock mass unit is a polyhedron surrounded by triangular mesh facets.

[0041] Furthermore, the dangerous rock mass judgment module filters out dangerous rock mass units k from the rock mass unit set and establishes a dangerous body set. The methods include:

[0042] 1) Judging by joint surfaces:

[0043] Identify the set of all joint strips belonging to the rock mass unit from the point cloud on the surface or inside of the rock and soil unit. N k Indicates the total number of joints;

[0044] If the first in the rock mass unit Joints satisfy:

[0045]

[0046] Where, φ min φ max These represent the minimum and maximum tilt angle thresholds, typically 30° and 70°, respectively. Indicates joints The angle of inclination, Indicates joints The direction, α k Indicates the main slope aspect of the rock mass unit, Δθ max This indicates the threshold for determining the direction difference, which is typically 20°.

[0047] Therefore, if condition φ exists min ≤φ k , Then the rock mass unit is identified as a dangerous rock mass unit;

[0048] 2) Judging by the terrain:

[0049] The topographic features of each rock mass unit are determined as follows:

[0050] slope

[0051] in, This represents the DEM gradient at point p, reflecting the slope in the steepest direction of the terrain at that location.

[0052] This indicates the slope angle to be calculated; therefore, the steepest point within the rock mass unit is taken to measure the overall steepness.

[0053] Maximum normal curvature

[0054] Among them, κ n (p) represents the curvature along the normal direction at point p, reflecting the intensity of the curvature or turning of the terrain surface. It should be noted that high normal curvature is usually found at cliff corners or protrusions, and is used to distinguish between cliffs and steep cliffs.

[0055] Secondly, root growth generates wedging pressure, which significantly impacts crack propagation. Assessing the vegetation status and root growth potential of rock mass units includes:

[0056] NDVI calculations are performed on rock mass units to measure vegetation activity and coverage; a higher NDVI value indicates denser vegetation.

[0057] Determining the root growth rate category based on NDVI value ρ k (This may include multiple ranges, such as low, medium, and high). Typically, a root rate threshold is set and compared with the NDVI value. This will not be elaborated here.

[0058] If condition s exists k ≥s env ∨c k ≥c env ∨ρ k If the value is high, then the rock mass unit is identified as a dangerous rock mass unit.

[0059] The unstable rock mass index analysis module deploys corresponding sensors to collect time-series data for hazardous rock mass units. Environmental indicators are calculated from this time-series data to construct feature vectors, which are then used to predict the instability of the hazardous rock mass units. The details are as follows:

[0060] When calculating environmental indicators, we have:

[0061] cumulative rainfall r i Let R represent the rainfall intensity at time i, and let R represent the length of the time window.

[0062] Maximum hourly rainfall intensity r t max =max i∈[t-L+1] r i ;

[0063] Daily temperature difference ΔT t =max i∈[t-1,t] T i -min i∈[t-1,t] T i T i Indicates the temperature at time i;

[0064] Freeze-thaw cycle count 1(·) represents an indicator function;

[0065] Thermal stress index α represents the coefficient of thermal expansion, d represents the depth of thermal diffusion, and I t Indicates irradiance;

[0066] Dynamic root pressure index P root,k (t)=P 0,k +β k,1 t+β k,2 SM k,t+ β k,3 Sap k,t P 0,k β represents the initial wedging pressure of the root system. k,1 β represents the time sensitivity coefficient. k,2β represents the soil moisture coefficient. k,3 SM represents the sap activity coefficient. k,t Indicates soil moisture content, Sap k,t Indicates the sap flow rate;

[0067] Deformation rate P k,t Let represent the three-dimensional spatial coordinate vector of the k-th rock mass element at time t, and Δt represent the sampling time interval;

[0068] displacement acceleration v k,t Represents the displacement rate at time t;

[0069] When calculating wind pull:

[0070] Considering that in actual bare rock monitoring, wind force does not act uniformly on the entire rock surface, local topography (such as grooves and protrusions) and vegetation (such as shrubs and vines) can produce "wind speed acceleration" or "wind speed inhibition" effects. Traditional mechanical models that calculate wind speed across the entire field are unable to capture this effect, leading to prediction distortion. Consequently:

[0071] Multiple wind speed or wind direction sensor arrays are deployed around each dangerous rock mass unit, along with structured light scanning, to reconstruct the instantaneous turbulent structure between wind and vegetation.

[0072] A physics-based graph convolutional neural network, by inputting sensor array data and a 3D mesh, predicts the local surface wind speed gain, resulting in:

[0073] Input the 3D geometric model M0 and the hazardous rock mass elements;

[0074] Based on the dangerous rock mass element grid subset g k =M 0|k The triangular facets and vertices of the rock mass unit constituting the hazard are preserved.

[0075] Construct a graph structure: Nodes are grid vertices n = 1, ..., F k Corresponding to three-dimensional coordinates X n F k This represents the number of vertices. For an edge, if vertices n and m share an edge on the grid, then there exists an undirected edge (n, m) in the graph.

[0076] Predicting magnification factor using graph convolutional neural networks:

[0077] Node feature vectors n n Represents the normal vector of a mesh vertex. These represent the average wind speed and average wind direction around n (with a radius of 5, 10, or 15 m).

[0078] Edge features: Selectable options include geometric distance between vertices or area of ​​a face;

[0079] Multi-layer graph convolution: γ = 0,…,L-1

[0080] in, When the layer is γ+1, the eigenvector of node n in the graph, when γ = 0, That is, the initial node feature vector. When representing layer γ, the feature vector of neighbor node m, b (γ) Let σ(·) represent the bias vector of the γ-th layer, L represent the total number of layers in the network, σ(·) represent the activation function, such as ReLU or sigmoid, which performs a non-linear mapping on the linear transformation result, acting element-wise, N(n) represent the set of neighbors of node n, and W (γ) This represents the weight matrix of the γth layer, used for linear transformation of the feature vectors from all neighbors;

[0081] Therefore, after L layers of graph convolution, each node n obtains a high-order feature vector.

[0082] Global pooling (usually max pooling or average pooling) is used to extract high-order feature vectors at the node level. Aggregates into a graph-level feature vector h graph ;

[0083] Graph-level feature vector h graph Based on the wind amplification factor α output by the multilayer sensor network (MLP) (two-layer MLP) k,t ,in:

[0084] If α k,t A value greater than 1 indicates that local terrain or vegetation causes wind speed to increase; a value less than 1 indicates that obstacles or depressions cause wind speed to decrease; and a value equal to 1 indicates that there is no significant attenuation.

[0085] Then, wind force correction calculations are performed:

[0086] F wind,k (t) represents the corrected wind load of the k-th hazardous rock mass element at time t, ρ air C represents air density. d A represents the drag coefficient. k The windward area of ​​a dangerous rock mass unit, v t This indicates the wind speed recorded by the weather station (environmental data: i.e., the aforementioned air density, drag coefficient, etc.).

[0087] The rock mass instability prediction module constructs a feature vector x based on environmental indicators. k,t By using the feature vector x k,tRelevant feature indicators, such as cumulative rainfall, daily temperature range, and number of freeze-thaw cycles, are extracted as exogenous variables to construct a time series model to predict the instability probability of dangerous rock mass units at future times, as detailed below:

[0088] Obtain time-series data affecting the instability probability of rock mass units, including:

[0089] Data on instability events (the probability of instability of a rock mass unit or whether an actual instability event has occurred), displacement data of the rock mass unit, deformation rate, wind speed, air temperature, stress, strain, crack width, etc., can be used to construct a complete time series data of a rock mass unit by aligning the timestamps.

[0090] The time series model is constructed by introducing the feature indicators calculated above:

[0091]

[0092] in, Let c represent the instability probability (objective variable) of rock mass element k at time t, and φ represent the constant term. i θ represents the autoregressive coefficient, indicating the influence of historical values ​​of the target variable on the current value; p represents the coefficient of the autoregressive term; θ j ε represents the moving average coefficient. t -j represents the error term at time tj, q represents the order of the moving average term, and β u Represents the coefficients of exogenous variables. The exogenous variable u related to rock mass unit k at time t includes cumulative rainfall, daily temperature range, number of freeze-thaw cycles, etc., where U represents the number of exogenous variables, ε t Indicates the noise term;

[0093] By combining time series data and exogenous variables, a time series model is established to accurately predict the instability probability of rock mass units at future moments, thereby providing an effective early warning for rock mass stability monitoring. Through the autoregressive part of historical time series data, the model captures the temporal pattern of rock mass instability. The introduction of external environmental factors (such as cumulative rainfall, daily temperature difference, wind speed, etc.) as exogenous variables enables the model to adapt to environmental changes and reflect the real-time impact of environmental factors such as cumulative rainfall and daily temperature difference on rock mass stability.

[0094] Secondly, the addition of the moving average component further optimizes the model's accuracy, effectively eliminating short-term fluctuations and errors, thereby improving the accuracy of predictions and enabling precise monitoring of the risk status of areas such as mine rock masses and slopes.

[0095] It should be noted that:

[0096] The above provides the instability probability of each hazardous rock mass unit at different times. However, in actual rock mass environments, the following needs to be emphasized:

[0097] Interactions in space:

[0098] The instability of a rock mass may be affected by the state of neighboring rock mass units. The instability of a certain rock mass unit is not only affected by its own historical behavior, but also closely related to the stability state of the surrounding rock mass. Therefore, the instability of rock mass units may have spatial correlation.

[0099] Time delay and dynamic effects:

[0100] The instability of a rock mass unit is often a dynamic process, affected by time delay. For example, the instability of a certain rock mass unit may lag behind the instability of its neighboring units in time. Such time delay and cumulative effect are difficult to capture and reflect by time series models.

[0101] Nonlinear relationships and complex propagation states:

[0102] The propagation of instability between rock mass units is usually not linear and is highly complex. For example, the propagation of instability between rock mass units may have threshold effects, superposition effects, etc., and time series models have difficulty capturing these nonlinear relationships.

[0103] Therefore, the following steps are also included:

[0104] Considering the spatial relationships between multiple rock mass units, establish an adjacency matrix A. k,k ′ This describes which rock mass units are spatially adjacent or may influence each other:

[0105]

[0106] Where, ω k,k′ The weights of physical properties between rock mass elements, including hardness, cracks, and rock type, are represented by λ. k,k′ Indicating the influence of geological transmission paths, such as fractures and faults, α k,k′ β represents the vibration propagation coefficient. k The weighting factor represents the external environmental factors (daily temperature range, wind speed, etc.), α1 represents the adjustment parameter, and d k,k ′ represents the spatial distance between rock mass elements k and k′ (calculated using Euclidean distance). By introducing an adjacency matrix, various factors such as physical properties, spatial distance, vibration effects, geological structure, and external environment between rock mass elements are considered, which can more accurately simulate the mutual influence and propagation effects between rock mass elements instability.

[0107] Define the instability propagation equation:

[0108]

[0109] Among them, y k,t This represents the instability probability after propagation, reflecting the instability of rock mass unit k at time t, taking into account the influence of neighboring rock mass units and environmental factors. This represents the dynamic influence of the external environment on the stability of the rock mass unit, such as diurnal temperature range, x. k,t x k′,t Let f(·) represent the characteristic vectors of rock mass units k and k′ at time t, such as cumulative rainfall, daily temperature difference, etc., respectively. f(·) represents the propagation function, which can be linearly weighted (established by weighted summation of instability probability and physical characteristic difference between rock mass units k and k′, and the physical differences between them, such as stress) or nonlinear, without specific limitations.

[0110] Furthermore, considering that the instability of multiple rock mass units may affect each other, the instability propagation process is simulated through an instability propagation equation. The impact of the instability of one rock mass unit on adjacent rock mass units is transmitted through a weighted matrix. The instability probability of each rock mass unit is not only affected by its own historical data, but also by the instability probability of neighboring rock mass units. Therefore, the instability probability of the rock mass unit is corrected.

[0111] Furthermore, based on the modified instability probability, it can be determined whether the instability threshold has been reached. If the instability threshold is reached, it is determined that the rock mass unit has a significant risk of instability. Subsequently, early warning and implementation of instability-related emergency control measures can be carried out for the rock mass unit to improve the accurate monitoring and control of instability risks in the rock mass area. At the same time, it enables the implementation of appropriate resource management plans for rock mass units with urgent or general risk levels, further balancing the difficulty of emergency handling of rock mass units and reducing the risk expansion degree when subsequent mine rock mass disasters occur.

[0112] In addition, this scheme also considers the actual physical environment of the rock mass unit to analyze the vibration propagation coefficient α. k,k′ The following examples illustrate the effects of different scenarios on the vibration propagation coefficient α. k,k′ Impact:

[0113] Example 2: The vibration propagation capability of a rock mass unit is affected by the environment, especially factors such as temperature, humidity, and precipitation, which can cause changes in the physical properties of the rock mass (such as hardness and brittleness), thereby affecting vibration propagation. Therefore, an environmental change coefficient is introduced to dynamically construct the vibration propagation coefficient α. k,k′ Then we have:

[0114] α k,k′ (t)=α0·exp(γ·ΔE(t))

[0115] α0 represents the vibration propagation coefficient when there is no external environmental influence, γ represents the environmental change adjustment factor, ΔE(t) represents the environmental change at time t, such as the change in temperature difference, precipitation, etc., and exp(·) represents the indicator function.

[0116] Example 3: Environmental changes may cause drastic changes in the vibration propagation capability of rock mass elements near certain critical values. For example, temperature changes may cause cracks in rock mass elements to propagate, thereby increasing the vibration propagation capability. A nonlinear exponential function is introduced to reflect the abrupt effect of environmental changes, and the vibration propagation coefficient α is defined. k,k′ Then we have:

[0117]

[0118] Example 4: In arid environments, cracks in rock masses may shrink, thus reducing vibration propagation capacity. Conversely, in humid environments, rock masses may become more brittle, increasing vibration propagation efficiency. An adaptive mechanism, considering the physical properties of the rock mass and abrupt changes, is introduced to comprehensively define the vibration propagation coefficient. This ensures that an accurate and comprehensive vibration propagation coefficient can be established based on the actual influences on the rock mass, as detailed below:

[0119]

[0120] Among them, E k and E k′ Let FRF(f) represent the elastic moduli of rock mass elements k and k′, respectively. k ,ω k ,ζ k The expression f represents the response of rock mass element k to vibrations of different frequencies during vibration propagation. It is used to consider the response characteristics of rock mass element k to external vibrations, thus affecting the intensity of vibration propagation. k The frequency ω represents the frequency response. k ζ represents the angular frequency of the vibration. k The damping coefficient of the rock mass element is represented by φ(t), which represents a time-varying factor describing the adaptability of the rock mass element k at time t. For example, as time progresses, changes in the adaptability of the rock mass may lead to changes in the vibration propagation coefficient. Furthermore, by adjusting the vibration propagation coefficient through multiple factors, including the physical properties of the rock mass (such as elastic modulus and distance), the frequency response of vibration, environmental changes (temperature, humidity, etc.), and time-related adaptability factors, the propagation process of vibration between rock mass elements can be described more accurately. Through these comprehensive factors, the complexity of the mutual influence between rock mass elements can be captured, and a more dynamic vibration propagation coefficient related to changes in environmental and physical conditions can be provided.

[0121] Finally, the present invention also provides:

[0122] A risk assessment method is implemented based on the aforementioned rockfall monitoring system;

[0123] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the system described above;

[0124] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the system described above; for details, please refer to the aforementioned rockfall monitoring system, which will not be repeated here.

[0125] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the protection scope of the technical solutions of the embodiments of the present invention.

Claims

1. A rockfall monitoring system, characterized in that, include: The regional positioning module collects data on dangerous rock areas and establishes a three-dimensional geometric model, constructing a set of rock mass units. The dangerous rock mass identification module filters dangerous rock mass units based on joint surfaces or terrain. The dangerous rock mass index analysis module calculates environmental indicators based on the dangerous rock mass units, and determines the wind load capacity among the environmental indicators as follows: A graph convolutional neural network is introduced to predict the local surface wind speed gain. Node feature vectors and edge features are established, and multi-layer graph convolution is performed to output graph-level feature vectors. The wind amplification factor is calculated based on the graph-level feature vectors, and the wind load is established based on the wind amplification factor and environmental data. The rock mass instability prediction module inputs characteristic indicators from the environmental indicators to establish exogenous variables, and preliminarily predicts the instability probability of rock mass units based on a time series model. An adjacency matrix is ​​established in response to the spatial relationship between soil and rock elements. An instability propagation equation is established based on the adjacency matrix to simulate the instability propagation process and to re-predict the instability probability of the rock mass elements. The method of multi-layer graph convolution to output graph-level feature vectors is as follows: ; in, Presentation layer Nodes in the diagram eigenvectors, Presentation layer At that time, neighboring nodes eigenvectors, Indicates the first Layer bias vector, Indicates the total number of network layers. This represents the activation function. Represents a node The neighborhood group, Indicates the first Layer weight matrix; go through After layer graph convolution, each node Obtaining high-order feature vectors ; High-order feature vectors at the node level are pooled using global pooling. Aggregate into graph-level feature vectors ; The method for determining wind load capacity is as follows: Graph-level feature vectors Based on the wind amplification factor output by the multilayer sensor network (MLP) ; Perform the calculation: ; Indicates the first A dangerous rock mass unit at time Corrected wind load, Indicates air density, Indicates the drag coefficient. The windward area of ​​a dangerous rock mass unit. This indicates the wind speed recorded by the weather station.

2. The rockfall monitoring system according to claim 1, characterized in that, The method for the time series model to initially predict the instability probability of rock mass units is as follows: Obtain time-series data and exogenous variables that affect the instability probability of rock mass units; Building a time series model: ; in, Indicates the first Rock mass unit at time The probability of instability, Represents a constant term. Represents the autoregressive coefficient. The coefficients of the autoregressive term are represented. Represents the moving average coefficient. Indicates in Error term at time, This indicates the order of the moving average term. Represents the coefficients of exogenous variables. Indicates time With rock mass unit Related exogenous variables , Indicates the number of exogenous variables. This indicates the noise term.

3. The rockfall monitoring system according to claim 1, characterized in that, The adjacency matrix simulates the mutual influence and propagation effects of rock mass unit instability based on the physical properties, spatial distance, vibration propagation coefficient, geological structure, and external environmental factors between rock mass units.

4. The rockfall monitoring system according to claim 3, characterized in that, When the vibration propagation coefficient is determined: Considering that the vibration propagation of rock mass units is affected by the environment and affects the changes in the physical properties of the rock mass, a vibration propagation coefficient is dynamically constructed by introducing an environmental change coefficient.

5. The rockfall monitoring system according to claim 3, characterized in that, When the vibration propagation coefficient is determined: Considering the changes in vibration propagation of rock mass elements caused by environmental changes, a nonlinear exponential function is introduced to reflect the abrupt effect of environmental changes, and a vibration propagation coefficient is defined.

6. The rockfall monitoring system according to claim 3, characterized in that, When the vibration propagation coefficient is determined: Considering that in arid environments, the cracks in the rock mass shrink, reducing the intensity of vibration propagation, while in humid environments, the rock mass may become more brittle, increasing the intensity of vibration propagation; The vibration propagation coefficient is defined by introducing an adaptive mechanism, the physical properties of the rock mass, and the abrupt change effect.

7. The rockfall monitoring system according to claim 1, characterized in that, The method for identifying dangerous rock mass units based on terrain is as follows: Determine the topographic features of the rock mass unit: slope ; in, Point DEM gradient at the location, Indicates the calculation of the slope angle; Maximum normal curvature ; in, Point The curvature along the normal direction; Perform NDVI calculations on the rock mass elements; Determining the root growth rate category based on NDVI value ; If conditions exist If so, the rock mass unit is identified as a dangerous rock mass unit.

8. A risk assessment method, applied to the rockfall monitoring system according to any one of claims 1-7, characterized in that, The methods and steps include the following: Collect data on the dangerous rock area and establish a three-dimensional geometric model to construct a set of rock mass units; The dangerous rock mass identification module filters dangerous rock mass units based on joint surfaces or terrain. The dangerous rock mass index analysis module calculates environmental indicators based on the dangerous rock mass units. Exogenous variables are established, and the instability probability of rock mass units is preliminarily predicted based on time series models; An adjacency matrix is ​​established in response to the spatial relationship between rock and soil elements. An instability propagation equation is established based on the adjacency matrix to simulate the instability propagation process, and the instability probability of rock mass elements is re-predicted to achieve risk assessment of rock mass elements.

Images