Data risk analysis method and system based on Internet of Things

Abstract

The invention discloses a data risk analysis method based on the Internet of Things, and relates to the field of risk analysis, and the method comprises the steps: obtaining laser radar point cloud data of a target mining area; acquiring sensor information of a target mining area by using multiple types of sensors; constructing a mining area three-dimensional model of the target mining area by using the laser radar point cloud data, and obtaining slope crack information of the target mining area; generating a crack monitoring point distribution strategy of the target mining area in combination with the mining area three-dimensional model and the slope crack information; deploying a displacement sensor according to the crack monitoring point distribution strategy, and monitoring local displacement information of the slope crack by using the displacement sensor; constructing a slope risk prediction model, and obtaining slope risk information of the target mining area by using the slope risk prediction model; and carrying out visual display on the slope risk information through the mining area digital twinborn model. According to the invention, the slope area of the target mining area can be effectively and accurately monitored in all directions.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of data analysis, and in particular, to a data risk analysis method and system based on the Internet of Things. Background Art

[0002] With the continuous acceleration of the industrialization process in modern society, the demand for minerals, as an important guarantee for modernization construction, is also increasing day by day. However, in the process of mineral mining, there are often various safety hazards, such as gas explosions and mine collapses. Among them, mine collapse is one of the frequent accidents in mining areas, and one of the areas prone to collapse is the mine slope area. The factors that lead to collapse accidents include but are not limited to cracks in the slope, unstable rock structure, improper mining technology, etc. Once a collapse accident occurs, it will lead to serious safety accidents and threaten the personal safety of mining workers. Therefore, it is necessary to conduct risk analysis on the slope area of ​​the mining area to check for safety hazards.

[0003] Traditional risk monitoring methods are relatively simple, mainly relying on manual inspections and risk monitoring based on multiple types of sensors installed in mining areas. However, manual inspections are easily affected by subjective factors and cannot monitor risks 24 hours a day. In addition, for mining areas with larger areas, the slope areas are also larger. Manual and multiple types of sensors cannot perform all-round monitoring of the slope areas of mining areas, and therefore cannot conduct a comprehensive risk assessment of the slope areas. As a result, failure to promptly discover the potential collapse hazards in the slope areas will lead to collapse accidents in the slope areas, threatening the lives of workers in the mining areas. Therefore, how to achieve all-round and high-precision risk monitoring of the slope areas of mining areas is a major challenge facing the mining industry. Summary of the invention

[0004] The embodiment of the present application provides a data risk analysis method based on the Internet of Things, which is used to solve the problem that the existing technology cannot achieve high-precision risk monitoring of the slope area of ​​the mining area in all directions.

[0005] To achieve the above objectives, the embodiments of the present application adopt the following technical solutions:

[0006] In a first aspect, a data risk analysis method based on the Internet of Things is provided, the method comprising:

[0007] Obtain the LiDAR point cloud data collected by the drone in the target mining area;

[0008] Collecting sensor information of the target mining area based on multiple types of sensors preset in the target mining area, wherein the multiple types of sensors include stress sensors and groundwater level sensors;

[0009] Using the laser radar point cloud data to construct a three-dimensional model of the target mining area, and based on the three-dimensional model of the mining area and using a point cloud computing method to calculate slope crack information of the slope cracks in the target mining area;

[0010] Combining the three-dimensional model of the mining area and the slope crack information to generate a crack monitoring point distribution strategy for the slope area of ​​the target mining area;

[0011] Deploy a plurality of displacement sensors in the target mining area according to the monitoring point distribution strategy, and use the displacement sensors to perform local displacement monitoring on all the slope cracks in the slope area to obtain local displacement information;

[0012] Building a slope risk prediction model based on the sensor information, the slope crack information and the local displacement information, and using the slope risk prediction model to perform slope risk analysis on the slope area to obtain slope risk information of the target mining area;

[0013] A mining area digital twin model of the target mining area is constructed by combining the mining area three-dimensional model and the slope risk prediction model, and the slope risk information is visualized through the mining area digital twin model.

[0014] Optionally, the step of constructing a three-dimensional mining area model of the target mining area using the laser radar point cloud data, and obtaining slope crack information of the slope cracks of the target mining area by calculating the three-dimensional mining area model and using a point cloud computing method includes the following steps:

[0015] Preprocessing the laser radar point cloud data of each target area in the target mining area, wherein the target area includes a slope area;

[0016] Performing surface fitting on the preprocessed laser radar point cloud data to obtain a plurality of fitting surfaces, and combining the plurality of fitting surfaces to construct a three-dimensional mining area model of the target mining area, wherein the fitting surfaces include a slope fitting surface corresponding to the slope area;

[0017] Calculating residual values ​​between all slope point clouds belonging to the slope area in the laser radar point cloud data and the slope fitting surface, and integrating all the slope point clouds whose residual values ​​are greater than a preset first residual threshold into a plurality of suspected crack point cloud areas;

[0018] For each of the suspected crack point cloud areas, respectively calculating the normal vector angles between each of the slope point clouds and the neighborhood point clouds in the suspected crack point cloud area, wherein the neighborhood point clouds are all point clouds within a preset neighborhood range around the slope point cloud;

[0019] If the normal vector angle is greater than a preset angle threshold and the residual value is less than a preset second residual threshold, the corresponding slope point cloud is used as the regional edge point cloud of the suspected crack point cloud area, and the second residual threshold is less than the first residual threshold;

[0020] Calculate the average curvature of all edge point clouds of the region using a curvature calculation formula;

[0021] If the average curvature is a positive curvature, the suspected crack point cloud area is used as an initial crack point cloud area;

[0022] Using a region growing algorithm to filter out non-crack point clouds in the initial crack point cloud region to obtain a target crack point cloud region;

[0023] Based on the three-dimensional model of the mining area, multi-directional two-dimensional mapping is performed on the crack point cloud in the target crack point cloud area to obtain the two-dimensional point cloud coordinates of the crack point cloud in different directions, and the slope crack information of the slope cracks in the target mining area is calculated according to the two-dimensional point cloud coordinates.

[0024] Optionally, the step of using a region growing algorithm to filter out non-crack point clouds in the initial crack point cloud region to obtain a target crack point cloud region comprises the following steps:

[0025] Performing curvature estimation calculation on the initial crack point cloud according to the normal vector angles of all the initial crack point clouds in the initial crack point cloud region to obtain the principal curvatures of all the initial crack point clouds;

[0026] Constructing an empty seed point sequence and a non-crack region point set, and adding the initial crack point cloud with the smallest principal curvature to the seed point sequence as a seed point of the curvature region growing algorithm;

[0027] Based on a preset neighborhood order and using a K-neighborhood algorithm, a plurality of seed neighborhood points of the seed point are identified, and the normal angle between the seed point and each of the seed neighborhood points is calculated respectively;

[0028] If all the normal angles are greater than the normal angle threshold, it is determined that the region growing is completed;

[0029] If any one or more of the normal angles are smaller than a preset normal angle threshold, the corresponding seed neighborhood point is added to the non-crack area point set;

[0030] Calculating the curvature value of each of the seed neighborhood points in the non-crack area point set respectively;

[0031] If any one or more of the curvature values ​​are less than a preset curvature value threshold, the corresponding seed neighborhood point is added to the seed point sequence as a new seed point for regional growth, and the original seed point in the seed point sequence is removed;

[0032] If all the curvature values ​​are greater than the curvature value threshold or the seed point sequence is an empty set, it is determined that the region growing is completed;

[0033] If it is determined that the region growth is completed, the non-crack region point set is removed from the initial crack point cloud region to obtain a target crack point cloud region.

[0034] Optionally, performing multi-directional two-dimensional mapping on the crack point cloud in the target crack point cloud area based on the three-dimensional model of the mining area to obtain the two-dimensional point cloud coordinates of the crack point cloud in different directions, and calculating the slope crack information of the slope cracks in the target mining area according to the two-dimensional point cloud coordinates includes the following steps:

[0035] For each of the target fracture point cloud regions, a point cloud three-dimensional coordinate set of all fracture point clouds in the target fracture point cloud region is obtained according to the three-dimensional model of the mining area;

[0036] Based on the point cloud three-dimensional coordinate set, all the crack point clouds in the target crack point cloud area are two-dimensionally mapped in three mapping directions: the x-axis-y-axis plane direction, the y-axis-z-axis plane direction, and the x-axis-z-axis plane direction, to obtain two-dimensional point clouds in the three mapping directions;

[0037] Calculating the crack width, crack length and crack depth of the local cracks on the slope corresponding to the target crack point cloud area according to the two-dimensional point cloud coordinates of the two-dimensional point cloud in the three mapping directions respectively;

[0038] The crack position of the local crack of the slope corresponding to the target crack point cloud area is calculated by the point cloud three-dimensional coordinate set;

[0039] The crack width, the crack length, the crack depth and the crack position of the local cracks in the slope corresponding to all the target crack point cloud areas are integrated into the slope crack information of the slope cracks in the target mining area.

[0040] Optionally, constructing a slope risk prediction model based on the sensor information, the slope crack information and the local displacement information, and performing slope risk analysis on the slope area using the slope risk prediction model to obtain the slope risk information of the target mining area includes the following steps:

[0041] Acquire mining blasting information of the target mining area according to a preset mining area mining plan;

[0042] Performing a change trend analysis on the slope crack information according to the mining area blasting information and the local displacement information to obtain crack prediction information of the target mining area in a future time period;

[0043] According to the crack position information and the three-dimensional model of the mining area, a stress distribution analysis is performed on the rock mass stress information collected by the stress sensor using a threshold method to obtain stress distribution information of the target mining area;

[0044] Estimate groundwater prediction information of the target mining area based on the precipitation information of the future time period acquired in advance and the groundwater information collected by the groundwater level sensor;

[0045] Based on the crack prediction information and the groundwater prediction information, the water seepage rate grade analysis of the target mining area is performed using a threshold method to obtain the slope water seepage information of the target mining area;

[0046] A slope risk prediction model for the target mining area is constructed based on the local displacement information, the stress distribution information, the slope seepage information and the crack prediction information, and the slope risk information of the target mining area is predicted based on the slope risk prediction model.

[0047] Optionally, performing a change trend analysis on the slope crack information according to the mining area blasting information and the local displacement information to obtain crack prediction information of the target mining area in a future time period comprises the following steps:

[0048] Calculating the blasting influence radius of the target mining area according to the mining area blasting information, evaluating the blasting influence degree of the slope cracks according to the blasting influence radius, and using the blasting influence degree evaluation result as the first influence factor;

[0049] According to the local displacement information and the pre-constructed displacement crack digital model, the displacement influence degree of all the slope cracks is evaluated, and the displacement influence degree evaluation result is used as the second influencing factor;

[0050] Analyzing the influence of the first influencing factor and the second influencing factor on the slope cracks by using a grey correlation model, and assigning weights to the first influencing factor and the second influencing factor according to the influence degrees;

[0051] Using the least squares method, a linear regression analysis is performed between all the first influencing factors, the second influencing factors and the mining area crack information after the weight allocation is completed, and a crack prediction model is constructed according to the linear regression analysis results;

[0052] The crack changes of the slope cracks are predicted according to the crack prediction model to obtain crack prediction information of all the slope cracks in the target mining area in the future time period.

[0053] Optionally, constructing a slope risk prediction model for the target mining area according to the local displacement information, the stress distribution information, the slope seepage information and the crack prediction information, and predicting the slope risk information of the target mining area according to the slope risk prediction model comprises the following steps:

[0054] According to the pre-acquired slope risk assessment index and using a scaling method, a direct impact degree is assigned to the impact degree between the local displacement information, the stress distribution information, the slope seepage information and the crack prediction information;

[0055] The local displacement information, the stress distribution information, the slope seepage information and the crack prediction information are used as the core influencing factors of the slope of the target mining area, and the slope direct influence matrix of the core influencing factors of the slope is constructed according to the direct influence degree assignment result. The slope direct influence matrix is ​​S=[s ij ] 4×4 , where s ij Indicates the influence degree of the core influencing factor i of the slope on the core influencing factor j of the slope;

[0056] Normalizing the slope direct impact matrix to obtain a standardized matrix;

[0057] Based on the standardized matrix, a plurality of decreasing matrix series are generated as slope indirect influence matrices;

[0058] Establishing a slope comprehensive impact matrix by combining the slope direct impact matrix and all the slope indirect impact matrices;

[0059] Calculate the influence degree and the influenced degree of the core influencing factor of the slope according to the comprehensive influence matrix of the slope, and calculate the centrality and the cause degree of the core influencing factor of the slope according to the influence degree and the influenced degree;

[0060] A slope risk prediction model is established based on the coupling coordination function in combination with the centrality and the causality, and slope risk information of the target mining area is obtained according to the slope risk prediction model.

[0061] Optionally, the combining the centrality and the causality and establishing a slope risk prediction model based on a coupling coordination function, and obtaining the slope risk information of the target mining area according to the slope risk prediction model comprises the following steps:

[0062] Assigning weights to the core influencing factors of the slope according to the centrality corresponding to the local displacement information, the stress distribution information, the slope seepage information and the crack prediction information, to obtain the weights of the core influencing factors of the slope;

[0063] Constructing a four-quadrant diagram according to the centrality and the cause degree of the core influencing factors of the slope, and dividing the core influencing factors of the slope into cause factors of the slope and result factors of the slope according to the four-quadrant diagram;

[0064] A slope risk prediction model is established according to the weights corresponding to the core influencing factors of the slope and the coupling coordination function. The slope risk prediction model is:

[0065]

[0066] Among them, γ is the risk coupling degree, is the weight of the i-th core influencing factor of the slope, is the nth core influencing factor of the slope;

[0067] The slope risk information of the target mining area is obtained by combining the slope cause factors and the slope result factors and utilizing the slope risk prediction model.

[0068] Optionally, the step of combining the slope cause factor and the slope result factor with the slope risk prediction model to obtain the slope risk information of the target mining area includes the following steps:

[0069] Divide the slope area according to the three-dimensional model of the mining area to obtain a plurality of slope sub-areas;

[0070] The slope sub-areas where the slope cause factors exist are screened out according to the three-dimensional model of the mining area to obtain the slope risk area;

[0071] Analyzing the slope risk area according to the slope result factors, and judging whether there is stress concentration and water seepage in the slope risk area according to the analysis results;

[0072] If the stress concentration phenomenon and / or the water seepage phenomenon does not exist in the slope risk area, the slope risk area is determined to be a low-risk area;

[0073] If the stress concentration phenomenon and the water seepage phenomenon exist simultaneously in the slope risk area, the weights corresponding to the local displacement information, the stress distribution information, the slope water seepage information and the crack prediction information are input into the slope risk prediction model to obtain the risk coupling degree of the slope risk area;

[0074] If the risk coupling degree is greater than a preset risk coupling degree threshold, marking the slope risk area as a high-risk area;

[0075] If the risk coupling degree is less than or equal to the risk coupling degree threshold, marking the slope risk area as the low risk area;

[0076] Acquire the location information of all the high-risk areas and the low-risk areas according to the three-dimensional model of the mining area, and obtain the location of the high-risk areas and the low-risk areas;

[0077] The locations of all the high-risk areas and the low-risk areas are integrated to obtain the slope risk information of the target mining area.

[0078] In a second aspect, the present application provides a data risk analysis system based on the Internet of Things, characterized in that it includes:

[0079] a memory configured to store instructions; and

[0080] The processor is configured to call the instruction from the memory and implement the method for data risk analysis based on the Internet of Things described in any one of the first aspects when executing the instruction.

[0081] Through the above technical scheme, by using the laser radar point cloud data of the target mine collected by the drone to construct a three-dimensional model of the mining area, the overall situation of the target mining area can be intuitively and comprehensively grasped, which is convenient for the subsequent risk monitoring of the slope area of ​​the target mine. The point cloud of the slope cracks is accurately screened out by calculating the residual value and curvature of the point cloud, and the length, width and depth of the slope cracks are accurately calculated based on the crack point cloud of the slope cracks and the two-dimensional mapping method, which is convenient for the subsequent evaluation of the risk level of the target mine. The slope risk prediction model is established by the decision laboratory method to realize the comprehensive evaluation of the risk of the slope area, and considering the large area of ​​the slope area of ​​the target mining area, the slope area is divided into multiple sub-areas to realize the accurate evaluation of the risk level of the slope area. In summary, a more comprehensive and accurate mining area risk monitoring method is provided, which can realize all-round monitoring of the slope area of ​​the target mining area efficiently and accurately, and provide protection for the personal safety of the mining area workers.

[0082] Other features and advantages of the embodiments of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 A flowchart of a data risk analysis method based on the Internet of Things provided in an embodiment of the present application;

[0084] Figure 2 A flow chart of a curvature-based region growing algorithm provided in an embodiment of the present application;

[0085] Figure 3A schematic diagram of the relationship between the normal vectors of a point cloud and its neighboring points provided in an embodiment of the present application;

[0086] Figure 4 A schematic diagram of the process of computing slope crack information based on crack point cloud computing provided in an embodiment of the present application. DETAILED DESCRIPTION

[0087] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present application, and are not used to limit the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0088] It should be noted that if the embodiments of the present application involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.

[0089] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the technical solutions between the various embodiments can be combined with each other, but they must be based on the ability of ordinary technicians in the field to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0090] Figure 1 The following schematically shows a flow chart of a data risk analysis method based on the Internet of Things according to an embodiment of the present application. Figure 1 As shown, the embodiment of the present application provides a data risk analysis method based on the Internet of Things, which may include the following steps:

[0091] S101. Obtaining laser radar point cloud data collected by the drone in the target mining area.

[0092] In this embodiment, the drone uses its own laser radar equipment to emit laser pulses to the entire target mining area, and generates a three-dimensional point cloud of the target mine based on the received laser emission signal. These three-dimensional point clouds with three-dimensional coordinates, laser reflection intensity and color information are the laser radar point cloud data. Since the laser point cloud scanning accuracy of the drone is very high, reaching the millimeter level, it can reflect the real situation of the target mining area with high accuracy. At the same time, the laser point cloud data is dense and contains a large amount of point data, which makes the subsequent three-dimensional model of the mining area more refined and accurate.

[0093] S102. Collect sensor information of the target mining area based on multiple types of sensors preset in the target mining area. The multiple types of sensors include stress sensors and groundwater level sensors.

[0094] In this embodiment, before predicting the risk information of the mining area, the mining area staff installed various types of sensors in the target mining area, especially in the slope area of ​​the target mining area, and used these sensors to collect sensor information of the target mining area. Among them, the stress sensor is mainly used to monitor the stress changes inside the rock body of the mining area. The stress changes will affect the stability of the target mining area, and in severe cases, may cause collapse accidents in the target mining area. The groundwater sensor is mainly used to monitor the water level changes of the groundwater in the target mining area. This is because the groundwater will also affect the stability of the target mining area. When the groundwater level rises, the wetting range of the rock and soil body increases, the wetting degree intensifies, and the rock and soil body is saturated and softened by water, resulting in a decrease in shear strength, and it is easy to deform, slip, collapse and other adverse geological phenomena. This effect is particularly obvious in river valley terraces, mining slopes and shore areas.

[0095] S103, constructing a three-dimensional model of the target mining area using the laser radar point cloud data, and obtaining slope crack information of the slope cracks of the target mining area based on the three-dimensional model of the mining area and by using a point cloud computing method.

[0096] In this embodiment, after the collected laser radar point cloud data is denoised and the abnormal points are removed, the point cloud in the laser radar point cloud data is used for surface fitting. The methods that can be used for surface fitting include polynomial regression, spline interpolation, etc. The point cloud is constructed into a mathematical surface, and the surface is made as close as possible to all point clouds to obtain multiple fitting surfaces, and the fitting surfaces include slope fitting surfaces. The multiple fitting surfaces are fused to obtain a three-dimensional model of the target mining area. After obtaining the slope fitting surface, the residual value between all point clouds and the surface is calculated, and the point cloud with a residual value greater than the preset first residual threshold is screened out to obtain multiple suspected crack point cloud areas. This is because the residual value of the area where the slope crack appears is large, and the residual value of the area without cracks and obstacles on the slope is almost zero. In this way, the suspected crack point cloud area can be preliminarily screened out, but the screened suspected crack area is not accurate. There may be raised obstacles in some areas, such as stones, and these areas may not have slope cracks, so it is necessary to continue screening.

[0097] The point cloud in the suspected crack point cloud area is divided into neighborhoods according to a preset neighborhood range threshold, and then the normal vector angle between each point cloud in the suspected crack point cloud area and the neighborhood point cloud corresponding to the point cloud is calculated, and the point clouds whose normal vector angle is greater than the preset angle threshold and whose residual value is less than the preset second residual threshold are screened out to obtain the regional edge point cloud of the suspected crack point cloud area. This is because the normal vector angle suddenly increases at the junction between the slope crack area and the raised area on the slope surface and the flat ground on the slope, and the residual value of the point cloud at the junction is smaller than the residual value at the bottom of the crack or the top of the raised area. Therefore, a second residual threshold less than the first residual threshold is set to screen out the point cloud at the junction to obtain the regional edge point cloud of the suspected crack point cloud area.

[0098] Then, the regional edge point cloud is fitted into a surface and the curvature calculation formula is used to calculate the average curvature of the regional edge point cloud. If the calculated average curvature is positive, it means that the suspected crack point cloud area is the initial crack point cloud area. This is because the surface fitted by the regional edge point cloud in the slope crack area is convex, so the calculated average curvature is positive, and the surface fitted by the regional edge point cloud in the convex area of ​​the slope surface is concave, and the calculated average curvature should be negative. The point cloud in the convex area of ​​the suspected crack point cloud area can be screened out and eliminated through the positive and negative curvatures to obtain the initial crack point cloud area.

[0099] In order to ensure that the information such as crack width, crack length and crack depth calculated later is more accurate and reliable, it is necessary to use the regional growing algorithm again to screen out the non-crack point clouds in the initial crack point cloud area to obtain the target crack point cloud area. The regional growing algorithm extracts the crack area by selecting the point with the smallest curvature as the seed point, and uses the normal angle and curvature value as the threshold to stop growing for regional growth, thereby obtaining the target crack point cloud area. This method can effectively, accurately and completely extract the point cloud of the crack area to obtain the target crack point cloud area. After segmenting the precise target crack point cloud area, the target crack point cloud area is mapped from the x-axis-y-axis plane direction, the y-axis-z-axis plane direction and the x-axis-z-axis plane direction to obtain three different directions of two-dimensional point clouds. According to the two-dimensional coordinates of these two-dimensional point clouds, the crack width, crack length and crack depth of the local cracks in the slope corresponding to the target crack point cloud area can be calculated. The two-dimensional point cloud mapped in the xy direction can calculate the width of the local cracks in the slope, the two-dimensional point cloud mapped in the yz direction can calculate the length of the local cracks in the slope, and the two-dimensional point cloud mapped in the xz direction can calculate the depth of the local cracks in the slope. In addition, since the segmented target crack point cloud area is very accurate, the calculated crack width, crack length and crack depth are also accurate. Finally, according to the three-dimensional coordinates of the target crack point cloud area and the three-dimensional model of the mining area, the specific location of all local cracks in the slope in the target mining area can be determined to obtain the crack location. After integrating the crack width, crack length, crack depth and crack location of the local cracks in the slope corresponding to all target crack point cloud areas, the slope crack information of the slope cracks in the target mining area can be obtained.

[0100] S104, generating a crack monitoring point distribution strategy for the slope area of ​​the target mining area by combining the mining area three-dimensional model and the slope crack information.

[0101] In this embodiment, the crack monitoring point distribution strategy is generated based on the distribution of slope cracks on the three-dimensional model of the mining area and the length, width and depth of the slope cracks. Specifically, one or more displacement sensors are deployed at each slope crack, and the number of displacement sensors is determined based on the crack width, crack length and crack depth. That is, for slope cracks whose crack length, crack length and crack depth exceed the corresponding preset thresholds, multiple displacement sensors will be set to facilitate more accurate acquisition of rock movement information at the slope cracks and timely control of the degree of slope rock deformation.

[0102] S105. Deploy multiple displacement sensors in the target mining area according to the monitoring point distribution strategy, and use the displacement sensors to perform local displacement monitoring on all slope cracks in the slope area to obtain local displacement information.

[0103] In this embodiment, one or more displacement sensors are deployed at the slope cracks of the target mining area according to the generated monitoring point distribution strategy, and these sensors are used to perform local displacement monitoring on all slope cracks in the slope area to obtain local displacement information. Commonly used displacement sensors are GPS / GNSS (satellite navigation system) receivers, which can monitor millimeter-level displacement. Local displacement information includes displacement degree and displacement direction. The displacement degree refers to the distance of rock mass movement around the slope crack, and the displacement direction refers to the direction of rock mass movement. For example, when the rock mass sinks, the direction of rock mass movement is vertically downward.

[0104] S106: construct a slope risk prediction model based on sensor information, slope crack information and local displacement information, and use the slope risk prediction model to perform slope risk analysis on the slope area to obtain slope risk information of the target mining area.

[0105] In this embodiment, the change trend of the slope cracks in the future event segment is first analyzed according to the mining area blasting information and the local displacement information, that is, the crack prediction information. Specifically, the influence of the mining area blasting information on all the slope cracks is evaluated according to the amount of explosives in the mining area blasting information and the distance between the blasting position and the slope cracks to obtain the blasting influence level, and the influence of the local displacement information on all the slope cracks is evaluated according to the local displacement in the local displacement information to obtain the displacement influence level, and the slope crack information of all the slope cracks is subjected to linear regression analysis with the blasting influence level and the displacement influence level of all the slope cracks to fit a linear regression equation, and a crack prediction model is constructed according to the fitted linear regression equation, and the crack prediction information of the future time period is predicted according to the blasting influence level and the displacement influence level of the current event node, and the crack prediction information obtained includes the crack prediction width, crack prediction length and crack prediction depth of all the slope cracks in the future time period.

[0106] The foot and shoulder positions of the slope are determined according to the three-dimensional model of the mining area. Since the stress distribution at the foot, shoulder and crack positions is more concentrated, the location information of the area with a high probability of stress concentration in the target mining area is analyzed according to the foot, shoulder and crack positions, as well as the stress magnitude at the location of the stress sensor. All the location information is integrated and marked in the three-dimensional model of the mining area to obtain the stress distribution information of the target mining area. The seepage level of the target mining area is analyzed according to the crack prediction information and the precipitation information in the future time period to obtain the seepage level of each slope crack area. The seepage level of all slope crack areas is integrated to obtain the slope seepage information of the target mining area.

[0107] Finally, based on the acquired local displacement information, stress distribution information, slope seepage information and crack prediction information, the decision laboratory method was used to construct a slope risk prediction model for the target mining area. The risk coupling degree of each slope risk area in the target mining area was calculated according to the slope risk prediction model. The risk level of the slope risk area was divided according to the risk coupling degree. The location information and risk level of all slope risk areas were integrated to obtain the slope risk information.

[0108] S107. Construct a mining area digital twin model of the target mining area by combining the mining area three-dimensional model and the slope risk prediction model, and visualize the slope risk information through the mining area digital twin model.

[0109] In this embodiment, a digital twin model of the target mining area is constructed based on the three-dimensional model of the mining area and the slope risk prediction model. The digital twin model refers to a virtual model of a physical entity, process or system created on an information platform through digital modeling technology. The location information and risk level of the slope risk area are visualized in the digital twin model of the mining area, so that mining staff can eliminate risks in the mining area based on the digital twin model of the mining area, eliminate safety hazards in a timely manner, and give priority to high-risk areas.

[0110] In one embodiment, a three-dimensional model of a target mining area is constructed using laser radar point cloud data, and slope crack information of slope cracks in the target mining area is calculated based on the three-dimensional model of the mining area and by using a point cloud computing method, including the following steps:

[0111] Preprocessing the LiDAR point cloud data of each target area in the target mining area, including the slope area;

[0112] Perform surface fitting on the preprocessed LiDAR point cloud data to obtain multiple fitting surfaces, and combine the multiple fitting surfaces to construct a three-dimensional model of the target mining area, wherein the fitting surfaces include slope fitting surfaces corresponding to the slope area;

[0113] Calculate the residual values ​​between all slope point clouds belonging to the slope area in the laser radar point cloud data and the slope fitting surface, and integrate all slope point clouds with residual values ​​greater than a preset first residual threshold into a plurality of suspected crack point cloud areas;

[0114] For each suspected crack point cloud area, the normal vector angles between each slope point cloud and the neighborhood point cloud in the suspected crack point cloud area are calculated respectively, and the neighborhood point cloud is all point clouds within a preset neighborhood range around the slope point cloud;

[0115] If the normal vector angle is greater than the preset angle threshold and the residual value is less than the preset second residual threshold, the corresponding slope point cloud is used as the regional edge point cloud of the suspected crack point cloud area, and the second residual threshold is less than the first residual threshold;

[0116] The curvature calculation formula is used to calculate the average curvature of the edge point cloud of all regions;

[0117] If the average curvature is positive curvature, the suspected crack point cloud area is taken as the initial crack point cloud area;

[0118] The non-crack point cloud in the initial crack point cloud area is screened out by using the region growing algorithm to obtain the target crack point cloud area;

[0119] Based on the three-dimensional model of the mining area, the crack point cloud in the target crack point cloud area is mapped in two directions in multiple directions to obtain the two-dimensional point cloud coordinates of the crack point cloud in different directions. The slope crack information of the slope cracks in the target mining area is calculated according to the two-dimensional point cloud coordinates.

[0120] In this embodiment, the collected LiDAR point cloud data is first subjected to denoising and outlier removal. Commonly used denoising and outlier removal methods include radius filtering and Gaussian filtering, which can effectively remove noise points and outliers in LiDAR point cloud data. After preprocessing, the point cloud in the slope area of ​​the preprocessed LiDAR point cloud data is used for surface fitting. The methods that can be used for surface fitting include polynomial regression, spline interpolation, and triangulation algorithms. The point cloud is constructed into a mathematical surface, and the surface is made as close to all point clouds as possible. The surface is the slope fitting surface. Taking the triangulation algorithm as an example, the point cloud in the lidar point cloud data is first connected into triangles to construct a triangular mesh, and the adjacent triangular meshes are stitched to form a continuous surface. Then, the size, shape and distribution of the triangles are adjusted to ensure that the surface is smooth and has no obvious defects. Finally, the fitting surfaces of different areas in multiple target mining areas are obtained by stitching and optimizing the triangular meshes, including the slope fitting surfaces of the slope area. The multiple fitting surfaces obtained by fitting can be fused using modeling software to obtain a three-dimensional model of the target mining area. Commonly used modeling software includes CloudCompare software (three-dimensional point cloud processing software) and Meshlab software (open source mesh model processing software). After obtaining the slope fitting surface, the residual value between all point clouds and the surface is calculated. The residual value refers to the vertical distance between the point cloud and the fitting surface. The point cloud with a residual value greater than the preset first difference threshold is screened out to obtain a suspected crack point cloud area. This is because the area with slope cracks on the slope is concave, the distance between the point cloud and the slope fitting surface is large, and the residual value is also large, while the distance between the point cloud and the fitting surface in the gentle slope area is extremely small, and the residual of some point clouds is almost zero. Therefore, the threshold method can be set to preliminarily screen out the suspected crack point cloud area. However, since there may be raised obstacles in the slope area, such as stones, the residual values ​​of these areas may also be greater than the first residual threshold, so further screening is required.

[0121] The point cloud in the suspected crack point cloud area is divided into neighborhoods according to the preset neighborhood range threshold, and then the normal vector angle between each point cloud in the suspected crack point cloud area and the neighborhood point cloud corresponding to the point cloud is calculated, such as Figure 3 As shown, first, an osculating circle is constructed by connecting the point cloud and its corresponding neighborhood points, and then the tangents at the point cloud and the neighborhood points are obtained, and the normal vectors of the point cloud and the neighborhood points are obtained according to the vertical line of the tangent, and then the angle between the normal vectors, that is, the normal vector angle, is calculated. Similarly, the normal vector angle between each point cloud and its neighborhood points is calculated, and the suspected crack point cloud area is screened secondary according to the normal vector angle, that is, the point cloud with a normal vector angle greater than the preset angle threshold and a residual value less than the preset second residual value threshold is screened out to obtain the regional edge point cloud of the suspected crack point cloud area. This is because the normal vector angle suddenly increases at the junction between the crack area and the convex area and the flat land on the slope, and the residual value of the point cloud at the junction is smaller than the residual value at the bottom of the crack or the top of the convexity. Therefore, the point cloud at the junction can be screened out by setting a second residual threshold less than the first residual threshold to obtain the regional edge point cloud of the suspected crack point cloud area.

[0122] Then, the regional edge point cloud is fitted into an edge surface and the curvature calculation formula is used to calculate the average curvature of the regional edge point cloud. The method of fitting the edge surface is the same as the method of fitting the slope fitting surface. Both methods can be used, such as triangulation algorithm and polynomial regression. The formula for calculating the average curvature is as follows:

[0123]

[0124] Among them, E, F, G are the first basic invariants of the surface, and L, M, N are the second basic invariants of the surface.

[0125] If the calculated average curvature is positive, it means that the suspected crack point cloud area is the initial crack point cloud area. This is because the surface fitted by the regional edge point cloud of the slope crack area is convex, so the calculated average curvature is positive, while the surface fitted by the regional edge point cloud of the convex area of ​​the slope surface is concave, and the calculated average curvature should be negative. The point cloud of the convex area in the suspected crack point cloud area can be screened out and removed through positive and negative curvature to obtain the initial crack point cloud area. In order to ensure that the subsequent calculated information such as crack width, crack length and crack depth is more accurate and reliable, it is necessary to use the regional growing algorithm again to screen out the non-crack point cloud in the initial crack point cloud area to obtain the target crack point cloud area. After segmenting the precise target crack point cloud area, the target crack point cloud area is mapped from the x-axis-y-axis plane direction, the y-axis-z-axis plane direction and the x-axis-z-axis plane direction to obtain three different directions of two-dimensional point clouds. The crack width, crack length and crack depth of the local cracks on the slope corresponding to the target crack point cloud area can be calculated based on the two-dimensional coordinates of these two-dimensional point clouds. Finally, the specific location of all local cracks on the slope in the target mining area can be determined based on the three-dimensional coordinates of the target crack point cloud area and the three-dimensional model of the mining area to obtain the crack location. The crack width, crack length, crack depth and crack location of all local cracks on the slope corresponding to the target crack point cloud area are integrated to obtain the slope crack information of the slope cracks in the target mining area.

[0126] In one embodiment, Figure 2 As shown in the figure, the non-crack point cloud in the initial crack point cloud area is screened out by using the region growing algorithm to obtain the target crack point cloud area, which includes the following steps:

[0127] The curvature of the initial crack point cloud is estimated and calculated according to the normal vector angles of all the initial crack point clouds in the initial crack point cloud area to obtain the principal curvature of all the initial crack point clouds;

[0128] Construct an empty seed point sequence and a non-crack area point set, and add the initial crack point cloud with the smallest principal curvature to the seed point sequence as the seed point of the curvature region growing algorithm;

[0129] Based on the preset neighborhood order and using the K-neighborhood algorithm, multiple seed neighborhood points of the seed point are identified, and the normal angle between the seed point and each seed neighborhood point is calculated respectively;

[0130] If all normal angles are greater than the normal angle threshold, the region growth is considered complete;

[0131] If any one or more normal angles are less than the preset normal angle threshold, the corresponding seed neighborhood points are added to the non-crack area point set;

[0132] Calculate the curvature value of each seed neighborhood point in the non-crack area point set respectively;

[0133] If any one or more curvature values ​​are less than the preset curvature value threshold, the corresponding seed neighborhood point is added to the seed point sequence as a new seed point for regional growth, and the original seed point in the seed point sequence is removed;

[0134] If all curvature values ​​are greater than the curvature value threshold or the seed point sequence is an empty set, the region growing is considered complete;

[0135] If it is determined that the region growth is completed, the non-crack area point set is removed from the initial crack point cloud area to obtain the target crack point cloud area.

[0136] In this embodiment, the normal curvature of all initial point clouds is calculated according to the normal vector angles of all initial crack point clouds in the initial crack point cloud area. The normal curvature calculation formula is as follows:

[0137]

[0138] Among them, α is the angle between the straight line connecting the initial crack point cloud and its corresponding initial neighborhood point cloud and the normal vector of the initial crack point cloud, β is the initial normal vector angle, and pq i It is the straight-line distance between the initial crack point cloud p and the i-th initial neighborhood point cloud q.

[0139] The principal curvature of all initial crack point clouds is calculated based on the calculated normal curvature. The relationship between the principal curvature and the normal curvature is as follows:

[0140]

[0141] Among them, θ is the angle between the two main direction vectors of the initial crack point cloud p, θ i pq i With pq i The angle between the plane projections of , k1 and k2 are the two principal curvatures of the initial crack point cloud.

[0142] Construct an empty seed point sequence and a point set in the non-crack area, and add the initial crack point cloud with the smallest principal curvature as the seed point of the curvature region growing algorithm to the seed point sequence. First determine the neighborhood order (can be first order), calculate the distance between the point cloud and all other point clouds, and select the K points closest to it as the neighborhood point cloud of the point cloud. This is the K-neighbor algorithm. This method can be used to determine multiple seed points of the seed point, and then calculate the normal angle between the seed point and each seed neighborhood point. The normal line refers to a line on a plane that is perpendicular to the tangent of a curve at a certain point. The normal angle between the seed point and each seed neighborhood point can be calculated using geometry. If all calculated normal angles are greater than the normal angle threshold, regional growing is completed. If any one or more calculated normal angles are less than the preset normal angle threshold, the seed point is added to the non-crack area point set, and then the normal curvature calculation formula is used to calculate the curvature values ​​of all seed neighborhood points in the non-crack area point set. The seed neighborhood points whose curvature values ​​are less than the preset curvature value threshold are added to the seed point sequence as new seed points for regional growing, and the seed points are removed from the seed point sequence. The above steps are repeated until all seed points in the seed sequence are removed, indicating that regional growing is completed, and the non-crack area point set is removed from the initial crack point cloud area to obtain the target crack point cloud area.

[0143] In one embodiment, Figure 3 As shown, based on the three-dimensional model of the mining area, the crack point cloud in the target crack point cloud area is mapped in two dimensions in multiple directions to obtain the two-dimensional point cloud coordinates of the crack point cloud in different directions. The slope crack information of the slope cracks in the target mining area is calculated according to the two-dimensional point cloud coordinates, including the following steps:

[0144] S201, for each target fracture point cloud region, obtaining a point cloud three-dimensional coordinate set of all fracture point clouds in the target fracture point cloud region according to a three-dimensional model of the mining area;

[0145] S202, based on the point cloud three-dimensional coordinate set, two-dimensionally map all crack point clouds in the target crack point cloud area in three mapping directions: the x-axis-y-axis plane direction, the y-axis-z-axis plane direction, and the x-axis-z-axis plane direction to obtain two-dimensional point clouds in the three mapping directions;

[0146] S203, respectively calculating the crack width, crack length and crack depth of the local cracks on the slope corresponding to the target crack point cloud area according to the two-dimensional point cloud coordinates of the two-dimensional point clouds in the three mapping directions;

[0147] S204, calculating the crack position of the local crack of the slope corresponding to the target crack point cloud area through the point cloud three-dimensional coordinate set;

[0148] S205, integrating the crack width, crack length, crack depth and crack position of the local cracks in the slope corresponding to all target crack point cloud areas into slope crack information of the slope cracks in the target mining area.

[0149] In this embodiment, the point cloud three-dimensional coordinate set of all crack point clouds in each target crack point cloud area is obtained according to the three-dimensional model of the mining area. These point cloud three-dimensional coordinates contain information of the width, length and depth of the local cracks in the slope in three dimensions. This information can be stably and clearly reflected from the side view, top view and front view of all crack point clouds. Therefore, by mapping these three perspectives, the mapping diagrams in the three directions of xy, yz and xz can be obtained. According to the point cloud three-dimensional coordinate set, mapping is performed from the x-axis-y-axis plane direction, the y-axis-z-axis plane direction and the x-axis-z-axis plane direction of the target crack point cloud area to obtain three two-dimensional point clouds in different directions. According to the two-dimensional coordinates of these two-dimensional point clouds, the crack width, crack length and crack depth of the local cracks in the slope corresponding to the target crack point cloud area can be calculated. The two-dimensional point cloud mapped in the xy direction can calculate the local crack width of the slope, the two-dimensional point cloud mapped in the yz direction can calculate the local crack length of the slope, and the two-dimensional point cloud mapped in the xz direction can calculate the local crack depth of the slope. For example, in the mapping in the x-axis-y-axis plane direction, the two-dimensional coordinates of the nth two-dimensional point cloud are (Xn, Yn), so the crack length can be obtained by calculating the difference between the maximum and minimum values ​​of the X-axis coordinate value in the two-dimensional coordinates. Similarly, the crack width can be obtained by calculating the difference between the maximum and minimum values ​​of the Y-axis coordinate value in the two-dimensional coordinates. In addition, since the segmented target crack point cloud area is very accurate, the calculated crack width, crack length and crack depth are also accurate. Finally, the center point coordinates of the target crack point cloud area are determined according to the three-dimensional model of the mining area, and the center point coordinates are used as the crack position of the slope crack in the target mining area. The slope crack information of the slope cracks in the target mining area is obtained by integrating the crack width, crack length, crack depth and crack position of the local cracks in the slope corresponding to all target crack point cloud areas.

[0150] In one embodiment, a slope risk prediction model is constructed based on sensor information, slope crack information and local displacement information, and slope risk analysis is performed on the slope area using the slope risk prediction model to obtain slope risk information of the target mining area, including the following steps:

[0151] Obtain mining blasting information of the target mining area according to the preset mining plan;

[0152] According to the mining area blasting information and local displacement information, the slope crack information change trend analysis is carried out to obtain the crack prediction information of the target mining area in the future time period;

[0153] According to the crack location information and the three-dimensional model of the mining area, the stress distribution information of the rock mass collected by the stress sensor is analyzed by using the threshold method to obtain the stress distribution information of the target mining area;

[0154] Estimate groundwater prediction information of the target mining area based on the precipitation information of the future time period acquired in advance and the groundwater information collected by the groundwater level sensor;

[0155] Based on the crack prediction information and groundwater prediction information, the seepage velocity level of the target mining area is analyzed by using the threshold method to obtain the slope seepage information of the target mining area;

[0156] A slope risk prediction model for the target mining area is constructed based on local displacement information, stress distribution information, slope seepage information and crack prediction information, and the slope risk information of the target mining area is predicted based on the slope risk prediction model.

[0157] In this embodiment, the change trend of the slope cracks in the future event period is first analyzed according to the mining area blasting information and the local displacement information, that is, the crack prediction information. Specifically, the blasting influence radius of the blasting operation is calculated according to the explosive quality and type of explosives in the mining area blasting information and the influence radius calculation formula, the blasting influence range is determined according to the blasting influence radius and the blasting position, the blasting influence degree of the blasting operation on the slope cracks is measured according to the blasting influence range, the displacement influence degree of the local displacement information on all the slope cracks is evaluated according to the pre-constructed displacement crack digital model, the slope crack information of all the slope cracks is linearly regressed with the blasting influence degree and displacement influence degree of all the slope cracks, a linear regression equation is fitted, a crack prediction model is constructed according to the fitted linear regression equation, and the crack prediction information of the future time period is predicted according to the blasting influence degree and displacement influence degree of the current event node, and the crack prediction information obtained includes the crack prediction width, crack prediction length and crack prediction depth of all the slope cracks in the future time period.

[0158] The toe and shoulder positions of the slope are determined according to the three-dimensional model of the mining area. Since the stress distribution at the toe, shoulder and crack positions is more concentrated, the location information of the area with a higher probability of stress concentration in the target mining area is analyzed according to the stress magnitude of the toe, shoulder and crack positions and the location of the stress sensor. Specifically, the distance between the crack position and the toe position, the distance between the crack position and the shoulder position, and the distance between the crack position and the position where the stress is greater than the preset stress threshold are calculated respectively. The three distances are added together to obtain a comprehensive distance. The probability level of stress concentration is divided according to the size of the comprehensive distance. The probability level of stress concentration in all slope cracks is integrated with the slope positions corresponding to all slope cracks to obtain stress distribution information. For example, at the coordinate (20, 10, 20), the probability level of stress concentration is primary stress.

[0159] The precipitation information in the future time period refers to the total precipitation in the future time period. The groundwater information collected by the groundwater level sensor refers to the groundwater level of the target mining area collected by the groundwater sensor such as the water level meter. According to the total precipitation in the future time period and the groundwater level at the current time node, the predicted underground water level in the future time period can be estimated, that is, the groundwater prediction information. Then, the slope seepage information of the target mining area is predicted according to the crack prediction information and the groundwater prediction information. The slope seepage information refers to the seepage speed level of the area around all the slope cracks in the slope area. Since the crack width will affect the seepage path and speed of the groundwater, the groundwater seepage speed is measured according to the crack prediction width and the underground predicted water level in the crack prediction information. For example, the area around the slope cracks with a crack prediction width greater than a preset crack prediction width threshold and an underground predicted water level greater than a preset underground predicted water level threshold is divided into a first-level seepage speed. The crack positions of all slope cracks and the seepage speed level of the corresponding area around the slope cracks are integrated to obtain the slope seepage information of the target mining area.

[0160] Since local displacement information, stress distribution information, slope seepage information and crack prediction information will affect whether collapse, landslide and other accidents will occur in the target mining area, a slope risk prediction model for the target mining area can be constructed based on the acquired local displacement information, stress distribution information, slope seepage information and crack prediction information and the decision laboratory method. The risk coupling degree of each slope risk area in the target mining area is calculated according to the slope risk prediction model, and the risk level of the slope risk area is divided according to the risk coupling degree. The location information and risk level of all slope risk areas are integrated to obtain the slope risk information.

[0161] In one embodiment, performing a change trend analysis on slope crack information based on mining area blasting information and local displacement information to obtain crack prediction information of a target mining area in a future time period includes the following steps:

[0162] Calculate the blasting impact radius of the target mining area based on the blasting information of the mining area, evaluate the blasting impact degree of the slope cracks based on the blasting impact radius, and use the blasting impact degree evaluation result as the first influencing factor;

[0163] The displacement impact degree of all slope cracks is evaluated based on the local displacement information and the pre-built displacement crack digital model, and the displacement impact degree evaluation result is used as the second influencing factor;

[0164] The grey correlation model is used to analyze the influence of the first and second influencing factors on slope cracks, and weights are assigned to the first and second influencing factors according to the influence.

[0165] The least square method is used to perform linear regression analysis between all the first influencing factors, the second influencing factors and the crack information in the mining area after the weight distribution is completed, and a crack prediction model is constructed according to the results of the linear regression analysis;

[0166] The crack changes of slope cracks are predicted according to the crack prediction model to obtain the crack prediction information of all slope cracks in the target mining area in the future time period.

[0167] In this embodiment, the explosive equivalent is calculated based on the explosive quality and type in the mining area blasting information using the equivalent calculation formula, the explosion pressure is calculated based on the calculated explosive equivalent using the pressure calculation formula, and finally the blasting influence radius is calculated based on the calculated explosion equivalent and explosion pressure using the blasting influence radius calculation formula, the blasting range is calculated based on the blasting position and blasting influence radius in the mining area blasting information, and the overlap of the blasting range and the crack position in the slope crack information is used as the blasting influence degree of the blasting operation carried out in the target mining area on the slope cracks. The overlap refers to the number of slope cracks within the blasting range, and the obtained blasting influence degree is used as the first influencing factor. The equivalent calculation formula is as follows:

[0168]

[0169] Where W is the mass of explosive, Q is the energy released by explosive, Q TNT is the energy released by TNT.

[0170] The pressure calculation formula is as follows:

[0171]

[0172] Where P is the pressure at a distance R, K is a constant that depends on the type of explosive and environmental conditions, R is the distance from the center of the explosion, and n is the attenuation exponent, which is usually between 1.5 and 2.0.

[0173] The calculation formula of blasting impact radius is as follows:

[0174]

[0175] The displacement critical values ​​of slope cracks of different widths, depths and lengths are obtained by combining local displacement information and slope crack information using a pre-constructed displacement crack digital model. The difference between the local displacement corresponding to each slope crack and the displacement critical value is used as the influence of the local displacement on each slope crack. The displacement critical value is simulated experimentally. Exceeding this value will cause crack expansion. The displacement crack digital model is also constructed through pre-constructed simulation experiments. The obtained displacement influence degree is used as the second influencing factor.

[0176] The crack prediction information of all slope cracks is used as the reference sequence of the grey correlation model, and the first and second influencing factors are used as the comparison sequence. The reference sequence and the comparison sequence are dimensionless, and the grey correlation coefficient between the reference sequence and the comparison sequence is calculated using the calculation formula of the grey system correlation analysis. The average value of the grey correlation coefficient is calculated as the correlation degree value between the reference sequence and the comparison sequence. The correlation degree value is the influence degree of the first and second influencing factors on the slope cracks. The weights are assigned to the first and second influencing factors according to the obtained correlation degree value. After the weight assignment is completed, it is first assumed that there is a linear relationship between the first and second influencing factors and the crack information of the mining area. Then, the slope and intercept in the linear relationship are calculated using the least squares method to obtain the regression equation between the first and second influencing factors and the crack information of the mining area. The crack prediction model is constructed according to the obtained regression equation. In the future time period of the target mining area, there is no need to recalculate the crack information of the mining area. The crack prediction information of all slope cracks in the target mining area in the future time period can be directly predicted based on the blasting information of the mining area and the local displacement information collected by the displacement sensor and the crack prediction model.

[0177] In one embodiment, a slope risk prediction model for a target mining area is constructed based on local displacement information, stress distribution information, slope seepage information, and crack prediction information. Predicting the slope risk information of the target mining area based on the slope risk prediction model includes the following steps:

[0178] According to the pre-acquired slope risk assessment index, the direct impact degree is assigned to the impact degree between the local displacement information, stress distribution information, slope seepage information and crack prediction information by using the scaling method;

[0179] The local displacement information, stress distribution information, slope seepage information and crack prediction information are used as the core influencing factors of the slope in the target mining area, and the slope direct influence matrix of the core influencing factors of the slope is constructed according to the direct influence degree assignment results. The slope direct influence matrix is ​​S = [s ij ] 4×4 , where s ij It indicates the influence degree of slope core influencing factor i on slope core influencing factor j;

[0180] Normalize the slope direct impact matrix to obtain a standardized matrix;

[0181] Generate multiple decreasing matrix series as slope indirect impact matrix based on the standardized matrix;

[0182] The slope comprehensive impact matrix is ​​established by combining the slope direct impact matrix and all slope indirect impact matrices;

[0183] The influence and affected degree of the core influencing factors of the slope are calculated according to the comprehensive influence matrix of the slope, and the centrality and causality of the core influencing factors of the slope are calculated according to the influence and affected degree;

[0184] A slope risk prediction model is established based on the coupling coordination function by combining the centrality and causality, and the slope risk information of the target mining area is obtained according to the slope risk prediction model.

[0185] In this embodiment, the four core influencing factors of the slope are compared in pairs according to the pre-acquired slope risk assessment index to determine the direct influence relationship between them, and the direct influence degree between the four core influencing factors of the slope is divided into four levels and assigned using the judgment rule of the 0-5 scale method, where 0 represents no direct influence, 1 represents very weak direct influence, 2 represents relatively weak direct influence, 3 represents medium direct influence, 4 represents relatively strong direct influence, and 5 represents extremely strong direct influence. The direct influence degree is divided according to the proportion of the influence of the core influencing factor i of the slope on the core influencing factor j of the slope, and the score is 0 when the proportion is 0%, the score is 1 when the proportion is 1% to 20%, the score is 2 when the proportion is 21% to 40%, the score is 3 when the proportion is 41% to 60%, the score is 4 when the proportion is 61% to 80%, and the score is 5 when the proportion is 81% to 100%. According to the assignment results, the slope direct influence matrix of the four core influencing factors of the slope, namely local displacement information, stress distribution information, slope seepage information and crack prediction information, is established. The slope direct influence matrix is ​​as follows:

[0186]

[0187] The number filled in the blank of the i-th row and j-th column represents the influence of the i-th core influencing factor on the j-th core influencing factor of the slope, for example, S 12 It is the influence degree of the first slope core influencing factor on the second slope core influencing factor.

[0188] The slope direct impact matrix is ​​normalized using the standardized calculation formula to obtain a standardized matrix. The standardized calculation formula is as follows:

[0189] a=S / F

[0190] Among them, S is the slope direct impact matrix, and F is the maximum value of the sum of the elements in each row of the slope direct impact matrix.

[0191] Based on the standardized matrix, a plurality of decreasing matrix sequences are generated as the slope indirect influence matrix, and the plurality of decreasing matrix sequences are composed of a 2 、a 3 ...an, satisfying the following formula:

[0192]

[0193] The slope comprehensive influence matrix is ​​established based on the slope direct influence matrix and all slope indirect influence matrices. The calculation formula of the slope comprehensive influence matrix K is as follows:

[0194] K=a+a 2 +a 3 +......+a n =a(1-a) -1

[0195] Each row of the slope comprehensive influence matrix K is added to obtain the influence degree of the four core influencing factors of the slope. Each column of the slope comprehensive influence matrix K is added to obtain the influence degree of the four core influencing factors of the slope. The influence degree and the influence degree are added to obtain the centrality of the core influencing factors of the slope. The influence degree minus the influence degree is used to obtain the cause degree of the core influencing factors of the slope. The centrality indicates the degree of influence of the core influencing factor of the slope on other core influencing factors of the slope and the total degree of influence of the core influencing factor of the slope by other core influencing factors of the slope. The cause degree indicates the degree of influence of the core influencing factor of the slope on other core influencing factors of the slope or the total degree of influence of the core influencing factor of the slope by other core influencing factors of the slope. The centrality is used as the indicator weight required to construct the slope risk prediction model. The slope risk prediction model is established based on the coupling coordination function by combining the centrality and the cause degree. The risk coupling degree of the slope risk area is calculated according to the slope risk prediction model. The risk level of the slope risk area is divided according to the risk coupling degree. The location information and risk level of all slope risk areas are integrated to obtain the slope risk information.

[0196] In one embodiment, a slope risk prediction model is established based on the coupling coordination function in combination with the centrality and the causality, and the slope risk information of the target mining area is obtained according to the slope risk prediction model, including the following steps:

[0197] According to the centrality corresponding to the local displacement information, stress distribution information, slope seepage information and crack prediction information, weights are assigned to the core influencing factors of the slope to obtain the weights of the core influencing factors of the slope;

[0198] A four-quadrant diagram is constructed based on the centrality and cause degree of the core influencing factors of the slope, and the core influencing factors of the slope are divided into slope cause factors and slope result factors according to the four-quadrant diagram;

[0199] The slope risk prediction model is established based on the weights corresponding to the core influencing factors of the slope and the coupling coordination function. The slope risk prediction model is:

[0200]

[0201] Among them, γ is the risk coupling degree, is the weight of the core influencing factor of the i-th slope, is the nth slope core influencing factor;

[0202] The slope risk information of the target mining area is obtained by combining the slope cause factors and slope result factors and using the slope risk prediction model.

[0203] In this embodiment, weights of the four core influencing factors of the slope are allocated according to the centralities corresponding to the local displacement information, stress distribution information, slope seepage information and crack prediction information. Then, the centrality of the core influencing factors of the slope is used as the horizontal coordinate, the causal degree of the core influencing factors of the slope is used as the vertical coordinate, and the mean of the centrality and the causal degree are used as the horizontal and vertical dividing lines to construct a four-quadrant diagram. The local displacement information and crack prediction information in the first and second quadrants of the four-quadrant diagram are used as slope causal factors, which are the core influencing factors of the slope that will directly lead to slope collapse. The stress distribution information and slope seepage information in the third and fourth quadrants are used as slope result factors. Subsequently, the slope risk area is divided according to the slope cause factors, and the slope risk level is calculated according to the slope result factors.

[0204] The slope risk prediction model is established based on the weights corresponding to the core influencing factors of the slope and the coupling coordination function. The coupling coordination function is:

[0205]

[0206] Among them, Cn is the coupling degree of the n-element system; u1...un are the contribution rates of the first subsystem to the nth subsystem to the order of the total system.

[0207] In one embodiment, the slope risk prediction model is used to obtain the slope risk information of the target mining area by combining the slope cause factors and the slope result factors, including the following steps:

[0208] Divide the slope area according to the three-dimensional model of the mining area to obtain multiple slope sub-areas;

[0209] According to the three-dimensional model of the mining area, the slope sub-areas with slope causes are screened out to obtain the slope risk areas;

[0210] Analyze the slope risk area according to the slope result factors, and determine whether there is stress concentration and water seepage in the slope risk area based on the analysis results;

[0211] If there is no stress concentration phenomenon and / or water seepage phenomenon in the slope risk area, the slope risk area is judged to be a low-risk area;

[0212] If stress concentration and water seepage exist simultaneously in the risk area of ​​the slope, the weights corresponding to the local displacement information, stress distribution information, slope water seepage information and crack prediction information are input into the slope risk prediction model to obtain the risk coupling degree of the risk area of ​​the slope.

[0213] If the risk coupling degree is greater than the preset risk coupling degree threshold, the slope risk area is marked as a high-risk area;

[0214] If the risk coupling degree is less than or equal to the risk coupling degree threshold, the slope risk area is marked as a low-risk area;

[0215] Obtain the location information of all high-risk areas and low-risk areas according to the three-dimensional model of the mining area, and obtain the location of the high-risk area and the low-risk area;

[0216] The locations of all high-risk areas and low-risk areas are integrated to obtain the slope risk information of the target mining area.

[0217] In this embodiment, the slope area is first divided into multiple slope sub-areas according to the three-dimensional model of the mining area. Since the target mining area is large enough and the slope area is also large enough, multiple slope sub-areas can be divided. The slope sub-areas are monitored separately, which can make the monitored slope risk information more comprehensive and more accurate. Then, the slope sub-areas with slope cracks are divided according to the three-dimensional model of the mining area, and these slope sub-areas are determined as slope risk areas. This is because the slope area of ​​the target mining area is originally an area prone to landslides and collapses. Once cracks appear, it proves that the area is very likely to have a collapse accident, so it is directly divided into a slope risk area for key monitoring.

[0218] According to the stress distribution information and slope seepage information, it is judged whether stress concentration and seepage will occur in the current slope risk area. The judgment is mainly based on the probability level and seepage speed level in the stress distribution information and slope seepage information. If there is no stress concentration or seepage in the slope risk area, the current slope risk area can be directly judged as a low-risk area. The low-risk area will also be eliminated in the future, but the priority of the investigation work is lower than that of the high-risk area. If there are stress concentration and seepage in the current slope risk area at the same time, the weights corresponding to the local displacement information, stress distribution information, slope seepage information and crack prediction information are input into the slope risk prediction model. The output of the slope risk prediction model is the risk coupling degree. The risk coupling degree can measure the probability of dangerous accidents in the slope risk area. The higher the risk coupling degree, the greater the possibility of accidents such as collapse and landslides, and the higher the danger level of the slope risk area. Slope risk areas with risk coupling degrees greater than the preset risk coupling degree threshold will be marked as high-risk areas, and risk investigation will be carried out on high-risk areas first. Slope risk areas with risk coupling degrees less than or equal to the coupling degree threshold will be marked as low-risk areas. After completing the risk elimination work in the high-risk areas, risk investigation will be carried out on the low-risk areas immediately.

[0219] According to the three-dimensional model of the mining area, the location information of all high-risk areas and low-risk areas is obtained, and the locations of high-risk areas and low-risk areas are obtained, and the slope risk information of the target mining area is integrated. The locations of high-risk areas and low-risk areas will be directly marked in the three-dimensional model of the mining area and intuitively displayed in the digital twin model of the mining area, which is convenient for the mining area staff to arrange the risk investigation work of the slope area of ​​the target mine in the future, eliminate the safety hazards of the target mine in time, and ensure the personal safety of the staff of the target mining area.

[0220] The present application also discloses a data risk analysis system based on the Internet of Things, which is characterized by comprising:

[0221] a memory configured to store instructions; and

[0222] The processor is configured to call instructions from the memory and implement the above-mentioned method of data risk analysis based on the Internet of Things when executing the instructions.

[0223] Among them, the processor can adopt a central processing unit (CPU). Of course, according to actual usage, other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. can also be adopted. The general-purpose processor can adopt a microprocessor or any conventional processor, etc., and this application does not impose any restrictions on this.

[0224] Among them, the memory can be an internal storage unit of a computer device, such as a hard disk or memory of a computer device, or an external storage device of a computer device, such as a plug-in hard disk, a smart memory card (SMC), a secure digital card (SD) or a flash memory card (FC) equipped on the computer device, etc., and the memory can also be a combination of an internal storage unit and an external storage device of a computer device. The memory is used to store computer programs and other programs and data required by the computer device. The memory can also be used to temporarily store data that has been output or is to be output, and this application does not impose any restrictions on this.

[0225] An embodiment of the present application also provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the above-mentioned method of data risk analysis based on the Internet of Things.

[0226] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0227] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0228] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0229] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0230] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0231] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0232] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0233] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0234] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A data risk analysis method based on the Internet of Things, characterized in that: The method comprises the following steps: Obtain the LiDAR point cloud data collected by the drone in the target mining area; Collecting sensor information of the target mining area based on multiple types of sensors preset in the target mining area, wherein the multiple types of sensors include stress sensors and groundwater level sensors; Using the laser radar point cloud data to construct a three-dimensional model of the target mining area, and based on the three-dimensional model of the mining area and using a point cloud computing method to calculate slope crack information of the slope cracks in the target mining area; Combining the three-dimensional model of the mining area and the slope crack information to generate a crack monitoring point distribution strategy for the slope area of ​​the target mining area; Deploy a plurality of displacement sensors in the target mining area according to the monitoring point distribution strategy, and use the displacement sensors to perform local displacement monitoring on all the slope cracks in the slope area to obtain local displacement information; Building a slope risk prediction model based on the sensor information, the slope crack information and the local displacement information, and using the slope risk prediction model to perform slope risk analysis on the slope area to obtain slope risk information of the target mining area; A mining area digital twin model of the target mining area is constructed by combining the mining area three-dimensional model and the slope risk prediction model, and the slope risk information is visualized through the mining area digital twin model.

2. The method according to claim 1, characterized in that: The step of constructing a three-dimensional mining area model of the target mining area by using the laser radar point cloud data, and obtaining slope crack information of the slope cracks of the target mining area by calculating the three-dimensional mining area model and using a point cloud computing method, comprises the following steps: Preprocessing the laser radar point cloud data of each target area in the target mining area, wherein the target area includes a slope area; Performing surface fitting on the preprocessed laser radar point cloud data to obtain a plurality of fitting surfaces, and combining the plurality of fitting surfaces to construct a three-dimensional mining area model of the target mining area, wherein the fitting surfaces include a slope fitting surface corresponding to the slope area; Calculating residual values ​​between all slope point clouds belonging to the slope area in the laser radar point cloud data and the slope fitting surface, and integrating all the slope point clouds whose residual values ​​are greater than a preset first residual threshold into a plurality of suspected crack point cloud areas; For each of the suspected crack point cloud areas, respectively calculating the normal vector angles between each of the slope point clouds and the neighborhood point clouds in the suspected crack point cloud area, wherein the neighborhood point clouds are all point clouds within a preset neighborhood range around the slope point cloud; If the normal vector angle is greater than a preset angle threshold and the residual value is less than a preset second residual threshold, the corresponding slope point cloud is used as the regional edge point cloud of the suspected crack point cloud area, and the second residual threshold is less than the first residual threshold; Calculate the average curvature of all edge point clouds of the region using a curvature calculation formula; If the average curvature is a positive curvature, the suspected crack point cloud area is used as an initial crack point cloud area; Using a region growing algorithm to filter out non-crack point clouds in the initial crack point cloud region to obtain a target crack point cloud region; Based on the three-dimensional model of the mining area, multi-directional two-dimensional mapping is performed on the crack point cloud in the target crack point cloud area to obtain the two-dimensional point cloud coordinates of the crack point cloud in different directions, and the slope crack information of the slope cracks in the target mining area is calculated according to the two-dimensional point cloud coordinates.

3. The method according to claim 2, characterized in that The method of using a region growing algorithm to screen out non-crack point clouds in the initial crack point cloud region to obtain a target crack point cloud region comprises the following steps: Performing curvature estimation calculation on the initial crack point cloud according to the normal vector angles of all the initial crack point clouds in the initial crack point cloud region to obtain the principal curvatures of all the initial crack point clouds; Constructing an empty seed point sequence and a non-crack region point set, and adding the initial crack point cloud with the smallest principal curvature to the seed point sequence as a seed point of the curvature region growing algorithm; Based on a preset neighborhood order and using a K-neighborhood algorithm, a plurality of seed neighborhood points of the seed point are identified, and the normal angle between the seed point and each of the seed neighborhood points is calculated respectively; If all the normal angles are greater than the normal angle threshold, it is determined that the region growing is completed; If any one or more of the normal angles are smaller than a preset normal angle threshold, the corresponding seed neighborhood point is added to the non-crack area point set; Calculating the curvature value of each of the seed neighborhood points in the non-crack area point set respectively; If any one or more of the curvature values ​​are less than a preset curvature value threshold, the corresponding seed neighborhood point is added to the seed point sequence as a new seed point for regional growth, and the original seed point in the seed point sequence is removed; If all the curvature values ​​are greater than the curvature value threshold or the seed point sequence is an empty set, it is determined that the region growing is completed; If it is determined that the region growth is completed, the non-crack region point set is removed from the initial crack point cloud region to obtain a target crack point cloud region.

4. The method according to claim 2, characterized in that: The method of performing multi-directional two-dimensional mapping on the crack point cloud in the target crack point cloud area based on the three-dimensional model of the mining area to obtain the two-dimensional point cloud coordinates of the crack point cloud in different directions, and calculating the slope crack information of the slope cracks in the target mining area according to the two-dimensional point cloud coordinates comprises the following steps: For each of the target fracture point cloud regions, a point cloud three-dimensional coordinate set of all fracture point clouds in the target fracture point cloud region is obtained according to the three-dimensional model of the mining area; Based on the point cloud three-dimensional coordinate set, all the crack point clouds in the target crack point cloud area are two-dimensionally mapped in three mapping directions: the x-axis-y-axis plane direction, the y-axis-z-axis plane direction, and the x-axis-z-axis plane direction, to obtain two-dimensional point clouds in the three mapping directions; Calculating the crack width, crack length and crack depth of the local cracks on the slope corresponding to the target crack point cloud area according to the two-dimensional point cloud coordinates of the two-dimensional point cloud in the three mapping directions respectively; The crack position of the local crack of the slope corresponding to the target crack point cloud area is calculated by the point cloud three-dimensional coordinate set; The crack width, the crack length, the crack depth and the crack position of the local cracks in the slope corresponding to all the target crack point cloud areas are integrated into the slope crack information of the slope cracks in the target mining area.

5. The method according to claim 1, characterized in that The step of constructing a slope risk prediction model based on the sensor information, the slope crack information and the local displacement information, and performing slope risk analysis on the slope area using the slope risk prediction model to obtain the slope risk information of the target mining area comprises the following steps: Acquire mining blasting information of the target mining area according to a preset mining area mining plan; Performing a change trend analysis on the slope crack information according to the mining area blasting information and the local displacement information to obtain crack prediction information of the target mining area in a future time period; According to the crack position information and the three-dimensional model of the mining area, a stress distribution analysis is performed on the rock mass stress information collected by the stress sensor using a threshold method to obtain stress distribution information of the target mining area; Estimate groundwater prediction information of the target mining area based on the precipitation information of the future time period acquired in advance and the groundwater information collected by the groundwater level sensor; Based on the crack prediction information and the groundwater prediction information, the water seepage rate grade analysis of the target mining area is performed using a threshold method to obtain the slope water seepage information of the target mining area; A slope risk prediction model for the target mining area is constructed based on the local displacement information, the stress distribution information, the slope seepage information and the crack prediction information, and the slope risk information of the target mining area is predicted based on the slope risk prediction model.

6. The method according to claim 5, characterized in that The step of performing a change trend analysis on the slope crack information according to the mining area blasting information and the local displacement information to obtain crack prediction information of the target mining area in a future time period comprises the following steps: Calculating the blasting influence radius of the target mining area according to the mining area blasting information, evaluating the blasting influence degree of the slope cracks according to the blasting influence radius, and using the blasting influence degree evaluation result as the first influence factor; According to the local displacement information and the pre-constructed displacement crack digital model, the displacement influence degree of all the slope cracks is evaluated, and the displacement influence degree evaluation result is used as the second influencing factor; Analyzing the influence of the first influencing factor and the second influencing factor on the slope cracks by using a grey correlation model, and assigning weights to the first influencing factor and the second influencing factor according to the influence degrees; Using the least squares method, a linear regression analysis is performed between all the first influencing factors, the second influencing factors and the mining area crack information after the weight allocation is completed, and a crack prediction model is constructed according to the linear regression analysis results; The crack changes of the slope cracks are predicted according to the crack prediction model to obtain crack prediction information of all the slope cracks in the target mining area in the future time period.

7. The method according to claim 5, characterized in that The step of constructing a slope risk prediction model for the target mining area according to the local displacement information, the stress distribution information, the slope seepage information and the crack prediction information, and predicting the slope risk information of the target mining area according to the slope risk prediction model comprises the following steps: According to the pre-acquired slope risk assessment index and using a scaling method, a direct impact degree is assigned to the impact degree between the local displacement information, the stress distribution information, the slope seepage information and the crack prediction information; The local displacement information, the stress distribution information, the slope seepage information and the crack prediction information are used as the core influencing factors of the slope of the target mining area, and the slope direct influence matrix of the core influencing factors of the slope is constructed according to the direct influence degree assignment result. The slope direct influence matrix is ​​S=[s ij ] 4×4 , where s ij Indicates the influence degree of the core influencing factor i of the slope on the core influencing factor j of the slope; Normalizing the slope direct impact matrix to obtain a standardized matrix; Based on the standardized matrix, a plurality of decreasing matrix series are generated as slope indirect influence matrices; Establishing a slope comprehensive impact matrix by combining the slope direct impact matrix and all the slope indirect impact matrices; Calculate the influence degree and the influenced degree of the core influencing factor of the slope according to the comprehensive influence matrix of the slope, and calculate the centrality and the cause degree of the core influencing factor of the slope according to the influence degree and the influenced degree; A slope risk prediction model is established based on the coupling coordination function in combination with the centrality and the causality, and slope risk information of the target mining area is obtained according to the slope risk prediction model.

8. The method according to claim 7, characterized in that The step of establishing a slope risk prediction model based on the coupling coordination function by combining the centrality and the causality, and obtaining the slope risk information of the target mining area according to the slope risk prediction model comprises the following steps: Assigning weights to the core influencing factors of the slope according to the centrality corresponding to the local displacement information, the stress distribution information, the slope seepage information and the crack prediction information, to obtain the weights of the core influencing factors of the slope; Constructing a four-quadrant diagram according to the centrality and the cause degree of the core influencing factors of the slope, and dividing the core influencing factors of the slope into cause factors of the slope and result factors of the slope according to the four-quadrant diagram; A slope risk prediction model is established according to the weights corresponding to the core influencing factors of the slope and the coupling coordination function. The slope risk prediction model is: Among them, γ is the risk coupling degree, is the weight of the i-th core influencing factor of the slope, is the nth core influencing factor of the slope; The slope risk information of the target mining area is obtained by combining the slope cause factors and the slope result factors and utilizing the slope risk prediction model.

9. The method according to claim 8, characterized in that The step of combining the slope cause factors and the slope result factors with the slope risk prediction model to obtain the slope risk information of the target mining area comprises the following steps: Divide the slope area according to the three-dimensional model of the mining area to obtain a plurality of slope sub-areas; The slope sub-areas where the slope cause factors exist are screened out according to the three-dimensional model of the mining area to obtain the slope risk area; Analyzing the slope risk area according to the slope result factors, and judging whether there is stress concentration and water seepage in the slope risk area according to the analysis results; If the stress concentration phenomenon and / or the water seepage phenomenon does not exist in the slope risk area, the slope risk area is determined to be a low-risk area; If the stress concentration phenomenon and the water seepage phenomenon exist simultaneously in the slope risk area, the weights corresponding to the local displacement information, the stress distribution information, the slope water seepage information and the crack prediction information are input into the slope risk prediction model to obtain the risk coupling degree of the slope risk area; If the risk coupling degree is greater than a preset risk coupling degree threshold, marking the slope risk area as a high-risk area; If the risk coupling degree is less than or equal to the risk coupling degree threshold, marking the slope risk area as the low risk area; Acquire the location information of all the high-risk areas and the low-risk areas according to the three-dimensional model of the mining area, and obtain the location of the high-risk areas and the low-risk areas; The locations of all the high-risk areas and the low-risk areas are integrated to obtain the slope risk information of the target mining area.

10. A data risk analysis system based on the Internet of Things, characterized in that: include: a memory configured to store instructions; as well as A processor is configured to call the instructions from the memory and to implement a method for data risk analysis based on the Internet of Things according to any one of claims 1 to 9 when executing the instructions.

Images