Mountain stability evaluation method and device, computer equipment and storage medium

Abstract

The application relates to a mountain stability evaluation method, device, computer equipment and storage medium, and relates to the technical field of mountain stability. The evaluation method comprises the following steps: dividing a mountain into a plurality of voxels; constructing an adjacency structure graph and a density field posterior probability distribution of the mountain, wherein the adjacency structure graph is constructed by muon data recorded by a muon detector arranged in a grotto, and the density field posterior probability distribution is a probability distribution of all density models of the mountain under the muon data and the adjacency structure graph; sampling from the density field posterior probability distribution according to the adjacency structure graph to obtain a plurality of density field samples of the mountain; obtaining a density value of each voxel according to the plurality of density field samples; and obtaining a stability abnormal area in the mountain according to the density values of the plurality of voxels. The mountain stability evaluation method can accurately determine the stability abnormal area in the mountain, and improves the accuracy of mountain stability evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mountain stability, and in particular to a mountain stability evaluation method and device, computer equipment and a storage medium. BACKGROUND

[0002] Cave-like cultural relics are attached to mountains, and the stability of the mountains affects the safety of the cave-like cultural relics. Currently, methods such as geological radar detection, seismic exploration, drilling and numerical simulation are mainly used to evaluate the stability of the mountains attached to the caves. However, these methods have problems such as limited detection depth, insufficient resolution, only local detection, destructive, dependence on ideal parameters and prior assumptions, and difficulty in reflecting the true structure, making it difficult to accurately evaluate the stability of the mountains attached to the caves. SUMMARY

[0003] To overcome the problems in the related art, the present application provides a mountain stability evaluation method and device, computer equipment and a storage medium.

[0004] According to a first aspect of an embodiment of the present application, a mountain stability evaluation method is provided, which is used to evaluate the stability of a mountain attached to a cave. The evaluation method comprises:

[0005] dividing the mountain into a plurality of voxels;

[0006] constructing an adjacency structure graph and a density field posterior probability distribution of the mountain, wherein the adjacency structure graph is constructed by muon data recorded by a muon detector arranged in the cave, and the density field posterior probability distribution is a probability distribution of all density models of the mountain under the muon data and the adjacency structure graph;

[0007] sampling from the density field posterior probability distribution according to the adjacency structure graph to obtain a plurality of density field samples of the mountain;

[0008] obtaining a density value of each voxel according to a plurality of density field samples;

[0009] obtaining a stability abnormal area in the mountain according to the density values of a plurality of voxels.

[0010] In some exemplary embodiments of the present application, the construction of the adjacency structure graph and the density field posterior probability distribution of the mountain comprises:

[0011] record the muon data within a preset time period using the muon detector;

[0012] determine the muon transmission rate according to the muon data;

[0013] According to the point cloud data of the mountain and the muon transmission rate, the adjacency structure graph is obtained.

[0014] In some exemplary embodiments of the present application, the obtaining of the adjacency structure graph according to the point cloud data of the mountain and the muon transmission rate comprises:

[0015] According to the point cloud data and the muon transmission rate, a voxel graph model of the mountain is obtained, wherein the voxel graph model comprises an initial density value of each voxel;

[0016] Each voxel in the voxel graph model is taken as a node in the adjacency structure graph, wherein the attribute of the node comprises the initial density value;

[0017] The connection relationship between each two adjacent voxels in the voxel graph model is taken as an edge connecting two corresponding nodes in the adjacency structure graph, wherein the two corresponding nodes refer to the nodes corresponding to the two adjacent voxels, and each edge corresponds to a weight, which is used to reflect the structural similarity degree of the two nodes connected by the edge, and the greater the structural similarity degree, the greater the weight.

[0018] In some exemplary embodiments of the present application, the weight is the product of the geological structure main direction similarity weight, the joint density weight and the slope direction weight, or the weight is the product of the weighted geological structure main direction similarity weight, the weighted joint density weight and the weighted slope direction weight.

[0019] In some exemplary embodiments of the present application, after the adjacency structure graph and the density field posterior probability distribution of the mountain are constructed, the method further comprises:

[0020] A density field prior probability distribution of the mountain is constructed, wherein the density field prior probability distribution is an initial probability distribution of all the density models constructed according to the adjacency structure graph;

[0021] According to the muon transmission rate and the muon statistical law, a likelihood function is constructed, wherein the likelihood function represents the probability of observing the muon transmission rate under all the density models;

[0022] According to the density field prior probability distribution and the likelihood function, the density field posterior probability distribution is constructed, wherein the density field posterior probability distribution is proportional to the product of the density field prior probability distribution and the likelihood function.

[0023] In some exemplary embodiments of the present application, the sampling from the density field posterior probability distribution according to the adjacency structure graph comprises:

[0024] for each of the nodes, repeating the following operations until converging to the density field posterior probability distribution or reaching a maximum sampling round number:

[0025] sampling from all the density models of the density field posterior probability distribution and keeping the densities of the rest of the nodes in the density model as a sample except the current node unchanged;

[0026] in response to a weight of a target edge connected with the current node being less than a preset weight, deleting the target edge, and / or, in response to an absolute difference value of the density of a target node in a target density model and the current node being greater than a preset density difference value, deleting an edge connecting the target node and the current node, the target node being a node adjacent to the current node, the target density model being the density model as a sample.

[0027] In some exemplary embodiments of the present application, the obtaining of the density value of each voxel according to a plurality of the density field samples comprises:

[0028] averaging the densities of the node corresponding to each voxel in a plurality of the density field samples as the density value of the corresponding voxel; and / or,

[0029] the obtaining of the stability abnormal region in the mountain according to the density values of a plurality of the voxels comprises:

[0030] collecting all the voxels with a density value less than a preset density value as a voxel set;

[0031] regarding a region formed by connecting a plurality of the voxels in the voxel set as the stability abnormal region.

[0032] According to a second aspect of the embodiments of the present application, an evaluation device for evaluating the stability of a mountain attached to a grotto is provided, and the evaluation device comprises:

[0033] a division module configured to divide the mountain into a plurality of voxels;

[0034] a construction module configured to construct an adjacency structure graph of the mountain and a density field posterior probability distribution, wherein the adjacency structure graph is constructed by muon data recorded by muon detectors arranged in the grotto, and the density field posterior probability distribution is a probability distribution of all density models of the mountain under the muon data and the adjacency structure graph;

[0035] a sampling module configured to sample from the density field posterior probability distribution according to the adjacency structure graph to obtain a plurality of density field samples of the mountain;

[0036] A first determining module is configured to obtain a density value of each voxel according to a plurality of density field samples;

[0037] A second determining module is configured to obtain a stability abnormal region in the mountain according to the density values of the plurality of voxels.

[0038] According to a third aspect of the embodiments of the present application, a computer device is provided, comprising a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method according to the first aspect when executing the computer program.

[0039] According to a fourth aspect of the embodiments of the present application, a non-transitory computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method according to the first aspect.

[0040] The technical scheme provided by the embodiments of the present application can have the following beneficial effects: the mountain stability evaluation method provided by the present application combines the density field posterior probability distribution of the mountain constructed based on the adjacency structure diagram of the mountain, and samples under the constraint of the adjacency structure diagram of the mountain, so that the density inversion result of the mountain has structural consistency and physical rationality, the stability abnormal region in the mountain can be accurately determined, and the accuracy of the mountain stability evaluation is improved.

[0041] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0042] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present application and, together with the specification, serve to explain the principles of the present application.

[0043] Figure 1 is a flowchart of a mountain stability evaluation method according to the first exemplary embodiment of the present application;

[0044] Figure 2 is a flowchart of a mountain stability evaluation method according to the second exemplary embodiment of the present application;

[0045] Figure 3 is a flowchart of a mountain stability evaluation method according to the third exemplary embodiment of the present application;

[0046] Figure 4 is a flowchart of a mountain stability evaluation method according to the fourth exemplary embodiment of the present application;

[0047] Figure 5is a flow chart of a mountain stability evaluation method according to a fifth exemplary embodiment of the present application;

[0048] Figure 6 is a flow chart of a mountain stability evaluation method according to a sixth exemplary embodiment of the present application;

[0049] Figure 7 is a flow chart of a mountain stability evaluation method according to a seventh exemplary embodiment of the present application;

[0050] Figure 8 is a block diagram of a mountain stability evaluation device according to an exemplary embodiment of the present application;

[0051] Figure 9 is a block diagram of a computer device according to an exemplary embodiment of the present application.

[0052] In the drawings:

[0053] 81 - division module; 82 - construction module; 83 - sampling module; 84 - first determination module; 85 - second determination module; 900 - computer device; 901 - calculation unit; 902 - read-only memory (ROM); 903 - random access memory (RAM); 904 - bus; 905 - input / output (I / O); 906 - input unit; 907 - output unit; 908 - storage unit; 909 - communication unit. DETAILED DESCRIPTION

[0054] The exemplary embodiments will be described in detail herein with reference to the attached drawings. In the following description, the same numbers are used to designate the same elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not meant to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the present application as detailed in the appended claims.

[0055] Cave-like cultural relics are attached to mountains, and the stability of the mountains affects the safety of the cave-like cultural relics. Currently, methods such as geological radar detection, seismic exploration, drilling, and numerical simulation are mainly used to evaluate the stability of the mountains to which the caves are attached. However, these methods have problems such as limited detection depth, insufficient resolution, only partial detection, destructiveness, dependence on ideal parameters and prior assumptions, difficulty in reflecting the true structure, and difficulty in accurately evaluating the stability of the mountains to which the caves are attached.

[0056] To solve the above technical problems, the present application provides a mountain stability evaluation method for evaluating the stability of a mountain attached to a grotto. The mountain is divided into a plurality of voxels. An adjacency graph of the mountain is constructed from muon data recorded by muon detectors arranged in the grotto, and a density field posterior probability distribution is constructed. The density field posterior probability distribution is the probability distribution of all density models of the mountain under the muon data and the adjacency graph. A plurality of density field samples of the mountain are obtained by sampling from the density field posterior probability distribution according to the adjacency graph. The density values of each voxel are obtained according to the plurality of density field samples. The stability anomaly area in the mountain is obtained according to the density values of the plurality of voxels. The density inversion result of the mountain obtained by constructing the density field posterior probability distribution of the mountain under the constraint of the adjacency graph of the mountain and sampling under the constraint of the adjacency graph of the mountain has structural consistency and physical rationality, which can accurately determine the stability anomaly area in the mountain and improve the accuracy of the mountain stability evaluation.

[0057] The exemplary embodiments of the present application provide a mountain stability evaluation method for evaluating the stability of a mountain attached to a grotto. The grotto is a cave relying on a natural mountain or artificially excavated. The grotto cultural relics include not only the grotto itself, but also historical remains and cultural information in the grotto, such as murals, stone carvings, etc. The stability of the mountain refers to the ability of the mountain to maintain its original shape and balance state after resisting sliding, collapse, toppling and other forms of damage.

[0058] As shown in Figure 1 The mountain stability evaluation method according to the exemplary embodiments of the present application includes:

[0059] S101, dividing the mountain into a plurality of voxels.

[0060] The mountain is divided into a plurality of cubic regions, and each cubic region is a voxel.

[0061] In some examples, the mountain can be divided into a plurality of voxels according to the point cloud data of the mountain obtained at a historical time. The point cloud data of the mountain is obtained by using three-dimensional laser scanning technology to measure the surface of the mountain at high speed and high precision. The three-dimensional coordinate data of each point on the surface of the mountain is obtained. According to the point cloud data of the mountain, a digital surface model of the mountain is generated, and the region enclosed by the digital surface model of the mountain is the internal region of the mountain. The voxel resolution is set, and the range of the voxel resolution can be 0.5m x 0.5m x 0.5m to 2m x 2m x 2m, for example, the voxel resolution is 1m x 1m x 1m. The digital surface model of the mountain and the internal region of the mountain enclosed by the digital surface model of the mountain are divided into a plurality of cubic regions according to the voxel resolution, and each cubic region is a voxel of the mountain.

[0062] S102, construct an adjacency structure graph of the mountain and a density field posterior probability distribution.

[0063] The adjacency structure graph is constructed by muon data recorded by muon detectors arranged in the cave. The adjacency structure graph of the mountain represents the spatial connection relationship between a plurality of voxels of the mountain. The adjacency structure graph of the mountain is composed of a plurality of nodes and a plurality of edges, each node representing a voxel, and each edge representing a connection relationship between two nodes.

[0064] The density field posterior probability distribution is the probability distribution of all density models of the mountain under the muon data and the adjacency structure graph.

[0065] In an exemplary embodiment of the present application, as shown in Figure 2 The step S102 of constructing the adjacency structure graph of the mountain and the density field posterior probability distribution includes:

[0066] S102-1, record the muon data within a preset time length using the muon detector.

[0067] In some examples, the number of muon detectors arranged in the cave can be one or more. The position and pose of the muon detector are aligned with the coordinate system of the point cloud data, so that the muon data and the point cloud data are in the same three-dimensional space reference system. The preset time length can be 1 month to 12 months, for example, the preset time length is 6 months. One or more muon detectors arranged in the cave record the muon flux after the muon passes through the mountain attached to the cave within the preset time length as the muon data.

[0068] S102-2, determine the muon transmissivity according to the muon data.

[0069] In some examples, the muon transmissivity is determined according to the muon flux after the muon passes through the mountain attached to the cave recorded by the muon detector within the preset time length, and the muon flux of the region where the mountain attached to the cave is located under the open condition and within the preset time length obtained historically. The muon transmissivity is the ratio of the muon flux after the muon passes through the mountain attached to the cave recorded by the muon detector within the preset time length to the muon flux of the region where the mountain attached to the cave is located under the open condition and within the preset time length obtained at a historical time. The muon transmissivity can include a plurality of sub-muon transmissivities in a plurality of observation directions, and the observation direction is defined by the zenith angle and the azimuth angle, and the zenith angle and / or the azimuth angle of different observation directions are different.

[0070] S102-3, derive the adjacency structure graph according to the point cloud data of the mountain and the muon transmissivity.

[0071] In an exemplary embodiment of the present application, as shown in Figure 3As shown, the step S102-3 obtains the adjacency graph from the point cloud data of the mountain and the muon transmission rate, including:

[0072] S102-3-1, obtaining the voxel graph model of the mountain from the point cloud data and the muon transmission rate.

[0073] In some examples, each muon passes through multiple voxels, and the transmission rate of the muon is constructed using a variant of the Beer-Lambert law:

[0074] ;

[0075] wherein, is the muon transmission rate, is the number of any voxel passed by the muon, is the density of the voxel numbered , is the path length of the muon passing through the voxel numbered , .

[0076] Using the above formula reversely, taking the natural logarithm of the muon transmission rate obtained in step S102-2, and multiplying by -1, the , is the muon coarse imaging data of the mountain, and the muon coarse imaging data includes the average density thickness product of each muon path.

[0077] The point cloud data of the mountain and the muon coarse imaging data are input into a graph neural network, and the graph neural network is used to identify the geological structure inside the mountain, and the voxel graph model of the mountain is output. The graph neural network is a deep learning model for processing graph structure data. The voxel graph model of the mountain is a model that divides the mountain into multiple voxels and assigns one or more physical attribute values to each voxel, such as density, lithology category. In the exemplary embodiment, the voxel graph model includes an initial density value for each voxel.

[0078] S102-3-2, each voxel in the voxel graph model is taken as a node in the adjacency graph.

[0079] In the exemplary embodiment, the properties of the node include the initial density value.

[0080] Each voxel in the voxel graph model generated in step S102-3-1 is taken as a node in the adjacency graph, and the initial density value of each voxel in the voxel graph model is taken as the property of the corresponding node in the adjacency graph.

[0081] S102-3-3, the connection relationship between each two adjacent voxels in the voxel graph model is taken as an edge connecting the corresponding two nodes in the adjacency structure graph, and the corresponding two nodes refer to the nodes corresponding to the two adjacent voxels.

[0082] In the example embodiment, each edge corresponds to a weight, and the weight is used to reflect the structural similarity degree of the two nodes connected by the edge. The greater the structural similarity degree, the greater the weight.

[0083] In some examples, the connection relationship between each two adjacent voxels in the voxel graph model generated in step S102-3-1 is taken as an edge connecting the corresponding two nodes in the adjacency structure graph, and each edge is assigned a weight.

[0084] In the example embodiment of the present application, the weight is the product of the main direction similarity weight of the geological structure, the joint density weight, and the slope direction weight, or the weight is the product of the main direction similarity weight of the geological structure, the joint density weight, and the slope direction weight after weighting.

[0085] The main direction similarity weight of the geological structure is used to reflect whether the dominant direction of the internal structure surface of the two adjacent nodes is consistent, and the greater the consistency, the greater the main direction similarity weight of the geological structure.

[0086] The joint density weight is used to reflect the difference in the fragmentation degree of the rock mass in the two adjacent nodes, and the smaller the difference, the greater the joint density weight.

[0087] The slope direction weight is used to reflect whether the rock layer tendency of the geological structure of the two adjacent nodes is consistent with the slope direction, and the greater the consistency, the greater the slope direction weight.

[0088] In some examples, the point cloud data and the MuCI coarse imaging data of the mountain are input into the graph neural network, and the graph neural network is used to identify the geological structure inside the mountain. While outputting the voxel graph model of the mountain, the spatial and attribute correlation between voxels can also be obtained. According to the voxel graph model and the spatial and attribute correlation between voxels, the graph convolution network is used to automatically identify the structural weak surface of the mountain, i.e., the adjacent voxels with low consistency in the dominant direction of the internal structure surface, the adjacent voxels with large difference in the fragmentation degree of the rock mass, and the adjacent voxels with low consistency between the rock layer tendency of the geological structure and the slope direction, so as to calculate the main direction similarity weight of the geological structure, the joint density weight, and the slope direction weight of the edge between the adjacent nodes corresponding to the adjacent voxels.

[0089] In some examples, the product of the main direction similarity weight of the geological structure, the joint density weight, and the slope direction weight of the edge between the two adjacent nodes is taken as the weight of the edge.

[0090] In some examples, the weight of the edge between two adjacent nodes is the product of the weighted geological structure main direction similarity weight, the weighted joint density weight and the weighted slope direction weight of the edge. For example, the weighting coefficient of the geological structure main direction similarity weight is 0.5, the weighting coefficient of the joint density weight is 0.3, and the weighting coefficient of the slope direction weight is 0.2.

[0091] The weight of the edge is not a fixed constant, but is assigned according to the geological structure main direction similarity, the joint density difference and the slope direction consistency of the adjacent nodes, which can reflect the structural similarity degree of the two nodes connected by the edge, ensure that the density inversion of the mountain is smoothly constrained in the structural consistent area, and retain the necessary density mutation ability at the structural weak plane, thereby improving the accuracy and geological rationality of the density inversion of the mountain.

[0092] In the exemplary embodiments of the present application, the density field prior probability distribution and the likelihood function of the mountain are modeled by the Bayesian probability theory system to obtain the density field posterior probability distribution of the mountain. The Bayesian probability theory system takes the Bayesian theorem as the core and is used to update the probability judgment through prior knowledge and new evidence in uncertainty. As shown in Figure 4 The step S102 of constructing the adjacency structure graph and the density field posterior probability distribution of the mountain further includes:

[0093] S102-4, constructing the density field prior probability distribution of the mountain.

[0094] In the exemplary embodiments, the density field prior probability distribution is the initial probability distribution of all density models constructed according to the adjacency structure graph.

[0095] In some examples, the density field prior probability distribution of the mountain is constructed by the following formula.

[0096] ;

[0097] wherein, is the density field prior probability distribution, is all density models, is the adjacency structure graph of the mountain; is a regularization factor for controlling the strength of the density field smoothness, the greater the numerical value, the stronger the constraint of the density field prior probability distribution on the density continuity, which can be set as a fixed value, or can be adaptively learned by the Bayesian probability theory system to adapt to the density inversion requirements under different geological structures; and are the numbers of the two adjacent nodes in the adjacency structure graph; is all nodes in the adjacency structure graph; is the number of the node adjacent to the node with the number in the adjacency structure graph. The weight of the edges of the node; It is numbered in the adjacency structure diagram. The initial density value of the nodes; It is numbered in the adjacency structure diagram. The initial density value of the nodes.

[0098] The prior probability distribution of the density field constructed based on the adjacency structure graph has directionality and structural control capabilities compared to the Gaussian smooth density field prior probability distribution, and can highlight the influence of weak surfaces of the structure on density perturbations.

[0099] S102-5. Construct the likelihood function based on muon transmittance and muon statistical laws.

[0100] In this exemplary embodiment, the likelihood function represents the probability of observing muon transmittance under all density models.

[0101] In some examples, the likelihood function is... The statistical regularity of muons can be represented by a Poisson distribution. A Poisson distribution is the probability distribution of the number of occurrences of a random event within a fixed time interval or a fixed spatial region. The process of muons reaching a muon detector fits the application scenario of the Poisson distribution. Therefore, a likelihood function is constructed based on muon transmittance and the Poisson distribution, representing the probability of observing muon transmittance under all density models.

[0102] S102-6. Construct the posterior probability distribution of the density field based on the prior probability distribution and likelihood function of the density field.

[0103] In this exemplary embodiment, the posterior probability distribution of the density field is proportional to the product of the prior probability distribution of the density field and the likelihood function.

[0104] In some examples, the posterior probability distribution of the density field of a mountain is constructed using the following formula.

[0105] ;

[0106] in, It is the posterior probability distribution of the density field, which is the probability distribution of all density models of the mountain under the conditions of muon transmittance and adjacency structure diagram.

[0107] S103. Based on the adjacency structure diagram, sample from the posterior probability distribution of the density field to obtain multiple density field samples of the mountain.

[0108] In some examples, the Graph Structure Markov Chain Monte Carlo (MCMC) algorithm is used to sample from the posterior probability distribution of the density field, obtaining multiple density field samples of the mountain. The Markov Chain Monte Carlo algorithm is a statistical algorithm that estimates model parameters by sampling from a probability distribution. It constructs a Markov chain conforming to a specific probability distribution for sampling, gradually approximating the stable state of the target distribution using a stochastic process. For example, after modeling using Bayesian probability theory, samples are generated from a complex posterior probability distribution of the density field. The Graph Structure Markov Chain Monte Carlo algorithm, building upon the standard Markov Chain Monte Carlo algorithm, guides sampling based on the adjacency structure graph. It allows for larger movements in areas with higher edge weights and smaller movements in areas with lower edge weights, thus exploring high-probability regions more efficiently and accurately.

[0109] In an exemplary embodiment of the present invention, such as Figure 5 As shown, in step S103, based on the adjacency structure graph, multiple density field samples of the mountain are obtained by sampling from the posterior probability distribution of the density field, including:

[0110] For each node, repeat the following operation until convergence to the posterior probability distribution of the density field or the maximum number of sampling rounds is reached:

[0111] S103-1. Sample from all density models of the posterior probability distribution of the density field, and keep the density of all nodes in the sampled density models unchanged except for the current node.

[0112] In some examples, the Gibbs sampling algorithm can be used to sample from all density models of the posterior probability distribution of the density field, while keeping the densities of all nodes in the sampled density models unchanged except for the current node. The Gibbs sampling algorithm is a specific algorithm of the Markov chain Monte Carlo algorithm, which samples only one variable at a time, fixing the current values ​​of other variables.

[0113] The formula for sampling from all density models of the posterior probability distribution of the density field using the Gibbs sampling algorithm is as follows:

[0114] ;

[0115] in, In the adjacency structure graph, except for the one numbered... The density of nodes other than the node, It is the probability distribution that keeps the density of all nodes in the density model used as samples constant, except for the current node. Is with number The set of adjacent nodes of a given node.

[0116] S103-2, in response to the weight of the target edge connected with the current node being less than a preset weight, deleting the target edge, and / or, in response to the absolute difference of the densities of the current node and the target node in the target density model being greater than a preset density difference, deleting the edge connecting the target node and the current node.

[0117] The target node is a node adjacent to the current node, and the target density model is a density model as a sample.

[0118] The preset weight and the preset density difference can be set according to the actual situation of the mountain. In some examples, the preset weight can be 0.5. In other examples, the preset density difference can be 0.8 g / cm³.

[0119] The weight of the target edge connected with the current node is less than the preset weight, or the absolute difference of the densities of the current node and the target node in the target density model is greater than the preset density difference, which reflects that the structural difference between the target node and the current node is large. Deleting the edge between the target node and the current node can delete the target node from the set of nodes adjacent to the current node, so that the target node is not subjected to smoothing constraint, thereby improving the response capability to structural mutations.

[0120] The maximum sampling round number can be set according to the actual situation. In some examples, the maximum sampling round number can be 100,000 times.

[0121] S104, obtaining the density value of each voxel according to the plurality of density field samples.

[0122] In an exemplary embodiment of the present application, the average value of the densities of the node corresponding to each voxel in the plurality of density field samples is taken as the density value of the corresponding voxel.

[0123] In some examples, in addition to obtaining the density value of each voxel, the density confidence interval of each voxel can also be obtained. The densities of the node corresponding to each voxel in the plurality of density field samples are sorted from small to large, and the densities located between the first preset percentage and the second preset percentage are taken as the density confidence interval of the voxel. The first preset percentage is less than the second preset percentage, the first preset percentage can be 20%, and the second preset percentage can be 80%.

[0124] S105, obtaining the stability abnormal area in the mountain according to the density values of the plurality of voxels.

[0125] In an exemplary embodiment of the present application, as shown in Figure 6 obtaining the stability abnormal area in the mountain according to the density values of the plurality of voxels in step S105 includes:

[0126] S105-1, all voxels with a density value less than a preset density value are taken as a voxel set.

[0127] The preset density value can be specifically set according to the rock mass density of the mountain. In some examples, the rock mass density of the mountain can be 2.3 g / cm3 to 2.6 g / cm3, and the preset density value can be 1.5 g / cm3.

[0128] S105-2, a region formed by multiple voxels in the voxel set being connected is taken as a stability abnormal region.

[0129] The density value of a single voxel is less than the preset density value, and the size of the low-density region is small, so it will not affect the stability of the mountain.

[0130] The size of the region formed by multiple voxels with a density value less than the preset density value being connected is large, which can be a structure such as a crack, which can cause collapse, landslide and other disaster conditions that endanger the grotto cultural relics, so the region formed by multiple voxels in the voxel set being connected is taken as a stability abnormal region.

[0131] The exemplary embodiments of the present application provide a mountain stability evaluation method for evaluating the stability of a mountain attached to a grotto.

[0132] As shown in Figure 7 The mountain stability evaluation method shown in the exemplary embodiments includes:

[0133] S701, dividing the mountain into multiple voxels.

[0134] S702, using a muon detector to record muon data within a preset time length.

[0135] S703, determining muon transmission according to the muon data.

[0136] S704, obtaining a voxel graph model of the mountain according to the point cloud data and the muon transmission.

[0137] S705, taking each voxel in the voxel graph model as a node in an adjacency structure graph.

[0138] S706, taking the connection relationship between each two adjacent voxels in the voxel graph model as an edge connecting the corresponding two nodes in the adjacency structure graph, and the corresponding two nodes refer to the nodes corresponding to the two adjacent voxels.

[0139] S707, constructing a density field prior probability distribution of the mountain.

[0140] S708, constructing a likelihood function according to the muon transmission and the muon statistical law.

[0141] S709, constructing a density field posterior probability distribution according to the density field prior probability distribution and the likelihood function.

[0142] S710, for each node, repeating the following operations until converging to the density field posterior probability distribution or reaching a maximum sampling round number: sampling from all density models of the density field posterior probability distribution and keeping the densities of the remaining nodes except the current node in the density model as samples unchanged; in response to the weight of the target edge connected with the current node being less than a preset weight, deleting the target edge.

[0143] S711, averaging the densities of the node corresponding to each voxel in the multiple density field samples as the density value of the corresponding voxel.

[0144] S712, taking all voxels with a density value less than a preset density value as a voxel set.

[0145] S713, taking a region formed by multiple voxels in the voxel set being connected as a stability anomaly region.

[0146] In the exemplary embodiment, the density field posterior probability distribution of the mountain is constructed in combination with the adjacency structure diagram of the mountain, and sampling is performed under the constraint of the adjacency structure diagram of the mountain, so that the density inversion result of the mountain has structural consistency and physical rationality. The region formed by multiple voxels connected with a density value less than a preset density value is taken as a stability anomaly region, so that the stability anomaly region in the mountain can be accurately determined, and the accuracy of the stability evaluation of the mountain is improved.

[0147] The exemplary embodiment of the present application provides a mountain stability evaluation device for evaluating the stability of a mountain attached to a grotto. As shown in the figure, Figure 8 The exemplary embodiment of the present application provides a mountain stability evaluation device for evaluating the stability of a mountain attached to a grotto. As shown in the figure,

[0148] The division module 81 is configured to divide the mountain into multiple voxels.

[0149] The construction module 82 is configured to construct an adjacency structure diagram of the mountain and a density field posterior probability distribution, wherein the adjacency structure diagram is constructed by muon data recorded by a muon detector arranged in the grotto, and the density field posterior probability distribution is a probability distribution of all density models of the mountain under the muon data and the adjacency structure diagram.

[0150] The sampling module 83 is configured to sample from the density field posterior probability distribution according to the adjacency structure diagram, to obtain multiple density field samples of the mountain.

[0151] The first determination module 84 is configured to obtain a density value of each voxel according to the multiple density field samples.

[0152] The second determination module 85 is configured to obtain a stability abnormal area in the mountain according to the density values of the plurality of voxels.

[0153] In the example embodiment, the density field posterior probability distribution of the mountain is constructed in combination with the adjacency structure diagram of the mountain, and sampling is performed under the constraint of the adjacency structure diagram of the mountain, so that the density inversion result of the mountain has structural consistency and physical rationality, the stability abnormal area in the mountain can be accurately determined, and the accuracy of the mountain stability evaluation is improved.

[0154] In the example embodiment, the construction module 82 is further configured to:

[0155] record muon data within a preset time length using a muon detector;

[0156] determine muon transmission according to the muon data;

[0157] obtain an adjacency structure diagram according to the point cloud data of the mountain and the muon transmission.

[0158] In the example embodiment, the construction module 82 is further configured to:

[0159] obtain a voxel graph model of the mountain according to the point cloud data and the muon transmission, wherein the voxel graph model includes an initial density value of each voxel;

[0160] take each voxel in the voxel graph model as a node in the adjacency structure diagram, wherein the attribute of the node includes the initial density value;

[0161] take the connection relationship between each two adjacent voxels in the voxel graph model as an edge connecting the corresponding two nodes in the adjacency structure diagram, wherein the corresponding two nodes refer to the nodes corresponding to the two adjacent voxels, and each edge corresponds to a weight, and the weight is used to reflect the structural similarity degree of the two nodes connected by the edge, and the greater the structural similarity degree, the greater the weight.

[0162] In the example embodiment, the weight is the product of the geological structure main direction similarity weight, the joint density weight and the slope direction weight, or the weight is the product of the geological structure main direction similarity weight, the joint density weight and the slope direction weight after weighting.

[0163] In the example embodiment, the construction module 82 is further configured to:

[0164] construct a density field prior probability distribution of the mountain, wherein the density field prior probability distribution is an initial probability distribution of all density models constructed according to the adjacency structure diagram;

[0165] According to the muon transmissivity and the muon statistical law, a likelihood function is constructed, wherein the likelihood function represents a probability of observing the muon transmissivity under all density models;

[0166] According to the density field prior probability distribution and the likelihood function, a density field posterior probability distribution is constructed, wherein the density field posterior probability distribution is proportional to a product of the density field prior probability distribution and the likelihood function.

[0167] In an exemplary embodiment of the present application, the sampling module 83 is further configured to:

[0168] For each node, the following operation is repeated until the density field posterior probability distribution is converged or a maximum sampling round number is reached:

[0169] A density model is sampled from all density models of the density field posterior probability distribution, and densities of the remaining nodes except the current node in the density model as the sample are kept unchanged;

[0170] In response to a weight of a target edge connected with the current node being less than a preset weight, the target edge is deleted, and / or, in response to an absolute difference value of the density of the current node and a target node in a target density model being greater than a preset density difference value, an edge connecting the target node and the current node is deleted, the target node being a node adjacent to the current node, and the target density model being the density model as the sample.

[0171] In an exemplary embodiment of the present application, the first determining module 84 is further configured to:

[0172] An average value of the densities of the node corresponding to each voxel in the plurality of density field samples is taken as a density value of the corresponding voxel.

[0173] In an exemplary embodiment of the present application, the second determining module 85 is further configured to:

[0174] All voxels with a density value less than a preset density value are taken as a voxel set;

[0175] A region formed by connecting a plurality of voxels in the voxel set is taken as a stability anomaly region.

[0176] The above-mentioned modules in the mountain stability evaluation device can be realized by software, hardware, and combinations thereof, in whole or in part. The above-mentioned modules can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in a computer device in software form, so as to be called and executed by a processor to perform the operations corresponding to the above-mentioned modules.

[0177] An exemplary embodiment of the present application provides a computer device including a processor and a memory, the memory storing a computer program, and the processor implementing the steps of any of the above-mentioned mountain stability evaluation methods when executing the computer program.

[0178] The exemplary embodiments of the present application provide a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of any of the above mountain stability evaluation methods.

[0179] The exemplary embodiments of the present application provide a computer program product comprising a computer program which, when executed by a processor, implements the steps of any of the above mountain stability evaluation methods.

[0180] Reference Figure 9 A block diagram of the structure of a computer device that can be used as the computer device of the present application will now be described. The computer device 900 includes a computing unit 901 which can perform various appropriate actions and processes in accordance with a computer program stored in a read only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903. Various programs and data required for the operation of the computer device 900 can also be stored in the RAM 903. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other through a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0181] A plurality of components in the computer device 900 are connected to the I / O interface 905, including an input unit 906, an output unit 907, the storage unit 908, and a communication unit 909. The input unit 906 can be any type of device that can input information to the computer device 900, and can receive inputted numerical or character information, and generate key signal inputs related to user settings and / or function controls of the computer device 900, and can include, but is not limited to, a mouse, a keyboard, a touch screen, a track pad, a track ball, a joystick, a microphone, and / or a remote controller. The output unit 907 can be any type of device that can present information, and can include, but is not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 908 can include, but is not limited to, a magnetic disk, an optical disk. The communication unit 909 allows the computer device 900 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and can include, but is not limited to, a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0182] The computing unit 901 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 901 performs various methods and processes described above, such as the mountain stability evaluation method. For example, in some embodiments, the mountain stability evaluation method can be implemented as a computer software program that is tangibly embodied in a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed onto the computer device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded onto the RAM 903 and executed by the computing unit 901, one or more steps of the mountain stability evaluation method described above can be performed. Alternatively, in other embodiments, the computing unit 901 can be configured to perform the mountain stability evaluation method by other any appropriate means, such as by means of firmware.

[0183] The computer device 900 can be implemented with one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic components, for performing the mountain stability evaluation method described above.

[0184] Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the present application cover any and all variations of the application that come within the scope of the present application. It is intended that the specification and examples be considered exemplary only, with the true scope and spirit of the application being indicated by the following claims.

[0185] It is to be understood that the application is not limited to the precise construction described above and shown in the attached drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the present application. The scope of the application is to be determined only by the appended claims.

Claims

1. A method of evaluating the stability of a mountain, characterized by, The evaluation method is used for evaluating the stability of a mountain body attached to a grotto, and the evaluation method comprises: dividing the mountain body into a plurality of voxels; constructing an adjacency structure graph of the mountain body and a density field posterior probability distribution, wherein the adjacency structure graph is constructed by muon data recorded by muon detectors arranged in the grotto, and the density field posterior probability distribution is a probability distribution of all density models of the mountain body under the muon data and the adjacency structure graph; sampling from the density field posterior probability distribution according to the adjacency structure graph to obtain a plurality of density field samples of the mountain body; obtaining a density value of each voxel according to a plurality of the density field samples; obtaining a stability abnormal area in the mountain body according to the density values of a plurality of the voxels.

2. The method of evaluating stability of a mountain body according to claim 1, characterized by, The constructing of the adjacency structure graph of the mountain body and the density field posterior probability distribution comprises: recording the muon data by the muon detectors within a preset time length; determining muon transmission rates according to the muon data; obtaining the adjacency structure graph according to point cloud data of the mountain body and the muon transmission rates.

3. The method of evaluating stability of a mountain body according to claim 2, characterized by, The obtaining of the adjacency structure graph according to the point cloud data of the mountain body and the muon transmission rates comprises: obtaining a voxel graph model of the mountain body according to the point cloud data and the muon transmission rates, wherein the voxel graph model comprises an initial density value of each voxel; taking each voxel in the voxel graph model as a node in the adjacency structure graph, wherein an attribute of the node comprises the initial density value; taking a connection relationship between each two adjacent voxels in the voxel graph model as an edge connecting corresponding two nodes in the adjacency structure graph, wherein the corresponding two nodes refer to nodes corresponding to the two adjacent voxels, and each edge corresponds to a weight, which is used to reflect a structural similarity degree of the two nodes connected by the edge, and the greater the structural similarity degree, the greater the weight.

4. The method of claim 3, wherein the method comprises: The weight is a product of geological structure main direction similarity weight, joint density weight and slope direction weight, or the weight is a product of the geological structure main direction similarity weight, the joint density weight and the slope direction weight after weighting.

5. The method of claim 2, wherein the method comprises: The constructing of the adjacency structure graph of the mountain body and the density field posterior probability distribution further comprises: constructing a density field prior probability distribution of the mountain body, wherein the density field prior probability distribution is an initial probability distribution of all the density models constructed according to the adjacency structure graph; constructing a likelihood function according to the muon transmission rates and muon statistical rules, wherein the likelihood function represents a probability of observing the muon transmission rates under all the density models; constructing the density field posterior probability distribution according to the density field prior probability distribution and the likelihood function, wherein the density field posterior probability distribution is proportional to a product of the density field prior probability distribution and the likelihood function.

6. The method of claim 3, wherein the method comprises: The sampling from the density field posterior probability distribution according to the adjacency structure graph to obtain a plurality of density field samples of the mountain body comprises: For each of the nodes, the following operations are repeated until convergence to the density field posterior probability distribution or reaching a maximum sampling round number: sampling from all the density models of the density field posterior probability distribution and keeping the densities of the rest of the nodes in the density model as a sample except the current node unchanged; in response to the weight of a target edge connected with the current node being less than a preset weight, deleting the target edge, and / or, in response to the absolute difference of the densities of a target node and the current node in a target density model being greater than a preset density difference value, deleting an edge connecting the target node and the current node, the target node being a node adjacent to the current node, and the target density model being the density model as a sample.

7. The method of claim 6, wherein the method further comprises: The method further includes: averaging the densities of the node corresponding to each voxel in the plurality of density field samples to obtain the density value of the corresponding voxel; and / or, The method further includes: collecting all the voxels with density values less than a preset density value as a voxel set; regarding a region formed by connecting the voxels in the voxel set as the stability abnormal region.

8. A device for assessing mountain stability, characterized in that, The evaluation device is used for evaluating the stability of a mountain body attached to a grotto, and includes: a division module configured to divide the mountain body into a plurality of voxels; a construction module configured to construct an adjacency structure graph of the mountain body and a density field posterior probability distribution, wherein the adjacency structure graph is constructed based on muon data recorded by muon detectors arranged in the grotto, and the density field posterior probability distribution is a probability distribution of all density models of the mountain body under the muon data and the adjacency structure graph; a sampling module configured to sample from the density field posterior probability distribution according to the adjacency structure graph to obtain a plurality of density field samples of the mountain body; a first determination module configured to obtain a density value of each voxel according to the plurality of density field samples; a second determination module configured to obtain a stability abnormal region in the mountain body according to the density values of the plurality of voxels. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to implement the steps of the evaluation method of the stability of the mountain body according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the evaluation method of the stability of the mountain body according to any one of claims 1 to 7.

Images