A structural surface network sequence modeling method and system suitable for layered rock mass

Abstract

The present invention is a structural plane network sequence modeling method and system suitable for layered rock mass, which relates to the field of rock mass engineering technology, including obtaining layered rock mass outcrop data as basic data, and obtaining basic data interpretation representation after processing; determining the rock mass outcrop homogeneous zone division result according to the type attribute of the basic data; correcting the occurrence data in each homogeneous zone according to the interpreted occurrence data combined with the rock mass outcrop homogeneous zone division result by using a weight function, grouping the structural planes by using a K-means particle swarm algorithm, and obtaining the grouped structural planes; determining the density of each group of structural plane distribution space according to the spacing, occurrence, size and survey line strike data of each group of structural planes; determining the position and size of the plane in the plane structure space according to the confirmed structural plane distribution position; and randomly combining the parameters of each group of structural planes, combining sequence modeling and Monte Carlo simulation methods, to construct a layered rock mass structural space network model.

Description

Technical Field

[0001] The invention relates to the technical field of rock mass structure, and in particular to a structural surface network sequence modeling method and system suitable for layered rock mass. Background Art

[0002] Rock masses with typical layered development characteristics, such as sedimentary rocks, are widely exposed in my country, with a distribution area of ​​up to about 77.3%. Rock mass structure controls rock mass strength and deformation failure mode. Therefore, exploring the rock mass structure development characteristics of layered rock mass is of great significance in slope stability assessment, slope structure disaster prediction and disaster prevention. However, the development of structural surfaces inside the rock mass cannot be directly observed due to its closed nature. Therefore, rock mass structure modeling technology is currently the core means to simulate the rock mass structure development characteristics inside the rock mass.

[0003] The current mainstream rock mass structure modeling method follows the basic assumption that the disc model is used as the structural surface shape unit, and uses it to derive parameters such as occurrence, size, density and position. However, the structural surface of layered rock mass often develops in a rectangular shape and is dominated by large-scale layers. Therefore, the development of structural surfaces shows a strong dominant clustering, and there is a statistical relationship between the size of the structural surface and the layer thickness. In addition, the layered rock mass structure network has a unique topological structure feature, that is, the structural surface easily intersects with the layer to form a T-type node. However, the traditional rock mass structure modeling method often produces too many X- and I-type nodes and almost no T-type nodes. In addition, the disc cannot accurately represent the shape of the real layered rock mass structural surface. Therefore, there is a large deviation between the simulation of the traditional rock mass structure modeling method and the actual situation, which can no longer meet the actual engineering needs. It is urgent to propose a structural surface spatial network sequence modeling method suitable for layered rock mass. Summary of the invention

[0004] In order to overcome the shortcomings of the traditional rock structure modeling methods in the above-mentioned prior art that the simulation deviates greatly from the actual situation and can no longer meet the actual engineering needs, the main purpose of the present invention is to provide a structural surface network sequence modeling method and system suitable for layered rock masses.

[0005] To achieve the above object, the present invention adopts the following technical scheme, a structural surface network sequence modeling method suitable for layered rock mass, comprising the following steps:

[0006] Processing the acquired layered rock outcrop data to obtain data interpretation representation; wherein the layered rock outcrop data includes layer-limited structural planes and non-layer-limited structural planes, and the structural planes are represented by rectangles;

[0007] Determine the result of rock mass outcrop homogeneous zone division according to the type attribute of the layered rock mass outcrop data;

[0008] According to the interpreted occurrence data combined with the result of the homogeneous zone division of the rock mass outcrop, the occurrence data in each homogeneous zone is corrected by using a weight function to obtain the corrected occurrence data of the homogeneous zone, and the structural surfaces are grouped by using a K-means particle swarm algorithm to obtain the grouped structural surfaces;

[0009] According to the spacing, occurrence, size and survey line trend data of each group of structural surfaces, the density of the distribution space of each group of structural surfaces is determined; according to the confirmed distribution position of the structural surfaces, the position and size of the layers in the layer structure space are determined;

[0010] The occurrence, size, density and spatial position parameters of each group of structural planes are randomly combined, and a spatial network model of layered rock structure is constructed by combining sequence modeling and Monte Carlo simulation methods.

[0011] The determination of the results of the rock mass outcrop homogeneous zone division includes:

[0012] Determine homogeneous partition results based on geological domain and geotechnical domain according to geological attribute data and geotechnical property data;

[0013] The results of homogeneous zoning of rock mass outcrop structural domains based on occurrence were quantitatively determined using the improved Miller contingency table method;

[0014] Based on the zoning results of geological domain, geotechnical domain and structural domain, combined with the homogeneous zone division rules, the homogeneous zone division results of rock mass outcrop are determined

[0015] The layer-limited structural surface is defined as a structural surface whose development boundary at least one side ends at a layer, and the non-layer-limited structural surface is defined as a structural surface whose development boundary does not end at a layer;

[0016] The construction of the layered rock mass structure spatial network model comprises:

[0017] The occurrence, size, density and spatial position parameters of each group of structural surfaces are randomly combined to construct a spatial network of layered rock mass confined structural surfaces in groups. Then, the Monte Carlo simulation method is used to randomly combine the occurrence, size, density and spatial position parameters of each group of structural surfaces to construct a spatial network model of non-layered rock mass unconfined structural surfaces.

[0018] The confirmed structural plane distribution positions, including the positions of layer-limited structural planes and the positions of non-layer-limited structural planes in the structural space;

[0019] The position of the layer-limited structural surface is determined by the position of the center point of the long side of the rectangle on each layer;

[0020] The position of the non-layer-limited structural surface in the structural space is determined by Poisson point sampling in the structural subspace naturally divided at each layer based on an optimized Latin hypercube sampling method;

[0021] The probability distribution and termination condition of each Poisson point in the structural subspace are as follows:

[0022]

[0023] Among them, V is the three-dimensional structure space, k is the number of Poisson points, λ(x, y, z) is the intensity function at a certain point, and U is the subspace area.

[0024] The weight function is adjusted by using the occurrence observation frequency, which is to obtain the adjusted weight coefficient by multiplying the occurrence observation frequency by the weight. The formula is as follows:

[0025]

[0026] Among them, W i is the weight coefficient of the i-th structural surface, w and h are the width and height of the measurement window respectively, d i is the equivalent circle diameter of the i-th rectangular structure surface, α i and is the inclination and dip of the height of the measuring window, α r To measure the tendency of window width.

[0027] The structural plane network sequence modeling method applicable to layered rock mass also includes using von Mises distribution to represent the distribution of the major axis direction of each group of rectangular structural planes, and its probability density function formula is expressed as follows:

[0028]

[0029] Where x is the major axis rotation angle, δ is the mean rotation angle, K is the rotation angle dispersion, I 0 is the zero-order modified Bessel formula.

[0030] The density of each group of structural surface distribution space is determined by the following formula:

[0031]

[0032] Among them, P 30 is the number of structural surfaces per unit space, k is the ratio of the length to the short side of the rectangular structural surface, P 10 is the reciprocal of the average spacing between structural surfaces in the same group, a is the long side of the rectangular structural surface, m is the average unit normal vector of the structural surfaces in the same group, and n is the unit vector of the survey line strike.

[0033] A structural plane network sequence modeling system suitable for layered rock mass comprises the following steps:

[0034] The rock mass structural surface data processing module is used to interpret and characterize the outcrop data of the layer-limited structural surface and the non-layer-limited structural surface of the layered rock mass to obtain the interpreted occurrence data, wherein the structural surface is represented by a rectangle; according to the type attribute of the layered rock mass outcrop data, the homogeneous zone division result of the rock mass outcrop is determined; according to the interpreted occurrence data combined with the homogeneous zone division result of the rock mass outcrop, the weight function is adjusted by using the occurrence observation frequency to correct the occurrence data in each homogeneous zone to obtain the corrected occurrence data of the homogeneous zone; based on the corrected occurrence data of the homogeneous zone, the structural surfaces are grouped by using the K-means particle swarm algorithm to obtain the grouped structural surfaces; according to the spacing, occurrence, size and survey line strike data of each group of structural surfaces, the density of the distribution space of each group of structural surfaces is determined; the position of the simulated layer in space of the actual layer thickness data distribution is obtained to determine the spatial position parameters of each group of structural surfaces;

[0035] The model building module is used to randomly combine the occurrence, size, density and spatial position parameters of each group of structural surfaces, and use the Monte Carlo simulation method to construct a spatial network model of layered rock mass structure.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention uses a rectangular model as the basic morphological unit of the layered rock mass structural surface, completes the derivation of model parameters based on field outcrop interpretation data, and further uses sequence modeling technology and Monte Carlo simulation to establish a spatial network model of the layered rock mass structure. The rectangular model proposed by the present invention can more accurately characterize the morphology of the layered rock mass structural surface. The sequence modeling technology ensures the spatial distribution characteristics of the structural surface under the main control of the layer and forms a higher proportion of T-shaped topological nodes in the model, realizing high-fidelity simulation of the distribution and intersection of the layered rock mass structural surface. The present invention forms a complete working framework suitable for spatial network modeling of layered rock mass structures, optimizes the drawbacks of modeling in traditional modeling methods, and is of great significance to the stability analysis of layered rock mass slopes, connectivity assessment, and structural disaster prediction and prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings described herein are used to provide further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0038] Figure 1 It is a structural plane space network sequence modeling method process applicable to layered rock mass in the present invention;

[0039] Figure 2 It is a two-dimensional schematic diagram of the definition of the layer, layer-limited and non-layer-limited structural surface in the present invention;

[0040] Figure 3 It is a schematic diagram of the simulation of the occurrence distribution of the structural surface in the present invention;

[0041] Figure 4 It is a schematic diagram representing the distribution of outcrop trace length in the present invention;

[0042] Figure 5 It is a schematic diagram of the simulation of the spatial position points of the layer, layer-limited and non-layer-limited structural surfaces in the present invention;

[0043] Figure 6 This is a schematic diagram of the spatial network modeling results of the layered rock mass structure in the present invention.

[0044] Figure 7 It is a schematic diagram of the process structure in the present invention. DETAILED DESCRIPTION

[0045] The current mainstream rock mass structure modeling method follows the basic assumption that the disc model is used as the structural surface shape unit, and uses it to derive parameters such as occurrence, size, density and position. However, the structural surface of layered rock mass often develops in a rectangular shape and is dominated by large-scale layers. Therefore, the development of structural surfaces shows a strong dominant clustering, and there is a statistical relationship between the size of the structural surface and the layer thickness. In addition, the layered rock mass structure network has a unique topological structure feature, that is, the structural surface easily intersects with the layer to form a T-type node. However, the traditional rock mass structure modeling method often produces too many X- and I-type nodes and almost no T-type nodes. In addition, the disc cannot accurately represent the shape of the real layered rock mass structure surface. Therefore, there is a large deviation between the simulation of the traditional rock mass structure modeling method and the actual situation, and it can no longer meet the actual engineering needs. Therefore, it is urgent to propose a structural surface spatial network sequence modeling method suitable for layered rock mass.

[0046] The purpose of the present invention is to provide a structural plane spatial network sequence modeling method suitable for layered rock masses, which uses rectangular structural planes as basic shape units for modeling and derives other required parameters, thereby solving the problem in the prior art that there is a large error between the use of a disc model to simulate rock mass structure and the actual rock mass structure development.

[0047] A structural plane network sequence modeling method suitable for layered rock masses, see Figure 7 , including the following steps:

[0048] The outcrop data of the layer-limited structural plane and the non-layer-limited structural plane of the layered rock mass are interpreted and characterized to obtain the interpreted occurrence data, among which the structural plane is represented by a rectangle;

[0049] Determining the result of the rock mass outcrop homogeneous zone division according to the type attributes of the layered rock mass outcrop data;

[0050] According to the interpreted occurrence data combined with the homogeneous zone division result of the rock mass outcrop, the occurrence observation frequency is used to adjust the weight function, the occurrence data in each homogeneous zone is corrected to obtain the corrected occurrence data of the homogeneous zone, and based on the corrected occurrence data of the homogeneous zone, the structural planes are grouped using the K-means particle swarm algorithm to obtain the grouped structural planes;

[0051] According to the spacing, occurrence, size and survey line trend data of each group of structural surfaces, the density of the distribution space of each group of structural surfaces is determined; the position of the simulated layer in space by the actual layer thickness data distribution is obtained to determine the spatial position parameters of each group of structural surfaces;

[0052] The occurrence, size, density and spatial position parameters of each group of structural planes are randomly combined, and the Monte Carlo simulation method is used to construct a spatial network model of layered rock mass structure.

[0053] The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

[0054] Example

[0055] Taking a highway slope outcrop in Chongqing as an example, this paper further illustrates the proposed structural space network sequence modeling method suitable for layered rock mass. Figure 1-Figure 7 , the specific steps are as follows:

[0056] Step 1: Interpretation and characterization of rock outcrop data

[0057] Through field investigation and modern photogrammetry, combined with data processing technology, the geometric characteristic data of the outcrops in the study area are interpreted, and the plane fitting algorithm and traversal algorithm are used to obtain the structural surface data such as trace length, occurrence, spacing, trace direction and clustering. The basic geological attributes of the rock outcrop, such as lithology, and geotechnical properties, such as weathering degree, are recorded in detail. At the same time, the task parameters of data acquisition and interpretation, i.e. the direction of the survey line, the size and direction of the survey window, etc., are recorded. In particular, the proportion of each group of layer-limited and non-layer-limited structural surfaces is counted, such as Figure 2 As shown in the figure, the definition of a stratigraphically bounded structural surface is a structural surface whose development boundary ends at least on one side at a layer, and the definition of a non-stratigraphically bounded structural surface is a structural surface whose development boundary does not end at a layer. The structural surface is characterized by a rectangular shape. In this example, the outcrop attitude can be divided into two groups, with average attitudes of 203°∠80° and 126°∠60°, respectively, and the layer attitude is 108°∠18°. The ratio of stratigraphically bounded to non-stratigraphically bounded structural surfaces is approximately 1:5.

[0058] Step 2: Divide the rock outcrop into homogeneous areas

[0059] The rock mass in the same homogeneous zone should show similar properties in the geological domain, geotechnical domain and structural domain. The homogeneous zoning based on the geological domain and geotechnical domain is qualitatively determined according to the geological attributes and geotechnical properties of the rock mass outcrop. The homogeneous zoning of the rock mass outcrop structural domain based on the occurrence is quantitatively determined using the improved Miller contingency table method. The results of the zoning of the geological domain, geotechnical domain and structural domain are combined to achieve qualitative and quantitative comprehensive judgment and determine the final rock mass outcrop homogeneous zone division. In this example, it is verified that the selected outcrops belong to the same homogeneous zone.

[0060] Step 3: Spatial simulation of structural surface distribution

[0061] Based on the interpreted occurrence data, the idea of ​​determining the homogeneous area according to the improved Miller contingency table method is used to divide the stereographic projection results into several patches of equal area. In each patch, the occurrence sampling deviation is corrected by multiplying the occurrence observation frequency by the weight. The weight function is shown in formula (6). Based on the actual occurrence distribution results, the K-means particle swarm algorithm is used to group the structural surfaces. Furthermore, the occurrence distribution characteristics required for rock mass structure modeling are simulated through empirical distribution. In this example, Figure 3 As shown, the distributions of the two groups of structural surfaces are close to the Fisher distributions of k=18 and k=14, respectively.

[0062]

[0063] Among them, W i is the weight coefficient of the i-th structural surface, w and h are the width and height of the measurement window respectively, d i is the equivalent circle diameter of the i-th rectangular structure surface, α i and is the inclination and dip of the height of the measuring window, α r To measure the tendency of window width.

[0064] Step 4: Simulation of the long axis direction of the structural surface distribution

[0065] The rectangular structural surface has a certain rotation angle on its own plane, which can be characterized by the rotation angle of a straight line passing through the center point and parallel to the long side of the rectangular structural surface. The rotation angle can be 0° to 360°, which is exactly a complete circle, and the directions of the lines in the range of 0° to 180° and 180° to 360° are symmetrical about the center point. Therefore, the von Mises distribution is used to describe the distribution of the long axis direction of each group of rectangular structural surfaces, and its probability density function is formula (7). The mean of the probability density function parameters is determined by the mean of the strike direction of each group of structural surfaces obtained in step 1, and the dispersion is consistent with the occurrence of each group. In this example, the mean of the long axis direction of each group of rectangular structural surfaces is 293°∠90° and 216°∠90°, respectively.

[0066]

[0067] Where x is the major axis rotation angle, δ is the mean rotation angle, K is the rotation angle dispersion, I 0 is the zero-order modified Bessel formula.

[0068] Step 5: Rectangular structural surface model size simulation

[0069] Based on the interpreted trace length data, the distribution type of each group of trace lengths is estimated by the maximum likelihood method, and the correction of each group of average trace length data is realized by the multi-line method. The formula for correcting the average trace length by the multi-line method is shown in formula (8). Through the nth-order moment relationship between the true trace length and the long side of the rectangular structural surface, the statistical distribution of the long side and the ratio of the long side to the short side of each group of rectangular structural surfaces is derived. The short side can be obtained by performing simple arithmetic operations on the long side and the ratio of the long side to the short side. In this example, the trace length distribution of each group of structural surfaces is as follows: Figure 4 It has been verified that the major axis of each group of structural surfaces obeys the gamma distribution and log-normal distribution respectively. The specific parameters are shown in Table 1.

[0070]

[0071] Where μ is the mean value of the corrected trace length, C m is the mean of the intersections of all traces and the survey line, w and h are the width and height of the survey window respectively, l is the length of the survey line, α is the relative apparent inclination angle between the trace and the survey line, and θ is the relative apparent inclination angle of the trace in the survey window.

[0072] Step 6: Structural surface distribution spatial density simulation

[0073] According to the data of spacing, occurrence, size and survey line direction of each group of structural surfaces obtained in steps 1 and 5, the density of the distribution space of each group of structural surfaces is determined by the following formula. In this example, the density of each group of structural surfaces is 1.1m -3 and 1.8m -3 .

[0074]

[0075] Among them, P 30 is the number of structural surfaces per unit space, k is the ratio of the length to the short side of the rectangular structural surface, P 10 is the reciprocal of the average spacing between structural surfaces in the same group, a is the long side of the rectangular structural surface, m is the average unit normal vector of the structural surfaces in the same group, and n is the unit vector of the survey line strike.

[0076] Step 7: Spatial simulation of structural surface distribution position

[0077] A basic assumption is that the layer is infinite, so the position of the layer in the structural space is simulated based on the measured layer thickness data distribution. The layer divides the structural space into several subspaces, satisfying the characteristics of Latin hypercube sampling in multiple subintervals. Therefore, based on the optimized Latin hypercube sampling method, Poisson point sampling is performed in the structural subspace naturally divided by each layer to determine the position of the non-layer-bounded structural surface in the structural space. The position of the layer-bounded structural surface is determined by the position of the center point of the long side of the rectangle on each layer. The probability distribution and termination conditions of the Poisson point in each structural subspace are shown in the following formula. In this example, the preset rock mass structural space size is 20m×20m×20m, and the layer thickness obeys a normal distribution with a mean of 1.27. The spatial position simulation results of the layers, layer-bounded structural surfaces and non-layer-bounded structural surfaces are shown in the following formula. Figure 5 shown.

[0078]

[0079] Among them, V is the three-dimensional structure space, k is the number of Poisson points, λ(x, y, z) is the intensity function at a certain point, and U is the subspace area.

[0080] Step 8: Modeling of layered rock mass

[0081] Based on the results and assumptions of step 7, determine the position and size of the layer in the structural space. The mean of the measured layer orientation is used as the simulated layer orientation. Figure 6 As shown, 12 levels are generated.

[0082] Step 9: Modeling of layered rock mass confined structural surfaces

[0083] Based on steps 3, 5, 6, 7 and 8, in the structural subspace divided at each level, the parameters of each group of structural surfaces, namely the occurrence, size, density and spatial position parameters, are randomly combined to establish a spatial network of layer-limited structural surfaces. Among them, the density (P) of each group of layer-limited structural surfaces is determined according to the ratio of each group of layer-limited and non-layer-limited structural surfaces determined in step 1. 30 ). The long axis direction of the layer-limited structural plane is parallel to the layer, so there is no need to simulate its long axis direction. Figure 6 As shown, 3731 layer-limited structural surfaces were generated.

[0084] Step 10: Modeling of non-layer-confined structural surfaces in layered rock mass

[0085] Based on the parameter derivation results of steps 3-7, the Monte Carlo simulation method is used to randomly combine the long axis distribution, size distribution, density distribution and spatial position distribution parameters of each group of structural faces in each structural subspace to establish a non-layer-confined structural face space network. Among them, the density of the non-layer-confined structural face (P) is determined according to the ratio of layer-confined to non-layer-confined structural faces determined in step 1. 30 ). In this example, Figure 6 As shown, 19474 non-layer-confined structural surfaces were generated.

[0086] Table 1 Parameters of spatial network modeling of layered rock mass structure

[0087]

[0088] It should be noted that step 1 belongs to data acquisition, steps 2-7 belong to parameter derivation, which is carried out according to the structural surface group, and steps 8-10 belong to the rock mass structure spatial network sequence modeling method. Repeat steps 8-10, and finally select the model that best matches the actual situation as the final rock mass structure spatial network model.

[0089] The present invention uses a rectangular model as the basic morphological unit of the layered rock mass structural surface, completes the derivation of model parameters based on field outcrop interpretation data, and further uses sequence modeling technology and Monte Carlo simulation to establish a spatial network model of the layered rock mass structure. The rectangular model proposed by the present invention can more accurately characterize the morphology of the layered rock mass structural surface. The sequence modeling technology ensures the spatial distribution characteristics of the structural surface under the main control of the layer and forms a higher proportion of T-shaped topological nodes in the model, realizing high-fidelity simulation of the distribution and intersection of the layered rock mass structural surface. The present invention forms a complete working framework suitable for spatial network modeling of layered rock mass structures, optimizes the drawbacks of modeling in traditional modeling methods, and is of great significance to the stability analysis of layered rock mass slopes, connectivity assessment, and structural disaster prediction and prevention.

[0090] It should be noted that, in the present invention, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device.

[0091] The above embodiments are merely examples of the present invention and do not limit the protection scope of the present invention. All designs that are the same or similar to the present invention fall within the protection scope of the present invention.

Claims

1. A structural surface network sequence modeling method suitable for layered rock mass, characterized in that: The following steps are involved: The outcrop data of the layer-limited structural plane and the non-layer-limited structural plane of the layered rock mass are interpreted and characterized to obtain the interpreted occurrence data, among which the structural plane is represented by a rectangle; Determining the result of the rock mass outcrop homogeneous zone division according to the type attributes of the layered rock mass outcrop data; According to the interpreted occurrence data combined with the homogeneous zone division result of the rock mass outcrop, the occurrence observation frequency is used to adjust the weight function, the occurrence data in each homogeneous zone is corrected to obtain the corrected occurrence data of the homogeneous zone, and based on the corrected occurrence data of the homogeneous zone, the structural planes are grouped using the K-means particle swarm algorithm to obtain the grouped structural planes; According to the spacing, occurrence, size and survey line trend data of each group of structural surfaces, the density of the distribution space of each group of structural surfaces is determined; the position of the simulated layer in space by the actual layer thickness data distribution is obtained to determine the spatial position parameters of each group of structural surfaces; The occurrence, size, density and spatial position parameters of each group of structural planes are randomly combined, and the Monte Carlo simulation method is used to construct a spatial network model of layered rock mass structure.

2. The structural surface network sequence modeling method applicable to layered rock mass according to claim 1, characterized in that: The determination of the results of the rock mass outcrop homogeneous zone division includes: Determine homogeneous partition results based on geological domain and geotechnical domain according to geological attribute data and geotechnical property data; The results of homogeneous zoning of rock mass outcrop structural domains based on occurrence were quantitatively determined using the improved Miller contingency table method; Based on the zoning results of geological domain, geotechnical domain and structural domain, combined with the homogeneous zone division rules, the homogeneous zone division results of rock outcrop are determined.

3. The structural surface network sequence modeling method applicable to layered rock mass according to claim 1, characterized in that: The layer-limited structural surface is a structural surface whose development boundary at least one side ends at a layer, and the non-layer-limited structural surface is a structural surface whose development boundary does not end at a layer; The construction of the layered rock mass structure spatial network model comprises: The occurrence, size, density and spatial position parameters of each group of structural surfaces are randomly combined to construct a spatial network of layered rock mass confined structural surfaces in groups. Then, the Monte Carlo simulation method is used to randomly combine the occurrence, size, density and spatial position parameters of each group of structural surfaces to construct a spatial network model of non-layered rock mass unconfined structural surfaces.

4. The structural surface network sequence modeling method applicable to layered rock mass according to claim 3, characterized in that: Determining the spatial position parameters of each group of structural surfaces includes the position of the layer-limited structural surface and the position of the non-layer-limited structural surface in the structural space; The position of the layer-limited structural surface is determined by the position of the center point of the long side of the rectangle on each layer; The position of the non-layer-limited structural surface in the structural space is determined by Poisson point sampling in the structural subspace naturally divided at each layer based on an optimized Latin hypercube sampling method; The probability distribution and termination condition of each Poisson point in the structural subspace are as follows: Among them, V is the three-dimensional structure space, k is the number of Poisson points, λ(x, y, z) is the intensity function at a certain point, and U is the subspace area.

5. The structural plane network sequence modeling method applicable to layered rock mass according to claim 1, characterized in that: The weight function is adjusted by using the occurrence observation frequency, which is to obtain the adjusted weight coefficient by multiplying the occurrence observation frequency by the weight. The formula is as follows: Among them, W i is the weight coefficient of the i-th structural surface, w and h are the width and height of the measurement window respectively, d i is the equivalent circle diameter of the i-th rectangular structure surface, α i and is the inclination and dip of the height of the measuring window, α r To measure the tendency of window width.

6. The structural plane network sequence modeling method applicable to layered rock mass according to claim 1, characterized in that: It also includes using von Mises distribution to represent the distribution of the major axis direction of each group of rectangular structural surfaces after obtaining the grouped structural surfaces. The probability density function formula is as follows: Among them, x is the major axis rotation angle, δ is the rotation + angle mean, K is the rotation angle dispersion, and I0 is the zero-order modified Bessel formula.

7. The structural plane network sequence modeling method applicable to layered rock mass according to claim 1, characterized in that: The density of each group of structural surface distribution space is determined by the following formula: Among them, P 30 is the number of structural surfaces per unit space, k is the ratio of the length to the short side of the rectangular structural surface, P 10 is the reciprocal of the average spacing between structural surfaces in the same group, a is the long side of the rectangular structural surface, m is the average unit normal vector of the structural surfaces in the same group, and n is the unit vector of the survey line strike.

8. A structural surface network sequence modeling system suitable for layered rock mass, characterized in that: The following steps are involved: The rock mass structural surface data processing module is used to interpret and characterize the outcrop data of the layer-limited structural surface and the non-layer-limited structural surface of the layered rock mass to obtain the interpreted occurrence data, wherein the structural surface is represented by a rectangle; according to the type attribute of the layered rock mass outcrop data, the homogeneous zone division result of the rock mass outcrop is determined; according to the interpreted occurrence data combined with the homogeneous zone division result of the rock mass outcrop, the weight function is adjusted by using the occurrence observation frequency to correct the occurrence data in each homogeneous zone to obtain the corrected occurrence data of the homogeneous zone; based on the corrected occurrence data of the homogeneous zone, the structural surfaces are grouped by using the K-means particle swarm algorithm to obtain the grouped structural surfaces; according to the spacing, occurrence, size and survey line strike data of each group of structural surfaces, the density of the distribution space of each group of structural surfaces is determined; the position of the simulated layer in space of the actual layer thickness data distribution is obtained to determine the spatial position parameters of each group of structural surfaces; The model building module is used to randomly combine the occurrence, size, density and spatial position parameters of each group of structural surfaces, and use the Monte Carlo simulation method to construct a spatial network model of layered rock mass structure.

Images