Tunnel surrounding rock deformation probability prediction method and system

Abstract

This application relates to a method and system for predicting the probability of deformation of tunnel surrounding rock, comprising: constructing multiple joint distribution functions based on the acquired geographical data of joint groups of the target rock mass and multiple geometric attribute features affecting the deformation of the surrounding rock; extracting multiple feature points based on each joint distribution function and arranging and combining them to obtain multiple feature point combinations; modeling the joint groups and solving the deformation field using the finite element method to obtain multiple gridded tunnel surrounding rock deformation fields; extracting and regularizing the deformation data of the grid nodes of the multiple gridded tunnel surrounding rock deformation fields to obtain multiple regularized deformation field datasets; and obtaining prediction results including the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range under different confidence levels based on the probability model of the tunnel surrounding rock deformation field obtained through multiple regularized deformation fields, thereby making the prediction results reliable and the prediction process efficient.

Description

Technical Field

[0001] This application relates to the field of tunnel surrounding rock technology, and in particular to a method and system for predicting the deformation probability of tunnel surrounding rock. Background Technology

[0002] In water resource mobilization projects, long-distance TBM excavation of hydraulic tunnels often requires traversing rock masses with densely jointed zones and well-developed structural surfaces, meaning the rock mass to be traversed contains multiple sets of joints. The geometric properties of these joint sets within the rock mass have a decisive influence on the excavation deformation pattern, deformation magnitude, and long-term stability of the tunnel surrounding rock, making them a crucial factor that cannot be ignored in the safe construction and operation of the tunnel. The spatial distribution of joint sets within the rock mass exhibits significant variability, manifested in the random spatial variation of the geometric properties of each joint within a set. The geometric properties of a joint set include the attitude, spacing, trace length, and opening characteristics of each joint within the set.

[0003] Currently, existing methods for predicting the probability of surrounding rock deformation can reflect the spatial distribution variability of the geometric properties of each joint within a joint group, but they have limitations such as low repeatability of prediction results and large computational workload.

[0004] Therefore, there is an urgent need to provide a method for predicting the deformation probability of tunnel surrounding rock that has reliable prediction results and high efficiency in the prediction process. Summary of the Invention

[0005] This application provides a method and system for predicting the deformation probability of surrounding rock in tunnels, which makes the prediction results reliable and the prediction process efficient.

[0006] In a first aspect, this application provides a method for predicting the deformation probability of surrounding rock in tunnels, including:

[0007] Based on the obtained geographical data of joint groups of the target rock mass and multiple geometric property features that affect the deformation of the surrounding rock, multiple joint group distribution functions are constructed.

[0008] For each joint group distribution function, at least two feature points are extracted to form an initial feature point group. Then, one feature point is taken from each initial feature point group and arranged in combination to obtain multiple feature point combinations.

[0009] Based on the combination of the multiple feature points, the joint group is modeled and the deformation field is solved by the finite element method to obtain multiple gridded tunnel surrounding rock deformation fields; the deformation data of the grid nodes of the multiple gridded tunnel surrounding rock deformation fields are extracted and gridded and regularized to obtain multiple regularized deformation field datasets.

[0010] Using the same probability distribution function as the joint group distribution function, the multiple deformation field datasets are fitted to obtain a tunnel surrounding rock deformation field probability model; based on the tunnel surrounding rock deformation field probability model, the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range under different confidence levels are obtained.

[0011] Secondly, this application provides a tunnel surrounding rock deformation probability prediction system, comprising:

[0012] The joint distribution function construction module is used to: construct multiple joint distribution functions based on the acquired geographic data of joints in the target rock mass and multiple geometric property features that affect the deformation of the surrounding rock;

[0013] The feature point combination acquisition module is used to: extract at least two feature points to form an initial feature point group for each joint group distribution function, and sequentially take one feature point from each initial feature point group for arrangement and combination to obtain multiple feature point combinations.

[0014] The deformation field dataset acquisition module is used to: model joint groups and solve deformation fields using the finite element method based on the combination of multiple feature points to obtain multiple gridded tunnel surrounding rock deformation fields; extract deformation data of grid nodes and perform grid regularization on the multiple gridded tunnel surrounding rock deformation fields to obtain multiple regularized deformation field datasets.

[0015] The tunnel surrounding rock deformation field probability model construction and prediction module is used to: fit the multiple deformation field datasets with the same probability distribution function as the joint group distribution function to obtain the tunnel surrounding rock deformation field probability model; and based on the tunnel surrounding rock deformation field probability model, obtain the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range under different confidence levels.

[0016] Thirdly, this application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the tunnel surrounding rock deformation probability prediction method provided in the first aspect.

[0017] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the tunnel surrounding rock deformation probability prediction method provided in the first aspect.

[0018] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the tunnel surrounding rock deformation probability prediction method provided in the first aspect.

[0019] The tunnel surrounding rock deformation probability prediction method provided in this application is as follows:

[0020] Firstly, by extracting at least two statistically significant feature points from the joint distribution function constructed based on the joint distribution data of the target rock mass, and forming multiple feature point combinations, each feature point combination can characterize a spatial distribution pattern of the geometric attribute characteristics of the joint group affecting the deformation of the surrounding rock. Furthermore, multiple feature point combinations can characterize the spatial distribution variability scenario of the geometric attribute characteristics of the joint group affecting the deformation of the surrounding rock. Therefore, by forming multiple feature point combinations, it is equivalent to mathematically constructing a spatial distribution variability scenario that can cover representative joint geological conditions within the target rock mass. Thus, by forming multiple feature point combinations, the continuous, random, and multi-geometric attribute characteristics of the joint distribution function described by multiple joint distribution functions are systematically discretized into multiple representative, deterministic, and multi-geometric attribute characteristics of joint spatial distribution patterns. This achieves networked sampling of joint geological conditions within the target rock mass, avoids sampling bias, and thus ensures the reliability of subsequent deformation probability predictions for target prediction sites in the tunnel surrounding rock.

[0021] By combining multiple feature points using the finite element method, joint group modeling and deformation field solving are performed to obtain multiple gridded tunnel surrounding rock deformation fields. Since the finite element method includes a deterministic physics engine based on solid mechanics principles, each combination of feature points is transformed into a gridded tunnel surrounding rock deformation field with clear, deterministic, and reliable mechanical principles through strict mechanical laws, thereby ensuring the reliability of subsequent deformation probability predictions for target prediction sites of tunnel surrounding rock.

[0022] Secondly, for each geometric property feature affecting the deformation of the surrounding rock, the distribution function of the joint group is extracted by extracting only a few representative feature points. By arranging and combining these feature points, a limited number of feature point combinations are formed, thereby realizing the networked sampling of the geological conditions of the joint group in the target rock mass. This constructs a limited number of spatial distribution variability scenarios of the representative geological conditions of the joint group in the target rock mass, which greatly reduces the number of times the finite element method is used to model the joint group and solve the deformation field.

[0023] By extracting and regularizing the deformation data of the gridded tunnel surrounding rock obtained from the solution, multiple regularized deformation field datasets are obtained. These multiple regularized deformation field datasets are spatially aligned gridded structure databases, which allow the deformation values ​​of all gridded tunnel surrounding rock deformation fields at each grid node to be obtained based on the coordinates of each grid node. Since each grid node corresponds to a tunnel surrounding rock location, the deformation probability function of the target prediction location of the tunnel surrounding rock and the deformation range under different confidence levels can be obtained, which greatly reduces the amount of computation for deformation probability prediction and improves the computational efficiency of the prediction process.

[0024] Therefore, this application provides a method for predicting the deformation probability of tunnel surrounding rock. By using joint group geographic data, it obtains the deformation probability function of the target prediction site of the tunnel surrounding rock and the prediction results of the deformation range under different confidence levels. While ensuring the reliability of the prediction results, it significantly improves the computational efficiency of the prediction process, making the prediction results reliable and the prediction process highly efficient. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] To gain a more complete understanding of this application and its beneficial effects, the following description will be provided in conjunction with the accompanying drawings, wherein the same reference numerals in the following description denote the same parts.

[0027] Figure 1 This is a schematic diagram of the first process of the tunnel surrounding rock deformation probability prediction method provided in the embodiments of this application.

[0028] Figure 2 This is a schematic diagram of the second process of the tunnel surrounding rock deformation probability prediction method provided in the embodiments of this application.

[0029] Figure 3 This is a schematic diagram of the distribution statistics of the tendency features and the joint group distribution function corresponding to the tendency features provided in the embodiments of this application; wherein, 201 shows a bar chart of the tendency distribution of joint groups; and 202 shows the joint group distribution function corresponding to the tendency features.

[0030] Figure 4 This is a schematic diagram of the distribution statistics of dip angle characteristics and the joint group distribution function corresponding to the dip angle characteristics provided in the embodiments of this application; wherein, 203 shows a bar chart of joint dip angle distribution; 204 shows the joint group distribution function corresponding to the dip angle characteristics.

[0031] Figure 5 For based on Figure 3 The joint distribution function corresponding to the tendency characteristics in the middle and Figure 4The diagram illustrates the extraction of feature points from the joint distribution function corresponding to the dip angle feature, and the subsequent acquisition of feature point combinations. Specifically, 301 shows the first feature point of the joint distribution function corresponding to the dip angle feature; 302 shows the second feature point of the joint distribution function corresponding to the dip angle feature; 303 shows the first feature point of the joint distribution function corresponding to the dip angle feature; and 305 shows a combination of multiple feature points.

[0032] Figure 6 For based on Figure 5 A schematic diagram of a meshed finite element computation model generated by combining multiple feature points, wherein... Figure 6 In the diagram, 'a' represents the meshed finite element calculation model corresponding to feature point combination 1. Figure 6 In the diagram, b is a schematic diagram of the meshed finite element calculation model corresponding to feature point combination 2; Figure 6 In the middle, c is a schematic diagram of the meshed finite element calculation model corresponding to feature point combination 3; Figure 6 In the diagram, d represents the meshed finite element calculation model corresponding to feature point combination 4.

[0033] Figure 7 For based on Figure 6 The gridded finite element calculation model generates a schematic diagram of the gridded deformation field of the tunnel surrounding rock, in which... Figure 7 In the diagram, 'a' represents a gridded deformation field of the surrounding rock of the tunnel corresponding to feature point combination 1. Figure 7 In the diagram, b is a gridded deformation field of the tunnel surrounding rock corresponding to feature point combination 2; Figure 7 In the middle, c is a schematic diagram of the gridded tunnel surrounding rock deformation field corresponding to feature point combination 3; Figure 7 In the diagram, d represents the gridded deformation field of the surrounding rock of the tunnel corresponding to feature point combination 4.

[0034] Figure 8 This is a schematic diagram of the tunnel surrounding rock deformation probability prediction device provided in the embodiments of this application.

[0035] Figure 9 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0036] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the protection scope of this application.

[0037] Figure 1This is a schematic diagram of the first process of the tunnel surrounding rock deformation probability prediction method provided in the embodiments of this application. Figure 2 This is a schematic diagram of the second process of the tunnel surrounding rock deformation probability prediction method provided in the embodiments of this application. Please refer to... Figure 1 and Figure 2 The tunnel surrounding rock deformation probability prediction method provided in this application includes steps 110 to 140, which will be described in detail below.

[0038] Step 110: Based on the obtained geographical data of joint groups of the target rock mass and multiple geometric property features that affect the deformation of the surrounding rock, construct multiple joint group distribution functions.

[0039] Step 120: For each joint distribution function, extract at least two feature points to form an initial feature point group, and then take one feature point from each initial feature point group and arrange them in sequence to obtain multiple feature point combinations.

[0040] Step 130: Based on the combination of multiple feature points, the joint group is modeled and the deformation field is solved by the finite element method to obtain multiple gridded tunnel surrounding rock deformation fields; the deformation data of the grid nodes of the multiple gridded tunnel surrounding rock deformation fields are extracted and gridded and regularized to obtain multiple regularized deformation field datasets.

[0041] Step 140: Using the same probability distribution function as the joint group distribution function, fit the multiple deformation field datasets to obtain a tunnel surrounding rock deformation field probability model; based on the tunnel surrounding rock deformation field probability model, obtain the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range under different confidence levels.

[0042] The tunnel surrounding rock deformation probability prediction method provided in this application embodiment is based on joint group geographical data. It constructs a joint group distribution function and obtains a combination of feature points. Based on the combination of feature points, it uses finite element calculation software to model the joint group and solve the deformation field, thereby obtaining a regularized deformation field dataset that characterizes the deformation field of the tunnel surrounding rock. Based on the regularized deformation field dataset, it constructs a tunnel surrounding rock deformation field probability model. This method enables accurate and rapid prediction of the deformation probability of any target prediction point in the tunnel surrounding rock when joint group geographical data is limited, providing more reliable technical support for the deformation risk control of TBM tunnels crossing densely jointed zones.

[0043] It is understandable that step 110 yields multiple joint group distribution functions. Each joint group distribution function can characterize the spatial distribution variability statistical law of a geometric property characteristic of a joint group affecting the deformation of the surrounding rock. Step 120 yields multiple feature point combinations. Each feature point combination can characterize a spatial distribution pattern of the geometric property characteristic of a joint group affecting the deformation of the surrounding rock. Multiple feature point combinations can characterize the spatial distribution variability scenario of the geometric property characteristic of a joint group affecting the deformation of the surrounding rock. Step 130 yields multiple regularized deformation field datasets. Each regularized deformation field dataset can characterize a tunnel surrounding rock deformation field. Each tunnel surrounding rock deformation field corresponds to a spatial distribution pattern of the geometric property characteristic of a joint group affecting the deformation of the surrounding rock.

[0044] It is understood that the target rock mass is the rock mass to be traversed by the tunnel, specifically the rock mass with dense joints along the tunnel route. The acquisition methods for the geographical data of the joint groups of the target rock mass include one or more of the following: geological logging, borehole imaging, and 3D scanning. The geographical data of the joint groups includes multiple geometric attribute features that affect the deformation of the surrounding rock. These geometric attribute features affecting the deformation of the surrounding rock are one of the following characteristics of each joint in the joint group: dip direction, dip angle, spacing, trace length, and opening. Specifically, the dip direction refers to the direction of the joint surface's inclination; the dip angle refers to the maximum acute angle between the joint surface and the horizontal plane; the spacing refers to the vertical distance between two adjacent joint surfaces in the same joint group; the trace length refers to the length of the joint surface exposed on the outcrop; and the opening refers to the vertical distance between the rock walls on both sides of the joint surface.

[0045] For example, when the joint group has multiple geometric attribute features that affect the deformation of the surrounding rock, including five geometric attribute features such as dip direction, dip angle, spacing, size and opening, then five joint group distribution functions are constructed based on these five geometric attribute features, with each geometric attribute feature affecting the deformation of the surrounding rock corresponding to one joint group distribution function.

[0046] It is understood that the step of modeling joint groups and solving deformation fields using the finite element method based on the multiple feature point combinations to obtain multiple meshed tunnel surrounding rock deformation fields is used to transform the feature point combinations one by one into tunnel surrounding rock deformation fields. It is understood that the finite element method can be executed using finite element calculation software; for example, the finite element calculation software can be one of Rocscience RS2 finite element analysis software, PLAXIS 2D geotechnical finite element analysis software, ANSYS extended finite element method module, or ABAQUS extended finite element module.

[0047] It is understood that the tunnel surrounding rock deformation probability prediction method provided in the embodiments of this application is as follows:

[0048] Firstly, by extracting at least two statistically significant feature points from the joint distribution function constructed based on the joint distribution data of the target rock mass, and forming multiple feature point combinations, each feature point combination can characterize a spatial distribution pattern of the geometric attribute characteristics of the joint group affecting the deformation of the surrounding rock. Furthermore, multiple feature point combinations can characterize the spatial distribution variability scenario of the geometric attribute characteristics of the joint group affecting the deformation of the surrounding rock. Therefore, by forming multiple feature point combinations, it is equivalent to mathematically constructing a spatial distribution variability scenario that can cover representative joint geological conditions within the target rock mass. Thus, by forming multiple feature point combinations, the continuous, random, and multi-geometric attribute characteristics of the joint distribution function described by multiple joint distribution functions are systematically discretized into multiple representative, deterministic, and multi-geometric attribute characteristics of joint spatial distribution patterns. This achieves networked sampling of joint geological conditions within the target rock mass, avoids sampling bias, and thus ensures the reliability of subsequent deformation probability predictions for target prediction sites in the tunnel surrounding rock.

[0049] By combining multiple feature points using the finite element method, joint group modeling and deformation field solving are performed to obtain multiple gridded tunnel surrounding rock deformation fields. Since the finite element method includes a deterministic physics engine based on solid mechanics principles, each combination of feature points is transformed into a gridded tunnel surrounding rock deformation field with clear, deterministic, and reliable mechanical principles through strict mechanical laws, thereby ensuring the reliability of subsequent deformation probability predictions for target prediction sites of tunnel surrounding rock.

[0050] Secondly, for each geometric property feature affecting the deformation of the surrounding rock, the distribution function of the joint group is extracted by extracting only a few representative feature points. By arranging and combining these feature points, a limited number of feature point combinations are formed, thereby realizing the networked sampling of the geological conditions of the joint group in the target rock mass. This constructs a limited number of spatial distribution variability scenarios of the representative geological conditions of the joint group in the target rock mass, which greatly reduces the number of times the finite element method is used to model the joint group and solve the deformation field.

[0051] By extracting and regularizing the deformation data of the gridded tunnel surrounding rock obtained from the solution, multiple regularized deformation field datasets are obtained. These multiple regularized deformation field datasets are spatially aligned gridded structure databases, which allow the deformation values ​​of all gridded tunnel surrounding rock deformation fields at each grid node to be obtained based on the coordinates of each grid node. Since each grid node corresponds to a tunnel surrounding rock location, the deformation probability function of the target prediction location of the tunnel surrounding rock and the deformation range under different confidence levels can be obtained, which greatly reduces the amount of computation for deformation probability prediction and improves the computational efficiency of the prediction process.

[0052] Therefore, this application provides a method for predicting the deformation probability of tunnel surrounding rock. By using joint group geographic data, it obtains the deformation probability function of the target prediction site of the tunnel surrounding rock and the prediction results of the deformation range under different confidence levels. While ensuring the reliability of the prediction results, it significantly improves the computational efficiency of the prediction process, making the prediction results reliable and the prediction process highly efficient.

[0053] Step 110: Based on the obtained geographical data of joint groups of the target rock mass and multiple geometric property features that affect the deformation of the surrounding rock, construct multiple joint group distribution functions.

[0054] In some embodiments, the step of constructing multiple joint distribution functions based on the acquired geographic data of joint sets of the target rock mass and multiple geometric property features affecting the deformation of the surrounding rock includes:

[0055] Obtain geographical data of the joint groups of the target rock mass, which is the rock mass to be traversed by the tunnel;

[0056] The distribution statistics of multiple geometric attribute features affecting the deformation of the surrounding rock are obtained through the geographical data of the joint group; the multiple geometric attribute features affecting the deformation of the surrounding rock include at least two of the following: dip characteristics, dip angle characteristics, spacing characteristics, trace length characteristics, and opening characteristics of each joint in the joint group;

[0057] By traversing the distribution statistics of the multiple geometric property features affecting the deformation of the surrounding rock, multiple joint group distribution functions are obtained, and each joint group distribution function corresponds to a geometric property feature affecting the deformation of the surrounding rock.

[0058] It should be noted that, for ease of description, the "geometric property characteristics affecting the deformation of the surrounding rock" will be referred to as "influencing characteristics" in the following text. It should also be noted that "multiple" in this application means "at least two".

[0059] For example, please refer to Figure 3 and Figure 4 When multiple influencing features include dip characteristics and dip angle characteristics, the distribution statistics of the dip characteristics and the distribution statistics of the dip angle characteristics are obtained from the geographical data of the joint groups. Based on the distribution statistics of the dip characteristics, the distribution function of the joint group corresponding to the dip characteristics is obtained. Based on the distribution statistics of the dip angle characteristics, the distribution function of the joint group corresponding to the dip angle characteristics is obtained. Figure 3 The diagram below illustrates the distribution statistics of the tendency features and the distribution function of the joint group corresponding to the tendency features, provided in the embodiments of this application. 201 shows a bar chart of the tendency distribution of the joint group, which is the distribution statistics of the tendency features; 202 shows the distribution function of the joint group corresponding to the tendency features. Figure 4The following diagram illustrates the distribution statistics of dip angle features and the joint group distribution function of dip angle features provided in this application embodiment. In this diagram, 203 shows a bar chart of joint dip angle distribution, which is the distribution function data of dip angle features; and 204 shows the joint group distribution function corresponding to the dip angle features. Figure 3 and Figure 4 The distribution functions of the joint groups in the samples are all normal distribution functions.

[0060] It is understandable that the joint distribution function corresponding to the dip characteristic is used to characterize the statistical law of the spatial distribution variability of the dip characteristic of the joint group; the joint distribution function corresponding to the dip angle characteristic is used to characterize the statistical law of the spatial distribution variability of the dip angle characteristic of the joint group; the joint distribution function corresponding to the spacing characteristic is used to characterize the statistical law of the spatial distribution variability of the spacing characteristic of the joint group; the joint distribution function corresponding to the trace length characteristic is used to characterize the statistical law of the spatial distribution variability of the size characteristic of the joint group; and the joint distribution function corresponding to the opening characteristic is used to characterize the statistical law of the spatial distribution variability of the opening characteristic of the joint group.

[0061] For example, the probability distribution model can be one of the following: normal distribution function, log-normal distribution function, and Student's t-distribution function.

[0062] Step 120: For each joint distribution function, extract at least two feature points to form an initial feature point group, and then take one feature point from each initial feature point group and arrange them in sequence to obtain multiple feature point combinations.

[0063] In some embodiments, the step of extracting at least two feature points to form an initial feature point group for each joint group distribution function, and sequentially taking one feature point from each initial feature point group and arranging them in combination to obtain multiple feature point combinations includes:

[0064] For each joint distribution function, a first feature point and a second feature point are extracted. The first feature point is a point where the geometric attribute feature value of the joint distribution function affecting the deformation of the surrounding rock is less than the mean value of the joint distribution function. The second feature point is a point where the geometric attribute feature value of the joint distribution function affecting the deformation of the surrounding rock is greater than the mean value of the joint distribution function.

[0065] By sequentially selecting any feature point from the first and second feature points of the distribution function of each joint group and combining them, we obtain... There are multiple combinations of feature points, where m is the number of joint distribution functions among the multiple joint distribution functions.

[0066] In some embodiments, at least two feature points include a first feature point and a second feature point, wherein the first feature point is ( The second feature point is ( Wherein, when the joint distribution function is a normal distribution function, The mean, The standard deviation is denoted as ; when the distribution function of the joint group is a log-normal distribution function, Log-scale mean The standard deviation is logarithmic; when the joint distribution function is the Student's t-distribution function, The mean, Standard deviation; In other words, the first feature point is the point where the standard deviation of the joint distribution function is less than a first multiple, and the second feature point is the point where the standard deviation of the joint distribution function is added to the mean of the joint distribution function, and the first multiple is greater than or equal to 1 and less than or equal to 3.

[0067] It is understandable that the first feature point is ( This means that the value of the influencing feature is... The second feature point is ( This means that the value of the influencing feature is... For example, for the joint distribution function corresponding to the dip characteristic, the first characteristic point is ( The meaning is that the value of the tendency trait is... The second feature point is ( The meaning is that the value of the tendency trait is... .

[0068] It is understandable that by extracting the first feature point ( Second feature point ( By locking down the feature interval that contributes most to deformation, the key information of the continuous probability distribution of joint groups is compressed into a finite number of representative spatial distribution patterns of joint groups. While ensuring the reliability of deformation probability prediction, this significantly reduces computational complexity and uncertainty propagation.

[0069] By extracting feature points within a range of at least one standard deviation on both sides of the mean, the variability of feature points is actively controlled to ensure the effective transmission of variations in the joint group's influence characteristics. The extracted feature points need sufficient variability to ensure that the joint group's influence characteristics generated from these feature point combinations do not lose their inherent variability due to rounding errors or simplification during calculation. Setting the lower limit of the first multiple to 1 means that the two feature points are at least distributed within a range of one standard deviation on both sides of the mean. This distance avoids oversimplification of the distribution of joint group's influence characteristics, ensuring that there are distinguishable and significant differences between different combinations of generated feature points. This guaranteed variability allows the natural variability of the joint group to be preserved and reflected in subsequent calculations, thereby directly improving the stability and reliability of probability prediction.

[0070] By extracting feature points within a range of at most three standard deviations around the mean, excessive feature dispersion is prevented, ensuring that the target joint group obtained through joint group modeling conforms to engineering reality. The distribution of joint group influence features typically does not exhibit extreme outliers. Setting the upper limit of the first multiple to 3 is based on engineering practice and statistical common sense. In a normal distribution, the probability of data points exceeding three standard deviations from the mean is extremely low. Setting the upper limit of the first multiple to 3 as the boundary for feature point extraction effectively prevents the generated joint group influence features from deviating excessively from the common range, avoiding the generation of unrealistic and extreme target joint groups. This ensures that all subsequent deformation probability predictions are based on a reasonable model that conforms to engineering understanding, fundamentally guaranteeing the practical significance and usability of the prediction results.

[0071] For example, please refer to Figure 5 ,based on Figure 3 The joint distribution function corresponding to the dip feature is used to extract the first feature point 301 corresponding to the joint distribution function corresponding to the dip feature, which is ( ) and the second feature point 302, which is ( ).based on Figure 4 The joint distribution function corresponding to the dip angle feature is used to extract the first feature point 303 corresponding to the joint distribution function corresponding to the dip angle feature, which is ( ) and the second feature point 304, which is ( ).in , These are the mean and standard deviation of the distribution function of the joint group corresponding to the tendency characteristics, respectively. , Let be the mean and standard deviation of the joint distribution function corresponding to the dip angle feature, respectively. Based on the first feature point 301 and the second feature point 302 corresponding to the dip feature, and the first feature point 303 and the second feature point 304 corresponding to the dip angle feature, feature point combination 1 is obtained: ( , Feature point combination 2: ( , Feature point combination 3: ( , Feature point combination 4: ( , ).

[0072] It is understood that the aforementioned influencing features can be geometric attribute features of joint groups that are correlated to a certain extent with the deformation of the tunnel surrounding rock, or geometric attribute features whose influence on the deformation of the tunnel surrounding rock reaches a first set threshold, or geometric attribute features whose influence weight on the deformation of the tunnel surrounding rock reaches a second set threshold. For example, the degree of influence or influence weight can be determined through engineering experience. Generally, the influencing features, ordered from largest to smallest according to their degree of influence or influence weight, are: dip characteristics, dip angle characteristics, spacing characteristics, trace length characteristics, and opening characteristics.

[0073] In some embodiments, the distribution functions of the plurality of joint groups are obtained by fitting the same probability distribution model; the probability distribution model is any one of the normal distribution function, the log-normal distribution function, and the Student's t-distribution function;

[0074] When the probability distribution model is a normal distribution function, for each joint group distribution function, the first feature point is the point where the standard deviation of the joint group distribution function is less than the first multiple, and the second feature point is the point where the standard deviation of the joint group distribution function is added to the mean of the joint group distribution function, and the first multiple takes any value of 1, 2, or 3.

[0075] When the probability distribution model is a log-normal distribution function, for each joint group distribution function, the mean is the log-scale mean, the first feature point is the point where the log-scale mean of the joint group distribution function is minus the log-scale standard deviation of a first multiple, and the second feature point is the point where the log-scale mean of the joint group distribution function is plus the log-scale standard deviation of a first multiple, where the first multiple takes any value from 1, 2, and 3;

[0076] When the probability distribution model is a student t-distribution function, for each joint group distribution function, the first feature point is the point where the standard deviation of the joint group distribution function is less than a first multiple, and the second feature point is the point where the standard deviation of the joint group distribution function is added to the mean of the joint group distribution function, and the first multiple takes any value from 1, 2, and 3.

[0077] Understandably, when the first multiplier is 1, the multiple target joint groups constructed based on the combination of multiple feature points describe the geological conditions of high-probability, typical joint groups within the target rock mass. The prediction results correspond to deformation predictions under these high-probability, typical joint group geological conditions, which can be used to assess typical risk levels and serve as a benchmark for tunnel construction early warning. When the first multiplier is 2, the multiple target joint groups constructed based on the combination of multiple feature points describe the geological conditions of the target rock mass that include less favorable joint groups. The prediction results correspond to deformation predictions under these less favorable joint group geological conditions, which can be used to assess conventional risk levels and serve as a basis for tunnel design that balances safety and economy. When the first multiplier is 3, the multiple target joint groups constructed based on the combination of multiple feature points describe the geological conditions of extremely rare joint groups within the target rock mass. The prediction results correspond to deformation predictions under these extremely rare joint group geological conditions, which can be used to assess extreme risk levels and serve as a basis for developing tunnel emergency plans. It is understandable that the "unfavorable" in the geological conditions of the joint group refers to the evaluation of the impact characteristics of the joint group. For example, for the spacing characteristic, the smaller the spacing, the more unfavorable it is, because the smaller the spacing, the greater the probability of deformation of the surrounding rock of the tunnel; for the trace length characteristic, the larger the trace length, the more unfavorable it is, because the larger the trace length, the greater the probability of deformation of the surrounding rock of the tunnel.

[0078] For example, when the probability distribution model is a normal distribution function, the extracted feature points can be selected as ( )and( ),in and The first The mean and standard deviation of the distribution function of each joint group. For any natural number less than or equal to m, The interval corresponding to the feature point ( , The probability of ).

[0079] For example, for the joint distribution function corresponding to the tendency feature, the feature point is ( )and( ), the interval corresponding to the feature point The main range of the distribution of joint dip characteristic values ​​is defined. For the distribution function of the joint group corresponding to the dip characteristic, the mean represents the average dip direction of the joint group, that is, the dominant dip direction; the standard deviation is used to measure the degree of concentration or dispersion of the dip characteristic data of each joint in the joint group around the average dip direction.

[0080] For example, the step involves sequentially selecting any feature point from the first and second feature points of the distribution function of each joint group and combining them to obtain... In the step of combining different feature points, where m is the number of joint group distribution functions in the multiple joint group distribution functions, when m=3, that is, in the joint group of the target rock mass, there are 3 influencing features, such as dip characteristics, dip angle characteristics, and spacing characteristics. The multiple joint group distribution functions include 3 joint group distribution functions, namely, the first joint group distribution function representing the spatial distribution variability of dip characteristics, the second joint group distribution function representing the spatial distribution variability of dip angle characteristics, and the third joint group distribution function representing the spatial distribution variability of spacing characteristics. These distribution functions are all normal distribution functions. The two feature points extracted from the first joint group distribution function are A1: ( A2: ( The two feature points extracted from the distribution function of the second group are B2: B2: ( The two feature points extracted from the distribution function of the third group are C1: ( C2: ( If ), then the number of feature point combinations is The numbers are A1B1C1, A1B2C1, A1B1C2, A1B2C2, A2B1C1, A2B2C1, A2B1C2, and A2B2C2, respectively.

[0081] The feature point combination A1B1C1 is used to characterize the dip characteristic value of the joint group. The dip angle characteristic value is Spacing feature value At that time, the spatial distribution pattern of the joint group; the feature point combination A1B2C1 is used to characterize the dip characteristic value of the joint group. The dip angle characteristic value is Spacing feature value At that time, the spatial distribution pattern of the joint group.

[0082] Step 130: Based on the combination of multiple feature points, the joint group is modeled and the deformation field is solved by the finite element method to obtain multiple gridded tunnel surrounding rock deformation fields; the deformation data of the grid nodes of the multiple gridded tunnel surrounding rock deformation fields are extracted and gridded and regularized to obtain multiple regularized deformation field datasets.

[0083] In some embodiments, the step of modeling joint groups and solving deformation fields using the finite element method based on the combination of multiple feature points to obtain multiple gridded tunnel surrounding rock deformation fields includes:

[0084] The geometric model generation steps include: taking each of the aforementioned feature point combinations as a target feature point combination, drawing a target joint group in finite element calculation software based on the target feature point combination, wherein the target joint group has the spatial distribution pattern of the joint group defined by the target feature point combination; and forming a target finite element geometric model containing the target joint group by fusing the geometric contour of the tunnel and the geometric conditions of the computational domain region based on the target joint group.

[0085] The computational model generation steps include: generating a meshed target finite element computational model containing the target joint group by dividing the target finite element geometric model into a mesh, adding boundary conditions, and setting material models and material properties in finite element computation software;

[0086] The solution steps include: solving the target finite element calculation model by calling the solver in the element calculation software to obtain the meshed tunnel surrounding rock deformation field corresponding to the target finite element calculation model;

[0087] By traversing the multiple combinations of feature points, multiple target finite element geometric models are obtained through the geometric model generation step. The multiple target finite element geometric models have the spatial distribution variation pattern of joint groups defined by the multiple combinations of feature points. Multiple gridded target finite element calculation models are generated through the calculation model generation step. The multiple gridded tunnel surrounding rock deformation fields are generated through the solution step.

[0088] It should be noted that, in the embodiments of this application, for the purpose of convenient description, the target finite element geometric model can also be referred to as the finite element geometric model; the target finite element calculation model can also be referred to as the finite element calculation model.

[0089] For example, the finite element calculation software can be one of Rocscience RS2 finite element analysis software, PLAXIS2D geotechnical finite element analysis software, ANSYS extended finite element method module, or ABAQUS extended finite element module. The solver is a module within the finite element calculation software.

[0090] It is understood that each combination of feature points corresponds to one spatial distribution pattern of joint groups; multiple combinations of feature points correspond to multiple spatial distribution patterns of joint groups; the multiple spatial distribution patterns of joint groups cover the range of scenarios with variability in the spatial distribution of joint groups. It is understood that one spatial distribution pattern of joint groups can also be called a spatial distribution of joint groups, or a configuration of joint groups.

[0091] It is also understandable that each generated finite element geometric model precisely contains all the location data of a specific joint group's spatial distribution pattern. This data is recorded and stored in detail using node coordinates and element attributes within the finite element calculation software. Node coordinates define the precise three-dimensional position of each point within the joint group, while element attributes define the geometric properties of the joint elements connecting these points, such as dip direction, dip angle, spacing, trace length, and opening characteristics. The purpose of this data is to accurately simulate the mechanical response of the tunnel surrounding rock under different joint group spatial distribution patterns, thereby revealing and quantifying how the spatial distribution variability of joint groups specifically affects the deformation behavior and stability of the surrounding rock.

[0092] Understandably, in the process of generating each finite element calculation model, the purpose of meshing is to discretize the continuous geometric model into a finite number of small, regularly shaped units, such as tetrahedrons and hexahedrons, so as to perform subsequent solution calculations and obtain the mechanical response of the tunnel surrounding rock under a specific joint group spatial distribution pattern, that is, to obtain the deformation field of the tunnel surrounding rock under a specific joint group spatial distribution pattern.

[0093] Please see Figure 5 ,based on Figure 3 The joint distribution function corresponding to the dip feature is used to extract the first feature point 301 corresponding to the joint distribution function corresponding to the dip feature, which is ( ) and the second feature point 302, which is ( ).based on Figure 4 The joint distribution function corresponding to the dip angle feature is used to extract the first feature point 303 corresponding to the joint distribution function corresponding to the dip angle feature, which is ( ) and the second feature point 304, which is ( ).in , These are the mean and standard deviation of the distribution function of the joint group corresponding to the tendency characteristics, respectively. , Let be the mean and standard deviation of the joint distribution function corresponding to the dip angle feature, respectively. Based on the first feature point 301 and the second feature point 302 corresponding to the dip feature, and the first feature point 303 and the second feature point 304 corresponding to the dip angle feature, feature point combination 1 is obtained: ( , Feature point combination 2: ( , Feature point combination 3: ( , Feature point combination 4: ( , ).

[0094] For feature point combination 1, feature point combination 2, feature point combination 3, and feature point combination 4, respectively, through the geometric model generation step and the computational model generation step, the corresponding results are as follows: Figure 6 In Figure 6 a, Figure 6 b, Figure 6 c, Figure 6 The meshed finite element calculation model is shown in Figure d. It can be understood that... Figure 6 a, Figure 6 b, Figure 6 c, Figure 6 In the gridded finite element calculation model shown in d, the gridded region represents the surrounding rock of the tunnel, the white circular area enclosed by the gridded region represents the tunnel, and each straight line corresponds to a joint in the joint group.

[0095] For feature point combination 1, feature point combination 2, feature point combination 3, and feature point combination 4, respectively, through the geometric model generation step, the computational model generation step, and the solution step, the corresponding results are as follows: Figure 7 In Figure 7 a, Figure 7 b, Figure 7 c, Figure 7 The gridded deformation field of the tunnel surrounding rock is shown in Figure d. It is understandable that... Figure 7 The TotalDisplacement in this context refers to the overall deformation value scale. This scale, by setting multiple color intervals and defining the deformation value range corresponding to each color interval, visually displays the deformation value range of different areas in the gridded tunnel surrounding rock. Figure 7 a, Figure 7 b, Figure 7 c, Figure 7 In the gridded deformation field of the tunnel surrounding rock shown in Figure d, the red gridded area represents the tunnel surrounding rock, and the white circular area enclosed by the red gridded area represents the tunnel. Each straight line corresponds to a joint in the joint group. For example, Figure 7 The region in 'a' that contains the second color interval in Total Displacement has a distortion value range of 3.33. Meters up to 6.67 rice.

[0096] In some embodiments, the step of meshing in the finite element calculation software includes: meshing based on the location of joints in the target finite element geometric model, and performing local mesh refinement at the joints to ensure that deformation near the joints can be accurately captured.

[0097] In some embodiments, the step of meshing, adding boundary conditions, and setting material models and properties in the finite element calculation software includes setting the material models and properties, which involves simulating the contact, slip, and opening behavior of joint surfaces in the target finite element geometric model using joint elements with no or varying thickness. For example, the joint elements may be Goodman elements.

[0098] It is understandable that contact behavior, slip behavior, and opening behavior are mechanical behaviors. The geometric properties corresponding to contact behavior and opening behavior are opening degree characteristics, while the geometric properties corresponding to slip behavior are dip characteristics and tilt angle characteristics.

[0099] In some embodiments, when the number of influencing features is m, where m is an integer greater than or equal to 2, extracting two feature points from the joint group distribution function corresponding to each influencing feature can yield the following results: A combination of feature points; based on this The combination of feature points, through the geometric model generation step, the computational model generation step, and the solution step, can yield... A gridded deformation field of the surrounding rock of the tunnel.

[0100] In some embodiments, the step of extracting and regularizing the deformation data of the grid nodes of the plurality of gridded tunnel surrounding rock deformation fields to obtain a plurality of regularized deformation field datasets includes:

[0101] The data extraction steps include: taking each of the gridded tunnel surrounding rock deformation fields as the target tunnel surrounding rock deformation field, extracting the coordinates and deformation values ​​of all grid nodes of the target tunnel surrounding rock deformation field, and obtaining the deformation field dataset corresponding to the target tunnel surrounding rock deformation field. The deformation field dataset contains the coordinates and deformation values ​​of all grid nodes of the target tunnel surrounding rock deformation field.

[0102] By traversing the multiple gridded tunnel surrounding rock deformation fields, multiple deformation field datasets are obtained through the data extraction steps.

[0103] A predefined array of regular grid cells covers the area of ​​the deformation field of the surrounding rock of the multiple gridded tunnels and can accurately describe the location of the tunnels. The array of regular grid cells includes multiple regularly arranged grid cells with unique numbers and coordinates.

[0104] The grid normalization step includes: taking each of the aforementioned deformation field datasets as the target deformation field dataset; and using an interpolation algorithm to map the deformation values ​​of the grid nodes in the target deformation field dataset to the grid nodes in the predefined regular grid cell array to obtain the normalized deformation field data corresponding to the target deformation field dataset.

[0105] By traversing the multiple deformation field datasets and performing the gridding and normalization step, the multiple normalized deformation field datasets are obtained.

[0106] It is understood that the grid node is the intersection point of the grid cells that make up the grid. The deformation value is the deformation value at the grid node. The deformation value of the grid node can be the deformation value in at least one of the X, Y, and Z directions of the grid node.

[0107] It is understood that the multiple regularized deformation field datasets can be structurally organized into a matrix, with each regularized deformation field dataset constituting a data matrix. It is also understood that each of the multiple regularized deformation field datasets uses the same predefined regular grid cell array. Since the grid cells have the same distribution and the grid node coordinates correspond to the same values, the deformation values ​​at the same grid node coordinates can be extracted from the multiple regularized deformation field datasets to form a deformation value sample set. This sample set can be used to describe the statistical regularity of the deformation values ​​at the midpoint of the tunnel surrounding rock corresponding to that grid node coordinate. Furthermore, this deformation value sample set can be used to estimate the deformation probability at the midpoint of the tunnel surrounding rock.

[0108] In some embodiments, the maximum size of the grid cells in the regular grid cell array is less than or equal to the maximum size of the grid cells in the plurality of gridded tunnel surrounding rock deformation fields, and the minimum size of the grid cells in the regular grid cell array is greater than or equal to the maximum size of the plurality of gridded tunnel surrounding rock deformation fields.

[0109] In some embodiments, the interpolation algorithm may be one of Kriging interpolation, inverse distance weighted interpolation, and linear interpolation.

[0110] In some embodiments, when the number of influencing features is m, where m is an integer greater than or equal to 2, extracting two feature points from the joint group distribution function corresponding to each influencing feature can yield the following results: A combination of feature points; based on this The combination of feature points, through geometric model generation steps, computational model generation steps, and solution steps, can yield... A gridded deformation field of the surrounding rock of the tunnel; based on this A gridded deformation field of the surrounding rock of the tunnel can be obtained through the gridding and regularization steps. A dataset of deformation fields i is set to 1~ The integer, after being processed through the gridding and normalization step, can be obtained A normalized deformation field dataset i is set to 1~ Integers.

[0111] Step 140: Using the same probability distribution function as the joint group distribution function, fit the multiple deformation field datasets to obtain a tunnel surrounding rock deformation field probability model; based on the tunnel surrounding rock deformation field probability model, obtain the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range under different confidence levels.

[0112] In some embodiments, the step of fitting the plurality of deformation field datasets with the same probability distribution function as the joint group distribution function to obtain a probability model of the tunnel surrounding rock deformation field includes:

[0113] For the multiple deformation field datasets, the deformation values ​​of the same grid node attached to the predefined regular grid cell array are regarded as a sample set. The deformation values ​​of the same grid node are extracted to obtain the deformation value sample set corresponding to the same grid node.

[0114] Based on the deformation value sample set corresponding to the same grid node, the same probability distribution function as the joint group distribution function is used to obtain the deformation probability function corresponding to the same grid node;

[0115] Traverse all grid nodes in the predefined regular grid cell array to obtain the deformation probability function corresponding to each grid node; integrate the deformation probability functions corresponding to all grid nodes to obtain the tunnel surrounding rock deformation field probability model, which includes the deformation probability function corresponding to all grid nodes.

[0116] In some embodiments, when the number of influencing features is m, where m is an integer greater than or equal to 2, extracting two feature points from the joint group distribution function corresponding to each influencing feature can yield the following results: A combination of feature points; based on this The combination of feature points, through geometric model generation steps, computational model generation steps, and solution steps, can yield... A gridded deformation field of the surrounding rock of the tunnel; based on this A gridded deformation field of the surrounding rock of the tunnel can be obtained through the gridding and regularization steps. A dataset of deformation fields i is set to 1~ The integer, after being processed through the gridding and normalization step, can be obtained A normalized deformation field dataset i is set to 1~ Integers; based on the same grid node in the predefined regular grid cell array, for A normalized deformation field dataset i is set to 1~ For integers, extract the deformation values ​​of the same mesh node to obtain... Each deformation value is combined to form the deformation sample set corresponding to the same grid node. ,in For the first The deformation value corresponding to the same grid node in the normalized deformation field dataset. For the first The deformation value corresponding to the same grid node in the normalized deformation field dataset. For the first The deformation value corresponding to the same grid node in the normalized deformation field dataset; based on all grid nodes in the predefined regular grid cell array, for A normalized deformation field dataset i is set to 1~ Using integers, extract the deformation sample set for all mesh nodes to obtain a deformation sample set corresponding to each mesh node. ,in The coordinates of different grid nodes in all grid nodes are given; by adopting the same probability distribution function as the joint group distribution function, the deformation probability function corresponding to all grid nodes is obtained, and the set of deformation probability functions corresponding to all grid nodes is taken as the probability model of the tunnel surrounding rock deformation field.

[0117] In some embodiments, the step of obtaining the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range at different confidence levels based on the tunnel surrounding rock deformation field probability model includes:

[0118] Obtain the coordinates of the target prediction point in the tunnel surrounding rock; determine the grid node in the predefined regular grid cell array that has the same coordinates as the target prediction point; find the deformation probability function corresponding to the grid node from the tunnel surrounding rock deformation field probability model to obtain the deformation probability function of the target prediction point;

[0119] Based on the deformation probability function of the target prediction site, the deformation range of the target prediction site at different confidence levels is predicted.

[0120] It is understood that the deformation parameters of the target prediction point can be calculated based on the deformation probability function corresponding to the target prediction point; based on the deformation parameters, the deformation range of the target prediction point at different confidence levels can be predicted.

[0121] For example, when the deformation probability function is a normal distribution function, the expression for the deformation probability function corresponding to the target prediction site p can be:

[0122]

[0123] in, The value of deformation at the target predicted location p. The deformation value at the target predicted location p is... The probability, The mean deformation at the target predicted site p. The standard deviation of the deformation at the target prediction site p.

[0124] For example, based on the deformation probability function corresponding to the target prediction site p, a confidence level F is determined according to the importance of the project, and a confidence equation is constructed. The expression can be:

[0125]

[0126] Where F is the confidence level; The deformation value at the target predicted location p is... The probability of; The mean deformation at the target predicted site p. denoted as the standard deviation of deformation at the target predicted site p; 'a' represents the standard deviation relative to the target predicted site p. The offset; Indicates the lower limit of integration. ; Indicates the maximum number of points. .

[0127] The deformation parameters of the target prediction point, including the mean deformation, can be calculated using the deformation probability function corresponding to the target prediction point p. and deformation standard deviation Based on the mean deformation and deformation standard deviation Confidence level can be obtained. When the target predicted site p is, the deformation range is ( , ); confidence level When the target predicted site p is, the deformation range is ( , ); confidence level When the target predicted site p is, the deformation range is ( , ).

[0128] For example, the target prediction point p can be any point in the area where the tunnel surrounding rock is located, such as a point at the tunnel surrounding rock arch or the sidewalls on both sides of the tunnel surrounding rock that are of interest in the engineering.

[0129] Figure 8This is a schematic diagram of the tunnel surrounding rock deformation probability prediction system provided in the embodiments of this application. Please refer to... Figure 8 The tunnel surrounding rock deformation probability prediction system may include a joint group distribution function construction module 801, a feature point combination acquisition module 802, a deformation field dataset acquisition module 803, and a tunnel surrounding rock deformation field probability model construction module 804. Wherein:

[0130] The joint distribution function construction module 801 is used to construct multiple joint distribution functions based on the acquired geographical data of joints of the target rock mass and multiple geometric property features that affect the deformation of the surrounding rock.

[0131] The feature point combination acquisition module 802 is used to: extract at least two feature points to form an initial feature point group for each joint group distribution function, and sequentially take one feature point from each initial feature point group for arrangement and combination to obtain multiple feature point combinations.

[0132] The deformation field dataset acquisition module 803 is used to: model joint groups and solve deformation fields using the finite element method based on the combination of multiple feature points to obtain multiple gridded tunnel surrounding rock deformation fields; and extract and regularize the deformation data of grid nodes in the multiple gridded tunnel surrounding rock deformation fields to obtain multiple regularized deformation field datasets.

[0133] The tunnel surrounding rock deformation field probability model construction module 804 is used to: fit the multiple deformation field datasets with the same probability distribution function as the joint group distribution function to obtain the tunnel surrounding rock deformation field probability model; and based on the tunnel surrounding rock deformation field probability model, obtain the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range under different confidence levels.

[0134] In practical applications, the above system can be a terminal device or a chip applied to a terminal device. In this application, the system can implement the functions of multiple units through software, hardware, or a combination of software and hardware, enabling the system to execute the steps of the tunnel surrounding rock deformation probability prediction method provided in any of the above embodiments. Furthermore, the technical effects of each technical solution of this system can be referenced to the technical effects of the corresponding technical solutions in the tunnel surrounding rock deformation probability prediction method, and this application will not elaborate on them further.

[0135] Figure 9 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application.

[0136] Based on the hardware implementation of each unit in the above system, embodiments of this application also provide an electronic device, such as... Figure 9As shown, the electronic device 900 includes a memory 910 and a processor 920. The memory 910 stores a computer program, and the processor 920 executes the computer program to implement the steps of the tunnel surrounding rock deformation probability prediction method provided in any of the above embodiments.

[0137] Of course, in practical applications, such as Figure 9 As shown, the various components in the electronic device 900 are coupled together via a bus system 930. It is understood that the bus system 930 is used to enable communication between these components. In addition to a data bus, the bus system 930 also includes a power bus, a control bus, and a status signal bus. However, for clarity, all buses are labeled as bus system 930 in the figure.

[0138] In practical applications, the aforementioned processor can be at least one of the following: Application-Specific Integrated Circuit (ASIC), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field-Programmable Gate Array (FPGA), controller, microcontroller, and microprocessor. It is understood that, for different devices, the electronic devices used to implement the functions of the aforementioned processor can also be other types, and the embodiments of this application do not specifically limit this.

[0139] The aforementioned memory can be volatile memory, such as random-access memory (RAM); or non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state drive (SSD); or a combination of the above types of memory, and provides instructions and data to the processor.

[0140] The electronic devices described in the embodiments of this application can be terminal devices or chips applied to terminal devices. The terminal devices described in the embodiments of this application can be computers.

[0141] In an exemplary embodiment, this application also provides a computer-readable storage medium, such as a memory including a computer program, which can be executed by a processor of an electronic device to perform the steps of the aforementioned method.

[0142] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any one of the embodiments of this application.

[0143] Optionally, the computer program product can be applied to the electronic device in the embodiments of this application, and the computer program instructions cause the computer to execute the corresponding processes implemented by the electronic device in the various methods of the embodiments of this application. For the sake of brevity, they will not be described in detail here.

[0144] This application also provides a computer program.

[0145] Optionally, the computer program can be applied to the electronic device in the embodiments of this application. When the computer program is run on a computer, it causes the computer to execute the corresponding processes implemented by the electronic device in the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0146] It should be understood that in the embodiments of this application, data such as user information are involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0147] It should be understood that the terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items. The expressions “having,” “may have,” “comprising,” and “including,” or “may include” and “may contain” used herein may be used to indicate the presence of a corresponding feature (e.g., an element such as a number, function, operation, or component), but do not exclude the presence of additional features.

[0148] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another, and are not necessarily used to describe a specific order or sequence. For example, without departing from the scope of this invention, first information may also be referred to as second information, and similarly, second information may also be referred to as first information.

[0149] The technical solutions described in the embodiments of this application can be combined arbitrarily without conflict.

[0150] In the several embodiments provided in this application, it should be understood that the disclosed methods, apparatus, and devices can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical, or other forms.

[0151] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0152] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0153] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for predicting the deformation probability of surrounding rock in tunnels, characterized in that, include: Based on the obtained geographical data of joint groups of the target rock mass and multiple geometric property features that affect the deformation of the surrounding rock, multiple joint group distribution functions are constructed. For each joint group distribution function, at least two feature points are extracted to form an initial feature point group. Then, one feature point is taken from each initial feature point group and arranged in combination to obtain multiple feature point combinations. Based on the combination of the multiple feature points, the joint group modeling and deformation field solution are performed by the finite element method to obtain multiple gridded tunnel surrounding rock deformation fields. For the multiple gridded tunnel surrounding rock deformation fields, the deformation data of the grid nodes are extracted and the grid is regularized to obtain multiple regularized deformation field datasets; Using the same probability distribution function as the joint group distribution function, the multiple deformation field datasets are fitted to obtain a tunnel surrounding rock deformation field probability model; based on the tunnel surrounding rock deformation field probability model, the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range under different confidence levels are obtained. The step of extracting at least two feature points to form an initial feature point group for each joint group distribution function, and then sequentially taking one feature point from each initial feature point group and arranging them in combination to obtain multiple feature point combinations includes: For each joint distribution function, a first feature point and a second feature point are extracted. The first feature point is a point where the geometric attribute value of the joint distribution function affecting the surrounding rock deformation is less than the mean value of the joint distribution function. The second feature point is a point where the geometric attribute value of the joint distribution function affecting the surrounding rock deformation is greater than the mean value of the joint distribution function. Then, any feature point is selected from the first and second feature points of each joint distribution function and combined to obtain... There are multiple combinations of feature points, where m is the number of joint distribution functions among the multiple joint distribution functions.

2. The method for predicting the deformation probability of tunnel surrounding rock according to claim 1, characterized in that, The step of constructing multiple joint distribution functions based on the acquired geographic data of joint groups in the target rock mass and multiple geometric property features affecting the deformation of the surrounding rock includes: Obtain geographical data of the joint groups of the target rock mass, which is the rock mass to be traversed by the tunnel; The distribution statistics of multiple geometric attribute features affecting the deformation of the surrounding rock are obtained through the geographical data of the joint group; the multiple geometric attribute features affecting the deformation of the surrounding rock include at least two of the following: dip characteristics, dip angle characteristics, spacing characteristics, trace length characteristics, and opening characteristics of each joint in the joint group; By traversing the distribution statistics of the multiple geometric property features affecting the deformation of the surrounding rock, multiple joint group distribution functions are obtained, and each joint group distribution function corresponds to a geometric property feature affecting the deformation of the surrounding rock.

3. The method for predicting the deformation probability of tunnel surrounding rock according to claim 1, characterized in that, The distribution functions of the multiple joint groups are obtained by fitting the same probability distribution model; the probability distribution model is any one of the normal distribution function, log-normal distribution function, and Student's t-distribution function. When the probability distribution model is a normal distribution function, for each joint group distribution function, the first feature point is the point where the standard deviation of the joint group distribution function is less than the first multiple, and the second feature point is the point where the standard deviation of the joint group distribution function is added to the mean of the joint group distribution function, and the first multiple takes any value of 1, 2, or 3. When the probability distribution model is a log-normal distribution function, for each joint group distribution function, the mean is the log-scale mean, the first feature point is the point where the log-scale mean of the joint group distribution function is minus the log-scale standard deviation of a first multiple, and the second feature point is the point where the log-scale mean of the joint group distribution function is plus the log-scale standard deviation of a first multiple, where the first multiple takes any value from 1, 2, and 3; When the probability distribution model is a student t-distribution function, for each joint group distribution function, the first feature point is the point where the standard deviation of the joint group distribution function is less than a first multiple, and the second feature point is the point where the standard deviation of the joint group distribution function is added to the mean of the joint group distribution function, and the first multiple takes any value from 1, 2, and 3.

4. The method for predicting the deformation probability of tunnel surrounding rock according to claim 1, characterized in that, The steps of modeling joint groups and solving deformation fields using the finite element method based on the combination of multiple feature points to obtain multiple gridded tunnel surrounding rock deformation fields include: The geometric model generation steps include: taking each combination of feature points as a target feature point combination, drawing a target joint group in finite element calculation software based on the target feature point combination, wherein the target joint group has the spatial distribution pattern of the joint group defined by the target feature point combination; and forming a target finite element geometric model containing the target joint group by fusing the geometric contour of the tunnel and the geometric conditions of the computational domain region based on the target joint group. The computational model generation steps include: generating a meshed target finite element computational model containing the target joint group by dividing the target finite element geometric model into a mesh, adding boundary conditions, and setting material models and material properties in finite element computation software; The solution steps include: solving the target finite element calculation model by calling the solver in the element calculation software to obtain the meshed tunnel surrounding rock deformation field corresponding to the target finite element calculation model; By traversing the multiple feature point combinations, multiple target finite element geometric models are obtained through the geometric model generation step. The multiple target finite element geometric models have the spatial distribution variation pattern of joint groups defined by the multiple feature point combinations. Multiple gridded target finite element calculation models are generated through the calculation model generation step. The multiple gridded tunnel surrounding rock deformation fields are generated through the solution step.

5. The method for predicting the deformation probability of tunnel surrounding rock according to claim 1, characterized in that, The step of extracting and regularizing the deformation data of the grid nodes of the multiple gridded tunnel surrounding rock deformation fields to obtain multiple regularized deformation field datasets includes: The data extraction steps include: taking each of the gridded tunnel surrounding rock deformation fields as the target tunnel surrounding rock deformation field, extracting the coordinates and deformation values ​​of all grid nodes of the target tunnel surrounding rock deformation field, and obtaining the deformation field dataset corresponding to the target tunnel surrounding rock deformation field. The deformation field dataset contains the coordinates and deformation values ​​of all grid nodes of the target tunnel surrounding rock deformation field. By traversing the multiple gridded tunnel surrounding rock deformation fields, multiple deformation field datasets are obtained through the data extraction steps. A predefined array of regular grid cells covers the area of ​​the deformation field of the surrounding rock of the multiple gridded tunnels and can accurately describe the location of the tunnels. The array of regular grid cells includes multiple regularly arranged grid cells with unique numbers and coordinates. The grid normalization step includes: taking each of the aforementioned deformation field datasets as the target deformation field dataset; and using an interpolation algorithm to map the deformation values ​​of the grid nodes in the target deformation field dataset to the grid nodes in the predefined regular grid cell array to obtain the normalized deformation field data corresponding to the target deformation field dataset. By traversing the multiple deformation field datasets and performing the gridding and normalization step, the multiple normalized deformation field datasets are obtained.

6. The method for predicting the deformation probability of tunnel surrounding rock according to claim 5, characterized in that, The step of fitting the multiple deformation field datasets with the same probability distribution function as the joint group distribution function to obtain the probability model of the tunnel surrounding rock deformation field includes: For the multiple deformation field datasets, the deformation values ​​of the same grid node attached to the predefined regular grid cell array are regarded as a sample set. The deformation values ​​of the same grid node are extracted to obtain the deformation value sample set corresponding to the same grid node. Based on the deformation value sample set corresponding to the same grid node, the same probability distribution function as the joint group distribution function is used to obtain the deformation probability function corresponding to the same grid node; Traverse all grid nodes in the predefined regular grid cell array to obtain the deformation probability function corresponding to each grid node; integrate the deformation probability functions corresponding to all grid nodes to obtain the tunnel surrounding rock deformation field probability model, which includes the deformation probability function corresponding to all grid nodes.

7. The method for predicting the deformation probability of tunnel surrounding rock according to claim 5, characterized in that, The step of obtaining the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range at different confidence levels based on the tunnel surrounding rock deformation field probability model includes: Obtain the coordinates of the target prediction point in the tunnel surrounding rock; determine the grid node in the predefined regular grid cell array that has the same coordinates as the target prediction point; find the deformation probability function corresponding to the grid node from the tunnel surrounding rock deformation field probability model to obtain the deformation probability function of the target prediction point; Based on the deformation probability function of the target prediction site, the deformation range of the target prediction site at different confidence levels is predicted.

8. A tunnel surrounding rock deformation probability prediction system, characterized in that, include: The joint distribution function construction module is used to: construct multiple joint distribution functions based on the acquired geographic data of joints in the target rock mass and multiple geometric property features that affect the deformation of the surrounding rock; The feature point combination acquisition module is used to: extract at least two feature points to form an initial feature point group for each joint group distribution function, and sequentially take one feature point from each initial feature point group for arrangement and combination to obtain multiple feature point combinations. The deformation field dataset acquisition module is used to: model joint groups and solve deformation fields using the finite element method based on the combination of the multiple feature points, and obtain multiple gridded tunnel surrounding rock deformation fields. For the multiple gridded tunnel surrounding rock deformation fields, the deformation data of the grid nodes are extracted and the grid is regularized to obtain multiple regularized deformation field datasets; The tunnel surrounding rock deformation field probability model construction module is used to: fit the multiple deformation field datasets with the same probability distribution function as the joint group distribution function to obtain the tunnel surrounding rock deformation field probability model; and based on the tunnel surrounding rock deformation field probability model, obtain the deformation probability function of the target prediction site of the tunnel surrounding rock and the deformation range under different confidence levels. The feature point combination acquisition module is further configured to: For each joint distribution function, a first feature point and a second feature point are extracted. The first feature point is a point where the geometric attribute value of the joint distribution function affecting the surrounding rock deformation is less than the mean value of the joint distribution function. The second feature point is a point where the geometric attribute value of the joint distribution function affecting the surrounding rock deformation is greater than the mean value of the joint distribution function. Then, any feature point is selected from the first and second feature points of each joint distribution function and combined to obtain... There are multiple combinations of feature points, where m is the number of joint distribution functions among the multiple joint distribution functions.

9. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the tunnel surrounding rock deformation probability prediction method provided in any one of claims 1 to 7.

Images