High-stress deep-buried soft rock tunnel large deformation dynamic grading control method and system

Abstract

The application provides a high ground stress deep buried soft rock tunnel large deformation dynamic grading control method and system, relates to the fault zone identification technical field, and selects a plurality of fault core samples along the tunnel axial direction, divides a first distance section according to the vertical distance from the fault core to the tunnel side wall and arranges a drilling point, determines the structural plane development density and the structural plane density calculation interval based on the coring result, respectively fits the structural plane attenuation function in each angle interval under the water-containing and water-free working conditions according to the vertical distance of the fault core sample and the structural plane development density, selects the fault structure main control fissure zone according to the fissure tendency, the dip angle and the trace length in the excavation stage, and obtains the measured value of the structural plane development density, when falling into the structural plane density calculation interval, the measured drilling is arranged, the structural plane development density, the water-containing working condition and the structural plane attenuation function are combined to invert the fault trend and the vertical distance, the risk assessment and the support discrimination are carried out in combination with the historical deformation rate, and the early warning is pushed when the deformation exceeds the threshold value.

Description

Technical Field

[0001] This invention relates to the field of fault zone identification technology, specifically to a dynamic classification control method and system for large deformation of deep-buried soft rock tunnels under high ground stress. Background Technology

[0002] Conventional tunnel support schemes are mostly based on engineering experience in hard or medium-hard rock under medium-low ground stress conditions, using a combination of anchor bolts, steel arch frames and shotcrete. Support parameters are determined by classifying the surrounding rock. However, deeply buried soft rock has inherent characteristics of weak lithology, low strength and poor self-stability. When superimposed with a high ground stress environment, the excavation response of the surrounding rock exhibits complex mechanical behaviors such as strong rheology, large deformation and nonlinearity. It is necessary to predict the large deformation of the tunnel to determine the support parameters.

[0003] In terms of geological exploration and deformation control during tunnel construction, existing advanced geological prediction technology at the tunnel face can predict adverse geological bodies such as faults ahead. Conventional monitoring equipment such as convergence monitoring and stress monitoring are also installed on the tunnel sidewalls to monitor the real-time working status of the support structure.

[0004] However, in soft rock formations with high water content and strong fracturing, elastic wave and electromagnetic wave geophysical exploration methods suffer from rapid energy decay and insufficient accuracy in identifying geological body boundaries. The precise location and strike of fault cores often require excavation to a relatively close distance for effective identification, resulting in insufficient precision in early warning. Furthermore, soft rock within the fault influence zone exhibits significant time-dependent creep characteristics. The deformation of the surrounding rock does not stabilize shortly after excavation but rather develops over a long period due to the continuous release of ground stress and the gradual deterioration of surrounding rock strength caused by water-rock interaction. Existing time-dependent calculation methods for support deformation mostly only consider the creep characteristics of the rock mass itself and do not fully incorporate the stress concentration effect caused by the fault zone, leading to deviations in the prediction of long-term support deformation and making it easy for the support deformation to exceed the safety control standards.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a dynamic graded control method and system for large deformation of deep-buried soft rock tunnels under high ground stress, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A dynamic graded control method for large deformation of deep-buried soft rock tunnels under high ground stress, the specific steps of which include: Step 1: Obtain the orientation of several fault core samples along the tunnel axis and determine the corresponding angle interval. Calculate the vertical distance from each fault core sample to the tunnel sidewall. Divide each fault core sample into a first distance segment along the tunnel axis according to the preset vertical distance interval, and set up boreholes in each first distance segment at equal intervals and in a fixed number. Step 2: Determine the location of each borehole based on the fault strike and fix the drilling direction. Determine the structural surface development density of each borehole point based on the core sampling results. Determine the structural surface density calculation interval based on the structural surface development density of all borehole points. Step 3: Based on the vertical distance from the fault core sample to the tunnel sidewall and the structural surface development density data, the power function attenuation formula is used to fit the structural surface attenuation function in each angle range under water-bearing and waterless conditions respectively. Step 4: During the tunneling stage, the dip direction, dip angle, and trace length of the fractures are input into the trained binary classification model to screen the main control fracture zone of the fault structure. Verification boreholes are arranged perpendicular to the tunnel axis in the main control fracture zone to obtain the measured value of the structural surface development density. Step 5: If the measured structural surface development density falls within the structural surface density calculation range, then set up measured boreholes in the main control fracture zone. Based on the measured structural surface development density and water-bearing conditions of the boreholes, the fault strike and vertical distance are inverted using the structural surface attenuation function in each angle range. Risk assessment and support judgment are carried out in combination with historical deformation rates. When the deformation exceeds the threshold, an early warning is pushed out to complete the closed-loop dynamic control of large deformation in soft rock tunnels.

[0008] Furthermore, the method for dividing each fault core sample along the tunnel axis into first distance segments based on a preset vertical distance interval is as follows: A predefined vertical distance range from the fault core along its normal direction to the tunnel sidewall is defined. For any fault core sample, the angle between its strike and the tunnel central axis on the horizontal projection plane is obtained and recorded as the strike angle. Based on the strike angle, the fault core positions corresponding to the two boundaries of the vertical distance range are determined respectively. The two fault core positions are then converted to the tunnel axis, and the range formed between the two corresponding axial positions is taken as the first distance segment corresponding to the fault core sample.

[0009] Furthermore, the method for determining the location of each borehole and fixing the drilling direction based on the fault strike is as follows: For any fault core sample, the position of each borehole point along the tunnel axis is determined according to a preset fixed number and equal spacing within its corresponding first distance segment. Each borehole adopts a horizontal drilling method, and the direction perpendicular to the tunnel axis and pointing into the surrounding rock in the horizontal projection plane is used as the fixed drilling direction for core sampling to obtain the surrounding rock core sample at the corresponding position.

[0010] Furthermore, the method for determining the structural surface development density at each borehole point based on the core sampling results is as follows: The obtained core is continuously divided into several core statistical segments along the drilling depth direction according to the preset fixed core segment length. The number of effective structural surfaces in each core statistical segment is counted. The ratio of the number of effective structural surfaces to the fixed core segment length is used as the structural surface development density of the corresponding core statistical segment. At the same time, the distance from the midpoint of each core statistical segment to the borehole opening is recorded and used as the midpoint borehole depth of the corresponding core statistical segment.

[0011] Furthermore, the method for fitting the structural surface attenuation function in each angular interval under both water-bearing and waterless conditions is as follows: The core statistical segments corresponding to each known fault core sample are grouped according to water-bearing conditions, water-free conditions, and the angle interval to which they belong. The structural surface development density, the vertical distance from the corresponding borehole opening to the fault core, and the midpoint borehole depth of each core statistical segment are obtained. The vertical distance is the distance from the borehole opening to the fault core along the normal direction of the fault core. Based on the known fault strike and fixed drilling direction, the corresponding normal angle parameter is determined. Combined with the midpoint borehole depth, the normal projection distance of the core statistical segment relative to the borehole opening is determined. The remaining distance from the corresponding core statistical segment to the fault core is obtained by subtracting the normal projection distance from the vertical distance corresponding to the borehole opening. Core statistical segments with a remaining distance greater than zero are determined as valid core statistical segments. For each angle interval under both water-bearing and waterless conditions, a power function-type structural surface attenuation relationship is established with the remaining distance of the corresponding effective core statistical segment as the independent variable and the structural surface development density as the dependent variable. The scale parameter and attenuation index are determined by nonlinear least squares method to obtain the structural surface attenuation function corresponding to each angle interval under both water-bearing and waterless conditions.

[0012] Furthermore, the method for determining the structural surface density calculation interval based on the structural surface development density of all borehole points is as follows: The structural surface development density of the first core statistical segment of each known fault core sample is summarized to construct a structural surface development density sample set. The first core statistical segment is the first core statistical segment obtained from the borehole opening according to a preset fixed core segment length. The density average and density standard deviation of the structural surface development density sample set are calculated, and the structural surface density calculation interval is defined based on the density average and density standard deviation.

[0013] Furthermore, based on the measured density of the structural surfaces at the borehole points and the water-bearing conditions, the method for inverting the fault strike and vertical distance using the structural surface attenuation function in each angular interval is as follows: At least two measured boreholes are set in the main control fracture zone. Each measured borehole adopts the same horizontal drilling method and fixed drilling direction as the structural surface attenuation function fitting stage. The borehole openings of at least two measured boreholes are staggered along the tunnel axis. The position of the borehole opening of one of the measured boreholes is used as the reference position. The vertical distance from the reference position to the core of the fault to be determined and the strike of the fault to be determined are used as the parameters to be determined. Adjust the fault strike and the vertical distance from the reference position to the fault core; determine the angle interval according to the current fault strike, and determine the vertical distance from the borehole opening to the fault core based on the known spatial position of each measured borehole relative to the reference position; determine the normal angle parameter based on the current fault strike and the fixed drilling direction, calculate the remaining distance from the corresponding core statistical segment to the fault core based on the midpoint borehole depth of each core statistical segment, determine the core statistical segments with a remaining distance greater than zero as valid core statistical segments, and call the corresponding structural surface attenuation function according to the water-bearing conditions and the angle interval of each valid core statistical segment to obtain the predicted value of the structural surface development density of each valid core statistical segment; With the goal of minimizing the sum of squared residuals between the predicted and measured values ​​of structural surface development density in each effective core segment of all measured boreholes, the fault strike and the vertical distance from the reference position to the fault core are continuously adjusted to determine the final fault strike and the vertical distance from the reference position to the fault core.

[0014] Furthermore, by combining historical data on the deformation rate of the surrounding rock under the same conditions, the method for determining whether the current support scheme requires an alarm is as follows: Based on the vertical distance from the fault core to the tunnel sidewall and the fault strike obtained from the inversion of the current cross section to be evaluated, historical cross section records are extracted from historical data. The distance difference between the vertical distance from the corresponding fault core to the tunnel sidewall of each historical cross section and the current vertical distance, as well as the angle difference between the strike angle between the corresponding fault strike of each historical cross section and the tunnel axis and the strike angle between the current fault strike and the tunnel axis, are calculated. Historical cross sections with a distance difference not greater than a preset distance matching threshold and an angle difference not greater than a preset angle matching threshold are identified as historical cross sections under the same conditions. Each historical cross section record includes the support stiffness of the corresponding cross section and the time series data of the surrounding rock deformation rate collected from the completion of the initial support to the construction of the secondary lining. For each historical cross section under the same conditions, the maximum deformation rate in the time series data of surrounding rock deformation rate is compared with the preset safety threshold. The historical cross sections with the maximum deformation rate exceeding the preset safety threshold are identified as historical large deformation risk sections. The actual support stiffness used in each historical large deformation risk section is extracted, and the largest support stiffness is identified as the risk support stiffness threshold. The support stiffness of the current support scheme for the section to be evaluated is compared with the risk support stiffness threshold. If the current support stiffness is less than or equal to the risk support stiffness threshold, it is determined that the current support scheme has a risk of large deformation of the surrounding rock and a support warning is issued. The support stiffness of the current section to be evaluated is increased until the adjusted support stiffness is greater than the risk support stiffness threshold. If the current support stiffness is greater than the risk support stiffness threshold, the current support scheme is maintained and the deformation rate of the surrounding rock of the corresponding section is continuously acquired. After the current section to be evaluated is supported, the real-time collected deformation rate of the surrounding rock is compared with the preset safety threshold. When the deformation rate of the surrounding rock exceeds the preset safety threshold, a support warning is pushed out, and the support stiffness of the current support scheme is increased until the deformation rate of the surrounding rock collected later does not exceed the preset safety threshold. After the deformation monitoring of the corresponding section from the initial support to the secondary lining construction stage is completed, the vertical distance from the fault core to the tunnel sidewall, the fault direction, the actual support stiffness adopted, and the corresponding time series data of the surrounding rock deformation rate obtained by the inversion of the section are added to the historical data for subsequent support risk assessment and support identification under the same working conditions.

[0015] Additionally, a dynamic hierarchical control system for large deformation of deep-buried soft rock tunnels under high ground stress is provided. This system is used to execute the dynamic hierarchical control method for large deformation of deep-buried soft rock tunnels under high ground stress described above, including: The reference data module is used to obtain the orientation of several fault core samples along the tunnel axis and determine the angle interval accordingly, calculate the vertical distance from each fault core sample to the tunnel sidewall, divide the first distance segment along the tunnel axis of each fault core sample according to the preset vertical distance interval, and set up drilling points in each first distance segment at equal intervals and in a fixed number. The feature calculation module is used to determine the location of each borehole and fix the drilling direction based on the fault strike, and to determine the structural surface development density of each borehole point based on the core sampling results, and to determine the structural surface density calculation interval based on the structural surface development density of all borehole points. The feature extraction module is used to fit the structural surface attenuation function in each angle range under water-bearing and waterless conditions based on the vertical distance from the fault core sample to the tunnel sidewall and the structural surface development density data, using the power function attenuation formula. The fracture classification module is used during the tunneling stage to input the fracture's dip direction, dip angle, and trace length into the trained binary classification model to screen the main fracture zone controlled by the fault structure, and to arrange verification boreholes perpendicular to the tunnel axis in the main fracture zone to obtain the measured value of the structural surface development density. The fault inversion module is used to set up measured boreholes in the main control fracture zone if the measured structural surface development density falls within the structural surface density calculation range. Based on the structural surface development density and water-bearing conditions of the measured boreholes, the fault strike and vertical distance are inverted by combining the structural surface attenuation function of each angle range. Risk assessment and support discrimination are carried out in combination with historical deformation rates. When the deformation exceeds the threshold, an early warning is pushed to complete the closed-loop dynamic control of large deformation of soft rock tunnels.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention inputs the dip direction, dip angle, and trace length of fractures into a trained binary classification model during the tunneling stage to screen the main fracture zones controlled by fault structures. When the measured values ​​of the structural surface development density fall within the structural surface density calculation range, measured boreholes are set up. Based on the structural surface development density and water-bearing conditions of the measured boreholes, combined with the pre-fitted structural surface attenuation functions for each angle range, the fault strike and vertical distance are inverted. This overcomes the blind spot caused by the rapid attenuation of elastic waves / electromagnetic waves in soft rock, improving the accuracy of early warning of the spatial location of the fault core. At the same time, risk assessment and support discrimination are carried out by combining the vertical distance obtained by inversion with historical deformation rates, and an early warning is pushed when the deformation exceeds the threshold, so that the support scheme can be dynamically adjusted according to the parameters obtained by inversion and the actual deformation feedback. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the dynamic hierarchical control method for large deformation of deep-buried soft rock tunnels under high ground stress according to the present invention. Figure 2 This is a statistical chart showing the degree of change in the overall statistics of this invention; Figure 3 This is a schematic diagram of the overall system structure of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0019] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0020] Example: Please see Figures 1 to 2 The present invention provides a technical solution: A dynamic graded control method for large deformation of deep-buried soft rock tunnels under high ground stress, the specific steps of which include: Step 1: Obtain the orientation of several fault core samples along the tunnel axis and determine the corresponding angle interval. Calculate the vertical distance from each fault core sample to the tunnel sidewall. Divide each fault core sample into a first distance segment along the tunnel axis according to the preset vertical distance interval, and set up boreholes in each first distance segment at equal intervals and in a fixed number. Based on the preliminary geological survey data of the tunnel, several fault cores with known locations and structural properties were selected along the tunnel axis as fault core samples. The spatial location and fault strike of each fault core sample were obtained to determine the extension state of the fault cores along the space surrounding the tunnel. Sample data on the variation of subsequent structural surface development density with the spatial distance of the fault cores were established. The fault strike of each fault core sample and the tunnel central axis were projected onto the same horizontal projection plane, and the angle between them was calculated and recorded as the strike angle. Several continuous angle intervals were preset. In one specific implementation process, in order to reduce the impact of different intersection angles on the structural surface counting results per unit borehole length, the angle intervals were divided into 15° intervals, and the angle interval to which each fault core sample belonged was determined according to the strike angle. The direction perpendicular to the tunnel axis and pointing into the surrounding rock in the horizontal projection plane was used as the fixed drilling direction for core sampling. Subsequent boreholes all used the same fixed drilling direction. Different angle intervals corresponded to different drilling and core sampling intersection conditions between the fixed drilling direction and different fault strikes, which were used to establish the structural surface attenuation function corresponding to each angle interval.

[0021] For any fault core sample, its strike and the tunnel's central axis are projected onto the same horizontal projection plane. Based on the known spatial location and strike of the fault core sample, its spatial relationship along the strike is determined. Since the fault core extends at an angle relative to the tunnel, the spatial distance between the fault core and the tunnel sidewall varies at different positions along the strike. That is, as the fault core's position changes along its strike, its vertical distance along the fault core's normal direction to the adjacent tunnel sidewall changes accordingly.

[0022] Specifically, several location points are determined along the direction of any fault core sample, and the vertical distance from each location point to the adjacent tunnel sidewall along the normal direction of the fault core is determined respectively, so that a fault core sample corresponds to multiple vertical distances that change continuously along its direction, which are used to characterize the spatial distance relationship between different positions of the fault core and the tunnel sidewall.

[0023] To ensure lateral comparability of structural surface data collected from different fault core samples, spatial benchmark constraints must be applied to the exploration area. This is achieved by determining the corresponding positions of the two boundaries of the vertical distance interval along the strike of the fault core sample, and then converting these two corresponding positions to the tunnel axis. Using these two positions along the tunnel axis as the boundary, the resulting axial range is defined as the first distance segment corresponding to the fault core sample. Thus, the first distance segment essentially represents the actual range covered along the tunnel axis when the vertical distance from the fault core to the tunnel sidewall falls within the preset vertical distance interval. Geometrically, this avoids oversampling or undersampling caused by artificially setting a fixed axial length.

[0024] The lower limit of the preset vertical distance interval is set to zero, while the upper limit is determined based on the effective influence range of fault structures on the development of surrounding rock structural surfaces as observed in previous geological surveys. The specific upper limit setting rules are as follows: First, in the previous geological survey, a normal section without fault structures and with consistent lithology is selected as the benchmark area. The background average value and background standard deviation of the structural surface development density within this area are extracted and statistically analyzed. Based on statistical principles, the sum of the mean and twice the standard deviation is set as the critical threshold for background fluctuation. For any fault core sample, along the direction of increasing vertical distance from the fault core normal to the tunnel sidewall, the decay data of structural surface development density with increasing vertical distance are statistically analyzed. When the structural surface development density of the sample is first less than or equal to the critical threshold for background fluctuation, it is determined that the location has approached the structural surface development level of the non-fault-affected area, and the corresponding vertical distance is recorded as the independent effective influence distance of the fault core sample. Finally, the independent effective influence distances of each fault core sample are summarized and arranged in ascending order. The independent effective influence distance corresponding to the preset high quantile is determined as the unified upper limit of the vertical distance interval, preferably using the 95th quantile. The significance of using quantile truncation is that it can both filter out the excessive expansion of the exploration boundary by individual abnormally large distances and ensure that a uniform vertical distance interval can cover the effective influence range of the vast majority of fault core samples.

[0025] The fixed number of boreholes within each first distance segment is further determined. Specifically, several fault cores with structural surface survey data obtained during geological exploration are used as quantity calculation samples. Within the corresponding vertical distance interval, sampling is performed at equal intervals according to different candidate numbers from fewest to most. The average and standard deviation of the obtained structural surface development density are calculated for each level of candidate boreholes. Following the order of candidate borehole numbers from fewest to most, the relative changes in the average and standard deviation of the structural surface development density corresponding to adjacent candidate numbers are calculated, and their sum is taken as the comprehensive statistical change degree between adjacent candidate numbers. The absolute value of the difference between two adjacent comprehensive statistical change degrees is further calculated. The position with the largest absolute value of the difference is determined as the maximum turning point where the statistical characteristic transitions from significant change to stable change. The number of intermediate-level candidate boreholes corresponding to the maximum turning point is determined as the fixed number of boreholes. Figure 2 As shown, the absolute value of the difference between two adjacent comprehensive statistical change degrees is calculated sequentially. The absolute value of the difference between the first two levels of comprehensive statistical change degrees, 0.92 and 0.63, is 0.29, which is greater than the absolute value of the difference between the other adjacent comprehensive statistical change degrees. Therefore, this position is determined as the maximum turning point where the statistical characteristics change from obvious change to stable change, and the corresponding number of intermediate first-level candidate borehole points, 10, is determined as the fixed number of borehole points in the first distance segment.

[0026] The core significance of using this stability optimization method to determine a fixed number of boreholes lies in the following: when the number of boreholes is too small, local high-density or low-density fracture zones can easily lead to distortion of the statistical results of individual fault core samples; while when the number of boreholes is too large, blindly increasing the number of boreholes when the statistical characteristics have already tended to stabilize will have limited gain in identifying attenuation patterns and will only increase the drilling workload unnecessarily. Through this method, the drilling workload can be controlled while ensuring the representativeness of spatial sampling.

[0027] After determining the first distance segment corresponding to each fault core sample and the fixed number of boreholes, the same number of boreholes were laid out at equal intervals within each first distance segment. Since the length of the first distance segment varies depending on the fault strike of different fault core samples, while the total number of boreholes in each first distance segment remains consistent, the actual axial spacing between adjacent boreholes adaptively scales with the length of the first distance segment. This ensures that each borehole can uniformly cover different attenuation gradients from near to far within a unified vertical distance range. Simultaneously, the same number of boreholes is used in each first distance segment to reduce the initial sampling quantity differences caused by varying first distance segment lengths, ensuring a relatively balanced sample contribution from different fault core samples in subsequent statistical analysis and structural surface attenuation function fitting.

[0028] Simultaneously record the vertical distance from the borehole opening at the axial position of each borehole point to the fault core along the normal direction of the fault core. This allows for the establishment of a correspondence between the vertical distance of each borehole point and the structural surface development density obtained from subsequent borehole drilling, providing basic data for fitting the attenuation relationship of structural surface development density with vertical distance.

[0029] Step 2: Determine the location of each borehole based on the fault strike and fix the drilling direction. Determine the structural surface development density of each borehole point based on the core sampling results. Determine the structural surface density calculation interval based on the structural surface development density of all borehole points. Based on the determined position of the borehole points along the tunnel axis, the borehole opening position is determined on the corresponding tunnel sidewall. The borehole opening position is used to characterize the starting position for actual core drilling. For the boreholes corresponding to each fault core sample, a horizontal drilling method is adopted, and the direction perpendicular to the tunnel axis and pointing towards the interior of the surrounding rock in the horizontal projection plane is used as a fixed drilling direction. Each borehole is drilled and cored according to the fixed drilling direction, and the drilling orientation is not adjusted according to different fault strikes, so that the structural surface development density corresponding to each known fault core sample is obtained under uniform drilling and cored conditions.

[0030] After drilling is completed, several core statistical segments are continuously divided along the borehole depth direction with a preset fixed core segment length. Natural effective structural surfaces formed by geological processes such as fault shearing and tectonic fracturing within each statistical segment are then identified and counted. The fixed core segment length aims to balance the statistical stability and spatial resolution of the structural surfaces. If the length is too small, the appearance or absence of a single sporadic fracture can cause drastic fluctuations in density values; if it is too large, it will produce a segment averaging effect, smoothing out the sensitive gradient of density decay with distance near the fault core.

[0031] This scheme determines the fixed core segment length based on the maximum inflection point between the candidate core segment length and the degree of change in structural surface density. The specific method is as follows: Several candidate core segment lengths are pre-set in ascending order. Each continuous core sample is segmented according to its candidate length, and the structural surface development density of each statistical segment is calculated. For any candidate length, the absolute value of the density difference between all adjacent statistical segments is calculated and averaged. This average is then compared with the global density average for that candidate length to characterize the degree of change in structural surface density. An evolutionary relationship between candidate length and density change degree is established according to the ascending order of candidate lengths. The rate of change of the degree of change between adjacent candidate lengths, and the difference between adjacent rates of change, are calculated sequentially. The candidate core segment length corresponding to the largest difference in rate of change is extracted and determined as the fixed core segment length.

[0032] For each pre-defined core segment, the ratio of the number of effective structural planes to the fixed core segment length is used as the structural plane development density for that segment. Simultaneously, the distance from the midpoint of the core segment to the borehole opening is extracted and recorded as the midpoint borehole depth. To establish a unified benchmark for structural plane development density under known fault structural influence conditions, the structural plane development densities of the first core segment of each borehole corresponding to a known fault core sample are compiled, constructing a structural plane development density sample set. The mean density and standard deviation are calculated, and the structural plane density calculation interval is defined by adding or subtracting two times the standard deviation from the mean, serving as a benchmark for subsequent verification of measured borehole structural plane development density values.

[0033] Step 3: Based on the vertical distance from the fault core sample to the tunnel sidewall and the structural surface development density data, the power function attenuation formula is used to fit the structural surface attenuation function in each angle range under water-bearing and waterless conditions respectively. For any given core statistical segment, the fault core sample, borehole, structural surface development density, midpoint borehole depth, and corresponding water-bearing conditions are recorded. The angle interval corresponding to the fault core sample and the vertical distance from the borehole opening to the fault core are also obtained. This vertical distance is the distance from the borehole opening along the normal direction of the fault core to the fault core. Thus, each core statistical segment corresponds to a set of basic sample data consisting of vertical distance, midpoint borehole depth, structural surface development density, water-bearing conditions, and the corresponding angle interval.

[0034] For core statistical sections where groundwater is present at the corresponding location, such as water inflow, continuous seepage, obvious dripping, or continuous wetting of the core, the core statistical section is classified as a water-bearing condition sample; for those where no obvious groundwater is present at the corresponding location, it is classified as a waterless condition sample.

[0035] For any known fault core sample, the normal direction of the fault core is determined in the horizontal projection plane according to its fault strike, and the angle parameter between the normal direction of the fault core and the fixed drilling direction is determined.

[0036] For any obtained core statistical segment, the midpoint borehole depth is extracted, and the normal projection distance of the midpoint of the core statistical segment relative to the borehole opening along the fault core normal direction is determined based on the midpoint borehole depth and the normal angle parameter. In this embodiment, all core boreholes used to establish the structural plane attenuation relationship are drilled horizontally; therefore, the product of the midpoint borehole depth and the cosine value of the normal angle parameter is used as the corresponding normal projection distance.

[0037] During drilling, each core borehole extends from the tunnel sidewall into the surrounding rock along a fixed drilling direction. Based on the known geometric relationship between the fault strike and the fixed drilling direction, the normal projection distance of the midpoint of the core statistical section relative to the borehole opening along the fault core normal direction is subtracted from the vertical distance from the borehole opening to the fault core to obtain the remaining distance from the corresponding core statistical section to the fault core. The formula for calculating the remaining distance is as follows: In the formula, This indicates the remaining distance from the core statistical segment to the fault core; This indicates the vertical distance from the borehole opening to the fault core along the normal direction of the fault core; This indicates the borehole depth at the midpoint of the corresponding core sample segment; This parameter represents the normal angle between the projection direction of the borehole axis in the horizontal projection plane and the normal direction of the fault core. The product of the cosine value of the included angle parameter represents the normal projection distance of the midpoint of the core statistical segment relative to the borehole opening along the normal direction of the fault core. It should be noted that when the remaining distance corresponding to the core statistical segment is greater than zero, the core statistical segment is determined as a valid core statistical segment and used for subsequent structural surface attenuation function fitting. When the remaining distance is less than or equal to zero, it indicates that the corresponding core statistical segment has reached or crossed the fault core position determined according to the geometric relationship, and the core statistical segment is not included in the fitting data of the power function type structural surface attenuation relationship.

[0038] For the same borehole, since the borehole depth at the midpoint of each core statistical segment increases sequentially, the remaining distance corresponding to each core statistical segment changes sequentially, allowing a borehole to generate multiple structural surface development density samples at different remaining distance positions.

[0039] After calculating the remaining distance, the remaining distance of each effective core statistical segment and the corresponding structural plane development density are combined to form a remaining distance-structural plane development density data pair. This pair is then grouped according to water-bearing conditions, water-free conditions, and their respective angle intervals, creating structural plane attenuation fitting datasets for each angle interval under water-bearing conditions and under water-free conditions. This results in a unified data scale using the remaining distance from the core statistical segment to the fault core as the independent variable.

[0040] The density of structural plane development gradually decreases as the core statistical section moves further away from the fault core. Therefore, a power function is used to describe the attenuation relationship between the density of structural plane development and the remaining distance. The formula for calculating the attenuation function of the structural plane is as follows: In the formula, This indicates the density of structural plane development in the core statistical segment. This represents a scale parameter used to characterize the overall intensity level of structural surface development in the region adjacent to the fault core. The decay index is used to characterize how quickly the density of structural surfaces decays as the remaining distance increases. For any angular interval under water-bearing conditions, a power-law-type structural surface attenuation relationship is established using the remaining distance of each effective core statistical segment within that interval as the independent variable and the corresponding structural surface development density as the dependent variable. The scale parameter and attenuation index corresponding to that angular interval are determined using the nonlinear least squares method. Specifically, the predicted structural surface development density of each effective core statistical segment is calculated based on the scale parameter and attenuation index. The residual between the predicted value and the measured structural surface development density is calculated. The objective is to minimize the sum of squares of the residuals of all effective core statistical segments within that angular interval. The scale parameter and attenuation index are iteratively adjusted to obtain the structural surface attenuation function corresponding to that angular interval under water-bearing conditions. In the formula, This indicates the density of structural surfaces under water-bearing conditions. Indicates the water content scale parameter, Indicates the water content decay index; For each angular interval under waterless conditions, the corresponding scale parameters and attenuation exponents are determined using the same method to obtain the structural surface attenuation function for each angular interval under waterless conditions. Thus, the structural surface attenuation function for each angular interval under waterless conditions is formed: In the formula, This indicates the density of structural surfaces under waterless conditions. Indicates the anhydrous scale parameter, This indicates the waterless decay index.

[0041] Both the water-bearing and waterless conditions employ the same power function structure, but for each angular interval, the scale parameters and attenuation exponents are independently determined using the corresponding effective core statistical segments. This results in the structural surface attenuation functions for each angular interval under both water-bearing and waterless conditions. The scale parameters and attenuation exponents of the obtained structural surface attenuation functions remain fixed during the subsequent fault inversion process of the main control fracture zone. During the inversion stage, the angular interval to which the fault belongs is determined based on the currently adjusted fault strike, and the structural surface attenuation functions for the corresponding water-bearing condition and angular interval are called. By continuously adjusting the fault strike and the vertical distance from the reference position to the fault core, the final fault strike and vertical distance are obtained through inversion.

[0042] Step 4: During the tunneling stage, the dip direction, dip angle, and trace length of the fractures are input into the trained binary classification model to screen the main control fracture zone of the fault structure. Verification boreholes are arranged perpendicular to the tunnel axis in the main control fracture zone to obtain the measured value of the structural surface development density. First, a binary classification model for selecting fault-controlled fractures is pre-trained. The dip direction, dip angle, and trace length of each historical fracture are extracted to construct a three-dimensional feature vector. The training sample set of the model strictly reuses two types of areas from previous explorations: one is fractures within designated normal sections without fault structures and with consistent lithology, exhibiting a disordered and random state, such as those generated by unloading or blasting; these are labeled as negative samples (label 0). The other is fractures within a predetermined vertical distance interval corresponding to each known fault core, such as those induced by shear slip, exhibiting a concentrated dominant attitude; these are labeled as positive samples (label 1). This labeled dataset is then input into algorithms such as Support Vector Machine or Random Forest for iterative training to obtain the trained binary classification model.

[0043] During the dynamic excavation phase of soft rock tunnels, as the surrounding rock is gradually exposed, surface fracture mapping is continuously carried out to obtain the dip direction, dip angle, trace length, and spatial location of each identifiable fracture. The three-dimensional feature vectors are then input into the binary classification model in real time, causing the model to exclude fractures identified as non-fault-controlled fractures and retain fractures identified as fault-controlled fractures and their spatial locations.

[0044] The retained fault-controlled fracture traces are mapped onto the tunnel surface to determine spatial continuity. When two fault-controlled fracture traces overlap, or the minimum spatial distance between the ends of the two traces is not greater than the length of the shorter fracture, they are considered spatially continuous fractures. Based on the transitivity of the spatial continuity, all continuously connected fault-controlled fractures are grouped into the same fracture set. For any fracture set, the foremost and last ends along the tunnel axis, as well as the two outermost positions along the tunnel circumference, are determined based on the spatial positions of the endpoints of all fracture traces within the set on the tunnel surface. The tunnel surface area enclosed by these positions is defined as the corresponding fault-controlled fracture zone. Isolated fractures that do not have spatial continuity with other fault-controlled fractures do not form separate fault-controlled fracture zones.

[0045] For any given fault structure controlling a fracture zone, the midpoints of the traces of the controlling fractures within the fault structure are extracted, and the spatial positions of the midpoints of each trace on the tunnel surface are obtained. Using the trace length of each controlling fracture as a weight, the midpoints of each trace are sequentially selected as candidate verification locations. For any candidate verification location, the spatial distance between it and the midpoints of the remaining traces within the controlling fracture zone is calculated, and then multiplied by the corresponding fracture trace length and summed to obtain the weighted distance sum of the candidate verification location. The midpoint of the trace with the smallest weighted distance sum is determined as the final verification location. When there are multiple candidate verification locations with the same weighted distance sum and all of them being the minimum value, the midpoint of the trace with the larger corresponding fracture trace length is selected first. When the corresponding fracture trace lengths are still the same, the midpoint of the trace reached first along the tunnel excavation direction is selected as the final verification location. Thus, a verification location located on the tunnel surface is determined. The significance of this algorithm lies in making the verification location closer to the main control fractures of each fault structure within the main control fracture zone, and increasing the influence of the main control fractures of fault structures with larger trace lengths on the verification location determination results.

[0046] The final verification location is used as the borehole opening location for verification drilling. Verification drilling is carried out using the same horizontal drilling method and fixed drilling direction. After drilling and coring are completed, the first core statistical segment with a fixed core length is cut off from the borehole opening. The ratio of the number of effective structural surfaces to the fixed core segment length is used as the measured value of the structural surface development density at the verification borehole point, which serves as the core pre-signal for subsequent inversion triggering.

[0047] Step 5: If the measured structural surface development density falls within the structural surface density calculation range, then set up measured boreholes in the main control fracture zone. Based on the measured structural surface development density and water-bearing conditions of the boreholes, combine the fault strike and vertical distance inverted by the structural surface attenuation function of each angle range, and combine the historical deformation rate to carry out risk assessment and support judgment. When the deformation exceeds the threshold, push the early warning to complete the closed-loop dynamic control of large deformation of soft rock tunnel. First, the measured values ​​of structural surface development density obtained from the verification borehole points are compared with the determined structural surface density calculation interval. When the measured values ​​of structural surface development density fall within the structural surface density calculation interval, it indicates that the near-hole structural surface development degree of the current fault structure-controlled fracture zone has the same density characteristics as the previously known fault structure-affected area. The fault structure-controlled fracture zone is then identified as the area to be inverted, and further measured borehole core sampling is carried out. When the measured values ​​of structural surface development density do not fall within the structural surface density calculation interval, dynamic screening of subsequently revealed fractures continues as the tunnel is excavated.

[0048] For the identified inversion area, the verification borehole is used as the first measured borehole, and drilling continues into the surrounding rock. Simultaneously, at least one new measured borehole is added within the main control fracture zone, ensuring a total of at least two measured boreholes participate in the joint inversion. Each measured borehole uses the same horizontal drilling method and fixed drilling direction as in the structural surface attenuation function fitting stage, and the borehole openings of at least two measured boreholes are staggered along the tunnel axis, meaning the corresponding borehole openings are at different positions along the tunnel axis. When multiple new measured boreholes are added, they are spaced apart within the main control fracture zone along the tunnel axis. The core samples obtained from each measured borehole are continuously divided according to the same fixed core segment length as in step 2. The midpoint borehole depth, structural surface development density, and water-bearing conditions of each core segment are obtained, and the borehole opening position of one of the measured boreholes is used as a reference position. The vertical distance from the reference position to the fault core and the fault strike are used as parameters to be determined, and the fault strike and the vertical distance from the reference position to the fault core are continuously adjusted.

[0049] For each adjusted fault strike, the corresponding angle interval is determined based on the current fault strike. Combined with the known spatial position of each measured borehole opening relative to the reference position and the vertical distance from the reference position to the fault core, the vertical distance from each measured borehole opening along the normal direction to the fault core is determined. The normal angle parameter is determined based on the current fault strike and the fixed drilling direction. Then, the remaining distance from the corresponding core statistical segment to the fault core is calculated based on the midpoint borehole depth of each core statistical segment. In the formula, Indicates the first The first measured borehole The remaining distance from each core segment to the fault core. Indicates the first The measured vertical distance from the borehole opening to the fault core along the normal direction of the current fault core. Indicates the first The first measured borehole The midpoint borehole depth of each core sample segment Indicates the first The angle between the fixed drilling direction of a measured borehole and the normal direction of the fault core determined based on the current fault strike. This indicates the sequence number of each measured borehole drilled within the main control fracture zone. This indicates the sequence number of a core statistical segment continuously divided along the borehole depth within any measured borehole. When the value is greater than zero, the corresponding core statistical segment is determined as a valid core statistical segment; when... When the value is less than or equal to zero, the corresponding core statistical segment is not included in the structural surface attenuation function prediction and residual calculation. Based on the water-bearing conditions of each effective core segment and the angle range of the current fault strike, the fitted structural surface attenuation function is called. The scale parameter and attenuation index are substituted as known constants to calculate the predicted structural surface development density of each effective core segment. With the goal of minimizing the sum of squared residuals between the predicted and measured values ​​of structural surface development density of each effective core segment in all measured boreholes, the fault strike and the vertical distance from the reference position to the fault core are continuously adjusted until the sum of squared residuals reaches the minimum. The final fault strike and the vertical distance from the reference position to the fault core are then determined.

[0050] The inversion vertical distance and the strike angle between the current fault strike and the tunnel axis are extracted from the current cross-section and used as historical condition matching conditions. Historical cross-sections are first extracted from the historical cross-section database. The distance difference between the vertical distance from the fault core to the tunnel sidewall and the current inversion vertical distance, as well as the angle difference between the historical strike angle and the current strike angle, are calculated for each historical cross-section. When the distance difference is less than or equal to a preset distance matching threshold and the angle difference is less than or equal to a preset angle matching threshold, the corresponding historical cross-section is determined as a historical cross-section under the same conditions. The distance matching threshold and angle matching threshold are determined based on the inversion error of this method on known fault core verification samples: the spatial parameters of the samples are extracted as true values, the vertical distance and strike angle are inverted, and the absolute errors of the vertical distance and strike angle are statistically analyzed based on the true values. Their 95th percentiles are used as the distance matching threshold and angle matching threshold, respectively. This method ensures that the allowable range for historical condition matching is consistent with the spatial parameter identification accuracy of this method itself. When the difference between two cross-sections does not exceed the uncertainty range of the inversion method itself, they are considered to have similar fault spatial conditions.

[0051] For each historical cross section with the same conditions selected, the time series data of the surrounding rock deformation rate from the completion of the initial support to the construction of the secondary lining is extracted, and the maximum surrounding rock deformation rate is extracted. When the maximum surrounding rock deformation rate exceeds the preset safety threshold, the corresponding historical cross section is identified as a historical large deformation risk cross section. The preset safety threshold is determined according to the actual large deformation treatment results of the historical cross section: cross sections that have undergone secondary reinforcement, support replacement or other large deformation treatments in the past are marked as large deformation event cross sections, and cross sections that have not undergone the above treatments are used as control cross sections. The maximum surrounding rock deformation rate of each historical cross section is extracted, and the set candidate deformation rate thresholds are traversed. The identification sensitivity of each candidate threshold for historical large deformation event cross sections and the identification specificity of the control cross sections are calculated. The candidate deformation rate with the largest Youden index is determined as the preset safety threshold.

[0052] For the combined shotcrete, steel frame, and anchor bolt support systems actually used in each historical section, the support pressure-radial deformation relationship of the corresponding combined support system is determined according to the elastic modulus and thickness of the shotcrete, the elastic modulus, cross-sectional parameters, and longitudinal spacing of the steel frame, and the elastic modulus, cross-sectional area, length, and arrangement spacing of the anchor bolts, based on a unified support mechanics model, and a support characteristic curve is formed. The radial deformation is uniformly based on the state when the initial support begins to bear the surrounding rock load as the deformation starting benchmark. For any combined support scheme, the radial deformation corresponding to the initial support beginning to bear the surrounding rock load is taken as the lower limit of the effective radial deformation range, and the radial deformation corresponding to the first time the support pressure on the support characteristic curve reaches the design allowable bearing pressure of the combined support scheme is taken as the upper limit of the effective radial deformation range. The radial deformation range between the lower and upper limits is determined as the effective radial deformation range of the combined support scheme. The intersection of each effective radial deformation range is extracted as the common effective deformation range, and the average value of the upper and lower limits of the common effective deformation range is determined as the unified benchmark radial deformation. The support pressure corresponding to the reference radial deformation is read from the support characteristic curve of each combined support scheme. The ratio of the support pressure to the reference radial deformation is determined as the support stiffness of the corresponding combined support scheme. All combined support schemes of historical sections and current sections to be evaluated use the same unified support mechanics model and the same reference radial deformation to calculate the support stiffness, so as to ensure the comparability of stiffness between different combined support schemes. At the same time, the median of the common effective deformation range is used to avoid the reference radial deformation from approaching the range boundary, thereby reducing the influence of the boundary state of the support characteristic curve on the support stiffness calculation results.

[0053] The maximum support stiffness among all historical high-deformation risk sections is determined as the risk support stiffness threshold. This threshold represents the maximum support stiffness at which the surrounding rock deformation rate exceeds a preset safety threshold under historical conditions similar to the current fault structure. For the proposed combined support scheme for the current section to be evaluated, the current support stiffness is calculated using the same support mechanics model and the same baseline radial deformation as for historical sections. When the current support stiffness is less than or equal to the risk support stiffness threshold, a support warning is issued, and a support scheme with higher stiffness is selected from the preset support scheme levels. The stiffness is recalculated and compared until the adjusted support stiffness exceeds the risk support stiffness threshold. When the current support stiffness exceeds the risk support stiffness threshold, initial support is implemented according to the current support scheme, and the surrounding rock deformation rate monitoring stage begins. When no historical cross-sections with the same conditions are obtained through screening, or when there are no historical cross-sections with large deformation risks among the historical cross-sections with the same conditions, no risk support stiffness threshold is generated. Initial support is implemented according to the current design support scheme, and the process directly enters the surrounding rock deformation rate monitoring stage.

[0054] After the initial support is implemented, the deformation rate of the surrounding rock at the current cross-section is continuously acquired. When the deformation rate of the surrounding rock exceeds the preset safety threshold in any monitoring period, a support warning is issued and support reinforcement is implemented. After the support reinforcement is completed, the deformation rate of the surrounding rock continues to be monitored until the subsequent deformation rate of the surrounding rock does not exceed the preset safety threshold. After the deformation monitoring from the completion of the initial support to the construction of the secondary lining is completed, the inversion vertical distance, fault strike, support stiffness of the final actual combined support scheme, and time series data of the surrounding rock deformation rate of the current cross-section are used to form a new historical cross-section record and supplement it to the historical cross-section database for subsequent cross-section support risk assessment and support identification.

[0055] Please see Figure 3 The present invention also provides a dynamic hierarchical control system for large deformation of deep-buried soft rock tunnels under high ground stress, used to execute the dynamic hierarchical control method for large deformation of deep-buried soft rock tunnels under high ground stress described in any of the above claims, comprising: The reference data module is used to obtain the orientation of several fault core samples along the tunnel axis and determine the angle interval accordingly, calculate the vertical distance from each fault core sample to the tunnel sidewall, divide the first distance segment along the tunnel axis of each fault core sample according to the preset vertical distance interval, and set up drilling points in each first distance segment at equal intervals and in a fixed number. The feature calculation module is used to determine the location of each borehole and fix the drilling direction based on the fault strike, and to determine the structural surface development density of each borehole point based on the core sampling results, and to determine the structural surface density calculation interval based on the structural surface development density of all borehole points. The feature extraction module is used to fit the structural surface attenuation function in each angle range under water-bearing and waterless conditions based on the vertical distance from the fault core sample to the tunnel sidewall and the structural surface development density data, using the power function attenuation formula. The fracture classification module is used during the tunneling stage to input the fracture's dip direction, dip angle, and trace length into the trained binary classification model to screen the main fracture zone controlled by the fault structure, and to arrange verification boreholes perpendicular to the tunnel axis in the main fracture zone to obtain the measured value of the structural surface development density. The fault inversion module is used to set up measured boreholes in the main control fracture zone if the measured structural surface development density falls within the structural surface density calculation range. Based on the structural surface development density and water-bearing conditions of the measured boreholes, the fault strike and vertical distance are inverted by combining the structural surface attenuation function of each angle range. Risk assessment and support discrimination are carried out in combination with historical deformation rates. When the deformation exceeds the threshold, an early warning is pushed to complete the closed-loop dynamic control of large deformation of soft rock tunnels.

[0056] The above formulas are all dimensionless calculations, and the preset parameters in the formulas should be set by those skilled in the art according to the actual situation.

[0057] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0058] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0059] 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 dynamic graded control method for large deformation of deep-buried soft rock tunnels under high ground stress, characterized in that, The specific steps include: Step 1: Obtain the orientation of several fault core samples along the tunnel axis and determine the corresponding angle interval. Calculate the vertical distance from each fault core sample to the tunnel sidewall. Divide each fault core sample into a first distance segment along the tunnel axis according to the preset vertical distance interval, and set up boreholes in each first distance segment at equal intervals and in a fixed number. Step 2: Determine the location of each borehole based on the fault strike and fix the drilling direction. Determine the structural surface development density of each borehole point based on the core sampling results. Determine the structural surface density calculation interval based on the structural surface development density of all borehole points. Step 3: Based on the vertical distance from the fault core sample to the tunnel sidewall and the structural surface development density data, the power function attenuation formula is used to fit the structural surface attenuation function in each angle range under water-bearing and waterless conditions respectively. Step 4: During the tunneling stage, the dip direction, dip angle, and trace length of the fractures are input into the trained binary classification model to screen the main control fracture zone of the fault structure. Verification boreholes are arranged perpendicular to the tunnel axis in the main control fracture zone to obtain the measured value of the structural surface development density. Step 5: If the measured structural surface development density falls within the structural surface density calculation range, then set up measured boreholes in the main control fracture zone. Based on the measured structural surface development density and water-bearing conditions of the boreholes, the fault strike and vertical distance are inverted using the structural surface attenuation function in each angle range. Risk assessment and support judgment are carried out in combination with historical deformation rates. When the deformation exceeds the threshold, an early warning is pushed out to complete the closed-loop dynamic control of large deformation in soft rock tunnels.

2. The method for dynamic graded control of large deformation in deep-buried soft rock tunnels under high ground stress according to claim 1, characterized in that: The method for dividing the first distance segment along the tunnel axis of each fault core sample according to the preset vertical distance interval is as follows: A predefined vertical distance range from the fault core along its normal direction to the tunnel sidewall is defined. For any fault core sample, the angle between its strike and the tunnel central axis on the horizontal projection plane is obtained and recorded as the strike angle. Based on the strike angle, the fault core positions corresponding to the two boundaries of the vertical distance range are determined respectively. The two fault core positions are then converted to the tunnel axis, and the range formed between the two corresponding axial positions is taken as the first distance segment corresponding to the fault core sample.

3. The method for dynamic graded control of large deformation in deep-buried soft rock tunnels under high ground stress according to claim 2, characterized in that: The method for determining the location of each borehole and fixing the drilling direction based on the fault strike is as follows: For any fault core sample, the position of each borehole point along the tunnel axis is determined according to a preset fixed number and equal spacing within its corresponding first distance segment. Each borehole adopts a horizontal drilling method, and the direction perpendicular to the tunnel axis and pointing into the surrounding rock in the horizontal projection plane is used as the fixed drilling direction for core sampling to obtain the surrounding rock core sample at the corresponding position.

4. The method for dynamic graded control of large deformation in deep-buried soft rock tunnels under high ground stress according to claim 3, characterized in that: The method for determining the structural surface development density at each borehole point based on core sampling results is as follows: The obtained core is continuously divided into several core statistical segments along the drilling depth direction according to the preset fixed core segment length. The number of effective structural surfaces in each core statistical segment is counted. The ratio of the number of effective structural surfaces to the fixed core segment length is used as the structural surface development density of the corresponding core statistical segment. At the same time, the distance from the midpoint of each core statistical segment to the borehole opening is recorded and used as the midpoint borehole depth of the corresponding core statistical segment.

5. The method for dynamic graded control of large deformation in deep-buried soft rock tunnels under high ground stress according to claim 4, characterized in that: The method for fitting the structural surface attenuation function in each angle range under water-containing and waterless conditions is as follows: The core statistical segments corresponding to each known fault core sample are grouped according to water-bearing conditions, water-free conditions, and the angle interval to which they belong. The structural surface development density, the vertical distance from the corresponding borehole opening to the fault core, and the midpoint borehole depth of each core statistical segment are obtained. The vertical distance is the distance from the borehole opening to the fault core along the normal direction of the fault core. Based on the known fault strike and fixed drilling direction, the corresponding normal angle parameter is determined. Combined with the midpoint borehole depth, the normal projection distance of the core statistical segment relative to the borehole opening is determined. The remaining distance from the corresponding core statistical segment to the fault core is obtained by subtracting the normal projection distance from the vertical distance corresponding to the borehole opening. Core statistical segments with a remaining distance greater than zero are determined as valid core statistical segments. For each angle interval under both water-bearing and waterless conditions, a power function-type structural surface attenuation relationship is established with the remaining distance of the corresponding effective core statistical segment as the independent variable and the structural surface development density as the dependent variable. The scale parameter and attenuation index are determined by nonlinear least squares method to obtain the structural surface attenuation function corresponding to each angle interval under both water-bearing and waterless conditions.

6. The method for dynamic graded control of large deformation in deep-buried soft rock tunnels under high ground stress according to claim 4, characterized in that: The method for determining the structural surface density calculation interval based on the structural surface development density of all borehole points is as follows: The structural surface development density of the first core statistical segment of each known fault core sample is summarized to construct a structural surface development density sample set. The first core statistical segment is the first core statistical segment obtained from the borehole opening according to a preset fixed core segment length. The density average and density standard deviation of the structural surface development density sample set are calculated, and the structural surface density calculation interval is defined based on the density average and density standard deviation.

7. The method for dynamic graded control of large deformation in deep-buried soft rock tunnels under high ground stress according to claim 5, characterized in that: Based on the measured density of the structural surfaces at the borehole points and the water-bearing conditions, the method for inverting the fault strike and vertical distance using the structural surface attenuation function in each angle interval is as follows: At least two measured boreholes are set in the main control fracture zone. Each measured borehole adopts the same horizontal drilling method and fixed drilling direction as the structural surface attenuation function fitting stage. The borehole openings of at least two measured boreholes are staggered along the tunnel axis. The position of the borehole opening of one of the measured boreholes is used as the reference position. The vertical distance from the reference position to the core of the fault to be determined and the strike of the fault to be determined are used as the parameters to be determined. Adjust the fault strike and the vertical distance from the reference position to the fault core; determine the angle interval according to the current fault strike, and determine the vertical distance from the borehole opening to the fault core based on the known spatial position of each measured borehole relative to the reference position; determine the normal angle parameter based on the current fault strike and the fixed drilling direction, calculate the remaining distance from the corresponding core statistical segment to the fault core based on the midpoint borehole depth of each core statistical segment, determine the core statistical segments with a remaining distance greater than zero as valid core statistical segments, and call the corresponding structural surface attenuation function according to the water-bearing conditions and the angle interval of each valid core statistical segment to obtain the predicted value of the structural surface development density of each valid core statistical segment; With the goal of minimizing the sum of squared residuals between the predicted and measured values ​​of structural surface development density in each effective core segment of all measured boreholes, the fault strike and the vertical distance from the reference position to the fault core are continuously adjusted to determine the final fault strike and the vertical distance from the reference position to the fault core.

8. The method for dynamic graded control of large deformation in deep-buried soft rock tunnels under high ground stress according to claim 7, characterized in that: The method for determining whether the current support scheme requires an alarm is as follows, based on the surrounding rock deformation rate data under the same conditions: Based on the vertical distance from the fault core to the tunnel sidewall and the fault strike obtained from the inversion of the current cross section to be evaluated, historical cross section records are extracted from historical data. The distance difference between the vertical distance from the corresponding fault core to the tunnel sidewall of each historical cross section and the current vertical distance, as well as the angle difference between the strike angle between the corresponding fault strike of each historical cross section and the tunnel axis and the strike angle between the current fault strike and the tunnel axis, are calculated. Historical cross sections with a distance difference not greater than a preset distance matching threshold and an angle difference not greater than a preset angle matching threshold are identified as historical cross sections under the same conditions. Each historical cross section record includes the support stiffness of the corresponding cross section and the time series data of the surrounding rock deformation rate collected from the completion of the initial support to the construction of the secondary lining. For each historical cross section under the same conditions, the maximum deformation rate in the time series data of surrounding rock deformation rate is compared with the preset safety threshold. The historical cross sections with the maximum deformation rate exceeding the preset safety threshold are identified as historical large deformation risk sections. The actual support stiffness used in each historical large deformation risk section is extracted, and the largest support stiffness is identified as the risk support stiffness threshold. The support stiffness of the current support scheme for the section to be evaluated is compared with the risk support stiffness threshold. If the current support stiffness is less than or equal to the risk support stiffness threshold, it is determined that the current support scheme has a risk of large deformation of the surrounding rock and a support warning is issued. The support stiffness of the current section to be evaluated is increased until the adjusted support stiffness is greater than the risk support stiffness threshold. If the current support stiffness is greater than the risk support stiffness threshold, the current support scheme is maintained and the deformation rate of the surrounding rock of the corresponding section is continuously acquired. After the current section to be evaluated is supported, the real-time collected deformation rate of the surrounding rock is compared with the preset safety threshold. When the deformation rate of the surrounding rock exceeds the preset safety threshold, a support warning is pushed out, and the support stiffness of the current support scheme is increased until the deformation rate of the surrounding rock collected later does not exceed the preset safety threshold. After the deformation monitoring of the corresponding section from the initial support to the secondary lining construction stage is completed, the vertical distance from the fault core to the tunnel sidewall, the fault direction, the actual support stiffness adopted, and the corresponding time series data of the surrounding rock deformation rate obtained by the inversion of the section are added to the historical data for subsequent support risk assessment and support identification under the same working conditions.

9. A dynamic hierarchical control system for large deformation of deep-buried soft rock tunnels under high ground stress, characterized in that: The system is used to execute the dynamic graded control method for large deformation of deep-buried soft rock tunnels under high ground stress as described in any one of claims 1-8: The reference data module is used to obtain the orientation of several fault core samples along the tunnel axis and determine the angle interval accordingly, calculate the vertical distance from each fault core sample to the tunnel sidewall, divide the first distance segment along the tunnel axis of each fault core sample according to the preset vertical distance interval, and set up drilling points in each first distance segment at equal intervals and in a fixed number. The feature calculation module is used to determine the location of each borehole and fix the drilling direction based on the fault strike, and to determine the structural surface development density of each borehole point based on the core sampling results, and to determine the structural surface density calculation interval based on the structural surface development density of all borehole points. The feature extraction module is used to fit the structural surface attenuation function in each angle range under water-bearing and waterless conditions based on the vertical distance from the fault core sample to the tunnel sidewall and the structural surface development density data, using the power function attenuation formula. The fracture classification module is used during the tunneling stage to input the fracture's dip direction, dip angle, and trace length into the trained binary classification model to screen the main fracture zone controlled by the fault structure, and to arrange verification boreholes perpendicular to the tunnel axis in the main fracture zone to obtain the measured value of the structural surface development density. The fault inversion module is used to set up measured boreholes in the main control fracture zone if the measured structural surface development density falls within the structural surface density calculation range. Based on the structural surface development density and water-bearing conditions of the measured boreholes, the fault strike and vertical distance are inverted by combining the structural surface attenuation function of each angle range. Risk assessment and support discrimination are carried out in combination with historical deformation rates. When the deformation exceeds the threshold, an early warning is pushed to complete the closed-loop dynamic control of large deformation of soft rock tunnels.

Images