Water-rich composite stratum tunnel surrounding rock deformation identification method

Abstract

The invention discloses a water-rich composite stratum tunnel surrounding rock deformation identification method, and relates to the technical field of tunnel engineering, and the method comprises the following steps: collecting stratum information and hydrological data in a geological exploration stage, and on-site monitoring data in a construction stage, carrying out the cleaning and normalization processing of the data, constructing a training set of a random forest model, and carrying out the recognition of the deformation of the tunnel surrounding rock. Training a random forest model by using the training set to obtain a trained surrounding rock deformation classification model; and by means of the trained surrounding rock deformation classification model, current to-be-identified surrounding rock data are analyzed, first deformation data are obtained, and the first deformation data represent the preliminary tendency of surrounding rock deformation. Through the strong classification capability of the random forest model, the preliminary judgment of the surrounding rock deformation tendency is realized, the judgment accuracy and efficiency are improved, the weighted scoring method and the Bayesian network algorithm are adopted to carry out preliminary judgment and detailed judgment, the multi-source data are effectively integrated, and the judgment precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel engineering, and in particular to a method for identifying deformation of surrounding rock of a tunnel in a water-rich composite stratum. Background Art

[0002] Due to the unique geological structure and hydrological conditions of water-rich composite strata, surrounding rock deformation is highly likely to occur during tunnel construction. This deformation not only poses a serious threat to the lives of construction workers and causes failure of tunnel support structures, but also significantly increases project costs and delays the construction schedule. During tunnel operation, deformation can cause lining cracks, water seepage, and other problems, impacting the tunnel's normal operation.

[0003] Currently, existing methods for identifying surrounding rock deformation often focus on analyzing a single factor, failing to fully account for the complexity of water-rich composite formations. Some methods rely on empirical formulas that fail to accurately reflect actual geological and hydrological conditions. Furthermore, traditional identification methods employ relatively simple data processing methods and lack effective integration and analysis of multi-source data. Furthermore, the coupling between groundwater flow and stress fields is insufficiently considered during the identification process, resulting in significant deviations between the identification results and actual conditions. This not only impacts the safe construction and operation of tunnel projects but also results in a waste of resources. Summary of the Invention

[0004] In order to solve the above technical problems, a method for identifying deformation of surrounding rock of tunnels in water-rich composite strata is provided. This technical solution solves the above problems.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] The method for identifying deformation of surrounding rock of a tunnel in a water-rich composite stratum comprises the following steps:

[0007] Collecting stratigraphic information and hydrological data from the geological survey phase, as well as on-site monitoring data from the construction phase, cleaning and normalizing these data, and constructing a training set for the random forest model. The random forest model is trained using the training set to obtain a trained surrounding rock deformation classification model.

[0008] By using the trained surrounding rock deformation classification model, the current surrounding rock data to be identified is analyzed to obtain first deformation data, wherein the first deformation data represents the initial tendency of surrounding rock deformation;

[0009] The second deformation data is calculated by the permeability calculation formula, combined with the permeability coefficient of the water-rich composite stratum and the groundwater level parameters, to reflect the effect of groundwater on the surrounding rock.

[0010] Combining the first deformation data and the second deformation data, and using the preset preliminary judgment rules, the deformation possibility and preliminary deformation level of the surrounding rock are judged;

[0011] Conduct hydraulic fracturing tests and acoustic wave tests at the tunnel construction site. Based on the test and measurement results, calculate the stress intensity ratio and deformation influence range of the surrounding rock, and use them as the third deformation data.

[0012] According to the third deformation data and the deformation influence range, use the detailed judgment algorithm to accurately determine the deformation possibility and deformation level of the surrounding rock.

[0013] Preferably, the formula for calculating the seepage force is:

[0014] F = γ w i

[0015] Where, F is the seepage force, γ w is the specific weight of water, and i is the hydraulic gradient.

[0016] Preferably, the deformation levels are divided into four levels: stable, slight deformation, moderate deformation, and severe deformation.

[0017] Preferably, the on-site monitoring data during the construction stage includes tunnel convergence data, surrounding rock pressure data, and water inflow data.

[0018] Preferably, the preliminary judgment rule adopts the weighted scoring method. According to the importance of the first and second deformation data, assign corresponding weights, calculate the comprehensive score S to judge the deformation situation, and the formula for calculating the comprehensive score is:

[0019] S = w1D1 + w2D2

[0020] In the formula, the first deformation data is D1, the corresponding weight is w1, the second deformation data is D2, the corresponding weight is w2. Compare the calculated comprehensive score S with the pre-set threshold range to judge the deformation possibility and preliminary deformation level of the surrounding rock.

[0021] Preferably, the steps of comparing the calculated comprehensive score S with the pre-set threshold range to judge the deformation possibility and preliminary deformation level of the surrounding rock are as follows:

[0022] Determine three thresholds S1, S2, and S3, and S1 < S2 < S3, and divide the comprehensive score range into four intervals, (-∞, S1], (S1, S), (S2, S3], (S3, +∞);

[0023] Comparison and judgment:

[0024] When S ∈ (-∞, S1], it is determined that the surrounding rock is in a stable state;

[0025] When ∈ (S1, S2], it is determined that there is a possibility of slight deformation of the surrounding rock;

[0026] When ∈(S2, S3], it is determined that there is a high possibility of moderate deformation of the surrounding rock;

[0027] When ∈(S3, +∞), it is determined that there is a very high possibility of severe deformation of the surrounding rock.

[0028] Preferably, the detailed judgment algorithm is constructed based on a Bayesian network, comprehensively considering the probability relationships among various factors, and finally accurately determining the deformation possibility and deformation level of the surrounding rock. The specific steps are as follows:

[0029] Based on geological theory knowledge and historical data, determine the node variables in the network, including the surrounding rock stress data obtained from hydraulic fracturing tests, the rock mass structure characteristics obtained from acoustic wave tests, construction parameters, and hydrogeological parameters;

[0030] Collect multiple sets of tunnel surrounding rock deformation case data, and use the Bayesian estimation method to calculate the conditional probability of each node under given node conditions;

[0031] Input the third deformation data and deformation influence range data as evidence into the Bayesian network, and use the joint probability distribution formula Perform probability inference calculations to obtain the probability values of the surrounding rock being in different deformation levels, denoted as P stable 、P slight 、P moderate 、P severe ;

[0032] Preset probability thresholds for different deformation levels, compare the calculated probability values with these thresholds, and accurately determine the final deformation possibility and deformation level of the surrounding rock.

[0033] Preferably, the steps of comparing the calculated probability values with these thresholds and accurately determining the final deformation possibility and deformation level of the surrounding rock are as follows: ]]

[0034] Preset probability thresholds, stipulate the stable level threshold T1, the slight deformation level threshold T2, and the moderate deformation level threshold T3, and T1>T2>T3;

[0035] Compare the calculated probability values with the thresholds:

[0036] When P stable >T1, accurately determine that the surrounding rock is in a stable state;

[0037] When P slight ≥T2 and P stable <T1, accurately determine that there is slight deformation of the surrounding rock;

[0038] When P moderate ≥P3 and P slightWhen T2, it is accurately determined that there is a high probability that the surrounding rock will undergo moderate deformation;

[0039] When P severe > T3 and P moderate < T3, it is accurately determined that the surrounding rock is extremely likely to undergo severe deformation.

[0040] Preferably, in the hydraulic fracturing test, the maximum principal stress of the surrounding rock is calculated by a formula, and the specific formula is:

[0041]

[0042] Among them, σ H is the maximum principal stress of the surrounding rock, P b is the fracture pressure, V0 is the initial volume of the borehole, V1 is the volume of the borehole after fracture, and σ h is the minimum principal stress.

[0043] Preferably, the deformation influence range is determined by combining the acoustic wave test data with the rock mass wave velocity and attenuation law.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows: Through the powerful classification ability of the random forest model, the preliminary judgment of the deformation tendency of the surrounding rock is realized, and the accuracy and efficiency of the judgment are improved. Through quantitative analysis means such as the seepage force formula and the hydraulic fracturing test formula, the influence of groundwater and surrounding rock stress is fully considered, making the judgment result more scientific and reasonable. The weighted scoring method and the Bayesian network algorithm are used for preliminary judgment and detailed judgment, effectively integrating multi-source data and improving the accuracy of the judgment. In addition, this method comprehensively considers multi-source data in the geological exploration and construction stages, fully considers the characteristics of the water-rich composite stratum, provides a reliable decision-making basis for the design, construction and operation of tunnel engineering, reduces engineering risks, and saves engineering costs. s BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a schematic flow chart of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0046] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations.

[0047] Referring to Figure 1 as shown, the method for identifying the deformation of the surrounding rock of a tunnel in a water-rich composite stratum includes the following steps:

[0048] Collecting stratigraphic information and hydrological data from the geological survey phase, as well as on-site monitoring data from the construction phase, cleaning and normalizing these data, and constructing a training set for the random forest model. The random forest model is trained using the training set to obtain a trained surrounding rock deformation classification model.

[0049] By using the trained surrounding rock deformation classification model, the current surrounding rock data to be identified is analyzed to obtain first deformation data, wherein the first deformation data represents the initial tendency of surrounding rock deformation;

[0050] The second deformation data is calculated by the permeability calculation formula, combined with the permeability coefficient of the water-rich composite stratum and the groundwater level parameters, to reflect the effect of groundwater on the surrounding rock.

[0051] Combining the first deformation data and the second deformation data, and using the preset preliminary judgment rules, the deformation possibility and preliminary deformation level of the surrounding rock are judged;

[0052] Conduct hydraulic fracturing tests and acoustic wave tests at the tunnel construction site. Based on the test results, calculate the stress intensity ratio of the surrounding rock and the deformation influence range, which are used as the third deformation data;

[0053] Based on the third deformation data and deformation influence range, the detailed judgment algorithm is used to accurately determine the deformation possibility and deformation level of the surrounding rock.

[0054] Specifically, stratigraphic information and hydrological data from the geological survey phase can reflect the initial state of the water-rich composite strata, while field monitoring data from the construction phase demonstrates the real-time changes in the surrounding rock during tunnel construction. Cleaning this data removes outliers and erroneous data, and normalization unifies data of varying magnitudes to a common scale for easier analysis. The random forest model, composed of multiple decision trees, is trained on a training set to learn the relationship between surrounding rock deformation characteristics and various data types, thereby generating a trained surrounding rock deformation classification model. This model is used to analyze the surrounding rock data to be identified, and the resulting primary deformation data provides a preliminary indication of surrounding rock deformation propensity. Permeability is a key factor in groundwater's effect on surrounding rock. Using a specific calculation formula, combined with the permeability coefficient of the water-rich composite stratum and groundwater level parameters, secondary deformation data can be calculated to quantify the impact of groundwater on the surrounding rock. Combining the primary and secondary deformation data and applying pre-defined preliminary judgment rules, the likelihood and initial level of surrounding rock deformation can be determined. Hydraulic fracturing tests provide information on surrounding rock stress, while acoustic testing reveals rock mass structural characteristics. The surrounding rock stress intensity ratio and deformation influence range calculated based on these two test results serve as the third deformation data. Finally, using this third deformation data and the deformation influence range, a detailed judgment algorithm can be used to accurately determine the likelihood and degree of surrounding rock deformation.

[0055] This method integrates multi-stage and multi-type data, and gradually and deeply identifies surrounding rock deformation through two stages: preliminary and detailed identification. This improves the accuracy and comprehensiveness of the identification and provides a reliable decision-making basis for tunnel construction and operation.

[0056] The penetration calculation formula is:

[0057] F=γ w i

[0058] Among them, F penetration, γ w is the density of water, and i is the hydraulic gradient.

[0059] Specifically, permeability is the force exerted by groundwater on the surrounding rock as it flows through a water-rich composite formation. The gravity of water reflects the weight exerted on a unit volume of water, and the hydraulic gradient reflects the energy changes during groundwater flow. This formula quantifies permeability using these two parameters, intuitively reflecting the intensity of groundwater's effect on the surrounding rock. This clear permeability calculation formula quantifies the impact of groundwater on the surrounding rock, providing accurate data support for subsequent analysis and enhancing the scientific nature of the identification method.

[0060] The deformation levels are divided into four levels: stable, slight deformation, moderate deformation, and severe deformation.

[0061] Specifically, the surrounding rock deformation level is divided into four levels: stable, slight deformation, moderate deformation, and severe deformation, which provides a clear classification standard for the identification results. In actual projects, different deformation levels correspond to different response measures. This classification is convenient for construction personnel and management personnel to understand and handle.

[0062] The on-site monitoring data during the construction phase includes tunnel convergence data, surrounding rock pressure data, and water inflow data.

[0063] Specifically, tunnel convergence data reflects the displacement of the surrounding rock mass into the tunnel, surrounding rock pressure data reflects the force exerted by the surrounding rock on the support structure, and water inflow data indicates the amount of groundwater outflow. These data reflect the mechanical state of the surrounding rock and groundwater activity during the tunnel construction phase from different angles, providing rich real-time information for identifying surrounding rock deformation.

[0064] The initial judgment rule adopts a weighted scoring method, assigning corresponding weights according to the importance of the first and second deformation data, and calculating a comprehensive score S to judge the deformation situation. The calculation formula of the comprehensive score is:

[0065] S=w1D1+w2D2

[0066] In the formula, the first deformation data is D1, the corresponding weight is w1, the second deformation data is D2, and the corresponding weight is w2. The calculated comprehensive score S is compared with a pre-set threshold range to judge the deformation possibility of the surrounding rock and the preliminary deformation level.

[0067] Specifically, the weighted scoring method assigns corresponding weights according to the importance of the first and second deformation data on the deformation of the surrounding rock, and calculates the comprehensive score. The first deformation data reflects the preliminary tendency of the surrounding rock deformation, and the second deformation data reflects the influence of groundwater on the surrounding rock. The effective integration of the two avoids the one-sidedness of single data identification, improves the accuracy of the preliminary judgment result, and provides a reliable basis for the subsequent judgment.

[0068] The steps of comparing the calculated comprehensive score S with the pre-set threshold range to judge the deformation possibility of the surrounding rock and the preliminary deformation level are as follows:

[0069] Determine three thresholds S1, S2 and S3, and S1 < S2 < S3, and divide the comprehensive score range into four intervals, (-∞, S1], (S1, S2], (S2, S3], (S3, +∞);

[0070] Comparison and judgment:

[0071] When S ∈ (-∞, S1], it is determined that the surrounding rock is in a stable state;

[0072] When ∈ (S1, S2], it is determined that there is a possibility of slight deformation of the surrounding rock;

[0073] When ∈ (S2, S3], it is determined that there is a greater possibility of moderate deformation of the surrounding rock;

[0074] When ∈ (S3, +∞), it is determined that the surrounding rock is very likely to undergo severe deformation.

[0075] Specifically, by determining three thresholds and dividing the comprehensive score range into four intervals, and comparing the calculated comprehensive score with the thresholds, the deformation possibility of the surrounding rock and the preliminary deformation level are judged. Different comprehensive score intervals correspond to different surrounding rock deformation states. This comparison method is based on a large number of engineering practices and data analyses, and has scientificity and rationality. Through the clear threshold comparison steps, the preliminary judgment process is standardized and normalized, reducing the subjectivity of human judgment and improving the accuracy and consistency of the preliminary judgment result.

[0076] The detailed judgment algorithm is constructed based on the Bayesian network, comprehensively considering the probability relationships among various factors, and making a final accurate judgment on the deformation possibility and deformation level of the surrounding rock. The specific steps are as follows:

[0077] Based on geological theory and historical data, node variables in the network are determined, including surrounding rock stress data obtained from hydraulic fracturing tests, rock mass structural characteristics obtained from acoustic wave testing, construction parameters, and hydrological parameters;

[0078] Collect multiple sets of tunnel surrounding rock deformation case data and use the Bayesian estimation method to calculate the conditional probability of each node under given node conditions;

[0079] The third deformation data and deformation influence range data are input into the Bayesian network as evidence, and the joint probability distribution formula is used. Probabilistic reasoning calculations are performed to obtain the probability values ​​of the surrounding rock at different deformation levels, which are denoted as P stable 、P slight 、P moderate 、P severe ;

[0080] The probability thresholds of different deformation levels are set in advance, and the calculated probability values ​​are compared with these thresholds to accurately determine the final deformation possibility and deformation level of the surrounding rock.

[0081] Specifically, the Bayesian network is based on probability theory and can comprehensively consider the probabilistic relationship between multiple factors. It determines the node variables in the network based on geological theoretical knowledge and historical data. These variables include surrounding rock stress data obtained from hydraulic fracturing tests, rock structure characteristics obtained from acoustic tests, construction parameters, hydrological parameters, etc., which are all key factors affecting surrounding rock deformation. Using the Bayesian estimation method, multiple sets of tunnel surrounding rock deformation case data are collected, and the conditional probability of each node under given node conditions is calculated to construct a Bayesian network. The third deformation data and deformation influence range data are input into the network as evidence, and the joint probability distribution formula is used for inference calculation to obtain the probability values ​​of the surrounding rock at different deformation levels. The Bayesian network detailed judgment algorithm fully considers the interaction of multiple factors. Through probabilistic reasoning calculations, it can more accurately determine the possibility and deformation level of surrounding rock deformation.

[0082] The steps for comparing the calculated probability values ​​with these thresholds to accurately determine the final deformation possibility and deformation level of the surrounding rock are as follows:

[0083] The probability thresholds are pre-set, and the stability level threshold T1, the slight deformation level threshold T2, and the moderate deformation level threshold T3 are specified, and T1>T2>T3;

[0084] Compare the calculated probability value with the threshold:

[0085] When P stable When >T1, the surrounding rock is accurately determined to be in a stable state;

[0086] When P slight ≥T2 and Pstable When T1, it is accurately determined that there is slight deformation of the surrounding rock;

[0087] When P moderate ≥ T3 and P slight < T2, it is accurately determined that there is a high probability of moderate deformation of the surrounding rock;

[0088] When P severe > T3 and P moderate < T3, it is accurately determined that there is a very high probability of severe deformation of the surrounding rock.

[0089] Specifically, probability thresholds for different deformation levels are preset in advance, and the probability values of the surrounding rock at different deformation levels obtained by calculation are compared with these thresholds; the stability level threshold, slight deformation level threshold, and moderate deformation level threshold are set in descending order. By comparing the magnitude relationship between the probability value and the threshold, the final deformation possibility and deformation level of the surrounding rock are accurately determined. This makes the detailed judgment result more accurate and intuitive, provides a clear basis for engineering decision-making, and helps to take targeted measures to deal with different degrees of surrounding rock deformation.

[0090] In the hydraulic fracturing test, the maximum principal stress of the surrounding rock is calculated by a formula, and the specific formula is:

[0091]

[0092] Among them, σ H is the maximum principal stress of the surrounding rock, P b is the fracture pressure, V0 is the initial volume of the borehole, V1 is the volume of the borehole after fracture, and σ h is the minimum principal stress.

[0093] Specifically, the hydraulic fracturing test is an important method for obtaining the stress information of the surrounding rock. This formula calculates the maximum principal stress of the surrounding rock based on parameters such as the fracture pressure, initial volume of the borehole, volume of the borehole after fracture, and minimum principal stress during the test. These parameters can reflect the mechanical response of the surrounding rock during the test, and an accurate maximum principal stress value can be obtained through formula calculation.

[0094] The deformation influence range is determined by acoustic wave test data in combination with the wave velocity and attenuation law of the rock mass.

[0095] Specifically, when acoustic waves propagate in the rock mass, their wave velocity and attenuation law are closely related to the properties and structure of the rock mass. By acoustic wave test data in combination with the wave velocity and attenuation law of the rock mass, the deformation influence range can be determined. Different rock mass structures and deformation conditions will cause changes in the acoustic wave propagation characteristics. Using this relationship, the deformation influence range of the surrounding rock can be inferred, thus enriching the identification means and improving the accuracy of the identification result.

[0096] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. Various changes and improvements are possible without departing from the spirit and scope of the present invention, and such changes and improvements fall within the scope of the invention as claimed.

Claims

1. A method for identifying deformation of surrounding rock in a tunnel in a water-rich composite stratum, characterized in that: It includes the following steps: Collect the formation information and hydrological data in the geological exploration stage, as well as the on-site monitoring data in the construction stage. After cleaning and normalizing these data, construct a training set for the random forest model, and use the training set to train the random forest model to obtain a trained surrounding rock deformation classification model; With the help of the trained surrounding rock deformation classification model, analyze the current surrounding rock data to be identified to obtain the first deformation data, and the first deformation data characterizes the preliminary tendency of the surrounding rock deformation; Through the seepage force calculation formula, combined with the seepage coefficient and groundwater level parameters of the water-rich composite formation, calculate the second deformation data to reflect the influence of groundwater on the surrounding rock; Integrate the first deformation data and the second deformation data, and use the preset preliminary judgment rule to judge the deformation possibility and preliminary deformation level of the surrounding rock; Carry out hydraulic fracturing tests and acoustic wave tests at the tunnel construction site. Based on the test and test results, calculate the stress intensity ratio and deformation influence range of the surrounding rock, and use them as the third deformation data; Based on the third deformation data and the deformation influence range, use the detailed judgment algorithm to accurately determine the deformation possibility and deformation level of the surrounding rock.

2. The method for identifying deformation of surrounding rock in a tunnel in a water-rich composite stratum according to claim 1, characterized in that: The seepage force calculation formula is: F=γ w I Among them, F penetration, γ w is the density of water, and i is the hydraulic gradient.

3. The method for identifying deformation of surrounding rock in a tunnel in a water-rich composite stratum according to claim 1, characterized in that: The deformation level is divided into four levels: stable, slight deformation, moderate deformation, and severe deformation.

4. The method for identifying deformation of surrounding rock in a tunnel in a water-rich composite stratum according to claim 1, characterized in that: The on-site monitoring data in the construction stage includes tunnel convergence data, surrounding rock pressure data, and water inflow data.

5. The method for identifying deformation of surrounding rock in a tunnel in a water-rich composite stratum according to claim 1, characterized in that: The preliminary judgment rule adopts the weighted scoring method. According to the importance of the first and second deformation data, assign corresponding weights, calculate the comprehensive score S to judge the deformation situation, and the calculation formula of the comprehensive score is: S = w1D1 + w2D2 In the formula, the first deformation data is D1, the corresponding weight is w1, the second deformation data is D2, the corresponding weight is w2, and compare the calculated comprehensive score S with the pre-set threshold range to judge the deformation possibility and preliminary deformation level of the surrounding rock.

6. The method for identifying deformation of surrounding rock in a tunnel in a water-rich composite stratum according to claim 5, characterized in that: The steps of comparing the calculated comprehensive score S with the pre-set threshold range to judge the deformation possibility and preliminary deformation level of the surrounding rock are as follows: Determine three thresholds S1, S2, and S3, and S1 < S2 < S3, and divide the comprehensive score range into four intervals, (-∞, S1], (S1, S2], (S2, S3], (S3, +∞); Comparison and judgment: When S ∈ (-∞, S1], it is determined that the surrounding rock is in a stable state; When ∈ (S1, S2], it is determined that there is a possibility of slight deformation of the surrounding rock; When ∈ (S2, S3], it is determined that there is a greater possibility of moderate deformation of the surrounding rock; When ∈ (S3, +∞), it is determined that the surrounding rock is very likely to have severe deformation.

7. The method for identifying deformation of surrounding rock in a tunnel in a water-rich composite stratum according to claim 1, characterized in that: The detailed judgment algorithm is constructed based on the Bayesian network, comprehensively considering the probability relationship between various factors, and finally accurately determining the deformation possibility and deformation level of the surrounding rock. The specific steps are as follows: Based on geological theory knowledge and historical data, determine the node variables in the network, including the surrounding rock stress data obtained from the hydraulic fracturing test, the rock mass structure characteristics obtained from the acoustic wave test, construction parameters, and hydrological parameters; Collect multiple sets of tunnel surrounding rock deformation case data and use the Bayesian estimation method to calculate the conditional probability of each node under given node conditions; The third deformation data and deformation influence range data are input into the Bayesian network as evidence, and the joint probability distribution formula is used. Probabilistic reasoning calculations are performed to obtain the probability values ​​of the surrounding rock at different deformation levels, which are denoted as P stable 、P slight 、P moderate 、P severe ; The probability thresholds of different deformation levels are set in advance, and the calculated probability values ​​are compared with these thresholds to accurately determine the final deformation possibility and deformation level of the surrounding rock.

8. The method for identifying deformation of surrounding rock in a tunnel in a water-rich composite stratum according to claim 7, characterized in that: The steps for comparing the calculated probability values ​​with these thresholds to accurately determine the final deformation possibility and deformation level of the surrounding rock are as follows: The probability thresholds are pre-set, and the stability level threshold T1, the slight deformation level threshold T2, and the moderate deformation level threshold T3 are specified, and T1>T2>T3; Compare the calculated probability value with the threshold: When P stable When >T1, the surrounding rock is accurately determined to be in a stable state; When P slight ≥ T2 and P stable < T1, it is accurately determined that there is slight deformation of the surrounding rock; When P moderate ≥ T3 and P slight < T2, it is accurately determined that there is a high possibility that the surrounding rock will undergo moderate deformation; When P severe > T3 and P moderate < T3, it is accurately determined that the surrounding rock is very likely to undergo severe deformation.

9. The method for identifying deformation of surrounding rock of a tunnel in a water-rich composite stratum according to claim 1, characterized in that In the hydraulic fracturing test, the maximum principal stress of the surrounding rock is calculated by the formula, which is specifically: Among them, σ H is the maximum principal stress of the surrounding rock, P b is the fracture pressure, V0 is the initial borehole volume, V1 is the borehole volume after fracture, σ h is the minimum principal stress.

10. The method for identifying deformation of surrounding rock in a tunnel in a water-rich composite stratum according to claim 1, characterized in that: The deformation influence range is determined by acoustic wave test data in combination with rock mass wave velocity and attenuation law.

Images