A three-dimensional dynamic quantitative evaluation method for stability of deep hard rock tunnel surrounding rock

Abstract

The application provides a kind of deep hard rock tunnel surrounding rock stability three-dimensional dynamic quantification evaluation method, it is related to deep tunnel surrounding rock stability evaluation and intelligent construction technical field, the method comprises: based on the dynamic determination of tunnel geology and surrounding rock parameters based on digital means and in-situ testing means;Based on the rapid calculation of stress distribution in the failure zone considering three-dimensional complex high ground stress based on GZZ three-dimensional strength criterion;Based on the rapid quantitative solution method of failure zone distribution form based on artificial intelligence algorithm.The application takes stress accurate control and fine analysis as the tunnel design concept, based on considering the joint, fracture, excavation disturbance of deep tunnel surrounding rock, and the strength characteristics and mechanical mechanism of surrounding rock, using the powerful global optimization ability of intelligent algorithm and the high efficiency calculation characteristics of traditional analysis method, a programmed process is obtained for calculating the surrounding rock failure zone, realizing the accurate, rapid and dynamic analysis of surrounding rock stability in the process of tunneling.

Description

Technical Field

[0001] This invention belongs to the field of tunnel excavation engineering technology, and in particular relates to a three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels. Background Technology

[0002] Tunnels (or tunnels) are among the most common structures in underground engineering, playing a crucial role in many of the country's mega-projects. With the westward shift of my country's railway (highway) network construction focus, the exploitation of deep mineral resources, and the construction of the "Transportation Powerhouse" strategic projects under the 14th Five-Year Plan, tunnel engineering is rapidly developing towards deeper and longer tunnels. Advancing into deeper underground engineering is a major theme of the 21st century. Due to the complex geological conditions of deep tunnels, tunnel construction faces numerous challenges. Deep rock masses differ significantly from shallow rock masses in terms of mechanical properties such as structure, deformation, and strength. The disaster mechanisms of deep tunnels are also more complex. Among these, the collapse and instability of tunnels caused by hard rock masses under complex high-stress conditions, and high-intensity and high-frequency rockbursts, directly affect the safety, progress, and quality of tunnel construction, and are both hot topics and difficult points in deep engineering research.

[0003] Tunnel excavation disturbs the rock mass, which was originally in equilibrium. This causes stress redistribution in the surrounding rock, and under the influence of high three-dimensional ground stress, the rock is highly susceptible to exceeding its elastic limit, leading to relaxation, yielding, and failure, forming a so-called failure zone (or loosening zone) around the tunnel. The distribution pattern of this failure zone plays a crucial role in assessing the stability of the tunnel surrounding rock and in the support design during subsequent construction. The formation of the failure zone is a dynamic process; its shape and size change as the tunnel face advances. During construction, the thickness of the failure zone can generally be measured using on-site devices, allowing for adjustments and improvements to the pre-designed support structure to achieve dynamic design. However, for large-section tunnels, the thickness of the failure zone will far exceed the maximum detection range of the devices, making accurate measurement impossible. The large thickness of the failure zone also prevents effective reinforcement of the surrounding rock using anchor bolts of limited length; often, protection can only be achieved through (multiple) applications of steel arches and concrete linings. For hard surrounding rock, there is often no obvious deformation before failure. Traditional displacement (rate) monitoring methods are insufficient for real-time early warning of such tunnel disasters, making it difficult to guarantee tunnel safety. Therefore, developing reasonable and effective rapid prediction methods for surrounding rock failure zones will be of great significance in guiding the dynamic design of deep tunnels.

[0004] Tunnel projects are often situated in concealed and complex geological environments, making it impossible to fully understand the geological body to be excavated using existing exploration techniques or methods. This leads to potential risks in tunnel engineering. The current construction system, which separates design and construction, disrupts the relationship between the two, making it difficult to dynamically modify and improve pre-design plans based on exposed geological conditions and various changes occurring during construction. The concept of digitalized dynamic tunnel design emphasizes timely evaluation of the design's rationality, adjustment of support parameters and construction plans during construction, making the support structure more adaptable to the actual surrounding rock conditions, and improving the investment efficiency and overall benefits of tunnel engineering.

[0005] Digital dynamic design typically relies on integrated analysis using digital platforms and numerical analysis programs. However, the complexity of current digital and numerical modeling technologies, along with the significant time commitment associated with numerical computation, is insufficient to meet the real-time requirements of dynamic design. Traditional empirical analogy methods lack objectivity and struggle to account for complex factors, generally suitable only for shallow tunnel construction. Empirical formulas cannot explain the complex mechanical behavior of surrounding rock in deep tunnels. In most cases, accurate analytical methods for calculating the failure zone of surrounding rock are extremely difficult to obtain and require a high level of professional knowledge from practitioners, hindering method expansion. Currently, the integration of modern information technology, represented by artificial intelligence, with traditional construction industries such as tunneling is a major development trend, injecting new vitality into the fundamental theory of digital intelligent tunnel construction. However, intelligent algorithms often predict and analyze the temporal trends of tunnel structures and surrounding rock through numerous engineering measurement samples, lacking consideration of mechanical mechanisms. In conclusion, addressing this challenge requires not only utilizing new technologies and intelligent methods to meet the new challenge but also exploring the underlying physical mechanisms and abstracting reasonable mechanical models to propose accurate, efficient, and convenient methods for analyzing the stability of tunnel surrounding rock. Summary of the Invention

[0006] The purpose of this invention is to provide a three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels, characterized by the following steps:

[0007] S1: Collect parameters of tunnel geology and surrounding rock through in-situ testing and digital methods;

[0008] S2: Construct a three-dimensional mechanical analysis model of tunnel surrounding rock based on the GZZ three-dimensional rock mass strength criterion using the collected parameters, and give the control equation and objective function of the failure zone;

[0009] S3: Encapsulate the relevant parameters and objective function into an artificial intelligence algorithm, and use the artificial intelligence algorithm to solve the objective function;

[0010] S4: Utilize encapsulated artificial intelligence algorithms to automatically perform stability assessments after dynamic parameter updates during tunnel excavation.

[0011] Furthermore, S1 specifically includes the following steps:

[0012] S11: Utilize digital means (non-contact measurement technologies such as three-dimensional laser scanning or binocular digital photography) to accurately and rapidly acquire detailed digital information such as traces and structural surface attitude on the tunnel excavation surface in situ, and determine the geological strength index GSI and disturbance coefficient D;

[0013] S12: Obtain the rock hardness m through methods such as rebound dynamics testing or empirical statistical models. i Combining the parameters GSI and D obtained in S11, the rock mass strength parameter m is further calculated. b s and a, parameter m b The formulas for calculating s and a are expressed as follows:

[0014]

[0015] S13: Obtain the uniaxial compressive strength σ of rock using in-situ true triaxial tests. c Geological survey methods (such as hydraulic fracturing and acoustic emission) are used to conduct in-situ tests of horizontal geostress p and vertical geostress q.

[0016] Furthermore, S2 specifically includes the following steps:

[0017] S21: Based on the static equilibrium condition and the GZZ three-dimensional strength criterion, the governing equations for obtaining the stress distribution within the surrounding rock failure zone are constructed, where the GZZ three-dimensional strength criterion is expressed as:

[0018]

[0019] in,

[0020] S22: The stress distribution in the undamaged area of ​​the surrounding rock is given by using two elastic complex potential functions;

[0021] S23: Based on the principle of static equilibrium, the governing equations for determining the failure zone are given, and the objective function for solving the failure zone is given.

[0022] Furthermore, in S21, the stress distribution within the surrounding rock failure zone is achieved through the following steps:

[0023] S211: Establish the uncorrelated plastic potential function Q, expressed as:

[0024]

[0025] Where β is the volume correction factor reflecting the dilatation characteristics of the rock mass;

[0026] Based on the non-associated plastic potential function Q, the plastic flow law, and the tunnel axial strain ε z Obtain the out-of-plane plastic stress σ of the tunnel z The relation is expressed as:

[0027]

[0028] Where dλ is a positive scaling factor;

[0029] S212: Determine the out-of-plane plastic stress σ in S211 z Substituting the relational expression into the GZZ three-dimensional strength criterion, we obtain the in-plane stress control equation for the surrounding rock perpendicular to the tunnel excavation direction, which is expressed as:

[0030]

[0031] in, α is based on the tunnel axial strain ε z The size of ε is determined, and in the most general case, it is ε. z The plane strain model when α = 0, where α is taken as 0.5.

[0032] S213: By combining the governing equations with the equilibrium differential equations, the stress distribution of the surrounding rock in the failure zone is obtained using the slip line method.

[0033] Furthermore, in S22, the stress distribution within the undamaged area of ​​the surrounding rock is achieved through the following steps:

[0034] S221: The boundary of the failure zone in the surrounding rock is represented by a mapping function, which is in the form of a Taylor series, expressed as:

[0035]

[0036] Among them, R and C k Let be the coefficients of the mapping function to be determined, and all of them are real numbers; m≥6; solving for the coefficients of the mapping function is equivalent to determining the distribution pattern of the damaged area, that is, parameterizing and quantitatively representing the shape and size of the damaged area;

[0037] S222: The stress distribution within the undamaged surrounding rock area is represented by two complex potential functions related to geostress, which are in the form of Taylor series, expressed as:

[0038]

[0039]

[0040] Among them, a k and bk All numbers are real numbers; n ≥ 40;

[0041] S223: Given the coefficients of the mapping function, the stress distribution within the undamaged area of ​​the surrounding rock is given by the following equation:

[0042]

[0043]

[0044] in, This represents the polar coordinates of the in-plane stress components in the undamaged area of ​​the surrounding rock.

[0045] Furthermore, in S23, the objective function for the damaged zone is solved through the following steps:

[0046] S231: Based on the static equilibrium conditions at the boundary of the failure zone, the following equations are derived:

[0047]

[0048] in, This represents the polar coordinates of the in-plane stress components in the surrounding rock failure zone.

[0049] S232: Based on the above equation, construct the following matrix:

[0050]

[0051]

[0052]

[0053] in, j = 1, 2, 3, ..., n;

[0054] S233: Based on the above matrix, list the solutions for a. k and b k The system of linear equations:

[0055] AX = B

[0056] in

[0057]

[0058] B = [Re(B1) Re(B2)...Re(B n Im(B1) Im(B2)…Im(B n )] T

[0059] X = [a1 a2…a] n b1 b2…b n] T

[0060] Among them, Re(A) j,k ) and Im(A j,k ) are respectively A j,k The real and imaginary parts; Re(B j ) and Im(B j ) are respectively B j The real and imaginary parts;

[0061] S234: The objective function for solving the damaged zone is given as follows:

[0062]

[0063] Where f(X) has the same unit as the stress component; the magnitude of f(X) reflects the degree to which the equilibrium conditions are satisfied in the undamaged zone of the surrounding rock. When the value of f(X) is 0, the true solution of the damaged zone is obtained.

[0064] Furthermore, S3 specifically includes the following steps:

[0065] S31: Set known quantities that can be dynamically assigned to the tunnel geology and surrounding rock parameters related to the damaged area, and set the objective function f(X) as the fitness function of the artificial intelligence algorithm;

[0066] S32: The coefficients R and C of the mapping function characterizing the shape and size of the damaged area. k Given the range of design variables in the algorithm, set the algorithm parameters appropriately, initialize the population, and calculate the fitness of each individual in the population.

[0067] S33: Based on the idea of ​​biological genetic evolution, preserve the variant individuals that are beneficial to adapting to the environment, continuously reproduce offspring to obtain the individuals most beneficial to survival, and obtain the latest parameters of the most advantageous individuals, that is, continuously approach the shape and size of the destruction zone in this invention;

[0068] S34: Determine whether the latest mutated individual has reached the maximum genetic generation or satisfactory accuracy. If yes, output the shape and size of the surrounding rock failure zone. If not, repeat steps S32-S34 based on the latest parameters until the determination result is yes.

[0069] Furthermore, in S32, the artificial intelligence algorithm is the differential evolution algorithm, and the parameters of the differential evolution algorithm are set as follows:

[0070]

[0071] The value ranges of each design variable are shown in the table below:

[0072]

[0073] Where R0 is the tunnel radius.

[0074] Furthermore, S4 specifically includes the following steps:

[0075] S41: Encapsulate the above procedural process into the differential evolution algorithm;

[0076] S42: Solve the objective function using the differential evolution algorithm;

[0077] S43: Based on the solution obtained in S42, guide the design and construction of the support structure. Repeat steps S1-S42 during tunnel excavation to achieve the purpose of dynamic design.

[0078] Compared with the prior art, the beneficial effects of the present invention are mainly reflected in:

[0079] 1. The present invention proposes a three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels, which combines traditional analysis methods with artificial intelligence algorithms to achieve the assessment of surrounding rock stability during construction.

[0080] 2. This invention takes into account the tunnel geometric model and the mechanical properties of the surrounding rock through traditional analysis methods, and can accurately, completely and quantitatively predict and analyze the temporal change trend of the surrounding rock failure zone.

[0081] 3. This invention utilizes the powerful global optimization capabilities of artificial intelligence algorithms and the advantages of rapid analysis by traditional analysis methods to elucidate the dynamic failure mechanism of surrounding rock and better serve the digital dynamic design of tunnel support structures. Attached Figure Description

[0082] Figure 1 The flowchart of a three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels provided by the present invention is shown.

[0083] Figure 2 The flowchart shows the construction of the objective function in step S2 of the three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels provided by this invention.

[0084] Figure 3 The flowchart of step S3, which is based on an artificial intelligence algorithm, for solving the failure zone in a three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels provided by the present invention. Detailed Implementation

[0085] The following will describe in more detail, with reference to the schematic diagram, a three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels according to the present invention, which illustrates the preferred embodiments of the present invention. It should be understood that those skilled in the art can modify the present invention described herein while still achieving the advantageous effects of the present invention. Therefore, the following description should be understood as being of general knowledge to those skilled in the art and is not intended to limit the present invention.

[0086] like Figure 1 As shown, a three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels includes the following steps:

[0087] Step S1: Collect parameters of tunnel geology and surrounding rock through on-site in-situ testing and digital methods.

[0088] Step S1 specifically includes:

[0089] Step S11: Use digital means (non-contact measurement technology such as three-dimensional laser scanning or binocular digital photography) to accurately and quickly acquire detailed digital information such as traces and structural surface attitude on the tunnel excavation surface in situ, and determine the geological strength index GSI and disturbance coefficient D.

[0090] Step S12: Obtain the rock hardness mi through rebound dynamics testing or empirical statistical models; combine the parameters GSI and D obtained in S11 to further calculate the rock mass strength parameters mb, s and a.

[0091] Step S13: Obtain the uniaxial compressive strength σ of the rock using a true triaxial test in the field. c Geological survey methods (such as hydraulic fracturing and acoustic emission) are used to conduct in-situ tests of horizontal geostress p and vertical geostress q.

[0092] refer to Figure 2 Step S2: Construct a three-dimensional mechanical analysis model of the tunnel surrounding rock based on the GZZ three-dimensional rock mass strength criterion using the collected parameters, and give the control equation and objective function of the failure zone.

[0093] The mechanical analysis model includes a three-dimensional non-axisymmetric tunnel excavation model considering axial stress in the tunnel; the governing equations include the GZZ three-dimensional strength criterion.

[0094]

[0095]

[0096]

[0097] The GZZ criterion shares parameters with the Hoek-Brown criterion.

[0098] Uncorrelated plastic potential function:

[0099]

[0100] Wherein, β is the volume correction factor reflecting the dilatation characteristics of the rock mass.

[0101] Step S2 specifically includes:

[0102] Step S21: Based on the static equilibrium condition and the GZZ three-dimensional strength criterion, construct the governing equations for obtaining the stress distribution in the surrounding rock failure zone.

[0103] Step S22: Use two elastic complex potential functions to give the stress distribution in the undamaged area of ​​the surrounding rock.

[0104] Step S23: Based on the principle of static equilibrium, give the set of governing equations for determining the failure zone, and give the objective function for solving the failure zone.

[0105] Step S21 can be implemented according to the following steps:

[0106] Step S211: Obtain the relationship between the non-associated plastic potential function Q, the plastic flow law, and the tunnel axial strain εz to determine the out-of-plane plastic stress σz of the tunnel.

[0107] Step S212: Substitute the relationship for determining the out-of-plane plastic stress σz in S211 into the GZZ three-dimensional strength criterion to obtain the in-plane stress control equation for the surrounding rock perpendicular to the tunnel excavation direction, expressed as:

[0108]

[0109] in, α is based on the tunnel axial strain ε z The size of ε is determined, and in the most general case, it is ε. z The plane strain model when α = 0, where α is taken as 0.5.

[0110] Step S22 can be implemented according to the following steps:

[0111] Step S221: Represent the boundary of the failure zone in the surrounding rock using a mapping function, which is in the form of a Taylor series.

[0112]

[0113] Among them, R and c k The coefficients of the mapping function to be determined in this invention are all real numbers, and it is recommended that m be no less than 6. Solving for the mapping function coefficients is equivalent to determining the distribution pattern of the damaged area, that is, to parameterize and quantitatively represent the shape and size of the damaged area.

[0114] Step S222: The stress distribution in the undamaged area of ​​the surrounding rock can be represented by two complex potential functions related to geostress, which are in the form of Taylor series.

[0115]

[0116]

[0117] Among them, a k and b k All numbers are real numbers, and it is recommended that n be no less than 40.

[0118] Step S223: Given the coefficients of the mapping function, the stress distribution in the undamaged area of ​​the surrounding rock can be given by the following equation.

[0119]

[0120]

[0121] in, This represents the polar coordinates of the in-plane stress components in the undamaged area of ​​the surrounding rock.

[0122] Step S230 can be implemented according to the following steps:

[0123] Step S231: Based on the static equilibrium conditions at the boundary of the failure zone, the following equations are derived:

[0124]

[0125] Step S232: Based on the above equation, construct the following matrix:

[0126]

[0127]

[0128]

[0129] in, j = 1, 2, 3, ..., n;

[0130] Step S233: Based on the above matrix, list the solutions for a. k and b k The system of linear equations is as follows:

[0131] AX = B

[0132] in

[0133]

[0134] B = [Re(B1) Re(B2)...Re(Bn Im(B1) Im(B2)…Im(B n )] T

[0135] X = [a1 a2…a] n b1 b2…b n ] T

[0136] Among them, Re(A) j,k ) and Im(A j,k ) are respectively A j,k The real and imaginary parts; Re(B j ) and Im(B j ) are respectively B j The real and imaginary parts.

[0137] Step S234: The objective function for solving the damaged zone is given as follows:

[0138]

[0139] Here, f(X) has the same unit as the stress component. The magnitude of f(X) reflects the degree to which the equilibrium conditions are satisfied in the undamaged zone of the surrounding rock. When its value is 0, the true solution of the damaged zone can be obtained.

[0140] refer to Figure 3 Step S3: Encapsulate the relevant parameters and objective function into an artificial intelligence algorithm, and use the artificial intelligence algorithm to solve the objective function quickly, effectively and accurately.

[0141] Step S3 specifically includes:

[0142] Step S31: Set known quantities that can be dynamically assigned to the tunnel geology and surrounding rock parameters related to the damaged area, and set the objective function f(X) as the fitness function of the artificial intelligence algorithm (differential evolution algorithm).

[0143] Step S32: Use the mapping function coefficients R and ck, which characterize the shape and size of the damaged area, as design variables. Given the range of design variables in the algorithm, set the differential evolution algorithm parameters appropriately, initialize the population, and calculate the fitness of each individual in the population.

[0144] Step S33: Based on the idea of ​​biological genetic evolution, through operations such as crossover, mutation, and selection, individuals with advantageous mutations that are adapted to the environment are preserved, and offspring are continuously reproduced to obtain individuals that are most conducive to survival, that is, continuously approaching the shape and size of the destruction zone in this invention.

[0145] The recommended parameter settings for the differential evolution algorithm are shown in Table 1.

[0146]

[0147] Table 1

[0148] The suggested value ranges for each design variable are shown in Table 2.

[0149]

[0150] Table 2

[0151] Where R0 is the tunnel radius.

[0152] Step S4: The encapsulated artificial intelligence algorithm program is used to automatically perform a stability assessment after the parameters are dynamically updated during the tunnel excavation process.

[0153] Step S4 specifically includes:

[0154] Step S41: Encapsulate the above procedural process into an artificial intelligence algorithm.

[0155] Step S42: Use artificial intelligence algorithms to solve the objective function quickly, efficiently and accurately.

[0156] Step S43: Use the obtained results to guide the design and construction of the support structure. Repeat steps S1-S42 during tunnel excavation to achieve the purpose of dynamic design.

[0157] The above are merely preferred embodiments of the present invention and do not constitute any limitation on the present invention. Any equivalent substitutions or modifications made by those skilled in the art to the technical solutions and content disclosed in the present invention without departing from the scope of the present invention shall be deemed to have remained within the protection scope of the present invention.

Claims

1. A three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels, characterized in that, Includes the following steps: S1: Collect parameters of tunnel geology and surrounding rock through in-situ testing and digital methods; S2: Construct a three-dimensional mechanical analysis model of tunnel surrounding rock based on the GZZ three-dimensional rock mass strength criterion using the collected parameters, and give the control equation and objective function of the failure zone; S3: Encapsulate the relevant parameters and objective function into an artificial intelligence algorithm, and use the artificial intelligence algorithm to solve the objective function; S4: Utilize encapsulated artificial intelligence algorithms to automatically perform stability assessments after dynamic parameter updates during tunnel excavation; S2 specifically includes the following steps: S21: Based on the static equilibrium condition and the GZZ three-dimensional strength criterion, the governing equations for obtaining the stress distribution within the surrounding rock failure zone are constructed, where the GZZ three-dimensional strength criterion is expressed as: ； in, ; ; S22: The stress distribution in the undamaged area of ​​the surrounding rock is given by using two elastic complex potential functions; S23: Based on the principle of static equilibrium, the set of governing equations for determining the failure zone is given, and the objective function for solving the failure zone is given. In step S21, the stress distribution within the surrounding rock failure zone is achieved through the following steps: S211: Establish the uncorrelated plastic potential function Q 、 is represented as: ； in, β A volume correction factor reflecting the dilatation characteristics of rock mass; Based on the non-associated plastic potential function Q Plastic Flow Laws and Axial Strain in Tunnels e z Obtain the determined out-of-plane plastic stress of the tunnel s z The relation is expressed as: ； in, dl It is a positive scaling factor; S212: Determine the out-of-plane plastic stress in S211 s z Substituting the relational expression into the GZZ three-dimensional strength criterion, we obtain the in-plane stress control equation for the surrounding rock perpendicular to the tunnel excavation direction, which is expressed as: ； in, ; a Based on tunnel axial strain e z The size is determined when The plane strain model at this time, a Take 0.5; S213: Combine the governing equations with the equilibrium differential equations and use the slip line method to obtain the stress distribution of the surrounding rock in the failure zone; In step S22, the stress distribution within the undamaged area of ​​the surrounding rock is achieved through the following steps: S221: The boundary of the failure zone in the surrounding rock is represented by a mapping function, which is in the form of a Taylor series, expressed as: ； Among them, R and Let be the coefficients of the mapping function to be determined, and all of them are real numbers; Solving for the coefficients of the mapping function is equivalent to determining the distribution pattern of the damaged area, that is, to parameterize and quantitatively represent the shape and size of the damaged area. S222: The stress distribution within the undamaged surrounding rock area is represented by two complex potential functions related to geostress, which are in the form of Taylor series, expressed as: ； ； in, and All are real numbers; 40; S223: Given the coefficients of the mapping function, the stress distribution within the undamaged area of ​​the surrounding rock is given by the following equation: ； ； in, , , The polar coordinate representation of the in-plane stress components in the undamaged area of ​​the surrounding rock; In step S23, the objective function for the damaged area is solved through the following steps: S231: Based on the static equilibrium conditions at the boundary of the failure zone, the following equations are derived: ； in, , , The polar coordinate representation of the in-plane stress components in the surrounding rock failure zone; S232: Based on the above equation, construct the following matrix: ； ； ； in, ; ; ; S233: Based on the above matrix, list the solution. and The system of linear equations: ； in ； ； ； in, and They are respectively The real and imaginary parts; and They are respectively and The real and imaginary parts; S234: The objective function for solving the damaged zone is given as follows: ； in, The unit is the same as that of the stress component; The size reflects the degree to which the equilibrium conditions are satisfied in the undamaged zone of the surrounding rock. When the value of is 0, the true solution for the damaged area is obtained; S3 specifically includes the following steps: S31: Set known quantities that can be dynamically assigned to the tunnel geology and surrounding rock parameters related to the damaged area, and set the objective function. Set as the fitness function for the artificial intelligence algorithm; S32: The coefficients of the mapping function characterizing the shape and size of the damaged area. R , Given the range of design variables in the algorithm, set the algorithm parameters appropriately, initialize the population, and calculate the fitness of each individual in the population. S33: Based on the idea of ​​biological genetic evolution, preserve the variant individuals that are beneficial to adapting to the environment, continuously reproduce offspring to obtain the individuals most beneficial to survival, and obtain the latest parameters of the most advantageous individuals, that is, continuously approach the shape and size of the destruction zone in this invention; S34: Determine whether the latest mutated individual has reached the maximum genetic generation or satisfactory accuracy. If yes, output the shape and size of the surrounding rock failure zone. If not, repeat steps S32-S34 based on the latest parameters until the determination result is yes.

2. The three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels according to claim 1, characterized in that, S1 specifically includes the following steps: S11: Utilize digital methods to conduct in-situ acquisition of refined digital information and determine geological intensity indicators. GSI and disturbance coefficient D ; S12: Determine the rock hardness through rebound dynamics testing or empirical statistical models. m i Combined with the parameters obtained in S11 GSI and D Further calculation of rock mass strength parameters m b , s and a ,parameter , s and a The calculation formula is expressed as: ； S13: Obtain the uniaxial compressive strength of rock using in-situ true triaxial tests. ; In-situ testing of horizontal ground stress was obtained using hydraulic fracturing and acoustic emission methods. p and vertical stress q .

3. The three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels according to claim 1, characterized in that, In step S32, the artificial intelligence algorithm is a differential evolution algorithm, and the parameters of the differential evolution algorithm are set as shown in the table below: The value ranges of each design variable are shown in the table below: in, R 0 represents the tunnel radius.

4. The three-dimensional dynamic quantitative evaluation method for the stability of surrounding rock in deep hard rock tunnels according to claim 3, characterized in that, S4 specifically includes the following steps: S41: Encapsulate the above procedural process into the differential evolution algorithm; S42: Solve the objective function using the differential evolution algorithm; S43: Based on the solution obtained in S42, guide the design and construction of the support structure. Repeat steps S1-S42 during tunnel excavation to achieve the purpose of dynamic design.

Images