Method for determining mechanical parameters of broken surrounding rock grouting rock mass and storage medium

Abstract

The present application relates to a kind of broken surrounding rock grouting rock mass mechanical parameter determination method and storage medium, belong to mine roadway supporting technical field.It is obtained complete rock block basic mechanical parameter by indoor rock mechanics test;Rely on discrete fracture network modeling and numerical iterative inversion method, accurately restore the real fracture distribution characteristics of surrounding rock;Combined with fluid-structure interaction numerical simulation, complete the whole process of grouting, such as slurry seepage filling, slurry rock cementation solidification, and obtain the mechanical performance index of composite rock mass after grouting by standardized mechanical test.The present application realizes the effective combination of indoor test, numerical inversion, grouting simulation and engineering application, can accurately characterize the structure characteristics of broken surrounding rock and grouting improvement mechanism, and provide reliable technical basis and data support.

Description

Technical Field

[0001] This invention relates to the field of mine roadway support technology, and in particular to a method for determining the mechanical parameters of fractured surrounding rock grouting rock mass and a storage medium thereon. Background Technology

[0002] As mineral resource extraction in my country continues to extend deeper, deep mine roadways are generally situated in complex engineering environments characterized by high ground stress, strong disturbance, and low strength. The surrounding rock typically exhibits densely developed joints and fissures, is fragmented and loose, and has extremely poor integrity, making it highly susceptible to engineering disasters such as large deformations, spalling, roof falls, and even overall instability. Grouting reinforcement, as a core technical means to modify and strengthen fractured surrounding rock and improve its integrity and bearing capacity, has become a key process for the safe maintenance of deep fractured roadways. Grouting can effectively fill internal fissures in the rock mass, cement fractured rock blocks, and improve the rock mass structure, significantly enhancing key mechanical parameters such as the elastic modulus, cohesion, internal friction angle, and compressive strength of the surrounding rock. This provides a core basis for roadway support structure design, stability evaluation, and long-term safety control.

[0003] However, the rock mechanics parameters of fractured surrounding rock after grouting are difficult to obtain accurately, which has become a core technical bottleneck restricting the scientific and refined design of deep tunnel support. With the widespread application of numerical simulation technology in geotechnical engineering, some studies have attempted to use discrete element method and finite element method to study the parameters of grouting rock mass. However, existing technologies generally suffer from prominent problems such as low matching degree between modeling and on-site fractures, oversimplification of grouting effect simulation, lack of closed-loop verification of parameter inversion, and poor integration with engineering software. Most numerical models fail to perform refined fracture inversion based on actual geological conditions in the field, instead using simplified regular or random fracture generalization, which fails to reproduce the true structural characteristics of the rock mass in the field. Grouting simulations often employ solid element modeling or simple strength enhancement methods, neglecting micromechanical behaviors such as interfacial bonding, interlocking, and filling density between the grout and the fracture wall, which can easily lead to problems such as mesh distortion, non-convergence of calculations, and inaccurate simulation results. At the same time, parameter acquisition is disconnected from engineering applications; the micromechanical parameters obtained from numerical simulations cannot be directly converted into macromechanical parameters for commonly used engineering software such as FLAC3D, making it difficult to directly serve large-scale roadway stability analysis in the field.

[0004] The aforementioned technical defects directly result in a lack of reliable and accurate mechanical parameters to support the grouting support design of deep, high-stress, and fractured roadways. The selection of support parameters tends to be conservative or excessively reinforced, leading to resource waste and increased costs. Alternatively, the parameters may be too low, causing support structure failure and repeated roadway repairs, which seriously affects mining efficiency and production safety. This has become a technical problem that urgently needs to be solved. Summary of the Invention

[0005] To address the aforementioned technical problems, embodiments of the present invention provide a method and storage medium for determining the mechanical parameters of grouting rock masses in fractured surrounding rock. This invention can fully reproduce the fracture characteristics of fractured surrounding rock in the field, accurately simulate the microscopic effects of grouting cementation, and achieve high-precision acquisition and engineering application of parameters for determining the mechanical parameters of grouting rock masses in fractured surrounding rock. This has significant engineering value and theoretical significance for improving the design level of deep tunnel support and ensuring the safe and efficient mining of deep resources. The technical solution adopted by this invention is as follows:

[0006] A method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock, the method comprising:

[0007] Core sampling, screening, and standard processing were carried out on the fractured surrounding rock in the deep tunnel. Multiple mechanical property tests were conducted using a rigid servo mechanical testing system, and the baseline values ​​of the inherent mechanical parameters of the complete rock block were obtained through mathematical statistics.

[0008] Based on the obtained baseline values ​​of inherent mechanical parameters of the intact rock block, combined with the degree of fracture development and rock mass integrity coefficient of the surrounding rock at the site, and relying on various engineering rock mass classification indicators and using empirical formulas for comprehensive reduction calculation, the indoor parameters are transformed into mechanical parameters of fractured rock mass that fit the actual state of the site, and a dedicated parameter database is established to provide target true values ​​for fracture inversion.

[0009] Based on the baseline values ​​of the inherent mechanical parameters of the complete rock block and the target true values, the fracture distribution characteristics of the surrounding rock at the site are restored through controlled variable experiments, sensitivity analysis and multiple rounds of iterative optimization, so that the mechanical response of the numerical model and the deviation of the rock mass at the site meet the convergence requirements, and the fracture characteristics consistent with the mechanical behavior of the rock mass at the site are obtained.

[0010] A coupled calculation system for grout seepage and rock mass was constructed based on a three-dimensional fracture network numerical model. Combining the rheological properties of grout with the spatial distribution law of fractures, the entire process of grout diffusion, filling and cementation under grouting pressure was reproduced. The filling effect of grout on fractures and the cementation effect on rock blocks were quantitatively analyzed, and the mechanical modification mechanism of grouting reinforcement of fractured surrounding rock was restored.

[0011] Multiple sets of mechanical simulation tests were conducted to systematically determine the overall mechanical performance indicators of the rock mass after grouting reinforcement, compare the changes in the mechanical properties of the rock mass before and after grouting, quantitatively analyze the improvement effect of grouting filling and cementation reinforcement, and obtain the mechanical benchmark data of the reinforced rock mass.

[0012] By integrating multi-level mechanical data of intact rock blocks, naturally fractured rock masses, and grout-modified rock masses, and combining them with the on-site engineering geological conditions and stress environment of the surrounding rock, the obtained mechanical benchmark data of the reinforced rock mass is applied to the calculation of surrounding rock stability, support structure design, and disaster risk assessment.

[0013] Optionally, core sampling can be performed on the fractured surrounding rock in the deep roadway, specifically including: conducting on-site core sampling operations on three or more typical sections of the target deep mine roadway.

[0014] Optionally, based on the obtained baseline values ​​of the inherent mechanical parameters of the intact rock block, combined with the degree of fracture development in the surrounding rock and the rock mass integrity coefficient, specifically including:

[0015] Using the complete rock block mechanical parameters determined by indoor tests as the benchmark input data, and relying on the refined engineering geology of the surrounding rock on site, we can comprehensively obtain key information such as the distribution characteristics of joints and fissures in the surrounding rock, the integrity of the rock mass, the state of the structural planes, the groundwater occurrence conditions, and the degree of disturbance during mining and excavation.

[0016] Through on-site testing and calculation, the RMR rock mass geomechanical classification index, Q rock mass quality index, and GSI geological strength index were obtained. These three indices are used together to quantitatively characterize the overall quality and degree of fragmentation of the fractured surrounding rock on site.

[0017] Optionally, a comprehensive reduction calculation can be performed based on various engineering rock mass classification indicators and empirical formulas. Specifically, this includes using the Hoek-Brown rock mass strength criterion as the core theoretical basis for parameter reduction, and combining the material constants of intact rock with the tunnel excavation disturbance coefficient to calculate the characteristic constants of the rock mass.

[0018] Optionally, the fracture distribution characteristics of the surrounding rock in the field are reconstructed through controlled variable experiments, sensitivity analysis, and multiple rounds of iterative optimization, specifically including:

[0019] Based on the field investigation data of joints and fractures in the surrounding rock, a fracture stochastic generation model was constructed using the Discrete Fracture Network (DFN) technology and the Monte Carlo stochastic simulation principle.

[0020] Numerical experiments were conducted using the controlled single-variable method, fixing the probability distribution and initial mean of the other three fracture parameters and adjusting only the gradient value of the single target parameter. For each parameter combination, three parallel discrete fracture network models were generated using Monte Carlo simulation. Uniaxial compression and direct shear numerical mechanical tests were conducted on each model, with the loading method, loading rate, and boundary conditions being completely consistent with the laboratory tests. The arithmetic mean of the three parallel test results was taken as the simulated rock mechanics parameter value corresponding to that set of fracture parameters.

[0021] Using the actual mechanical parameters of the fractured rock mass obtained by parameter reduction as the target true values, and based on the results of single-factor tests, the grey relational method is used to calculate the sensitivity weights of fracture orientation, aperture, density, and penetration on four core parameters: elastic modulus, cohesion, internal friction angle, and uniaxial compressive strength.

[0022] Optionally, the relative deviations of the four core mechanical parameters are all less than the specified threshold as the convergence criterion; if the convergence requirement is not met, the iteration optimization step is returned to continue adjustment; if the convergence condition is met, the iteration is terminated, and the complete parameters and probability distribution characteristics of the fracture attitude, aperture, density, and penetration corresponding to the current model are output, which are determined as the real fracture characteristics of the fractured surrounding rock in the field.

[0023] Optionally, by combining the rheological properties of the grout with the spatial distribution of fractures, the entire process of grout diffusion, filling, and cementation under grouting pressure is reproduced. The filling effect of the grout on fractures and its cementing effect on rock blocks are quantitatively analyzed, and the mechanical modification mechanism of grouting reinforcement of fractured surrounding rock is reconstructed. Specifically, this includes:

[0024] The PFC three-dimensional fracture network numerical model and the corresponding DFN discrete fracture network file, which are converged and locked by fracture inversion, are directly imported into the computing platform, and the original microscopic contact parameters of the rock matrix, fracture spatial distribution characteristics, model boundary constraints and geometric parameters are completely preserved.

[0025] Optionally, based on the distribution of connected channels in the fracture network, the flow rate, diffusion range, and filling saturation of the slurry in fractures with different openings and penetrations can be solved in real time using a fluid-structure interaction calculation algorithm.

[0026] The simulation of high-pressure grouting shows the entire process of grout preferentially migrating and diffusing along large-aperture interconnected fractures, and gradually seeping and filling into small and isolated fractures. Simultaneously, the mechanical responses such as grouting pressure loss, fracture compression deformation, and small rock block displacement are calculated to reflect the compression and expansion effect of grouting pressure on surrounding rock fractures and the seepage driving law, thus fully reproducing the dynamic migration characteristics of grout in a complex fracture network.

[0027] Optionally, after the grout has filled the cracks, the bonding strength, stiffness and shear resistance of the grout inside the cracks are gradually improved according to the preset aging hardening parameters, so that a new type of cemented contact unit is formed at the contact position between the filling grout and the surrounding rock blocks.

[0028] By modifying the bonding model and contact strength parameters of the fracture contact surface to replace the original weak fracture contact properties, the mechanism by which scattered and broken rock blocks interlock and bond together after grout consolidation is simulated, ultimately forming a complete and continuous grout-rock composite structure.

[0029] A storage medium, which is a computer-readable storage medium, has computer-readable program instructions stored thereon, the computer-readable program instructions being used to execute the above-described method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock.

[0030] This invention provides a method for determining the mechanical parameters of grouting rock masses in fractured surrounding rock and a storage medium thereon. The technical solutions provided by the embodiments of this invention bring at least the following beneficial effects:

[0031] (1) High accuracy of parameter acquisition and true reproduction of rock mass and grouting effect: Through the closed loop path of "indoor test - parameter reduction - fracture inversion - grouting simulation", the true fracture distribution characteristics of the fractured surrounding rock and the microscopic effect of cementation of grouting body are accurately reproduced. The mechanical parameters deviate from the actual field by less than 5%, which fundamentally solves the problems of low accuracy and large dispersion of traditional methods.

[0032] (2) The numerical simulation is stable and efficient, with strong adaptability and versatility: The modeling adopts the grouting equivalent substitution approach to avoid grid distortion and non-convergence of calculations, and the crack inversion and grouting simulation are seamlessly connected; it can be adapted to deep fractured surrounding rock scenarios with different crack development degrees, with strong universality, and significantly improves the efficiency and reliability of grouting rock mass numerical simulation.

[0033] (3) The project has good feasibility and directly supports the support design and safety assurance. The obtained mechanical parameters can be directly used in engineering software such as FLAC3D to realize the full-process application from parameter determination to roadway stability analysis and support scheme optimization, which significantly improves the scientific nature of support design, reduces the risk of roadway instability, and ensures safe and efficient mining in deep mines.

[0034] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A flowchart illustrating the method for determining the mechanical parameters of grouting rock masses in fractured surrounding rock. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0038] Before describing the technical solution of the present invention in detail, the technical background and technical terms involved in the technical solution will be explained first:

[0039] Fractured surrounding rock: In the vicinity of deep mine roadways, under the influence of high ground stress, geological structure, and mining disturbance, primary and secondary joints and fissures are densely developed, severely damaging the integrity of the rock mass, resulting in fragmented, blocky, or even loose surrounding rock. Its low mechanical strength, high permeability, and poor self-stabilization capacity make it the direct research and treatment target of this invention. This invention addresses the engineering challenges of complex fissure distribution, large dispersion of mechanical parameters, and difficulty in accurately obtaining parameters after grouting in this type of rock mass. Through indoor experiments, parameter reduction, and fissure inversion, it accurately reconstructs the true fissure characteristics and mechanical state.

[0040] Grouting rock mass: A grout-rock composite structure formed after the fractured surrounding rock is reinforced by grouting. Under pressure, the grout fills the fissures and voids in the fractured surrounding rock, and after consolidation, it tightly bonds with the fractured rock blocks, forming a unified structure that restores continuity and load-bearing capacity to the loose, fragmented rock mass. This is the target research object of the method in this invention. This invention uses PFC numerical simulation to achieve equivalent modeling of the grouting process, accurately obtaining key mechanical parameters of the grouting rock mass such as elastic modulus, cohesion, and internal friction angle, and applying them to FLAC 3D engineering simulation to provide accurate data for tunnel support design.

[0041] FLAC 3D Engineering Simulation: Numerical simulation calculations in the fields of mining and geotechnical engineering, conducted using the FLAC 3D three-dimensional fast Lagrangian analysis program, are the core platform for the final engineering application of mechanical parameters in the technical solution of this invention. Based on the continuum mechanics method, this software can efficiently construct large-scale, realistic engineering-scale three-dimensional geological and engineering models, suitable for complex working conditions such as deep mine tunnel excavation, support, and surrounding rock stability analysis. In the technical solution of this invention, the mechanical parameters of grouting fractured rock mass, accurately inverted by the PFC discrete element software, are directly assigned to different surrounding rock regions in the FLAC 3D model. This allows for a complete simulation of the entire process of stress and deformation response during tunnel excavation and grouting support, ultimately achieving tunnel stability evaluation, support scheme design and optimization. This provides a key platform for the method of this invention to move from indoor experiments and numerical simulations to on-site engineering applications.

[0042] Figure 1 A flowchart illustrating the method for determining the mechanical parameters of grouting rock masses in fractured surrounding rock. The method includes the following steps:

[0043] Step S101: Core sampling, screening, and standard processing of the fractured surrounding rock in the deep tunnel are carried out. Multiple mechanical property tests are conducted using a rigid servo mechanical testing system, and the baseline values ​​of the inherent mechanical parameters of the complete rock block are obtained through mathematical statistics.

[0044] The above steps, used for indoor testing, are the core foundational steps in obtaining complete inherent mechanical reference parameters of the rock mass in this invention. The entire process is strictly performed in accordance with relevant engineering rock mass testing standards. Through standardized on-site sampling, specimen preparation, mechanical testing, and data processing procedures, a unique, stable, and accurate mechanical parameter reference is formed, providing reliable input for subsequent parameter reduction, calibration of rock mass matrix micro-parameters, and numerical simulation of fracture inversion. The specific implementation method is as follows:

[0045] 1. On-site coring and sample screening

[0046] Core sampling was conducted on three or more typical sections of the target deep mine roadway. Diamond drill bits were used for core sampling, and protective measures such as wall protection and vibration damping were implemented throughout the process to minimize mechanical damage to the core and ensure its original integrity. After the cores were transported back to the laboratory, they underwent meticulous screening. Core segments with uniform lithology, good integrity, no natural joints or fissures, no stratification, no weathering or alteration, and no visible defects were prioritized as raw materials for subsequent processing into complete standard test specimens, ensuring the quality of the test specimens from the outset.

[0047] 2. Standard specimen machining and precision control

[0048] Selected core segments were processed into standard cylindrical specimens according to relevant national standards. Strict control was exercised over the specimen's geometric dimensions and processing accuracy to ensure the height-to-diameter ratio met specifications. During processing, strict control was maintained over the specimen's end-face unevenness, perpendicularity deviation between the end face and the specimen axis, and diameter deviation along the specimen's height to ensure compliance with testing accuracy requirements. After processing, each specimen underwent visual and internal defect inspection. Specimens with microcracks or mechanical damage were discarded to ensure the number of valid specimens in each test group met statistical requirements, effectively reducing the dispersion of test data.

[0049] 3. Implementation of indoor mechanical tests

[0050] All mechanical tests were conducted in a temperature-stable environment to avoid interference from temperature and humidity fluctuations. A rock rigid servo mechanical testing system conforming to metrological standards was used to conduct uniaxial compression tests, Brazilian splitting tests, and direct shear tests. The uniaxial compression test employed a displacement-controlled loading mode, continuously loading until specimen failure, to determine the elastic modulus, Poisson's ratio, and uniaxial compressive strength of intact rock blocks. The Brazilian splitting test used a standard splitting fixture, loading in a force-controlled manner until the specimen split along its diameter, determining the tensile strength of the rock block. The direct shear test used a rock direct shear apparatus, applying vertical loads in multiple stages. After each load stage, the load was stabilized until deformation was stable, and then a uniform shear load was applied until shear failure of the specimen, determining the cohesion and internal friction angle of the rock block.

[0051] 4. Statistical analysis of experimental data and determination of benchmark parameters

[0052] The test results of each group of valid specimens underwent standardized statistical processing to remove outlier data with excessive dispersion, retaining only valid test data. The arithmetic mean of the test results of the valid specimens was taken as the benchmark value for the inherent mechanical parameters of the intact rock block, forming a benchmark report for the mechanical parameters of the intact rock block. This benchmark report provides a unique and stable basic input for subsequent field reduction of fractured rock mass parameters and calibration of the microscopic parameters of the rock mass matrix in numerical models, ensuring the accuracy and reliability of data in subsequent technical processes.

[0053] The above steps S101 provide accurate and reliable basic data for subsequent parameter reduction and numerical model calibration, eliminating experimental dispersion and ensuring the uniqueness and stability of parameter benchmarks.

[0054] Step S102: Based on the obtained baseline values ​​of inherent mechanical parameters of the intact rock block, combined with the degree of fracture development and rock mass integrity coefficient of the surrounding rock at the site, and relying on various engineering rock mass classification indicators and using empirical formulas for comprehensive reduction calculation, the indoor parameters are converted into mechanical parameters of fractured rock mass that fit the actual state of the site, and a dedicated parameter database is established to provide target true values ​​for fracture inversion.

[0055] The above steps are used for parameter reduction, which is the core step in this invention to convert the complete rock mass mechanical parameters obtained from laboratory tests into fractured rock mass mechanical parameters that conform to the actual conditions of deep tunnels. Through rock mass quality grading, empirical formula correction, weighted average integration, and database establishment, the accurate conversion of laboratory parameters to field parameters is achieved, providing a unique and reliable target true value for subsequent fracture inversion. The specific implementation method is as follows:

[0056] 1. Determination of reduced foundation and on-site measurement of rock mass quality indicators

[0057] The parameter reduction uses the mechanical parameters of intact rock blocks determined by laboratory tests as the baseline input data. Simultaneously, it relies on detailed on-site engineering geological surveys of the surrounding rock to comprehensively obtain key information such as the distribution characteristics of joints and fissures, rock mass integrity, structural surface condition, groundwater occurrence conditions, and the degree of disturbance during mining and excavation. Based on this, through on-site testing and calculation, the RMR (Rock Mechanics Rating), Q (Rock Mass Quality), and GSI (Geological Strength Index) indicators are obtained. These three indicators are used together to quantitatively characterize the overall quality and degree of fragmentation of the fractured surrounding rock in the field, providing an objective and comprehensive evaluation basis for parameter reduction.

[0058] 2. Calculation of characteristic constants of rock mass based on the Hoek-Brown strength criterion

[0059] Using the Hoek-Brown rock mass strength criterion as the core theoretical basis for parameter reduction, and combining the material constant of intact rock with the tunnel excavation disturbance coefficient, the characteristic constants of the rock mass are calculated. Based on the lithological properties of the surrounding rock at the site, the corresponding intact rock material constant is selected. Combined with the tunnel excavation method, the excavation disturbance coefficient is determined. These parameters, along with the GSI value, are substituted into the criterion formula to calculate the characteristic constants of the rock mass used for estimating rock mass strength and deformation parameters, providing standardized calculation parameters for subsequent mechanical parameter reduction. The specific implementation steps are as follows:

[0060] (1) Determine the basic input parameters

[0061] First, three basic calculation parameters are defined: the uniaxial compressive strength of the intact rock block, which is taken from the benchmark results of indoor tests; the material constant mi of the intact rock, which is selected based on the corresponding empirical value according to the lithology of the surrounding rock on site; and the excavation disturbance coefficient D, which is determined according to the tunnel excavation method, blasting intensity and the degree of disturbance to the surrounding rock.

[0062] At the same time, the GSI geological strength index obtained from the on-site back-calculation is used as the core classification parameter and substituted into the calculation.

[0063] (2) Calculate the characteristic constant a of the rock mass

[0064] Following the Hoek-Brown modified formula for fractured rock masses, the characteristic constant 'a' of the rock mass is calculated using the GSI value as the primary variable. This parameter describes the nonlinearity of the strength envelope after rock mass fracturing. The calculation uses GSI as the core independent variable and employs a modified expression suitable for fractured rock masses to obtain a constant value 'a' that matches the degree of fracturing in the field.

[0065] (3) Calculate the characteristic constant mb of the rock mass

[0066] mb reflects the degree of strength reduction caused by the rock mass structure. It is calculated based on the material constants mi and GSI of intact rock, and is converted using an exponential correction formula. The calculation process fully considers the reduction effect of rock mass fragmentation on strength indicators, and gradually corrects the material properties of intact rock blocks to the equivalent material properties of fragmented rock masses.

[0067] (4) Calculate the characteristic constants s of the rock mass

[0068] s is the rock mass integrity coefficient, reflecting the continuous change in the rock mass from intact to broken. This value decreases significantly for fractured surrounding rock. The calculation uses GSI as the core variable and adopts an exponential decay formula to ensure that the calculation results are consistent with the actual state of the rock mass in the field, where fractures are developed and the rock mass is loose and broken.

[0069] (5) Constant verification and determination

[0070] The three rock mass characteristic constants, a, mb, and s, are combined to form the Hoek-Brown parameter set applicable to fractured surrounding rock in the field. The rationality of these three constants is verified to ensure they conform to the strength variation law of fractured rock mass. Ultimately, this set of constants is used as fixed parameters for subsequent calculations of mechanical parameters such as elastic modulus, compressive strength, cohesion, and internal friction angle, and is used in subsequent empirical formula calculations.

[0071] 3. Use empirical formulas to reduce and estimate the mechanical parameters of the rock mass.

[0072] Based on the established characteristic constants and quality classification indicators of the rock mass, empirical formulas for elastic modulus, uniaxial compressive strength, and tensile strength, suitable for fractured rock masses, were used to progressively reduce and estimate parameters such as elastic modulus, Poisson's ratio, uniaxial compressive strength, tensile strength, cohesion, and internal friction angle obtained from laboratory tests. This resulted in multiple sets of independent estimations of rock mass mechanical parameters. Parallel calculation using multiple formulas improved the reliability of parameter estimation and avoided calculation errors caused by a single formula. The specific implementation process is as follows:

[0073] (1) Estimation of reduction in uniaxial compressive strength of rock mass

[0074] Using the uniaxial compressive strength of intact rock blocks as a benchmark, and combining it with the calculated rock mass characteristic constants a, mb, and s, the Hoek-Brown empirical formula for rock mass compressive strength is substituted into the formula for calculation. The formula correlates the strength of intact rock blocks with the structural planes and fracture development of fractured rock masses. Through exponential calculations and proportional conversions, the equivalent uniaxial compressive strength of fractured rock masses is obtained, achieving strength reduction from intact rock blocks to fractured rock masses.

[0075] (2) Estimation of rock mass elastic modulus reduction

[0076] An empirical formula for the deformation modulus of rock mass, matching the GSI value and uniaxial compressive strength, is adopted. Using the field-estimated uniaxial compressive strength and GSI value as input parameters, the elastic modulus of intact indoor rock blocks is corrected to the elastic modulus of fractured rock mass in the field through logarithmic or exponential correction relationships. This formula fully considers the weakening effect of rock mass fracturing and fracture penetration on rock mass stiffness, ensuring that the calculation results conform to the actual deformation characteristics of high-stress fractured surrounding rock.

[0077] (3) Estimation of reduction in tensile strength of rock mass

[0078] Based on the tensile strength of intact rock blocks, the uniaxial compressive strength of rock mass, and the characteristic constant s, the empirical formula for tensile strength recommended by Hoek-Brown was used for calculation. Since fractured rock masses have extremely low tensile strength, the formula significantly reduces the tensile strength by attenuating the value of s, thus obtaining tensile strength parameters consistent with those of fractured rock masses in the field.

[0079] (4) Estimation of rock mass cohesion and internal friction angle reduction

[0080] Substituting the obtained uniaxial compressive strength, elastic modulus, and characteristic constants of the rock mass into the corresponding empirical formulas, the cohesion and internal friction angle of the fractured rock mass are calculated respectively. The calculation process is based on the nonlinear strength criterion of the rock mass, and the high cohesion and high internal friction angle of the intact rock block are gradually corrected to equivalent parameters under the fractured and loose state, reflecting the significant weakening effect of fracture development on the shear resistance of the rock mass.

[0081] (5) Parallel computation of multiple parameters and integration of results

[0082] For all six key mechanical parameters—elastic modulus, Poisson's ratio, uniaxial compressive strength, tensile strength, cohesion, and internal friction angle—independent calculations were performed using corresponding empirical formulas, resulting in multiple sets of parameter estimation results. All calculations were constrained by the on-site rock mass quality indicators to ensure the mechanical consistency and engineering rationality among the parameters, providing reliable basic data for the subsequent weighted averaging.

[0083] 4. Determination of parameters by multi-index weighted average and actual parameters of fractured rock mass on site

[0084] Based on the fractured structural characteristics of the surrounding rock at the site, weight coefficients were assigned to the three rock mass quality grading indicators: GSI, RMR, and Q. A weighted average was then used to integrate the estimation results of multiple parameter sets. By allocating weights, the influence of the dominant evaluation indicators was highlighted, while the interference of secondary indicators was weakened, eliminating differences between different estimation methods. Ultimately, a unique and stable set of true mechanical parameters for the fractured rock mass at the site was obtained, including elastic modulus, Poisson's ratio, uniaxial compressive strength, tensile strength, cohesion, and internal friction angle. The specific implementation process is as follows:

[0085] (1) Determine the rock mass quality index system to be used for weighting

[0086] The three indicators—GSI geological strength index, RMR rock mass geomechanical classification index, and Q rock mass quality index—obtained from on-site measurements and calculations are used as the weighted basis. These three indicators quantitatively reflect the degree of surrounding rock fragmentation, the development state of structural planes, the integrity of rock mass, and engineering quality from different perspectives, and together constitute the basis for comprehensive evaluation, avoiding the one-sidedness of evaluation by a single indicator.

[0087] (2) Set weighting coefficients based on the characteristics of the rock mass on site.

[0088] Based on the actual characteristics of the surrounding rock in deep tunnels, such as fractured rock, dense fissures, and strong disturbance, a scientific weighting of the three indicators was assigned. The GSI indicator, which is more suitable for fractured rock masses, was used as the primary evaluation indicator and given a higher weight; the RMR and Q indicators were used as auxiliary verification indicators and given corresponding weights, ensuring that the weighting settings fit the on-site engineering geological conditions and highlight the contribution of indicators that are more sensitive to the fractured state.

[0089] (3) Calculate the weighted average of each mechanical parameter.

[0090] For each parameter—elastic modulus, Poisson's ratio, uniaxial compressive strength, tensile strength, cohesion, and angle of internal friction—a weighted calculation is performed. The estimated values ​​of the same parameter under the three index systems are multiplied by their corresponding weights and then summed to obtain the weighted average value of that parameter. During the calculation process, each parameter is calculated independently to ensure the accuracy and consistency of different mechanical parameters.

[0091] (4) Verification and correction of the reasonableness of the calculation results

[0092] The weighted calculation parameters are subjected to mechanical consistency verification to determine whether the parameter combination conforms to the mechanical laws of fractured rock mass, including the matching of strength and modulus, the rationality of shear parameters, and the consistency of integrity and parameter decay trends. If abnormal deviations are found, the weight settings and individual estimated values ​​are checked, abnormal data are removed, and the calculation is recalculated to ensure that the final parameters are true and reasonable.

[0093] (5) Determine the actual mechanical parameters of the fractured rock mass on site

[0094] After verification and correction, the weighted average results were determined as the true mechanical parameters of the fractured rock mass in the field, including elastic modulus, Poisson's ratio, uniaxial compressive strength, tensile strength, cohesion, and internal friction angle. This set of parameters provides a unique target true value and convergence criterion for subsequent fracture inversion, achieving accurate conversion from laboratory test parameters to true parameters of the rock mass in the field.

[0095] 5. Establish a database of surrounding rock parameters and provide target true values ​​for fracture inversion.

[0096] The weighted average of the actual mechanical parameters of the fractured rock mass determined on-site was systematically compiled. Based on the differences in fracture degree and geological conditions in different sections and areas of the tunnel, a mechanical parameter database corresponding to each zone was established. The parameters in this database directly serve as the target true values ​​and convergence criteria for subsequent fracture inversion, ensuring a high degree of matching between the numerical simulation results and the on-site rock mass mechanical state. This provides an accurate data foundation for grouting simulation, parameter testing, and engineering applications. The specific implementation process is as follows:

[0097] (1) Parameter organization and standardized archiving after reduction

[0098] First, the actual mechanical parameters of the fractured rock mass determined by weighted averaging of multiple indicators were comprehensively compiled. The six core parameters—elastic modulus, Poisson's ratio, uniaxial compressive strength, tensile strength, cohesion, and internal friction angle—were checked one by one to ensure numerical accuracy, unit uniformity, and logical consistency. All parameters were standardized in naming and classification, clearly defining the corresponding roadway location, surrounding rock type, fracture grade, rock mass quality indicators, and reduction calculation basis, forming a traceable and verifiable standardized parameter list to eliminate data confusion and the risk of misuse.

[0099] (2) Classify and categorize parameters according to engineering geological conditions.

[0100] Based on the on-site engineering geological conditions of deep tunnels, standardized parameters were categorized and classified according to differences in fracture development, rock mass integrity, structural plane distribution, and groundwater conditions in different sections, cross-sections, and surrounding rock areas. Surrounding rocks with similar geological conditions and degrees of fracturing were grouped into the same parameter zone, each zone was assigned a unique identifier and bound to corresponding mechanical parameters, achieving zoned and refined parameter management of the tunnel's surrounding rock. This allows for precise regional assignment in subsequent numerical simulations, improving the model's alignment with the actual site conditions.

[0101] (3) Construction of the database of surrounding rock mechanical parameters

[0102] The standardized parameters, after being categorized and classified, are entered into a dedicated database. This database uses the tunnel engineering location, surrounding rock zoning, and rock mass quality grade as its core indexes, and stores six core mechanical parameters. It also associates corresponding RMR values, Q values, GSI values, rock mass characteristic constants, and reduction calculation processes as auxiliary information. The database has storage, query, retrieval, and export functions, providing stable and reliable data support for subsequent fracture inversion, grouting simulation, and engineering applications, achieving centralized parameter management and efficient reuse.

[0103] (4) Set the target true value and convergence criterion for fracture inversion.

[0104] The actual mechanical parameters of the corresponding surrounding rock zones are extracted from the database and set as the target true values ​​for the fracture inversion step of the PFC three-dimensional discrete element numerical simulation, serving as the sole reference standard for iterative optimization of fracture parameters. Simultaneously, it is clearly defined that the relative deviations between the simulated values ​​and the target true values ​​of the four core parameters—elastic modulus, cohesion, internal friction angle, and uniaxial compressive strength—are all less than a specified threshold, which serves as the convergence criterion for fracture inversion, providing clear and quantifiable control standards for the iterative optimization of the numerical model.

[0105] (5) Achieve direct connection between parameter database and fracture inversion model.

[0106] By synchronously embedding the target ground truth values ​​locked in the database and the convergence criteria into the PFC numerical simulation process, the fracture inversion model can directly call the target ground truth values ​​for real-time comparison and iterative correction during operation. Data integration ensures that the entire fracture inversion process is guided by the actual rock mass mechanical state on site, avoiding deviations from reality and guaranteeing that the fracture distribution characteristics obtained in the final inversion accurately reflect the surrounding rock condition, providing an accurate and reliable basic model for subsequent grouting simulation.

[0107] Step S102 above can effectively reduce the deviation between numerical simulation and on-site rock mass, and improve the reliability and parameter accuracy of the overall technical solution.

[0108] Step S103: Based on the baseline values ​​of the inherent mechanical parameters of the complete rock block and the target true values, the fracture distribution characteristics of the surrounding rock at the site are restored through controlled variable experiments, sensitivity analysis and multiple rounds of iterative optimization, so that the mechanical response of the numerical model and the deviation of the rock mass at the site meet the convergence requirements, and the fracture characteristics consistent with the mechanical behavior of the rock mass at the site are obtained.

[0109] The above steps are used for fracture inversion, which is the core step in this invention to accurately restore the real fracture structure of deep fractured surrounding rock. Through PFC three-dimensional discrete element numerical simulation, discrete fracture network (DFN) generation, micro-parameter calibration, single-factor experiments, sensitivity analysis, and iterative convergence calculation, fracture characteristics that are highly consistent with the on-site rock mass mechanical behavior are finally obtained, providing a faithful model basis for grouting simulation. The specific implementation method is as follows:

[0110] 1. Numerical calculation framework and construction of proportional numerical samples

[0111] A numerical calculation framework was built based on the PFC three-dimensional discrete element numerical simulation software. The rock mass was discretized into a combination of rigid or deformable blocks. The forces between the blocks were transmitted through normal and tangential contact springs, realistically reflecting the discontinuous deformation and contact mechanics characteristics of jointed and fractured rock masses. A 1:1 scale cylindrical numerical specimen model was established according to the geometric dimensions of standard specimens used in laboratory tests. Rigid walls were used as the model boundaries, and rigid loading plates, consistent with the loading method of laboratory mechanical testing machines, were set at the ends to ensure that the simulated boundary conditions were completely consistent with those in laboratory tests. Simultaneously, based on the detailed investigation results of the surrounding rock structure in the field, the initial value ranges and upper and lower limits of the parameters for fracture attitude, aperture, density, and penetration were determined.

[0112] 2. Benchmark calibration of microscopic parameters of rock mass matrix

[0113] Using the results of laboratory mechanical tests on intact rock blocks as the calibration target, a baseline model of intact rock blocks without fractures was established in the numerical model. A trial-and-error method was used to iteratively adjust core micro-parameters such as normal contact stiffness, tangential contact stiffness, friction coefficient, tensile strength, and bond strength of the particles, ensuring that the relative deviation between the numerical simulation mechanical results and the laboratory test results of the intact rock block was less than a specified threshold. After calibration, the micro-parameters of the rock matrix remained constant throughout the fracture inversion process, using only fracture characteristic parameters as the sole variable to completely eliminate the interference of matrix parameter changes on the inversion results, thus ensuring the uniqueness and reliability of the inversion process.

[0114] 3. Construction of DFN Discrete Slit Network Model

[0115] Based on field survey data of joints and fractures in the surrounding rock, a stochastic fracture generation model was constructed using Discrete Fracture Network (DFN) technology and Monte Carlo stochastic simulation principles. The fracture attitude follows a Fisher distribution or a double normal distribution; the fracture trace length follows a negative exponential distribution or a power-law distribution; the fracture aperture follows a normal distribution; and the fracture spacing follows a negative exponential distribution. Based on this, the value ranges and gradient intervals of the four core variables were defined: the fracture attitude is controlled by the dip angle, with a gradient interval not exceeding a specified angle; the fracture aperture is taken at a fixed gradient within a specified range; the fracture density varies with a fixed number of fractures; and the fracture penetration varies with a fixed percentage gradient, providing a standardized parameter sequence for subsequent controlled variable experiments. The specific implementation is as follows:

[0116] (1) Determining the foundation and probability distribution type

[0117] Based solely on the detailed investigation results of joints and fractures in the surrounding rock at the site, a probability distribution pattern for the geometric and spatial distribution of fractures is determined to ensure that numerical fractures and in-situ fractures exhibit consistent statistical regularities. Specifically, the fracture orientation follows a Fisher distribution or a double normal distribution; the fracture trace length follows a negative exponential distribution or a power-law distribution; the fracture aperture follows a normal distribution; and the fracture spacing follows a negative exponential distribution. Once these probability distribution forms are fixed, they remain unchanged throughout the subsequent inversion process; only the distribution parameters and their value ranges are adjusted and optimized.

[0118] (2) Determine the core variables and initial value range of the fracture.

[0119] Based on field measurement data, four core control variables for fracture inversion were determined: fracture orientation, fracture aperture, fracture density, and fracture penetration. The initial value range and gradient interval of each variable were also defined to ensure that parameter changes are continuous, controllable, and iterative.

[0120] Fracture orientation: The dip angle is the core control index, and the value range covers the minimum to maximum value of the joint dip angle measured in the field, with a gradient interval of no more than 5°.

[0121] Crack aperture: The value range is set to 0.1mm~5.0mm, and the gradient interval is no greater than 0.2mm.

[0122] Crack density: The value range is set from 1 crack / m to 10 cracks / m, and the gradient interval is no greater than 1 crack / m.

[0123] Crack penetration: The value range is set to 20% to 100%, and the gradient interval is no more than 10%.

[0124] (3) Establishment of the fracture parameter variable system

[0125] The four core variables mentioned above are combined into a complete parameter variable system, with each variable containing three elements: distribution type, value range, and gradient step size. In subsequent iterative optimizations, only these parameters are adjusted without changing the basic logic of crack generation, ensuring the stability of the inversion process, the reproducibility of results, and the consistency of the comparison benchmark.

[0126] (4) Setting Monte Carlo random fracture generation rules

[0127] Monte Carlo stochastic simulation is used for fracture generation to ensure that each parameter combination can generate multiple fracture network models with consistent statistical characteristics and random spatial locations, thus eliminating the experimental discreteness caused by random distribution. The generation rules strictly follow the determined probability distribution and parameter gradient to ensure that the fracture models generated under the same parameter combination are statistically equivalent.

[0128] (5) Model embedding numerical specimen and boundary matching

[0129] The constructed DFN discrete fracture network model was embedded into a pre-established 1:1 scale cylindrical numerical specimen. Fractures were randomly generated within the model, and their location, orientation, aperture, density, and penetration completely conformed to the pre-defined distribution rules. The model used rigid wall boundaries, ensuring that the fracture distribution was not affected by the boundaries and remained consistent with the stress state, dimensional conditions, and loading method of the indoor test specimens, thus ensuring that the results of subsequent numerical mechanical tests could be directly compared with field parameters.

[0130] (6) Model validity verification

[0131] After the DFN model is constructed, the number of cracks, aperture, density, and penetration are statistically verified to confirm that they conform to the set parameter range and distribution characteristics, and that there are no unreasonable situations such as abnormal cracks, excessive penetration, or excessive density. This ensures that the model can be directly used for single-factor controlled variable numerical experiments, providing a qualified basic model for subsequent sensitivity analysis and iterative optimization.

[0132] 4. Numerical Experiment with Single-Factor Controlled Variables

[0133] Numerical experiments were conducted using the controlled single-variable method, fixing the probability distributions and initial means of the other three fracture parameters while adjusting only the gradient value of the single objective parameter. For each parameter combination, three parallel discrete fracture network models were generated using Monte Carlo simulation. Uniaxial compression and direct shear numerical mechanical tests were conducted on each model, with loading methods, loading rates, and boundary conditions identical to those in laboratory tests. The arithmetic mean of the three parallel test results was taken as the simulated rock mechanics parameter value corresponding to that set of fracture parameters, reducing the computational dispersion introduced by the stochastic model. The specific implementation is as follows:

[0134] (1) Experimental control principles and variable setting

[0135] This section employs a single-variable control method for numerical experiments. In each set of experiments, any three of the four core fracture parameters—fracture orientation, fracture aperture, fracture density, and fracture penetration—are fixed, maintaining their probability distribution characteristics, interval mean, and gradient values ​​unchanged. Only one target parameter is adjusted in a gradient manner to identify the influence of that parameter on the rock mass mechanical response, avoiding interference caused by simultaneous changes in multiple parameters and ensuring the uniqueness and interpretability of the experimental results.

[0136] (2) Parallel model generation (Monte-Carlo stochastic simulation)

[0137] For each set of fracture parameter combinations with gradient values, the Monte Carlo stochastic simulation principle is used to automatically generate three independent but statistically consistent discrete fracture network (DFN) models. The three models differ only in the specific spatial location and geometric details of the fractures, while the overall statistical parameters such as fracture density, aperture, attitude, and penetration are completely identical. This is used to offset the fluctuations in experimental results caused by random fracture generation and improve data reliability.

[0138] (3) Loading and execution of numerical mechanics experiments

[0139] For each set of parallel DFN models, uniaxial compression numerical tests and direct shear numerical tests were carried out. The loading mode, loading rate, boundary conditions, wall stiffness, and displacement rules of the loading plate were all strictly consistent with the indoor physical tests to ensure that the numerical tests could truly reproduce the stress process and failure characteristics of the indoor tests.

[0140] The model uses a rigid wall boundary, and the end loading plates are loaded at a uniform displacement rate. The stress, strain, and failure mode of the rock mass are recorded throughout the process until the model reaches the peak strength and undergoes stable failure. Key mechanical indicators such as elastic modulus, uniaxial compressive strength, cohesion, and internal friction angle are automatically output.

[0141] (4) Experimental data processing and arithmetic mean taking

[0142] The experimental results of three parallel models under the same parameter combination were statistically processed to remove abnormal discrete points caused by random differences in the models. The arithmetic mean of the three valid results was taken as the simulated value of the rock mechanics parameter corresponding to the fracture parameter of that group.

[0143] By averaging three parallel experiments, the computational dispersion caused by the random fracture network is significantly reduced, ensuring that each set of parameters corresponds to only one stable and reliable mechanical response result, thus providing a standard and regular experimental dataset for subsequent grey relational sensitivity analysis.

[0144] (5) Recording experimental data and establishing relationships between variables

[0145] The parameter gradient values ​​and mechanical response values ​​obtained from all single-factor experiments were systematically organized, and quantitative correspondences between fracture orientation, aperture, density, and penetration and elastic modulus, uniaxial compressive strength, cohesion, and internal friction angle were established to form a complete single-factor experiment database.

[0146] This database is directly used for the next step of calculating the sensitivity weights using the grey relational analysis method, providing direct experimental basis for iterative optimization of crack parameters in descending order of sensitivity.

[0147] 5. Sensitivity analysis and multi-round iterative optimization

[0148] Using the actual mechanical parameters of the fractured rock mass obtained by parameter reduction as the target true values, and based on the results of single-factor tests, the grey relational analysis method is used to calculate the sensitivity weights of fracture orientation, aperture, density, and penetration to four core parameters: elastic modulus, cohesion, internal friction angle, and uniaxial compressive strength. These parameters are then sorted from highest to lowest sensitivity. Following this order, the probability distribution and values ​​of the fracture parameters are iteratively adjusted multiple times. After each adjustment, the DFN model is regenerated and mechanical tests are conducted. The relative deviation between the simulated values ​​and the target true values ​​is calculated using a prescribed formula, achieving gradual approximation and optimization of the model. The specific implementation method is as follows:

[0149] (1) Constructing the basic dataset for sensitivity analysis

[0150] Based on all results obtained from single-factor controlled variable numerical experiments, rock mechanics simulation indices corresponding to different fracture orientations, fracture apertures, fracture densities, and fracture penetration gradients were integrated to form a standardized analysis dataset. The dataset uniformly incorporates four core mechanical response indices: elastic modulus, uniaxial compressive strength, cohesion, and internal friction angle, maintaining consistent data dimensions and calculation methods. Simultaneously, the field fractured rock mechanics parameters determined through parameter reduction are used as fixed reference values, providing a unified benchmark for subsequent correlation calculations and deviation comparisons.

[0151] (2) Quantitative sensitivity calculation was carried out using the grey relational analysis method.

[0152] Grey relational analysis was selected to conduct a multi-factor sensitivity quantitative evaluation. This method can effectively adapt to discrete data and nonlinear variation patterns in geotechnical engineering, and mitigate the data fluctuation interference caused by random fracture models. First, the sequence of fracture geometric parameters and mechanical response parameters were preprocessed to be dimensionless, eliminating calculation errors caused by differences in physical dimensions and numerical magnitudes. Then, the correlation coefficients between each fracture parameter and each mechanical index were calculated sequentially, and a comprehensive correlation coefficient value was obtained through weighted integration. Based on the magnitude of the comprehensive correlation coefficient, the sensitivity of fracture orientation, fracture aperture, fracture density, and fracture penetration was ranked, clarifying the primary and secondary relationships of the influence of each fracture parameter on the mechanical properties of the rock mass, and identifying the core variables with the strongest controlling effect on the overall mechanical performance.

[0153] (3) Formulate hierarchical iterative optimization logic based on sensitivity level

[0154] Based on the sensitivity ranking results, a hierarchical iterative adjustment mechanism is established, proceeding from high-sensitivity parameters to low-sensitivity parameters. Priority is given to gradient fine-tuning of fracture parameters with the highest overall correlation and most significant impact, making small adjustments to the mean and boundary ranges of parameter distribution within a reasonable value range. After the high-sensitivity parameters are optimized to a reasonable range, the moderately sensitive and low-sensitivity fracture parameters are then simultaneously adjusted. This hierarchical optimization approach avoids model disorder caused by simultaneous changes to multiple parameters, reduces the risk of iterative divergence, and improves overall inversion efficiency and correction accuracy.

[0155] (4) Real-time comparison of multiple rounds of iterative cycles and mechanical deviations

[0156] Following a predetermined optimization logic, continuous closed-loop iterative calculations are performed. After each adjustment of fracture parameters, a new DFN discrete fracture network model is generated using the Monte Carlo method, replicating the unified boundary conditions and loading methods. Numerical mechanics tests, such as uniaxial compression and direct shear, are then conducted again to obtain updated rock mechanics simulation parameters in real time. The simulation results are compared with the actual field values ​​in each iteration to quantify and statistically analyze the relative deviations, dynamically tracking the convergence trends of various mechanical indicators. The parameter adjustments, model status, and deviation data for each iteration are continuously recorded to form a complete iteration log.

[0157] (5) Parameter constraint control and model rationality verification

[0158] Throughout the iterative optimization process, the adjustment range of each fracture parameter was strictly limited, always adhering to the fracture statistical range obtained from on-site geological surveys, and avoiding extreme parameter values ​​that deviated from engineering reality. After each round of parameter correction, the spatial distribution morphology of the fracture network, the fracture interweaving state, and the filling structure characteristics were simultaneously verified to ensure that the fracture development law within the numerical model remained consistent with the actual structural characteristics of the deep fractured surrounding rock. While minimizing the deviation of mechanical parameters, the model also considered the rationality of fracture geometry and geological laws, preventing model distortion caused by solely pursuing numerical convergence.

[0159] (6) Dynamic correction of iterative direction and gradual convergence

[0160] Based on the deviation calculation results of each round, the parameter adjustment direction and step size are dynamically corrected. When the simulated mechanical parameters are higher than the target true values, the values ​​of weakened fracture parameters are appropriately increased; when the simulated mechanical parameters are lower than the target true values, the fracture development-related indicators are reasonably tightened, and the convergence speed is accelerated through adaptive step size adjustment. Through multiple rounds of repeated correction, calculation, comparison, and fine-tuning, the simulated values ​​of various core mechanical indicators gradually approach the target true values ​​of the fractured rock mass in the field, continuously compressing the data deviation and laying a solid foundation for subsequent convergence judgment and final determination of fracture characteristics.

[0161] 6. Convergence determination and output of true fracture characteristics

[0162] The convergence criterion is that the relative deviations of the four core mechanical parameters are all less than a specified threshold. If the convergence requirement is not met, the iteration optimization step is returned for further adjustment; if the convergence condition is met, the iteration is terminated, and the complete parameters and probability distribution characteristics of the fracture attitude, aperture, density, and penetration corresponding to the current model are output, which are then identified as the true fracture characteristics of the fractured surrounding rock in the field. Simultaneously, the PFC numerical model and DFN fracture network file in the converged state are saved and directly used as the base model for the subsequent grouting simulation, achieving seamless integration between fracture inversion and grouting simulation. The specific implementation is as follows:

[0163] 1. Core criteria and verification process for convergence determination

[0164] The convergence criterion takes "a high degree of matching between the numerical simulation results and the actual field conditions" as its core objective. It uses the actual mechanical parameters of the fractured rock mass determined in the parameter reduction process as the sole objective truth value. Combining experimental and iterative data from the entire fracture inversion process, and through quantitative deviation calculation and multi-index comprehensive verification, it determines whether the fracture inversion has met the convergence requirements. The specific process and standards are as follows:

[0165] Define the core indicators for convergence determination: Select four core mechanical parameters as key indicators for convergence determination: elastic modulus, uniaxial compressive strength, cohesion, and internal friction angle. These four indicators directly reflect the deformation, strength, and shear resistance of the rock mass. They are the core parameters that characterize the influence of fractures on the mechanical properties of the rock mass, and are also the key inputs for subsequent grouting simulation and engineering applications.

[0166] Quantitative deviation calculation method: The deviation between the simulated value and the target true value of each core mechanical parameter is calculated using the relative deviation formula, as follows:

[0167] Relative deviation = |simulated value - target true value| / target true value × 100%

[0168] The above calculation process retains two decimal places to ensure accurate deviation quantification and avoid misjudgment of convergence due to calculation errors.

[0169] Convergence threshold setting: Combining the accuracy requirements of deep fractured surrounding rock engineering and the industry standards for numerical simulation, the convergence threshold is set as "the relative deviation of the four core mechanical parameters is less than 5%". This threshold ensures the fit between the simulation results and the actual situation on site, while also taking into account the rationality and operability of the numerical simulation. At the same time, it matches the accuracy requirements of the previous indoor tests and parameter reduction, forming a closed loop of accuracy throughout the process.

[0170] Multiple rounds of verification and result confirmation: After each round of iterative optimization, the above deviation calculation and verification process is repeated. If in a certain round of verification, all four core indicators meet the condition that the relative deviation is <5%, it is initially determined to be converged. Then, three sets of parallel DFN models are generated again to carry out verification experiments. If the verification experiment results still meet the convergence threshold and the deviation fluctuation range is less than 1%, the fracture inversion is officially confirmed to have met the convergence requirements, and the iteration is terminated. If the convergence threshold is not met, the process returns to the sensitivity analysis and iterative optimization stage to continue adjusting the fracture parameters until the convergence standard is met.

[0171] 2. Output of true fracture characteristics

[0172] Once the fracture inversion reaches convergence, all parameters and model states are immediately locked, and a complete set of real fracture characteristic results is output. All output results have undergone accuracy verification and logical checking, and can be directly used for subsequent grouting simulation without additional conversion or correction. The specific output content is divided into four categories, each with clear details, specifications, and application scenarios, which are in line with the overall requirements of the technical solution of this application:

[0173] (1) Geometric and distribution characteristics of actual fractures in the fractured surrounding rock at the site

[0174] This type of data is core data characterizing the fractured state of the surrounding rock at the site, accurately replicating the real distribution pattern of fractures at the site, and providing direct basis for "simulation of grout diffusion path and filling effect" in grouting simulation. Specifically, it includes the following detailed parameters, all of which are quantitative values ​​and can be directly used:

[0175] Fracture attitude parameters: Determine the specific values ​​and probability distribution characteristics of the dip angle, strike, and dip direction of the fracture. The dip angle is in units of precision of 1°, and the range of values ​​is consistent with the field measurements. The probability distribution follows the Fisher distribution or double normal distribution determined when the DFN model was constructed. At the same time, output the distribution parameters (such as concentration coefficient) to clarify the statistical law of fracture attitude. The strike and dip directions are accurately marked according to the geological azimuth (0°-360°) to ensure complete matching with the spatial orientation of the fractures in the surrounding rock of the roadway.

[0176] Fracture aperture parameters: Output the specific numerical range, mean, and standard deviation of fracture aperture, with an accuracy unit of 0.01mm. The value range is strictly limited to the range of 0.1mm-5.0mm measured on site. At the same time, the normal distribution parameters of the aperture are specified to distinguish the differences in the aperture of fractures in different areas (such as the difference in the aperture of fractures around the tunnel and fractures in the deep surrounding rock), providing data support for subsequent grouting slurry selection and grouting pressure setting.

[0177] The fracture density parameter is output in units of "fractures / m". The specific value, mean and gradient distribution characteristics of the fracture density are output. The value range is consistent with the field measurement of 1 fracture / m to 10 fractures / m. The difference in fracture density in different roadway areas and different fracture grades is clearly defined. At the same time, the negative exponential distribution parameter of fracture density is marked to reflect the spatial uniformity of fracture distribution, which provides a basis for the layout of grouting holes and the determination of grouting range.

[0178] Crack penetration parameter: Outputs the specific value and distribution characteristics of crack penetration in percentage (%), with a value range of 20%-100%. It accurately distinguishes the proportion of through cracks, semi-through cracks and non-through cracks, clarifies the connectivity of the crack network, and outputs the statistical distribution law of penetration, providing core parameters for simulating the diffusion path and filling efficiency of grout in cracks.

[0179] Additional feature parameters: additional output of fracture trace length (unit: m), spacing (unit: m) and probability distribution characteristics (trace length follows a negative exponential distribution or a power law distribution, spacing follows a negative exponential distribution), while also marking the fracture filling state (e.g., no filling, mud filling, sand filling) and filling material thickness, to completely replicate the real geometric and structural characteristics of the fractures on site and avoid missing key influencing factors.

[0180] (2) A convergent PFC three-dimensional fracture network numerical model (three-dimensional discrete element numerical model)

[0181] This model is the core entity for fracture inversion and seamlessly integrates with subsequent grouting simulations without requiring remodeling or parameter adjustments. Specific details are as follows:

[0182] Basic model information: The model is a 1:1 scale cylinder with dimensions completely consistent with the standard specimens in the indoor test. Boundary conditions, loading methods, wall stiffness, and loading plate parameters are all consistent with the indoor test and previous numerical tests to ensure that the mechanical response of the model is highly consistent with the field rock mass and indoor test results.

[0183] The model's core features include a converged DFN discrete fracture network. The spatial location, geometry, and distribution of the fractures perfectly match those in the field. The microscopic parameters of the rock matrix (normal contact stiffness, tangential contact stiffness, friction coefficient, etc.) have been calibrated and fixed through indoor tests. The fracture parameters have been locked to their true values ​​after convergence. The overall mechanical response of the model meets the convergence criterion of "relative deviation of the four core parameters < 5%".

[0184] Model operability: The model is saved in a format that can be directly read by PFC software. It contains complete particle information, contact relationships, fracture network data, boundary settings and loading parameters. After opening, grouting simulation can be carried out directly without additional parameter adjustment or supplementary modeling. This achieves a seamless connection between "fracture inversion and grouting simulation" and improves the efficiency of the overall technical solution.

[0185] (3) DFN Discrete Fracture Network File

[0186] This file is a digital record of fracture characteristics, comprehensively documenting detailed information about all fractures. It can be independently accessed, modified, and reused. The specific content is as follows:

[0187] File Format and Compatibility: The file is in a PFC software-specific format, which can be directly read, edited, and called by the PFC 3D Discrete Element Software. It can also be exported to a general data format, which facilitates subsequent data processing, analysis, and engineering report preparation.

[0188] The core content of the document includes: a complete record of each fracture's unique identifier, spatial coordinates (x, y, z), geometric parameters (dip angle, strike, dip direction, aperture, trace length, spacing), distribution characteristics (probability distribution parameters), continuity status, and filling information. It also marks the Monte Carlo stochastic simulation parameters used to generate the fractures, ensuring that the fracture data is traceable and verifiable.

[0189] The file's application value lies in its ability to serve as a foundational data file for subsequent grouting simulations, tunnel stability analyses, and support scheme optimizations. If subsequent engineering geological conditions undergo minor changes, the fracture parameters in the file can be directly modified without the need to perform a complete fracture inversion again, thus enhancing the flexibility and reusability of the technical solution.

[0190] (4) Fracture inversion convergence judgment report

[0191] This report serves as the official basis for confirming the reliability and validity of the fracture inversion results, and supports the rationality of subsequent grouting simulation and engineering applications. The specific content is as follows:

[0192] Detailed convergence verification data: Clearly record the crack parameter adjustment amount, simulated values ​​of the four core mechanical parameters, target true values ​​and relative deviations for each iteration, forming a complete iteration deviation record table, clearly presenting the entire process of parameter optimization and deviation reduction, and proving the rationality and rigor of the convergence process.

[0193] Convergence Qualified Conclusion: Clearly indicate the formal conclusion, such as "The relative deviations of the four core mechanical parameters, namely elastic modulus, uniaxial compressive strength, cohesion, and internal friction angle, are all less than 5%, and the deviation fluctuation range of the verification test is less than 1%, and the crack inversion meets the convergence requirements", confirming that the output crack characteristics and numerical model meet the actual engineering needs.

[0194] Results Description and Connection Suggestions: Briefly describe the core process of this fracture inversion, the technical methods used (PFC discrete element method, DFN technology, Monte Carlo simulation, etc.), clarify the application scenarios of the output results, propose the method of calling fracture parameters in subsequent grouting simulations, ensure seamless connection with the subsequent stages of the technical solution in this application, and indicate the applicable scope and accuracy level of the results to provide a reference for engineering applications.

[0195] 3. Validation and archiving of output results

[0196] Results Verification: After outputting all results, a double verification is carried out again. On the one hand, the consistency between the fracture parameters and the field geological survey data is checked to ensure that there are no parameter values ​​that deviate from the actual field conditions. On the other hand, the deviation between the mechanical response of the numerical model and the target true value is checked to ensure that the convergence criteria are met and to prevent the results from being distorted.

[0197] Results archiving: All output results (fracture parameter list, PFC numerical model, DFN file, convergence judgment report) are archived in a standardized format, and a result association index is established, marking the corresponding roadway area, fracture level and inversion time of the results, so as to facilitate subsequent query, reuse and engineering acceptance. At the same time, a complete fracture inversion result archive is formed to support the integrity and traceability of the technical solution of this application.

[0198] All outputs from step S103 above directly serve the subsequent grouting simulation stage: the DFN discrete fracture network file and the PFC numerical model are directly used as the basic models for grouting simulation; the fracture geometry and distribution parameters are used to set the grout diffusion path, filling range, and grouting parameters; and the convergence judgment report is used to confirm the reliability of the basic data for grouting simulation, achieving a seamless connection between "fracture inversion and grouting simulation," providing accurate and reliable basic support for subsequent post-grouting parameter testing and engineering applications, and ensuring the continuity and integrity of the technical solution of this application.

[0199] Step S104: Construct a grout seepage-rock mass coupled calculation system based on the three-dimensional fracture network numerical model. Combine the rheological properties of grout with the spatial distribution law of fractures to reproduce the entire process of grout diffusion, filling and cementation under grouting pressure. Quantitatively analyze the filling effect of grout on fractures and the cementation effect on rock blocks, and restore the mechanical modification mechanism of grouting reinforcement of fractured surrounding rock.

[0200] The above steps are used for grouting simulation, which is a key step in this invention to characterize the grouting reinforcement mechanism of fractured surrounding rock and quantify the formation law of magma-rock composite structure. Based on the qualified three-dimensional numerical model and discrete fracture network data output by fracture inversion, combined with the physical and mechanical properties of the grout and the on-site grouting construction conditions, fluid-structure interaction numerical simulation calculations are carried out to fully simulate the physical response and structural evolution of the entire grouting process, achieving a refined reproduction of the grouting reinforcement process of fractured surrounding rock. The specific implementation method is as follows:

[0201] 1. Importing and Preprocessing the Basic Simulation Model

[0202] The PFC 3D fracture network numerical model and its corresponding DFN discrete fracture network file, which were converged and locked by fracture inversion, were directly imported into the computing platform. This process fully preserved the original microscopic contact parameters of the rock matrix, the spatial distribution characteristics of fractures, the model boundary constraints, and the geometric dimensions. Preprocessing was performed on the fracture structure within the model to uniformly identify the spatial connectivity of continuous, discontinuous, and closed fractures. This process also delineated slurry flow channels and closed void regions, while eliminating distorted contact and invalid fracture elements. The model's computational mesh and contact operation logic were optimized to ensure the stability and convergence of subsequent slurry seepage calculations, guaranteeing complete consistency between the simulation model and the characteristics of the fractured surrounding rock structure in the field.

[0203] 2. Setting of slurry material parameters and rheological model

[0204] Based on the characteristics of commonly used grouting materials in deep tunnel grouting projects, fundamental physical properties such as grout density, viscosity, setting time, and cohesion are determined. A grout rheological constitutive model suitable for fracture seepage scenarios is selected, fully considering rheological effects such as shear thinning and viscosity time-varying within narrow fractures. According to actual grouting conditions on site, the grout consolidation and hardening law is established, defining the strength growth curve of the grout as it gradually transforms from a liquid flow state to a solid cementitious body. The material mechanical differences between the flow stage, initial setting stage, and final setting stage are distinguished, ensuring that the grout material properties conform to the mechanical evolution law of actual grouting materials on site.

[0205] 3. Assignment of grouting boundary conditions and construction parameters

[0206] Referring to the on-site grouting construction process, the numerical model was designed to rationally set the grouting hole layout, grouting inlet range, and grout application method. A pressure-controlled grouting loading mode was adopted, applying constant grouting pressure in stages to simulate the construction logic of segmented pressure-stabilized grouting on-site. Seepage boundaries and water-tight constraints were set around the model to limit ineffective grout seepage and boundary overflow, thus reproducing the influence of in-situ stress in deep surrounding rock and groundwater environment on grout diffusion. The seepage calculation step size and the mechanical solution iteration step size were unified to achieve synchronous coupling of fluid dynamics and rock mechanics calculations.

[0207] 4. Simulation of the seepage, diffusion, and filling process of grout in fractures

[0208] Based on the distribution of connected channels in a fracture network, a fluid-structure interaction algorithm is used to calculate in real time the flow rate, diffusion range, and filling saturation of grout in fractures with different apertures and penetration depths. The simulation demonstrates the entire process under high-pressure grouting, where the grout preferentially migrates and diffuses along large-aperture connected fractures, gradually permeating and filling smaller and isolated fractures. Simultaneously, the mechanical responses, such as grouting pressure loss, fracture compression deformation, and minor rock block displacement, are calculated to reflect the compression and expansion effect of grouting pressure on surrounding rock fractures and the seepage driving law, fully reproducing the dynamic migration characteristics of grout in a complex fracture network.

[0209] The calculation formula for grout seepage control is as follows:

[0210]

[0211] The above formula is the continuity equation for unsteady seepage of grout within a fracture network. Based on the laws of mass conservation and fluid motion, it describes the dynamic evolution of grout pressure within the fracture channel under grouting pressure, considering spatial position and grouting time. It is the core control formula for fluid-structure interaction calculations. Where k is the equivalent permeability of the fracture (m²), representing the ability of the fracture to allow grout to pass through; P is the instantaneous grouting pressure (Pa), representing the driving force for grout diffusion; and x is the spatial coordinate of the grout flow direction (m). ρ is the grout density, in kg / m³, and ρ is an inherent physical parameter of the grouting material; c is the grout compressibility coefficient, in Pa. -1 , reflects the volume compression characteristics of the grout under high pressure; t is the grouting duration, in seconds.

[0212] The formula for calculating fracture permeability is as follows:

[0213]

[0214] The above formula is based on the classical theory of crack seepage in smooth parallel plates, revealing that crack permeability is proportional to the cube of crack aperture, accurately quantifying the physical mechanism of "preferential diffusion of grout in large-aperture cracks," which is completely consistent with the field grouting law. Wherein, k is the equivalent permeability of the crack, in m²; b is the aperture of a single crack, in meters, taken from the crack inversion convergence result; This represents the dynamic viscosity of the slurry, measured in Pa•s, and varies dynamically with slurry type and setting time. The specific implementation method is as follows:

[0215] First, based on the already constructed DFN discrete fracture network, all fracture surfaces, fracture voids, and particle contact gaps within the model are identified and divided in their entirety. Through fractures, branch fractures, and micro fractures are defined as independent fluid channel units. The coupling correspondence between the fracture fluid mesh and the rock mass particle solid unit is established, and the fluid flow path, closed and blocked areas, and boundary seepage restriction range are clarified, thus constructing the basic calculation carrier for slurry flow.

[0216] Secondly, pre-calibrated grout rheological parameters are imported, and a fluid constitutive model adapted to flow in narrow cracks is adopted. The time-dependent viscosity of the grout, shear thinning characteristics, and frictional resistance of the crack wall are considered. The fluid calculation time step and the mechanical solution step are set to be synchronously coupled and iterated. Using the actual grouting pressure on site as a constant driving load, a constant pressure boundary condition is applied at the preset grouting hole location to form a stable pressure gradient field, which serves as the core driving force for grout diffusion.

[0217] During the calculation iteration process, the software solves in real time the seepage velocity, pressure loss, friction resistance, and flow distribution within a single fracture. Based on the different fracture opening sizes, connectivity, spatial tortuosity, and roughness, it automatically allocates the grout flow rate. Main fractures with large openings and high connectivity experience lower flow resistance, allowing the grout to migrate and diffuse rapidly. Small fractures, isolated fractures, and low-connectivity structural surfaces are limited by resistance, only slowly permeating and wetting under continuous grouting pressure, accurately reflecting the non-uniform diffusion characteristics of the grout.

[0218] Simultaneously, fluid-structure interaction calculations are initiated. The grouting pressure acts on the fracture wall, generating normal compression. Real-time calculations are performed on micro-deformations such as fracture expansion under pressure, rock block micro-slippage, and changes in contact gaps. The dynamic changes in fracture morphology, in turn, correct the fluid channel size and seepage resistance, forming a two-way dynamic feedback mechanism between fluid pressure and rock mass deformation. The system statistically analyzes the grout filling saturation of each fracture unit hourly, distinguishing between fully filled, partially filled, and unfilled areas, and dynamically updates the fluid domain range until the grout diffusion range stabilizes and the pressure field distribution tends to balance, completing the numerical simulation of the entire fracture filling process.

[0219] 5. Simulation of Cementation and Solidification of Magmatic Rock and Generation of Composite Structures

[0220] After the grout fills the fissures, the bonding strength, stiffness, and shear strength of the grout inside the fissures are gradually improved according to the preset aging hardening parameters, so that a new type of cemented contact unit is formed at the contact point between the filling grout and the surrounding rock blocks. By modifying the bonding model and contact strength parameters of the fissure contact surface to replace the original weak fissure contact properties, the mechanism of the grout consolidating and interlocking of scattered and broken rock blocks is simulated, ultimately forming a complete and continuous grout-rock composite structure. The formation process and structural modification mechanism of the grouted rock mass are restored from a microscopic level.

[0221] The constitutive model for pre-cemented weak fractures in contact with the cementation process is calculated using the following formula:

[0222]

[0223] This calculation formula describes the mechanical behavior of primary fractures in fractured surrounding rock. Primary fractures lack cohesion and cannot withstand tensile stress, relying solely on friction between the fracture surfaces to transmit shear force. This is the fundamental reason for the fractured rock mass and its weak mechanical properties. Among other things, The normal stress along the fracture surface is expressed in Pa, with compressive stress being negative and tensile stress being positive. The shear stress at the crack surface is expressed in Pa. θ is the natural internal friction angle of the fracture wall, in degrees (°); c is the fracture cohesion, in Pa, and the cohesion of the original weak fracture is 0.

[0224] The constitutive model for cemented magma-rock composite bonded contact is calculated using the following formula:

[0225]

[0226] This calculation formula completely replaces the original weak fracture contact with the bonding contact formed by the solidified grout, achieving tensile, shear, and interlocking bonding of the fractured rock blocks, and reconstructing the continuity and integrity of the rock mass at the microscopic level. Wherein, Fn is the normal bond force at the cementing surface, in N; kn is the normal contact stiffness at the cementing surface, in N / m; un is the normal relative displacement at the cementing surface, in m; Fs is the tangential bond force at the cementing surface, in N; ks is the tangential contact stiffness at the cementing surface, in N / m; and us is the tangential relative displacement at the cementing surface, in m. cbond represents the ultimate normal tensile strength of the cemented surface, in Pa; cbond represents the magma-rock cement cohesion, in Pa. The internal friction angle of the magma-rock cementation, in degrees (°). The normal stress of the bonded surface is expressed in Pa.

[0227] The rules for determining equivalent substitution of grouting are as follows:

[0228] If the crack filling rate is ≥85%, the slurry curing strength parameter will be directly used for assignment.

[0229] If the fracture filling rate is 40%-85%, the weighted values ​​of the grout and rock mass mechanical parameters are used.

[0230] If the crack filling rate is less than 40%, the original weak crack contact properties remain unchanged.

[0231] 6. Model status locking and data recording after grouting is completed

[0232] After the grout has completely solidified and hardened, and the stress and displacement fields of the model have reached a stable equilibrium, the grouting coupling calculation is terminated, and the microscopic contact properties, fracture filling state, grout distribution range, and composite structure contact parameters of the overall model are locked. Key data such as the grout diffusion range, fracture filling rate, changes in contact bonding parameters, and rock mass stress-strain response are recorded throughout the entire grouting process, forming a complete grouting simulation dataset. The overall numerical model after reinforcement can be directly used for subsequent back-calculation of grouting rock mass mechanical parameters and performance testing, achieving effective connection between grouting simulation results and the next technical step. The specific implementation method is as follows:

[0233] After the grout has completed its full-area diffusion and crack filling, and the fluid pressure field has stabilized, the seepage calculation module is shut down, and the grout consolidation and hardening calculation logic is started. Based on the hydration and hardening law of the grouting material, the time sequence parameters and mechanical evolution curves for the three stages of initial setting, final setting, and later strength growth are set in segments. A progressive parameter assignment method is adopted to gradually increase the bulk modulus, shear modulus, apparent viscosity, and solidification strength of the grout unit inside the crack, simulating the physical change process of the grout smoothly transitioning from a liquid state to a plastic state and then to a solid state.

[0234] For all fracture contact surfaces filled with grout, the original weak fracture contact constitutive models were replaced in batches, and the low-strength contact properties of the original structural surfaces of the fractured rock mass were deleted. A new grout-rock composite bonded contact model was then constructed. High-strength cemented contact bonds were generated at the interface between the rock blocks and the solidified grout. The normal bond strength, tangential bond strength, friction coefficient, and contact stiffness of the composite structure were assigned values ​​sequentially, thereby forming reliable bond constraints between the discrete fractured rock blocks and limiting their mutual slippage, separation, and rotational deformation.

[0235] As the hardening process continues, the mechanical parameters of the cemented contact are continuously strengthened, gradually eliminating the structural discontinuities and mechanical weakening defects caused by the original fissures. Relying on the filling, encapsulation, and overlapping effects of the solidified slurry within the fissures, the originally loose, discrete, and relatively independent fragmented rock blocks are linked together into a whole, completing the reorganization and improvement of the surrounding rock structure. The distribution of contact forces, the number of bond formations, the overall stress level, and deformation characteristics within the model are monitored synchronously throughout the process, and the mechanical stability of the composite structure is checked in real time.

[0236] Once the grout has fully hardened and all cementation contact parameters have stabilized, and the overall stress and displacement fields of the model have reached a long-term equilibrium, all microscopic contact parameters and structural unit states are locked. This results in a numerical model of the grout-rock composite rock mass that is structurally continuous, homogeneous in performance, and significantly enhanced in strength. This model provides a standardized and directly applicable basic calculation model for subsequent testing of rock mass mechanical parameters after grouting and for quantitative analysis of reinforcement effects.

[0237] Step S105: Conduct multiple sets of mechanical simulation tests, systematically determine the overall mechanical performance indicators of the rock mass after grouting reinforcement, compare the changes in the mechanical properties of the rock mass before and after grouting, quantitatively analyze the improvement effect of grouting filling and cementation reinforcement, and obtain the mechanical benchmark data of the reinforced rock mass.

[0238] The above steps are used for post-grouting parameter testing, providing a scientific and accurate quantitative basis for subsequent surrounding rock stability evaluation, support optimization, and engineering parameter application. The specific implementation method is as follows:

[0239] 1. Preparation of the basic test model and unification of experimental conditions.

[0240] Post-grouting parameter testing focuses on a three-dimensional numerical model of the magma-rock composite, where grouting simulation is complete, the grout is fully consolidated and hardened, and the stress state tends towards equilibrium. The model fully preserves the grout filling distribution, magma-rock bonding relationship, microscopic parameters of the rock matrix, and the modified fracture structure characteristics. Strict adherence to the unified model dimensions, boundary constraints, loading control methods, and computational step sizes used in previous indoor physical experiments and fracture inversion numerical experiments ensures complete consistency of experimental conditions before and after grouting, eliminating comparison errors caused by boundary and loading differences and ensuring the comparability of all test results. Simultaneously, the model is preprocessed to remove abnormal contact elements, balance the initial stress field, and avoid interference from initial stress disturbances on mechanical test results, laying the foundation for standardized mechanical testing.

[0241] 2. Multiple types of mechanical numerical tests were carried out simultaneously.

[0242] Based on the discrete element method (DEM) numerical computing platform, uniaxial compression numerical tests, splitting numerical tests, and shear numerical tests were carried out sequentially, comprehensively covering the detection of key mechanical properties of rock mass deformation, compressive strength, tensile strength, and shear strength. The uniaxial compression test adopted a displacement-controlled loading mode, applying axial load uniformly through a rigid loading plate at the end, monitoring the entire process of stress-strain evolution of the model until the composite rock mass reached its peak strength and underwent overall failure. The splitting test used a standard loading arrangement to accurately obtain the tensile performance indicators of the rock mass after grouting. The direct shear test applied normal constraint loads in stages, and after the deformation stabilized at each load level, shear loads were applied, completely collecting data on the entire shear failure process, systematically characterizing the shear mechanical characteristics of the magmatic-rock composite rock mass.

[0243] 3. Extraction of mechanical parameters of composite rock mass after grouting

[0244] By combining monitoring data from the entire mechanical testing process, multiple core mechanical indicators after grouting reinforcement are extracted simultaneously, including key parameters such as overall deformation modulus, compressive strength, tensile strength, internal friction angle, and cohesion. Based on the numerical model's microscopic contact changes, crack closure states, and cemented bond damage evolution, the deformation characteristics, peak strength, residual strength, and failure modes of the composite rock mass under load are accurately captured. Microscopic data such as stress-strain curves, contact force changes, and cemented structure damage and fracture processes are fully recorded, forming a multi-dimensional, full-process set of raw test data to ensure the completeness and accuracy of parameter extraction.

[0245] 4. Comparison of parameters before and after grouting and analysis of reinforcement mechanism

[0246] The mechanical parameters of the composite rock mass obtained after grouting testing were compared layer by layer with the parameters of the natural fractured rock mass obtained by parameter reduction and the baseline parameters of the intact rock blocks in the laboratory. The improvement rate and proportion of each mechanical index were quantitatively calculated. Combining structural modification characteristics such as fracture filling rate, magma cementation range, and degree of closure of primary fractures, the improvement mechanism of grout filling density and cementation effect on the overall integrity and structural continuity of the rock mass was analyzed. The strengthening effect of grouting on weakened structures was clarified. From the dual perspectives of macroscopic mechanical response and microscopic structural evolution, the intrinsic law of grouting improving the overall bearing capacity and deformation resistance of fractured surrounding rock was elucidated.

[0247] 5. Data filtering, statistics, and final parameter determination

[0248] Multiple sets of parallel numerical test results were uniformly organized and screened to eliminate abnormal data caused by computational fluctuations. The arithmetic mean method was used to statistically process the valid test data, reducing the result deviation caused by numerical dispersion. Combined with the overall stability verification results of the composite rock mass structure, the rationality of various mechanical parameters was comprehensively judged, and a standardized comprehensive mechanical parameter system for the grouting rock mass was finally determined, forming complete test result data. This set of parameters can be directly used as input conditions for engineering design, surrounding rock support calculation, and stability analysis, realizing the effective transformation of numerical simulation results into practical engineering parameters and improving the data closed loop of the entire process of the technical solution in this application.

[0249] Step S106: Integrate multi-level mechanical data of intact rock blocks, naturally fractured rock masses, and grout-modified rock masses, and combine them with the on-site engineering geological conditions and stress environment of the surrounding rock in the tunnel. Apply the obtained mechanical benchmark data of the reinforced rock mass to the calculation of surrounding rock stability, support structure design, and disaster risk assessment.

[0250] The above steps are used for parameter application. Based on the composite rock mass mechanical indices calibrated after grouting reinforcement, the entire refined parameter system is applied to surrounding rock stability calculations, support structure design, and disaster risk assessment. This effectively reflects the mechanical evolution characteristics of reinforced fractured surrounding rock, optimizes engineering treatment schemes, and enhances the scientific rigor, rationality, and overall safety reserve of surrounding rock control in deep roadways, forming a complete closed-loop technology system for rock mass parameter acquisition and engineering application. The specific implementation method is as follows:

[0251] 1. Integration and analysis of multi-stage rock mass parameter system

[0252] In the early stages of parameter application, the system integrated the layered rock mechanics data generated throughout the entire application process. It uniformly included the complete rock block foundation parameters measured by laboratory tests, the in-situ parameters of naturally fractured rock mass after reduction and correction, and the reinforcement parameters of the magmatic-rock composite rock mass finally determined by grouting simulation and post-grouting parameter testing. Various parameters under different working conditions, different fracture levels, and different reinforcement conditions were categorized, sorted, and uniformly verified to ensure that all mechanical indicators had consistent calculation methods, continuous value logic, and matched data accuracy. The applicable scenarios and boundary conditions of each set of parameters were clarified, constructing a complete surrounding rock parameter system that is layered, graded, and can be used differentiatedly, providing a standardized data foundation for subsequent engineering applications.

[0253] 2. Secondary assignment of refined numerical model of surrounding rock

[0254] Based on the overall numerical calculation model of the tunnel project, differentiated parameter assignments were implemented according to the degree of rock fragmentation and the grouting reinforcement range in different areas. For areas with unreinforced original surrounding rock, mechanical parameters of naturally fractured rock mass were imported; for reinforced areas covered by grouting, measured mechanical parameters of composite rock mass after grouting were substituted in batches, accurately reproducing the structural and mechanical property differences of different sections of the tunnel surrounding rock. Simultaneously, the model's internal structural surface contact properties, rock mass deformation control indicators, and strength yield criteria were updated, enabling the overall calculation model to realistically reflect the heterogeneous mechanical characteristics of the surrounding rock before and after grouting reinforcement, thus improving the simulation accuracy of large-scale numerical calculations.

[0255] 3. Quantitative analysis and calculation of the stability of the surrounding rock in the tunnel

[0256] Based on the integrated set of rock mass parameters, a systematic calculation and analysis of the stress field, displacement field, and plastic zone distribution of the surrounding rock in deep roadways was conducted. Combining in-situ stress conditions, the impact of excavation disturbance, and the synergistic effect of the support structure, the stress concentration degree, surface convergence deformation, plastic failure range, and potential instability zone distribution patterns of the surrounding rock before and after reinforcement were determined. By comparing the calculation results under multiple working conditions, the actual effect of grouting reinforcement on suppressing surrounding rock deformation, shrinking the plastic zone, and improving stress concentration was quantitatively evaluated. This accurately identified key weak points and long-term stability control priorities in the roadway, providing a quantitative basis for determining the safety status of the surrounding rock.

[0257] 4. Optimization of support scheme and adaptation to engineering design

[0258] Graded rock mass parameters are directly applied to the selection of roadway support structures, the design of component parameters, and the verification of support strength. Based on the improved overall strength and deformation resistance of the rock mass after grouting reinforcement, the spacing, anchoring depth, and material strength indicators of support components such as anchor bolts, anchor cables, and shotcrete are rationally optimized, taking into account both engineering safety and economic rationality. Using real and effective rock mechanics parameters, stress calculations of the support structure are conducted to correct deviations in traditional experience-based designs, avoiding excessive support redundancy or insufficient support strength, thus forming a customized surrounding rock control scheme suitable for fractured grouting conditions.

[0259] 5. Engineering disaster prevention and long-term operation and maintenance guidance

[0260] Based on the evolution of rock mass mechanical parameters after reinforcement and the deformation and failure mechanism of deep fractured surrounding rock, risk classification assessments of geological hazards such as roadway surrounding rock collapse, spalling, and floor heave are conducted. High-risk control areas are delineated based on the rock mass bearing capacity and deformation characteristics reflected by the parameters, and targeted on-site monitoring plans and early warning threshold standards are formulated. Simultaneously, the entire set of experimental, inversion, simulation, and testing parameters are incorporated into the engineering operation and maintenance database, providing a reference for subsequent roadway expansion, repeated reinforcement, and parameter correction after changes in geological conditions, achieving a deep integration of experimental research, numerical simulation, and on-site engineering practice.

[0261] 6. Closed-loop system of technological achievements and promotion and application of methods

[0262] This step, through the engineering implementation of parameters throughout the entire process, comprehensively connects all technical aspects of indoor testing, parameter reduction, fracture inversion, grouting simulation, and post-grouting testing, forming an integrated technical closed loop from basic testing, numerical inversion, reinforcement simulation to engineering application. The complete set of parameter acquisition and application methods is adaptable to deep tunnel surrounding rock with varying degrees of fracture and lithological conditions. It can provide standardized reference ideas and parameter value bases for similar fractured rock mass grouting reinforcement projects, possessing good engineering applicability and promotional value.

[0263] In summary, this invention takes deep fractured surrounding rock as the research object and forms a complete integrated system for the refined acquisition and reinforcement evaluation of rock mass parameters, encompassing indoor testing, parameter correction, fracture inversion, grouting simulation, performance testing, and engineering applications. Through standard indoor rock mechanics tests, the basic mechanical parameters of intact rock blocks are determined to construct an original mechanical benchmark. Combining the fracture characteristics and grading indicators of the rock mass at the site, parameter reduction corrections are performed based on strength criteria and empirical formulas to obtain the true mechanical parameters of the fractured rock mass that conform to the in-situ environment. Secondly, based on discrete fracture network technology and discrete element numerical methods, sensitivity analysis and multi-round iterative inversion of fracture parameters are conducted to accurately reconstruct the spatial distribution and structural characteristics of fractures in the surrounding rock at the site, establishing a high-fidelity rock mass numerical model.

[0264] Based on this, the rheological properties of the grout are coupled with fluid-structure interaction calculations to fully simulate the entire process of grouting seepage diffusion, fissure filling, and magma-rock cementation and solidification, thus reproducing the microstructural modification mechanism of grouting reinforcement. Through mechanical simulation tests after grouting under unified conditions, the mechanical indices of the reinforced composite rock mass are quantitatively obtained, and the differences in rock mass mechanical properties before and after grouting are compared and analyzed to clarify the reinforcement and improvement effect. Finally, multi-stage graded rock mass parameters are applied to surrounding rock stability analysis, support scheme optimization, and engineering risk control, achieving an effective combination of basic experiments, numerical simulation, and engineering practice. This reliably solves the technical problems of inaccurate mechanical parameter values ​​for fractured surrounding rock, difficulty in quantifying fissure characteristics, and inability to accurately evaluate the effect of grouting reinforcement, providing scientific and precise technical support for the treatment of fractured surrounding rock tunnels and the design of similar projects.

[0265] The present invention provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the above-described method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock.

[0266] The present invention may also provide a storage medium, which may be a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the above-described method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock.

[0267] The computer-readable storage medium provided by this invention may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0268] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock, characterized in that, The method includes: Core sampling, screening, and standard processing were carried out on the fractured surrounding rock in the deep tunnel. Multiple mechanical property tests were conducted using a rigid servo mechanical testing system, and the baseline values ​​of the inherent mechanical parameters of the complete rock block were obtained through mathematical statistics. Based on the obtained baseline values ​​of inherent mechanical parameters of the intact rock block, combined with the degree of fracture development and rock mass integrity coefficient of the surrounding rock at the site, and relying on various engineering rock mass classification indicators and using empirical formulas for comprehensive reduction calculation, the indoor parameters are transformed into mechanical parameters of fractured rock mass that fit the actual state of the site, and a dedicated parameter database is established to provide target true values ​​for fracture inversion. Based on the baseline values ​​of the inherent mechanical parameters of the complete rock block and the target true values, the fracture distribution characteristics of the surrounding rock at the site are restored through controlled variable experiments, sensitivity analysis and multiple rounds of iterative optimization, so that the mechanical response of the numerical model and the deviation of the rock mass at the site meet the convergence requirements, and the fracture characteristics consistent with the mechanical behavior of the rock mass at the site are obtained. A coupled calculation system for grout seepage and rock mass was constructed based on a three-dimensional fracture network numerical model. Combining the rheological properties of grout with the spatial distribution law of fractures, the entire process of grout diffusion, filling and cementation under grouting pressure was reproduced. The filling effect of grout on fractures and the cementation effect on rock blocks were quantitatively analyzed, and the mechanical modification mechanism of grouting reinforcement of fractured surrounding rock was restored. Multiple sets of mechanical simulation tests were conducted to systematically determine the overall mechanical performance indicators of the rock mass after grouting reinforcement, compare the changes in the mechanical properties of the rock mass before and after grouting, quantitatively analyze the improvement effect of grouting filling and cementation reinforcement, and obtain the mechanical benchmark data of the reinforced rock mass. By integrating multi-level mechanical data of intact rock blocks, naturally fractured rock masses, and grout-modified rock masses, and combining them with the on-site engineering geological conditions and stress environment of the surrounding rock, the obtained mechanical benchmark data of the reinforced rock mass is applied to the calculation of surrounding rock stability, support structure design, and disaster risk assessment.

2. The method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock according to claim 1, characterized in that, Core sampling of fractured surrounding rock in deep mine roadways specifically includes: conducting on-site core sampling operations on three or more typical cross sections of the target deep mine roadway.

3. The method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock according to claim 1, characterized in that, Based on the obtained baseline values ​​of the inherent mechanical parameters of the intact rock block, combined with the degree of fracture development in the surrounding rock and the rock mass integrity coefficient, the specific parameters include: Using the complete rock block mechanical parameters determined by indoor tests as the benchmark input data, and relying on the refined engineering geology of the surrounding rock on site, we can comprehensively obtain key information such as the distribution characteristics of joints and fissures in the surrounding rock, the integrity of the rock mass, the state of the structural planes, the groundwater occurrence conditions, and the degree of disturbance during mining and excavation. Through on-site testing and calculation, the RMR rock mass geomechanical classification index, Q rock mass quality index, and GSI geological strength index were obtained. These three indices are used together to quantitatively characterize the overall quality and degree of fragmentation of the fractured surrounding rock on site.

4. The method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock according to claim 1, characterized in that, Based on various engineering rock mass classification indicators and empirical formulas, a comprehensive reduction calculation is performed. Specifically, the Hoek-Brown rock mass strength criterion is used as the core theoretical basis for parameter reduction. The characteristic constants of the rock mass are calculated by combining the material constants of intact rock and the tunnel excavation disturbance coefficient.

5. The method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock according to claim 1, characterized in that, Through controlled variable experiments, sensitivity analysis, and multiple rounds of iterative optimization, the fracture distribution characteristics of the surrounding rock in the field were reconstructed, specifically including: Based on the field investigation data of joints and fractures in the surrounding rock, a fracture stochastic generation model was constructed using the Discrete Fracture Network (DFN) technology and the Monte Carlo stochastic simulation principle. Numerical experiments were conducted using the controlled single-variable method, fixing the probability distribution and initial mean of the other three fracture parameters, and adjusting only the gradient value of the single target parameter. For each parameter combination, three parallel discrete fracture network models were generated through Monte Carlo simulation. Uniaxial compression and direct shear numerical mechanical tests were conducted on each model, with the loading method, loading rate, and boundary conditions completely consistent with the laboratory tests. The arithmetic mean of the three parallel test results was taken as the simulated rock mechanics parameter value corresponding to that set of fracture parameters. Using the actual mechanical parameters of the fractured rock mass obtained by parameter reduction as the target true values, and based on the results of single-factor tests, the grey relational method is used to calculate the sensitivity weights of fracture orientation, aperture, density, and penetration on four core parameters: elastic modulus, cohesion, internal friction angle, and uniaxial compressive strength.

6. The method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock according to claim 5, characterized in that, The convergence criterion is that the relative deviations of the four core mechanical parameters are all less than the specified thresholds. If the convergence requirement is not met, the iteration optimization step is returned to continue adjustment. If the convergence condition is met, the iteration is terminated, and the complete parameters and probability distribution characteristics of the fracture attitude, aperture, density, and penetration corresponding to the current model are output, which are determined as the real fracture characteristics of the fractured surrounding rock in the field.

7. The method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock according to claim 1, characterized in that, Combining the rheological properties of grout with the spatial distribution of fractures, the entire process of grout diffusion, filling, and cementation under grouting pressure is reproduced. The filling effect of grout on fractures and its cementing effect on rock blocks are quantitatively analyzed, and the mechanical modification mechanism of grouting reinforcement of fractured surrounding rock is reconstructed, specifically including: The PFC three-dimensional fracture network numerical model and the corresponding DFN discrete fracture network file, which are converged and locked by fracture inversion, are directly imported into the computing platform, and the original microscopic contact parameters of the rock matrix, fracture spatial distribution characteristics, model boundary constraints and geometric parameters are completely preserved.

8. The method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock according to claim 7, characterized in that, Based on the distribution of connected channels in the fracture network, the flow rate, diffusion range and filling saturation of slurry in fractures with different openings and penetrations are calculated in real time using a fluid-structure interaction calculation algorithm. The simulation of high-pressure grouting shows the entire process of grout preferentially migrating and diffusing along large-aperture interconnected fractures, and gradually seeping and filling into small and isolated fractures. Simultaneously, the mechanical responses such as grouting pressure loss, fracture compression deformation, and small rock block displacement are calculated to reflect the compression and expansion effect of grouting pressure on surrounding rock fractures and the seepage driving law, thus fully reproducing the dynamic migration characteristics of grout in a complex fracture network.

9. The method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock according to claim 8, characterized in that, After the grout has filled the cracks, the bonding strength, stiffness and shear resistance of the grout inside the cracks are gradually improved according to the preset time hardening parameters, so that a new type of cemented contact unit is formed at the contact position between the filling grout and the surrounding rock blocks. By modifying the bonding model and contact strength parameters of the fracture contact surface to replace the original weak fracture contact properties, the mechanism by which scattered and broken rock blocks interlock and bond together after grout consolidation is simulated, ultimately forming a complete and continuous grout-rock composite structure.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium having computer-readable program instructions stored thereon, which are used to perform the method for determining the mechanical parameters of grouting rock mass in fractured surrounding rock as described in any one of claims 1 to 9.

Images