Construction method of fully-weathered granite stratum tunnel grouting reinforcement model

Abstract

The invention discloses a construction method of a fully weathered granite stratum tunnel grouting reinforcement model, and relates to the technical field of tunnel engineering. The method comprises the steps of tunnel model acquisition, stratum parameter acquisition and strength analysis, demand strength calculation, intelligent recommendation of a grouting reinforcement method, tunnel grouting reinforcement simulation and effect evaluation, and result verification and optimization. According to the method, a three-dimensional digital tunnel model is constructed, geometric characteristics such as the section shape, the spatial position and the burial depth of a tunnel are accurately described, and a stratum strength distribution diagram and a required strength distribution diagram are generated by combining stratum physical, mechanical and hydrological parameters; a reinforcement scheme recommendation model based on a deep neural network is introduced, a stratum strength distribution diagram and a demand strength distribution diagram are integrated, and the economy, the construction efficiency and the reinforcement effect of the scheme are further balanced through a multi-objective optimization algorithm.

Description

Technical Field

[0001] This invention relates to the field of tunnel engineering technology, and in particular to a method for constructing a grouting reinforcement model for tunnels in completely weathered granite strata. Background Technology

[0002] In the field of tunnel engineering, completely weathered granite strata, due to their loose rock structure and low mechanical strength, struggle to meet the safety requirements of tunnel construction and operation. Therefore, grouting reinforcement technology has become a widely used method, improving the mechanical properties and permeability of the strata by injecting grout. However, existing technologies still have the following problems in their application: Insufficient integration of grouting reinforcement design with tunnel model: In the existing technology, grouting reinforcement design fails to fully integrate the three-dimensional spatial model of the tunnel, and does not adequately consider the cross-sectional shape, spatial orientation, burial depth and distribution characteristics of stratum strength of the tunnel; it is impossible to directly observe the grouting reinforcement requirements of the tunnel through an intuitive model.

[0003] The lack of dynamic adjustment of demand intensity distribution: The bearing capacity and seepage prevention requirements of the surrounding rock of the tunnel vary depending on the spatial location and the purpose of the tunnel. However, traditional methods make it difficult to map these requirements into the tunnel model, which may lead to the reinforcement design being too conservative or insufficient in some key areas, affecting the overall construction efficiency and effect.

[0004] The optimization and evaluation process for grouting schemes is lagging behind: Grouting reinforcement design and evaluation typically rely on static geological data and engineering experience, lacking dynamic simulation methods. For example, the diffusion range of grouting materials and the effect of surrounding rock reinforcement cannot be predicted and adjusted in real time during construction, resulting in a long optimization cycle and low utilization of construction resources.

[0005] By combining the three-dimensional digital model of the tunnel with the analysis of geological parameters, the strength distribution characteristics of the strata can be reflected more accurately, and the effect of the grouting scheme can be dynamically simulated. Summary of the Invention

[0006] This invention proposes a method for constructing a grouting reinforcement model based on a tunnel model. It incorporates information such as the tunnel cross-sectional shape, spatial location, and required intensity distribution map into the design process, and combines intelligent recommendation and simulation evaluation technologies to solve existing technical problems.

[0007] A method for constructing a grouting reinforcement model for tunnels in completely weathered granite strata includes the following steps: Step S1: Tunnel Model Acquisition: Obtain the tunnel cross-sectional shape based on tunnel usage data and construction environment data, and generate a three-dimensional digital tunnel model; or generate a tunnel model by inputting an existing tunnel design scheme. Step S2: Stratigraphic parameter acquisition and strength analysis: Physical, mechanical and hydrological parameters of the completely weathered granite strata are obtained through borehole sampling and / or in-situ testing. The strata strength is calculated and mapped into the tunnel model to form a strata strength distribution map that reflects the mechanical properties of the strata before reinforcement. Step S3: Demand Intensity Calculation: Calculate the demand intensity based on tunnel usage data and design requirements. The demand intensity includes bearing capacity and seepage prevention capacity. Map the demand intensity data onto the tunnel model to form a demand intensity distribution map that reflects the mechanical properties of the demand. Step S4 Intelligent Recommendation of Grouting Reinforcement Method: Input the tunnel model containing the stratum strength distribution map and the required strength distribution map into the reinforcement scheme recommendation model for calculation to obtain the recommended grouting reinforcement method; the grouting reinforcement method includes grouting material, grouting type, grouting density and construction method; Step S5: Simulation and Effect Evaluation of Tunnel Grouting Reinforcement: Simulate the recommended grouting reinforcement scheme in the tunnel model, analyze the grout diffusion range and the reinforcement effect on the tunnel structure; generate the load-bearing capacity distribution map and deformation prediction map of the tunnel after grouting reinforcement. Step S6 Result Verification and Optimization: Verify the effect of the deformation prediction map. If it is qualified, the recommended grouting reinforcement scheme will be used as the final reinforcement scheme; otherwise, simulate the recommended grouting reinforcement scheme again, update the bearing capacity distribution map, optimize the grouting reinforcement method according to the updated bearing capacity distribution map, and simulate, evaluate and verify the optimized grouting reinforcement scheme.

[0008] As a preferred embodiment of the present invention, the three-dimensional digital tunnel model includes: Tunnel cross-sectional shapes: including but not limited to circular, elliptical, horseshoe-shaped, and rectangular cross-sections; Tunnel spatial location: including the tunnel's orientation, gradient, burial depth, and length; Construction environment data includes the distribution of rock masses, location of fracture zones, distribution of groundwater, and geological interfaces in the strata; Design purpose data: including the tunnel's functional type, as well as design loads and seepage prevention requirements.

[0009] As a preferred embodiment of the present invention, the steps of obtaining the physical, mechanical, and hydrological parameters of the completely weathered granite strata and calculating the strata strength include the following: The obtained physical parameters include porosity and density, mechanical parameters include uniaxial compressive strength, rock mass integrity index and elastic modulus, and hydrological parameters include permeability coefficient and saturation. Porosity was obtained through rock sample experiments or nuclear magnetic resonance (NMR) measurements. The calculation formula is: ,in The pore volume of the rock sample. This represents the total volume of the rock sample. Density was obtained by rock sample dry density testing. The calculation formula is: ,in For the quality of rock sample drying, This represents the total volume of the rock sample. Uniaxial compressive strength is obtained through uniaxial compressive strength test. The calculation formula is: ,in To destroy the load, The cross-sectional area of ​​the rock sample; Rock mass integrity index obtained from drill cores The calculation formula is: ,in The total length of the core sample with a length greater than 10cm. This represents the total length of the borehole. The elastic modulus is obtained from the stress-strain curve. The calculation formula is: ,in For stress increment, For strain increment; Permeability coefficient was obtained through on-site pumping tests. The calculation formula is: ,in The infiltration flow rate per unit time. For the cross-sectional area through which the flow passes, For hydraulic gradient; Saturation was obtained through rock sample saturation tests. The calculation formula is: ,in For the volume of water, This represents the pore volume of the rock sample. The formation strength value is calculated by comprehensively considering the physical, mechanical, and hydrological parameters of the formation. The calculation formula is: In the formula These are the empirical coefficients corresponding to each parameter, used to balance the contribution of different parameters to the overall strength; This represents the contribution of permeability resistance to strength, and is inversely proportional to the permeability coefficient.

[0010] As a preferred embodiment of the present invention, the method for generating the formation intensity distribution map includes the following steps: Spatial interpolation algorithms are used to generate continuous distributions of physical, mechanical, and hydrological parameters of various strata in the tunnel using the acquired discrete data; the spatial interpolation algorithms include, but are not limited to, inverse distance weighting, kriging interpolation, or spline interpolation. The interpolated parameters are input into the formation strength calculation formula to generate the formation strength value for each grid cell, wherein the grid cells are divided according to the space around the tunnel. Using 3D modeling software, the geological strength values ​​are mapped to the tunnel model; the geological strength distribution at different spatial locations is displayed, and the strength differences are presented intuitively using color gradients or contour lines; key areas are marked in the model, including high-strength areas, low-strength areas, and fracture zones; a 3D strength distribution map is output for subsequent analysis and calculation.

[0011] As a preferred embodiment of the present invention, the demand intensity calculation includes: The required bearing capacity of the surrounding rock of the tunnel is obtained based on the tunnel's design purpose and construction load. The calculation formula is: ,in The unit weight of the surrounding rock. For tunnel burial depth, To design a safety factor; Based on the tunnel's waterproofing requirements, obtain the required seepage prevention strength. The calculation formula is: ,in To allow infiltration traffic, Permeability coefficient; The calculated required bearing capacity and seepage prevention strength data are mapped onto the tunnel's three-dimensional model; a required strength distribution map is generated, and color gradients are used to represent the strength requirements of different areas; areas with concentrated required strength and weak points are marked as key references for grouting reinforcement design.

[0012] As a preferred technical solution of the present invention, the recommended model of the reinforcement scheme includes: The recommended model for the hardening scheme is a multi-layered deep neural network, the structure of which includes: Input layer: Receives tunnel model data containing geological intensity distribution maps and demand intensity distribution maps; Hidden layer: Features are extracted using a convolutional neural network, and long short-term memory networks are used to analyze the nonlinear behavior of the formation; Output layer: Outputs recommended grouting reinforcement methods, including grouting materials, grouting type, grouting density, and construction method; The selection of convolutional kernels and activation functions in the hidden layers is based on the distribution characteristics of the input data, and an adaptive optimization mechanism is used for adjustment; The extracted features include: numerically gridded data of the stratum intensity distribution map; numerically gridded data of the demand intensity distribution map; tunnel cross-sectional shape and spatial location parameters; tunnel design purpose and load requirements; input features are processed through standardization and normalization. The grouting reinforcement methods output by the model include: Grouting materials include, but are not limited to, cement grout, ultrafine cement grout, chemical grout, and mixed grout; Grouting types include, but are not limited to, curtain grouting, consolidation grouting, and fracturing grouting; Grouting density: the amount of grout per unit volume; Construction methods include, but are not limited to, continuous grouting and segmented grouting; The output results are optimized using a multi-objective optimization algorithm, with optimization objectives including economy, construction efficiency, and reinforcement effect.

[0013] As a preferred embodiment of the present invention, the training of the reinforcement scheme recommendation model includes: Initial reinforcement scheme recommended model structure initialization: set the convolutional kernel size, number of layers, and stride of the convolutional layers; configure the number of units in the LSTM layer; set the number of neurons in the fully connected layer; Dataset partitioning: Label the historical data of the tunnel model, construct a dataset, and partition the dataset into a training set and a validation set; Training process: Input the training set into the initial reinforcement scheme recommendation model; use a stochastic gradient descent optimizer to adjust the learning rate to avoid overfitting or underfitting; use a loss function to evaluate the deviation between the prediction and the actual value. The formula for the loss function is: ,in Indicates the loss value. Represents the mean squared error function. To predict the classification accuracy of the results, For loss weighting coefficients, For predicted values, The model's hyperparameters are adjusted using Bayesian optimization to obtain the optimized reinforcement scheme recommendation model through training. The recommended model for optimized reinforcement schemes was validated, and the validation indicators included numerical indicators, classification indicators, and comprehensive indicators. Validation method: The performance of the optimized reinforcement scheme recommendation model is evaluated using validation set data, and ROC curves and confusion matrices are generated to analyze the classification effect. The predicted values ​​are compared with the true label values, and the numerical accuracy of the optimized reinforcement scheme recommendation model is evaluated by goodness of fit and root mean square error. The optimized reinforcement scheme recommendation model that meets the numerical accuracy standard is output as the reinforcement scheme recommendation model.

[0014] As a preferred technical solution of the present invention, the tunnel grouting reinforcement simulation and effect evaluation includes the following steps: Tunnel grouting reinforcement simulation: A fluid-structure interaction model of the strata surrounding the tunnel was established using numerical simulation software. Parameters such as grouting pressure, grout viscosity, stratum permeability coefficient, and porosity were set in the model to simulate the diffusion range of the grout in the strata. A three-dimensional contour map of the grout diffusion range was generated, and the diffusion range limit value was marked. Tunnel loads are applied to the reinforcement simulation model to analyze the stress distribution and deformation characteristics of the tunnel structure after grouting reinforcement, and to simulate the stress and deformation of the tunnel structure. The deformation of the surrounding rock and the location of the maximum stress are calculated, and stress distribution maps and deformation prediction maps are generated. Based on the simulation results, adjust the grouting material parameters and construction parameters to optimize the reinforcement scheme; Evaluation of Grouting Reinforcement Effect: Calculate the bearing capacity improvement rate based on the stress distribution of the tunnel surrounding rock before and after reinforcement; calculate the maximum deformation of the tunnel surrounding rock after grouting reinforcement and compare it with the design allowable deformation to evaluate the deformation control effect; compare the changes in the permeability coefficient of the tunnel surrounding rock before and after grouting reinforcement to evaluate the improvement in permeability performance; comprehensively analyze the amount of grouting material used, construction time, and project cost to evaluate the economy and construction efficiency of the reinforcement scheme. If the evaluation results meet the design objectives, the grouting reinforcement scheme is confirmed to be effective; otherwise, return to adjust the simulation parameters and re-evaluate.

[0015] As a preferred technical solution of the present invention, the result verification and optimization includes the following steps: Verification phase: Deformation prediction verification: Verify the tunnel deformation prediction map generated after grouting reinforcement, obtain the actual tunnel deformation using the on-site monitoring system, compare the deviation between the predicted deformation and the monitored deformation, and calculate the goodness of fit. Bearing capacity verification: Compare the bearing capacity distribution map of the reinforced tunnel with the results of on-site load tests; verify that the bearing capacity improvement rate meets the design requirements; if it does not meet the requirements, mark the weak areas. Permeability verification: The permeability coefficient of the reinforced surrounding rock is obtained using the permeability test results after grouting, and compared with the simulated permeability coefficient; if the deviation exceeds the set range, the formation permeability model parameters are corrected. Optimization phase: Parameter optimization: If the verification results are unqualified, adjust the following parameters: grouting pressure, grouting material properties, and construction method; Based on the optimized parameters, the grouting diffusion range and tunnel reinforcement effect are re-simulated; according to the optimized simulation results, new bearing capacity distribution maps, deformation prediction maps, and permeability distribution maps are generated; the updated distribution maps are mapped to the three-dimensional tunnel model, and the optimized key areas are marked; the optimized scheme is verified again, and the verification process of deformation, bearing capacity, and permeability is repeated. If the verification result is satisfactory, the optimized solution is confirmed as the final reinforcement solution; if it is still unsatisfactory, the optimization phase will be carried out again. Output the final reinforcement scheme, confirming that the reinforcement scheme meets the design requirements, including the optimized grouting materials, grouting type, construction parameters and grouting strategy; output the final three-dimensional reinforcement distribution map, including the diffusion range, the area of ​​increased bearing capacity and the area of ​​improved permeability; provide a reinforcement construction guidance report for use in project implementation.

[0016] As a preferred embodiment of the present invention, the grouting reinforcement model for tunnels in completely weathered granite strata includes the following modules: Tunnel model generation module: Generates a three-dimensional digital tunnel model based on tunnel usage data and construction environment data; Formation parameter acquisition and analysis module: Collects physical, mechanical and hydrological parameters of the formation; generates a formation strength distribution map using interpolation algorithms to intuitively reflect the mechanical properties of the formation before reinforcement. Demand Intensity Calculation Module: Calculates required bearing capacity and seepage prevention strength based on tunnel design requirements; generates demand intensity distribution map for analyzing areas requiring tunnel reinforcement; The reinforcement scheme recommendation module, based on the reinforcement scheme recommendation model, takes as input the stratum strength distribution map and the required strength distribution map, and outputs grouting reinforcement methods. The recommended grouting methods include grouting materials, grouting type, grouting density, and construction method; and provides preliminary optimized reinforcement schemes. Grouting reinforcement simulation module: uses a fluid-structure interaction model to simulate the diffusion range of grout in the formation; generates a bearing capacity distribution map and deformation prediction map after reinforcement; evaluates the permeability improvement effect after grouting; Verification and Optimization Module: Verify the grouting reinforcement effect by checking deformation, bearing capacity improvement rate, and permeability improvement rate through on-site monitoring data; if it does not meet the design requirements, adjust the grouting materials, construction methods, and parameters, and iteratively optimize the reinforcement scheme; update the distribution map and conduct multiple rounds of verification until it meets the requirements; Visualization output module: Visualizes the stratum strength distribution map, required strength distribution map, grouting diffusion range and reinforcement effect in three-dimensional form; provides an intuitive tunnel model display, including annotation of key areas; and outputs optimized reinforcement schemes and construction guidance reports.

[0017] The present invention has the following advantages: This invention constructs a three-dimensional digital tunnel model to accurately describe the tunnel's cross-sectional shape, spatial location, burial depth, and other geometric characteristics. It then combines geological physical, mechanical, and hydrological parameters to generate geological strength distribution maps and required strength distribution maps, enabling precise and targeted reinforcement scheme design. Furthermore, this invention introduces a reinforcement scheme recommendation model based on deep neural networks. This model integrates the geological strength distribution map and the required strength distribution map for intelligent recommendations. Through a multi-objective optimization algorithm, it further balances the economy, construction efficiency, and reinforcement effect of the scheme.

[0018] This invention, based on a fluid-structure interaction model, simulates the grout diffusion range during the grouting process and its reinforcement effect on the tunnel structure, generating a load-bearing capacity distribution map and a deformation prediction map. The scheme can be fully evaluated before construction, reducing later adjustment costs and improving the reliability of the scheme. This invention constructs a closed-loop process from design, simulation, evaluation to verification and optimization. By comparing and verifying with on-site monitoring data, grouting parameters and construction methods are dynamically adjusted, and iterative optimization is repeatedly performed until the design requirements are met, ensuring the scientific validity and practicality of the reinforcement scheme.

[0019] This invention presents the distribution of ground strength, required strength, grout diffusion range, and reinforcement effect in a three-dimensional format. Through color gradients and key area markings, it facilitates understanding and adjustments by engineering designers, providing clear and intuitive guidance for tunnel construction. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a method for constructing a grouting reinforcement model for a tunnel in a completely weathered granite stratum, as used in an embodiment of the present invention. Figure 2 This is a schematic diagram of the module of the tunnel grouting reinforcement model for completely weathered granite strata used in an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions of this invention, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this invention.

[0022] Example 1: A method for constructing a grouting reinforcement model for tunnels in completely weathered granite strata. (See [link]) Figure 1 As shown, it includes the following steps: Step S1: Tunnel Model Acquisition: Obtain the tunnel cross-sectional shape based on tunnel usage data and construction environment data, and generate a three-dimensional digital tunnel model; or generate a tunnel model by inputting an existing tunnel design scheme. The three-dimensional digitized tunnel model includes: Tunnel cross-sectional shapes: including but not limited to circular, elliptical, horseshoe-shaped, and rectangular cross-sections; Tunnel spatial location: including the tunnel's orientation, gradient, burial depth, and length; Construction environment data includes the distribution of rock masses, location of fracture zones, distribution of groundwater, and geological interfaces in the strata; Design purpose data: including the tunnel's functional type, as well as design loads and seepage prevention requirements.

[0023] Step S2: Stratigraphic parameter acquisition and strength analysis: Physical, mechanical and hydrological parameters of the completely weathered granite strata are obtained through borehole sampling and / or in-situ testing. The strata strength is calculated and mapped into the tunnel model to form a strata strength distribution map that reflects the mechanical properties of the strata before reinforcement. The process of obtaining the physical, mechanical, and hydrological parameters of the completely weathered granite strata and calculating the strata strength includes the following steps: The obtained physical parameters include porosity and density, mechanical parameters include uniaxial compressive strength, rock mass integrity index and elastic modulus, and hydrological parameters include permeability coefficient and saturation. Porosity was obtained through rock sample experiments or nuclear magnetic resonance (NMR) measurements. The calculation formula is: ,in The pore volume of the rock sample. This represents the total volume of the rock sample. Density was obtained by rock sample dry density testing. The calculation formula is: ,in For the quality of rock sample drying, This represents the total volume of the rock sample. Uniaxial compressive strength is obtained through uniaxial compressive strength test. The calculation formula is: ,in To destroy the load, The cross-sectional area of ​​the rock sample; Rock mass integrity index obtained from drill cores The calculation formula is: ,in The total length of the core sample with a length greater than 10cm. This represents the total length of the borehole. The elastic modulus is obtained from the stress-strain curve. The calculation formula is: ,in For stress increment, For strain increment; Permeability coefficient was obtained through on-site pumping tests. The calculation formula is: ,in The infiltration flow rate per unit time. For the cross-sectional area through which the flow passes, For hydraulic gradient; Saturation was obtained through rock sample saturation tests. The calculation formula is: ,in For the volume of water, This represents the pore volume of the rock sample. The formation strength value is calculated by comprehensively considering the physical, mechanical, and hydrological parameters of the formation. The calculation formula is: In the formula These are the empirical coefficients corresponding to each parameter, used to balance the contribution of different parameters to the overall strength; This represents the contribution of permeability resistance to strength, and is inversely proportional to the permeability coefficient.

[0024] The method for generating the formation intensity distribution map includes the following steps: Spatial interpolation algorithms are used to generate continuous distributions of physical, mechanical, and hydrological parameters of various strata in the tunnel using the acquired discrete data; the spatial interpolation algorithms include, but are not limited to, inverse distance weighting, kriging interpolation, or spline interpolation. The interpolated parameters are input into the formation strength calculation formula to generate the formation strength value for each grid cell, wherein the grid cells are divided according to the space around the tunnel. Using 3D modeling software, the geological strength values ​​are mapped to the tunnel model; the geological strength distribution at different spatial locations is displayed, and the strength differences are presented intuitively using color gradients or contour lines; key areas are marked in the model, including high-strength areas, low-strength areas, and fracture zones; a 3D strength distribution map is output for subsequent analysis and calculation.

[0025] Step S3: Demand Intensity Calculation: Calculate the demand intensity based on tunnel usage data and design requirements. The demand intensity includes bearing capacity and seepage prevention capacity. Map the demand intensity data onto the tunnel model to form a demand intensity distribution map that reflects the mechanical properties of the demand. The demand intensity calculation includes: The required bearing capacity of the surrounding rock of the tunnel is obtained based on the tunnel's design purpose and construction load. The calculation formula is: ,in The unit weight of the surrounding rock. For tunnel burial depth, To design a safety factor; Based on the tunnel's waterproofing requirements, obtain the required seepage prevention strength. The calculation formula is: ,in To allow infiltration traffic, Permeability coefficient; The calculated required bearing capacity and seepage prevention strength data are mapped onto the tunnel's three-dimensional model; a required strength distribution map is generated, and color gradients are used to represent the strength requirements of different areas; areas with concentrated required strength and weak points are marked as key references for grouting reinforcement design.

[0026] Step S4 Intelligent Recommendation of Grouting Reinforcement Method: Input the tunnel model containing the stratum strength distribution map and the required strength distribution map into the reinforcement scheme recommendation model for calculation to obtain the recommended grouting reinforcement method; the grouting reinforcement method includes grouting material, grouting type, grouting density and construction method; The recommended model for the reinforcement scheme includes: The recommended model for the hardening scheme is a multi-layered deep neural network, the structure of which includes: Input layer: Receives tunnel model data containing geological intensity distribution maps and demand intensity distribution maps; Hidden layer: Features are extracted using a convolutional neural network, and long short-term memory networks are used to analyze the nonlinear behavior of the formation; Output layer: Outputs recommended grouting reinforcement methods, including grouting materials, grouting type, grouting density, and construction method; The selection of convolutional kernels and activation functions in the hidden layers is based on the distribution characteristics of the input data, and an adaptive optimization mechanism is used for adjustment; The extracted features include: numerically gridded data of the stratum intensity distribution map; numerically gridded data of the demand intensity distribution map; tunnel cross-sectional shape and spatial location parameters; tunnel design purpose and load requirements; input features are processed through standardization and normalization. The grouting reinforcement methods output by the model include: Grouting materials include, but are not limited to, cement grout, ultrafine cement grout, chemical grout, and mixed grout; Grouting types include, but are not limited to, curtain grouting, consolidation grouting, and fracturing grouting; Grouting density: the amount of grout per unit volume; Construction methods include, but are not limited to, continuous grouting and segmented grouting; The output results are optimized using a multi-objective optimization algorithm, with optimization objectives including economy, construction efficiency, and reinforcement effect.

[0027] The training of the reinforcement scheme recommendation model includes: Initialization of the recommended model structure for the initial reinforcement scheme: Set the kernel size, number of layers, and stride of the convolutional layers to extract three-dimensional spatial features; configure the number of LSTM layer units to capture the temporal changes in stratum characteristics; set the number of neurons in the fully connected layers and map them to the output space of the grouting reinforcement method. Dataset partitioning: Label the historical data of the tunnel model, construct a dataset, and partition the dataset into a training set and a validation set; Training process: Input the training set into the initial reinforcement scheme recommendation model; use a stochastic gradient descent optimizer to adjust the learning rate to avoid overfitting or underfitting; use a loss function to evaluate the deviation between the prediction and the actual value. The formula for the loss function is: ,in Indicates the loss value. Represents the mean squared error function. To predict the classification accuracy of the results, For loss weighting coefficients, For predicted values, The model's hyperparameters are adjusted using Bayesian optimization to obtain the optimized reinforcement scheme recommendation model through training. The recommended model for optimized reinforcement schemes was validated, and the validation metrics included: Numerical indicators: the deviation between the predicted grout diffusion range and the actual diffusion range; the degree of agreement between the reinforced tunnel bearing capacity distribution map and the field measured data; Classification index: The accuracy rate of recommending grouting reinforcement methods, including the matching degree of materials, grouting type and construction method; Comprehensive indicators: A comprehensive evaluation of the economic efficiency and construction feasibility of the reinforcement scheme is conducted to obtain an optimization score; Validation method: The performance of the optimized reinforcement scheme recommendation model is evaluated using validation set data, and ROC curves and confusion matrices are generated to analyze the classification effect. The predicted values ​​are compared with the true label values, and the numerical accuracy of the optimized reinforcement scheme recommendation model is evaluated by goodness of fit and root mean square error. The optimized reinforcement scheme recommendation model that meets the numerical accuracy standard is output as the reinforcement scheme recommendation model.

[0028] Under the new tunnel strata conditions, the existing model is applied to the new data scenario through transfer learning technology to reduce training costs; the weights of some convolutional layers are frozen and only the fully connected layers are fine-tuned to improve the model's adaptability.

[0029] During actual construction, the model input is adjusted by combining real-time monitoring data (such as grouting diffusion radius and tunnel deformation data) to dynamically update the prediction results; new construction feedback data is added to the training set, and the model is retrained regularly to improve long-term performance.

[0030] Step S5: Simulation and Effect Evaluation of Tunnel Grouting Reinforcement: Simulate the recommended grouting reinforcement method in the tunnel model, analyze the grout diffusion range and the reinforcement effect on the tunnel structure; generate the load-bearing capacity distribution map and deformation prediction map of the tunnel after grouting reinforcement. The simulation and effect evaluation of tunnel grouting reinforcement includes the following steps: Tunnel grouting reinforcement simulation: A fluid-structure interaction model of the strata surrounding the tunnel was established using numerical simulation software. Parameters such as grouting pressure, grout viscosity, stratum permeability coefficient, and porosity were set in the model to simulate the diffusion range of the grout in the strata. A three-dimensional contour map of the grout diffusion range was generated, and the diffusion range limit value was marked. Apply tunnel loads (such as construction loads or operational loads) to the reinforcement simulation model, analyze the stress distribution and deformation characteristics of the tunnel structure after grouting reinforcement, simulate the stress and deformation of the tunnel structure, calculate the deformation of the surrounding rock and the location of the maximum stress, and generate stress distribution maps and deformation prediction maps. Based on simulation results (such as gel time and viscosity) and construction parameters (such as grouting pressure and grouting time), the reinforcement scheme is optimized.

[0031] Evaluation of Grouting Reinforcement Effect: Calculate the bearing capacity improvement rate based on the stress distribution of the tunnel surrounding rock before and after reinforcement; calculate the maximum deformation of the tunnel surrounding rock after grouting reinforcement and compare it with the design allowable deformation to evaluate the deformation control effect; compare the changes in the permeability coefficient of the tunnel surrounding rock before and after grouting reinforcement to evaluate the improvement in permeability performance; comprehensively analyze the amount of grouting material used, construction time, and project cost to evaluate the economy and construction efficiency of the reinforcement scheme. If the evaluation results meet the design objectives, the grouting reinforcement scheme is confirmed to be effective; otherwise, return to adjust the simulation parameters and re-evaluate.

[0032] Step S6 Result Verification and Optimization: Verify the effect of the deformation prediction map. If it is qualified, the recommended grouting reinforcement method will be used as the final reinforcement scheme; otherwise, simulate the recommended grouting reinforcement method again, update the bearing capacity distribution map, optimize the grouting reinforcement method according to the updated bearing capacity distribution map, and simulate, evaluate and verify the optimized grouting reinforcement method.

[0033] The result verification and optimization includes the following steps: Verification phase: Deformation prediction verification: The predicted tunnel deformation map generated after grouting reinforcement is verified. The actual tunnel deformation is obtained using the on-site monitoring system. The deviation between the predicted deformation and the monitored deformation is compared, and the goodness of fit is calculated. The calculation formula is: ,in This represents the actual deformation. To predict the amount of deformation, This represents the average value of the actual deformation. Bearing capacity verification: Compare the bearing capacity distribution map of the reinforced tunnel with the results of on-site load tests; verify that the bearing capacity improvement rate meets the design requirements; if it does not meet the requirements, mark the weak areas. Permeability verification: The permeability coefficient of the reinforced surrounding rock is obtained using the permeability test results after grouting, and compared with the simulated permeability coefficient; if the deviation exceeds the set range, the formation permeability model parameters are corrected. Optimization phase: Parameter optimization: If the verification result is unsatisfactory, adjust the following parameters: Grouting pressure: Adjust to optimize grout diffusion range; Grouting material performance: Optimize grout viscosity, gel time, and flowability; Construction method: Change the location of grouting sections or the grouting sequence; Based on the optimized parameters, the grouting diffusion range and tunnel reinforcement effect are re-simulated; according to the optimized simulation results, new bearing capacity distribution maps, deformation prediction maps, and permeability distribution maps are generated; the updated distribution maps are mapped to the three-dimensional tunnel model, and the optimized key areas are marked; the optimized scheme is verified again, and the verification process of deformation, bearing capacity, and permeability is repeated. If the verification result is satisfactory, the optimized solution is confirmed as the final reinforcement solution; if it is still unsatisfactory, the optimization phase will be carried out again. Output the final reinforcement scheme, confirming that the reinforcement scheme meets the design requirements, including the optimized grouting materials, grouting type, construction parameters and grouting strategy; output the final three-dimensional reinforcement distribution map, including the diffusion range, the area of ​​increased bearing capacity and the area of ​​improved permeability; provide a reinforcement construction guidance report for use in project implementation.

[0034] Example 2: Grouting reinforcement model for tunnels in completely weathered granite strata (see [reference]). Figure 2 As shown, it includes the following modules: Tunnel model generation module: Generates a three-dimensional digital tunnel model based on tunnel usage data and construction environment data; Formation parameter acquisition and analysis module: Collects physical, mechanical and hydrological parameters of the formation; generates a formation strength distribution map using interpolation algorithms to intuitively reflect the mechanical properties of the formation before reinforcement. Demand Intensity Calculation Module: Calculates required bearing capacity and seepage prevention strength based on tunnel design requirements; generates demand intensity distribution map for analyzing areas requiring tunnel reinforcement; The reinforcement scheme recommendation module, based on the reinforcement scheme recommendation model, takes as input the stratum strength distribution map and the required strength distribution map, and outputs grouting reinforcement methods. The recommended grouting methods include grouting materials, grouting type, grouting density, and construction method; and provides preliminary optimized reinforcement schemes. Grouting reinforcement simulation module: uses a fluid-structure interaction model to simulate the diffusion range of grout in the formation; generates a bearing capacity distribution map and deformation prediction map after reinforcement; evaluates the permeability improvement effect after grouting; Verification and Optimization Module: Verify the grouting reinforcement effect by checking deformation, bearing capacity improvement rate, and permeability improvement rate through on-site monitoring data; if it does not meet the design requirements, adjust the grouting materials, construction methods, and parameters, and iteratively optimize the reinforcement scheme; update the distribution map and conduct multiple rounds of verification until it meets the requirements; Visualization output module: Visualizes the stratum strength distribution map, required strength distribution map, grouting diffusion range and reinforcement effect in three-dimensional form; provides an intuitive tunnel model display, including annotation of key areas; and outputs optimized reinforcement schemes and construction guidance reports.

[0035] It should be understood that those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims. Parts not described in detail in this specification are prior art known to those skilled in the art.

Claims

1. A method for constructing a grouting reinforcement model for tunnels in completely weathered granite strata, characterized in that, Includes the following steps: Step S1: Tunnel Model Acquisition: Based on tunnel usage data and construction environment data, obtain the tunnel cross-sectional shape and generate a three-dimensional digital tunnel model; Alternatively, a tunnel model can be generated by inputting an existing tunnel design scheme; Step S2: Stratigraphic parameter acquisition and strength analysis: Physical, mechanical and hydrological parameters of the completely weathered granite strata are obtained through borehole sampling and / or in-situ testing. The strata strength is calculated and mapped into the tunnel model to form a strata strength distribution map that reflects the mechanical properties of the strata before reinforcement. Step S3 Calculation of Demand Strength: Calculate the demand strength based on the tunnel's intended use and design requirements. The demand strength includes bearing capacity and impermeability. Demand intensity data is mapped onto the tunnel model to form a demand intensity distribution map that reflects the mechanical properties of demand. Step S4 Intelligent Recommendation of Grouting Reinforcement Method: Input the tunnel model containing the stratum strength distribution map and the required strength distribution map into the reinforcement scheme recommendation model for calculation to obtain the recommended grouting reinforcement method; the grouting reinforcement method includes grouting material, grouting type, grouting density and construction method; Step S5: Simulation and Effect Evaluation of Tunnel Grouting Reinforcement: Simulate the recommended grouting reinforcement method in the tunnel model, and analyze the grout diffusion range and the reinforcement effect on the tunnel structure. Generate a load-bearing capacity distribution map and deformation prediction map of the tunnel after grouting reinforcement; Step S6 Result Verification and Optimization: Verify the effect of the deformation prediction map. If it is qualified, the recommended grouting reinforcement method will be used as the final reinforcement scheme; otherwise, simulate the recommended grouting reinforcement method again, update the bearing capacity distribution map, optimize the grouting reinforcement method according to the updated bearing capacity distribution map, and simulate, evaluate and verify the optimized grouting reinforcement method.

2. The method for constructing a grouting reinforcement model for a tunnel in a completely weathered granite strata according to claim 1, characterized in that, The three-dimensional digitized tunnel model includes: Tunnel cross-sectional shapes: including but not limited to circular, elliptical, horseshoe-shaped, and rectangular cross-sections; Tunnel spatial location: including the tunnel's orientation, gradient, burial depth, and length; Construction environment data includes the distribution of rock masses, location of fracture zones, distribution of groundwater, and geological interfaces in the strata; Design purpose data: including the tunnel's functional type, as well as design loads and seepage prevention requirements.

3. The method for constructing a grouting reinforcement model for a tunnel in a completely weathered granite strata according to claim 1, characterized in that, The process of obtaining the physical, mechanical, and hydrological parameters of the completely weathered granite strata and calculating the strata strength includes the following steps: The obtained physical parameters include porosity and density, mechanical parameters include uniaxial compressive strength, rock mass integrity index and elastic modulus, and hydrological parameters include permeability coefficient and saturation. Porosity was obtained through rock sample experiments or nuclear magnetic resonance (NMR) measurements. The calculation formula is: ,in The pore volume of the rock sample. This represents the total volume of the rock sample. Density was obtained by rock sample dry density testing. The calculation formula is: ,in For the quality of rock sample drying, This represents the total volume of the rock sample. Uniaxial compressive strength is obtained through uniaxial compressive strength test. The calculation formula is: ,in To destroy the load, The cross-sectional area of ​​the rock sample; Rock mass integrity index obtained from drill cores The calculation formula is: ,in The total length of the core sample with a length greater than 10cm. This represents the total length of the borehole. The elastic modulus is obtained from the stress-strain curve. The calculation formula is: ,in For stress increment, For strain increment; Permeability coefficient was obtained through on-site pumping tests. The calculation formula is: ,in The infiltration flow rate per unit time. For the cross-sectional area through which the flow passes, For hydraulic gradient; Saturation was obtained through rock sample saturation tests. The calculation formula is: ,in For the volume of water, This represents the pore volume of the rock sample. The formation strength value is calculated by comprehensively considering the physical, mechanical, and hydrological parameters of the formation. The calculation formula is: In the formula These are the empirical coefficients corresponding to each parameter, used to balance the contribution of different parameters to the overall strength; This represents the contribution of permeability resistance to strength, and is inversely proportional to the permeability coefficient.

4. The method for constructing a grouting reinforcement model for a tunnel in a completely weathered granite strata according to claim 3, characterized in that, The method for generating the formation intensity distribution map includes the following steps: Spatial interpolation algorithms are used to generate continuous distributions of physical, mechanical, and hydrological parameters of various strata in the tunnel using the acquired discrete data; the spatial interpolation algorithms include, but are not limited to, inverse distance weighting, kriging interpolation, or spline interpolation. The interpolated parameters are input into the formation strength calculation formula to generate the formation strength value for each grid cell, wherein the grid cells are divided according to the space around the tunnel. Using 3D modeling software, the geological strength values ​​are mapped to the tunnel model; the geological strength distribution at different spatial locations is displayed, and the strength differences are presented intuitively using color gradients or contour lines; key areas are marked in the model, including high-strength areas, low-strength areas, and fracture zones; a 3D strength distribution map is output for subsequent analysis and calculation.

5. The method for constructing a grouting reinforcement model for tunnels in completely weathered granite strata according to claim 1, characterized in that, The demand intensity calculation includes: The required bearing capacity of the surrounding rock of the tunnel is obtained based on the tunnel's design purpose and construction load. The calculation formula is: ,in The unit weight of the surrounding rock. For tunnel burial depth, To design a safety factor; Based on the tunnel's waterproofing requirements, obtain the required seepage prevention strength. The calculation formula is: ,in To allow infiltration traffic, Permeability coefficient; The calculated required bearing capacity and seepage prevention strength data are mapped onto the tunnel's three-dimensional model; a required strength distribution map is generated, and color gradients are used to represent the strength requirements of different areas; areas with concentrated required strength and weak points are marked as key references for grouting reinforcement design.

6. The method for constructing a grouting reinforcement model for a tunnel in a completely weathered granite strata according to claim 1, characterized in that, The recommended model for the reinforcement scheme includes: The recommended model for the hardening scheme is a multi-layered deep neural network, the structure of which includes: Input layer: Receives tunnel model data containing geological intensity distribution maps and demand intensity distribution maps; Hidden layer: Features are extracted using a convolutional neural network, and long short-term memory networks are used to analyze the nonlinear behavior of the formation; Output layer: Outputs recommended grouting reinforcement methods, including grouting materials, grouting type, grouting density, and construction method; The selection of convolutional kernels and activation functions in the hidden layers is based on the distribution characteristics of the input data, and an adaptive optimization mechanism is used for adjustment; The extracted features include: numerically gridded data of the stratum intensity distribution map; numerically gridded data of the demand intensity distribution map; tunnel cross-sectional shape and spatial location parameters; tunnel design purpose and load requirements; input features are processed through standardization and normalization. The grouting reinforcement methods output by the model include: Grouting materials include, but are not limited to, cement grout, ultrafine cement grout, chemical grout, and mixed grout; Grouting types include, but are not limited to, curtain grouting, consolidation grouting, and fracturing grouting; Grouting density: the amount of grout per unit volume; Construction methods include, but are not limited to, continuous grouting and segmented grouting; The output results are optimized using a multi-objective optimization algorithm, with optimization objectives including economy, construction efficiency, and reinforcement effect.

7. The method for constructing a grouting reinforcement model for a tunnel in a completely weathered granite strata according to claim 6, characterized in that, The training of the reinforcement scheme recommendation model includes: Initial reinforcement scheme recommended model structure initialization: set the convolutional kernel size, number of layers, and stride of the convolutional layers; configure the number of units in the LSTM layer; set the number of neurons in the fully connected layer; Dataset partitioning: Label the historical data of the tunnel model, construct a dataset, and partition the dataset into a training set and a validation set; Training process: Input the training set into the initial reinforcement scheme recommendation model; use a stochastic gradient descent optimizer to adjust the learning rate to avoid overfitting or underfitting; use a loss function to evaluate the deviation between the prediction and the actual value. The formula for the loss function is: ,in Indicates the loss value. Represents the mean squared error function. To predict the classification accuracy of the results, For loss weighting coefficients, For predicted values, The model's hyperparameters are adjusted using Bayesian optimization to obtain the optimized reinforcement scheme recommendation model through training. The recommended model for optimized reinforcement schemes was validated, and the validation indicators included numerical indicators, classification indicators, and comprehensive indicators. Validation method: The performance of the optimized reinforcement scheme recommendation model is evaluated using validation set data, and ROC curves and confusion matrices are generated to analyze the classification effect. The predicted values ​​are compared with the true label values, and the numerical accuracy of the optimized reinforcement scheme recommendation model is evaluated by goodness of fit and root mean square error. The optimized reinforcement scheme recommendation model that meets the numerical accuracy standard is output as the reinforcement scheme recommendation model.

8. The method for constructing a grouting reinforcement model for a tunnel in a completely weathered granite strata according to claim 1, characterized in that, The simulation and effect evaluation of tunnel grouting reinforcement includes the following steps: Tunnel grouting reinforcement simulation: A fluid-structure interaction model of the strata surrounding the tunnel was established using numerical simulation software. Parameters such as grouting pressure, grout viscosity, stratum permeability coefficient, and porosity were set in the model to simulate the diffusion range of the grout in the strata. A three-dimensional contour map of the grout diffusion range was generated, and the diffusion range limit value was marked. Tunnel loads are applied to the reinforcement simulation model to analyze the stress distribution and deformation characteristics of the tunnel structure after grouting reinforcement, and to simulate the stress and deformation of the tunnel structure. The deformation of the surrounding rock and the location of the maximum stress are calculated, and stress distribution maps and deformation prediction maps are generated. Based on the simulation results, adjust the grouting material parameters and construction parameters to optimize the reinforcement scheme; Evaluation of Grouting Reinforcement Effect: Calculate the bearing capacity improvement rate based on the stress distribution of the tunnel surrounding rock before and after reinforcement; calculate the maximum deformation of the tunnel surrounding rock after grouting reinforcement and compare it with the design allowable deformation to evaluate the deformation control effect; compare the changes in the permeability coefficient of the tunnel surrounding rock before and after grouting reinforcement to evaluate the improvement in permeability performance; comprehensively analyze the amount of grouting material used, construction time, and project cost to evaluate the economy and construction efficiency of the reinforcement scheme. If the evaluation results meet the design objectives, the grouting reinforcement scheme is confirmed to be effective; otherwise, return to adjust the simulation parameters and re-evaluate.

9. The method for constructing a grouting reinforcement model for a tunnel in a completely weathered granite strata according to claim 1, characterized in that, The result verification and optimization includes the following steps: Verification phase: Deformation prediction verification: Verify the tunnel deformation prediction map generated after grouting reinforcement, obtain the actual tunnel deformation using the on-site monitoring system, compare the deviation between the predicted deformation and the monitored deformation, and calculate the goodness of fit. Bearing capacity verification: Compare the bearing capacity distribution map of the reinforced tunnel with the results of on-site load tests; verify that the bearing capacity improvement rate meets the design requirements; if it does not meet the requirements, mark the weak areas. Permeability verification: The permeability coefficient of the reinforced surrounding rock is obtained using the permeability test results after grouting, and compared with the simulated permeability coefficient; if the deviation exceeds the set range, the formation permeability model parameters are corrected. Optimization phase: Parameter optimization: If the verification results are unqualified, adjust the following parameters: grouting pressure, grouting material properties, and construction method; Based on the optimized parameters, the grouting diffusion range and tunnel reinforcement effect are re-simulated; according to the optimized simulation results, new bearing capacity distribution maps, deformation prediction maps, and permeability distribution maps are generated; the updated distribution maps are mapped to the three-dimensional tunnel model, and the optimized key areas are marked; the optimized scheme is verified again, and the verification process of deformation, bearing capacity, and permeability is repeated. If the verification result is satisfactory, the optimized solution is confirmed as the final reinforcement solution; if it is still unsatisfactory, the optimization phase will be carried out again. Output the final reinforcement scheme, confirming that the reinforcement scheme meets the design requirements, including the optimized grouting materials, grouting type, construction parameters and grouting strategy; output the final three-dimensional reinforcement distribution map, including the diffusion range, the area of ​​increased bearing capacity and the area of ​​improved permeability; provide a reinforcement construction guidance report for use in project implementation.

10. A method for constructing a grouting reinforcement model for a tunnel in a completely weathered granite strata according to claims 1 to 9, characterized in that, The constructed grouting reinforcement model for tunnels in completely weathered granite strata includes the following modules: Tunnel model generation module: Generates a three-dimensional digital tunnel model based on tunnel usage data and construction environment data; Formation parameter acquisition and analysis module: Collects physical, mechanical and hydrological parameters of the formation; generates a formation strength distribution map using interpolation algorithms to intuitively reflect the mechanical properties of the formation before reinforcement. Demand strength calculation module: Calculates the required bearing strength and seepage prevention strength based on tunnel design requirements; Generate a demand intensity distribution map to analyze the areas requiring tunnel reinforcement; The reinforcement scheme recommendation module, based on the reinforcement scheme recommendation model, takes as input the stratum strength distribution map and the required strength distribution map, and outputs grouting reinforcement methods. The recommended grouting methods include grouting materials, grouting type, grouting density, and construction method; and provides preliminary optimized reinforcement schemes. Grouting reinforcement simulation module: uses a fluid-structure interaction model to simulate the diffusion range of grout in the formation; generates a bearing capacity distribution map and deformation prediction map after reinforcement; evaluates the permeability improvement effect after grouting; Verification and Optimization Module: Verify the grouting reinforcement effect by checking deformation, bearing capacity improvement rate, and permeability improvement rate through on-site monitoring data; if it does not meet the design requirements, adjust the grouting materials, construction methods, and parameters, and iteratively optimize the reinforcement scheme; update the distribution map and conduct multiple rounds of verification until it meets the requirements; Visualization output module: Visualizes the stratum strength distribution map, required strength distribution map, grouting diffusion range and reinforcement effect in three-dimensional form; provides an intuitive tunnel model display, including annotation of key areas; outputs optimized reinforcement scheme and construction guidance report.

Images