A method for controlling excavation face stability during cable shield tunneling in water-rich strata

Abstract

The present invention discloses a stable control method for excavation face of cable shield tunnel in water-rich strata, which relates to the field of shield tunnel excavation control. The method comprises: using virtual experiment technology to establish a tunnel excavation face model; using the tunnel excavation face model and discrete element method to simulate the deformation and settlement law of soil in the excavation face stratum of the support structure under different support force distribution conditions, and analyzing the ultimate support force of the support structure and the stability of the excavation face according to the soil deformation and settlement law; selecting the optimal support force distribution form, predicting the mechanical properties of the shield machine according to the shield machine structural parameters, and generating the excavation face stability control value in combination with the optimal support force distribution form; adjusting the excavation parameters of the shield machine according to the excavation face stability control value, and applying the adjusted shield machine to real-time tunnel excavation work. The present invention can effectively reduce the potential safety hazards such as excavation face instability and soil collapse during excavation, and further ensure the smooth progress of the project.

Description

Technical Field

[0001] The invention relates to the field of shield tunnel excavation control, and in particular to a method for stabilizing an excavation face of a cable shield tunnel excavation in a water-rich stratum. Background Art

[0002] As the urbanization process continues to accelerate, underground development has become an inevitable trend in the construction of urban power supply facilities. Urban cable tunnel projects have developed rapidly. At present, technology has implemented a variety of cable tunnel construction methods. Each method has its own characteristics and applicability. Among them, the shield method is increasingly used in cable tunnel construction due to its high degree of automation, fast construction speed, and little impact on the surrounding environment.

[0003] Water-rich strata refer to strata that contain a large amount of water underground. Such strata usually have a high porosity, and groundwater will accumulate in the pores, causing the strata to become saturated with water. Water-rich strata are significantly challenging in tunneling and underground engineering, because the increase in water content will cause changes in the mechanical properties of the soil, affecting the stability of the soil and the construction difficulty of the shield machine. Therefore, excavation face stability control is the key to ensuring the smooth and safe progress of the project. It not only helps prevent collapse and reduce settlement, but also effectively controls groundwater flow, protects the stability of cable lines, and improves the working efficiency and construction safety of the shield machine.

[0004] However, in the excavation face stability control of the existing technical group, it is impossible to pre-simulate the soil deformation and settlement laws of the support structure under different support force distribution conditions, and then it is impossible to select the optimal support force distribution form according to the law information to provide a basis for the excavation work, and it is impossible to set the working state of the shield machine according to the mechanical properties of the shield machine, and it is impossible to ensure the stability of the excavation face during the excavation process, which reduces the efficiency of the excavation work.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In view of the problems in the related art, the present invention proposes a method for stabilizing the excavation face of a cable shield tunnel in a water-rich stratum to overcome the above-mentioned technical problems existing in the existing related art.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] A method for controlling the stability of an excavation face during cable shield tunneling in a water-rich stratum, the method comprising:

[0009] The virtual experiment technology is used to calibrate the simulation parameters of water-rich strata, and the simulation parameters of water-rich strata are combined with the shield machine to establish the tunnel excavation face model;

[0010] The tunnel excavation face model and discrete element method are used to simulate the deformation and settlement of the soil in the excavation face under different support force distribution conditions, and the ultimate support force of the support structure and the stability of the excavation face are analyzed based on the deformation and settlement of the soil.

[0011] The optimal support force distribution form is selected based on the ultimate support force and the stability of the excavation face. The mechanical properties of the shield machine are predicted according to the strength of the water-rich stratum, the principal stress of the water-rich stratum and the structural parameters of the shield machine. The stability control value of the excavation face is generated in combination with the optimal support force distribution form.

[0012] The excavation parameters of the shield machine are adjusted according to the stability control value of the excavation face, and the adjusted shield machine is applied to real-time tunnel excavation work to judge the stress state of the excavation face after the excavation work is completed. The control ability of the excavation face stability control value is verified based on the stress state to achieve stable control of the shield tunnel excavation face.

[0013] Furthermore, the tunnel excavation face model and discrete element method are used to simulate the deformation and settlement law of the soil in the excavation face of the support structure under different support force distribution conditions, and the ultimate support force of the support structure and the stability of the excavation face are analyzed according to the soil deformation and settlement law, including:

[0014] A three-dimensional model of the support structure is established by using three-dimensional modeling technology, and a support stiffness coefficient and a bearing capacity coefficient are defined for the three-dimensional model of the support structure based on the support structure definition criteria;

[0015] The three-dimensional model of the support structure is integrated with the tunnel excavation face model to generate a discrete element simulation model, and the distribution state set of the three-dimensional model of the support structure in the tunnel excavation face model is defined in a linear distribution manner using the discrete element method;

[0016] Define the support force corresponding to the three-dimensional model of the support structure in each distribution form under the distribution state set, and use the discrete element method to simulate the excavation process after setting the initial stress state and boundary conditions;

[0017] The discrete element method is used to obtain the particle displacement of the water-rich stratum after excavation is completed, and the stratum movement and deformation of the excavation surface stratum are determined based on the particle displacement;

[0018] Based on the movement and deformation of the stratum, the deformation and settlement laws of the soil corresponding to the excavation surface stratum are analyzed, and the ultimate support force of the support structure and the stability of the excavation surface are determined in combination with curve analysis technology.

[0019] Furthermore, the optimal support force distribution form is selected based on the ultimate support force and stability, the mechanical properties of the shield machine are predicted according to the strength of the water-rich stratum, the principal stress of the water-rich stratum and the structural parameters of the shield machine, and the stability control value of the excavation face is generated in combination with the optimal support force distribution form, including:

[0020] The generalized Pareto distribution is used to fit the ultimate support force of the support structure and the stability of the excavation surface, and the test index is generated. The optimal support force distribution form is selected based on the test index.

[0021] Obtain water-rich stratum information including shear strength and principal stress, and collect shield machine structural parameters including cutter head diameter, thrust range, torque range and propulsion speed based on shield machine production information;

[0022] Based on the water-rich stratum information and the numerical application of shield machine structural parameters, the resistance of the shield machine during the excavation process is simulated and analyzed, and the mechanical performance of the shield machine is predicted based on the resistance results;

[0023] The support force control value of the shield machine at any excavation process point is determined according to the optimal support distribution form, and the shield machine control value is generated based on the mechanical performance of the shield machine. The support force control value and the shield machine control value are combined as the excavation face stability control value.

[0024] Furthermore, the method also includes monitoring the formation changes in real time, analyzing the real-time ultimate support force, and dynamically adjusting the distribution of the support force to obtain the optimal support force distribution form under the current formation state, including:

[0025] Using sensors to collect formation parameters in real time, the formation parameters include: soil deformation displacement, formation settlement, seepage pressure of water-rich formations, shield machine propulsion speed and shield machine thrust;

[0026] The influence coefficient of stratum support force is obtained by analyzing the stratum parameters;

[0027] Obtain the formation support force correction value corresponding to each influence coefficient interval preset in the database, and match the formation support force influence coefficient with each influence coefficient interval. If the formation support force influence coefficient belongs to a certain preset formation support force influence coefficient interval, obtain the formation support force correction value corresponding to the influence coefficient interval as the formation support force threshold adjustment factor;

[0028] The front support force, rear support force and shield machine displacement are obtained, and the support force distribution is obtained by analysis based on the stratum support force threshold adjustment factor.

[0029] The beneficial effects of the present invention are:

[0030] 1. The present invention can accurately calibrate the simulation parameters of water-rich strata through virtual experiment technology, and use the tunnel excavation face model and discrete element method to simulate the deformation and settlement law of soil under different support force distribution conditions of the support structure, so as to facilitate the identification of the relationship between support force distribution and stratum deformation and settlement, and avoid excessive or insufficient support. At the same time, the optimal support force distribution form is selected based on the ultimate support force and stability, and the mechanical properties are predicted according to the stratum strength, principal stress and shield machine parameters to generate the stability control value of the excavation face. Furthermore, through real-time detection, it is ensured that the working state of the shield machine can be dynamically adjusted during the actual excavation process, effectively reducing the safety hazards such as instability of the excavation face and soil collapse during the excavation process, and further ensuring the smooth progress of the project.

[0031] 2. The present invention establishes a three-dimensional model of the support structure and defines the support stiffness coefficient and the bearing capacity coefficient to quantify the deformation resistance and bearing capacity of the support structure, which is helpful to analyze its performance under different pressure conditions and ensure that the support structure can effectively maintain the stability of the excavation surface. At the same time, the three-dimensional model of the support structure is integrated with the tunnel excavation surface model to generate a discrete element simulation model, which can accurately reproduce the interaction between the support structure and the stratum, providing a more reliable basis for support design. Defining the distribution state set of the three-dimensional model of the support structure in the excavation surface according to the linear distribution method is helpful to find the best support structure layout, reduce material and cost waste, and ensure construction safety at the same time. The particle displacement of the stratum after the excavation is completed is obtained, and the movement and deformation of the stratum is judged to analyze the ultimate support force of the support structure and the stability of the excavation surface.

[0032] 3. The present invention can use the generalized Pareto distribution technology to perform a more scientific fitting analysis on the ultimate support force of the support structure and the stability of the excavation face to generate test indicators, which is convenient for accurately identifying the ultimate performance of the support structure and ensuring the optimization of the support force distribution during construction, thereby maximizing the stability of the excavation face. At the same time, combined with the shear strength and principal stress information of the water-rich stratum, the mechanical properties of the stratum are fully understood, ensuring that the support design and the operation of the shield machine can adapt to the complex stratum environment and reduce geological uncertainties in construction. The numerical simulation technology can also be used to predict the resistance of the shield machine under different geological conditions and excavation parameters, and the working state of the shield machine is predicted accordingly to generate the excavation face stability control value, ensuring that the excavation process can still maintain stability under complex conditions such as water-rich strata.

[0033] 4. The present invention obtains the formation support force influence coefficient by analyzing the formation parameters, and then obtains the formation support force threshold adjustment factor by matching it with the influence coefficient intervals preset in the database, and calculates the support force distribution in real time, so that the support force can accurately adapt to the support requirements of the current formation during the excavation process, thereby effectively enhancing the stability of the support structure and solving the problem in the prior art that the support force distribution cannot respond to formation changes in real time.

[0034] 5. The present invention feeds back the real-time calculated support force distribution to the excavation face support structure model, so that the support structure can dynamically respond to changes in formation conditions during the excavation process, thereby improving the support effect of the excavation face, ensuring the safety and continuity of the construction, and effectively solving the problem in the prior art that the support structure force distribution cannot be automatically adjusted according to real-time data. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 The present invention is a flowchart of a method for controlling the stability of an excavation face of a cable shield tunnel in a water-rich stratum according to an embodiment of the present invention. DETAILED DESCRIPTION

[0037] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments. They can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention.

[0038] According to an embodiment of the present invention, a method for controlling the stability of an excavation face during excavation of a cable shield tunnel in a water-rich stratum is provided.

[0039] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to the method for stabilizing the excavation face of a cable shield tunnel in a water-rich stratum according to an embodiment of the present invention, the method comprises:

[0040] Step S1, using virtual experiment technology to calibrate water-rich stratum simulation parameters, and combining the water-rich stratum simulation parameters with a shield machine to establish a tunnel excavation face model.

[0041] In this embodiment, it should be noted that virtual experiment technology is a technical means for calibrating simulation parameters of water-rich strata, used to simulate and reproduce the geological conditions of water-rich strata, so as to more accurately understand and predict the behavior of shield machines during excavation and the stability of excavation faces. Virtual experiment technology involves computer simulation and emulation, and researchers use computer models to test and verify theories, hypotheses or designs without the need for actual physical experiments. In the field of geological engineering and tunneling, virtual experiment technology can simulate tunneling processes under different geological conditions, including water-rich strata. This technology can simulate the physical properties of strata (such as strength, stress state, etc.), the excavation parameters of shield machines (such as thrust, torque, excavation speed, etc.), and the role of support structures (such as support force distribution, support stiffness, etc.). Discrete Element Method (DEM) is a numerical simulation method specifically used to solve discontinuous medium problems. This method regards the medium as consisting of discrete particles or blocks, allowing these particles or blocks to translate, rotate and deform after being subjected to force. By simulating the interactions and mechanical behaviors between these particles or blocks, the deformation and failure process of the entire medium can be predicted and analyzed.

[0042] When using virtual experiment technology to calibrate the simulation parameters of water-rich strata, and combining the simulation parameters of water-rich strata with the shield machine to establish a tunnel excavation face model, relevant geological data can be collected, including soil type, soil water content, soil particle size distribution, groundwater flow characteristics, etc.; the data can be analyzed to determine the main physical and chemical properties of the strata, which will serve as the basis for subsequent virtual experiments; the collected data can be used to establish an initial stratum model, and a series of virtual experiments can be designed to calibrate the behavior parameters of the stratum by simulating different groundwater flow and soil mechanical responses; according to the virtual experiments, The results adjusted the formation model parameters and performed multiple iterations until the model could accurately reflect the true behavior of the formation. The relevant parameters of the shield machine, such as cutterhead diameter, thrust range, torque range, and propulsion speed, were collected, and the adjusted formation parameters were integrated with the operating characteristics of the shield machine to establish a comprehensive tunnel excavation face model. Computer-aided design (CAD) software was used to establish a three-dimensional model of the tunnel excavation face, and the integrated formation and shield machine parameters were applied to the model to simulate various possible situations during tunneling, including formation deformation, stress distribution, and water flow changes.

[0043] Step S2: using the tunnel excavation face model and discrete element method, simulate the soil deformation and settlement law of the excavation face stratum under different support force distribution conditions of the support structure, and analyze the ultimate support force of the support structure and the stability of the excavation face according to the soil deformation and settlement law.

[0044] In this embodiment, when using the tunnel excavation face model and the discrete element method to simulate the soil deformation and settlement law of the excavation face stratum of the support structure under different support force distribution conditions, and analyzing the ultimate support force of the support structure and the stability of the excavation face according to the soil deformation and settlement law, the three-dimensional modeling technology can be used to establish a three-dimensional model of the support structure, and the support stiffness coefficient and the bearing capacity coefficient are defined for the three-dimensional model of the support structure based on the support structure definition criterion; the three-dimensional model of the support structure is integrated with the tunnel excavation face model to generate a discrete element simulation model, and the discrete element method is used to define the support structure in a linear distribution manner. The distributed state set of the three-dimensional model of the tunnel excavation face model is constructed; the supporting force corresponding to the three-dimensional model of each distribution form of the supporting structure under the distributed state set is defined, and the discrete element method is applied to simulate the excavation process after setting the initial stress state and boundary conditions; the discrete element method is used to obtain the particle displacement generated by the water-rich stratum after the excavation is completed, and the stratum movement and deformation conditions of the excavation face stratum are judged based on the particle displacement conditions; the soil deformation and settlement laws corresponding to the stratum of the excavation face are analyzed based on the stratum movement and deformation conditions, and the ultimate supporting force of the supporting structure and the stability of the excavation face are judged in combination with the curve analysis technology.

[0045] It should be noted that the definition criteria of the support structure include: safety and reliability, reasonable economy, full consideration of the environment and conditions, overall stability and deformation control. Curve analysis technology refers to the method of in-depth research and analysis of the curve shape, trend and characteristics obtained in experiments or actual data.

[0046] In this embodiment, when a three-dimensional model of a support structure is established by using a three-dimensional modeling technology, and a support stiffness coefficient and a bearing capacity coefficient are defined for the three-dimensional model of the support structure based on the support structure definition criteria, the design and construction parameters of the support structure can be collected, including the material type of the support structure (such as reinforced concrete, steel plate, etc.), structural geometric dimensions (such as thickness, length, width, etc.), and construction environment (such as soil type, groundwater level, etc.); according to relevant building or engineering standards and specifications, key indicators in the design of the support structure are determined, such as the allowable stress, deformation standard, and shear and compression resistance of the structure; and appropriate three-dimensional modeling software is selected, such as AutoCAD, SolidWorks, ANSYS, etc. are used to create the geometric model of the support structure; based on the collected geometric dimension data, such as the depth of the foundation pit, the diameter of the support piles, etc., an initial three-dimensional model of the support structure is established in the modeling software; based on the physical properties of the support structure material, such as elastic modulus, Poisson's ratio, and structural shape, the stiffness coefficient of the support structure is defined, and the stiffness coefficient usually reflects the ability of the support structure to resist deformation; based on engineering design specifications and geological conditions, such as the type, thickness, density, water content, strength and other characteristics of the soil layer, topography, geological structure, groundwater conditions, etc., the bearing capacity of the support structure is defined, and the bearing capacity coefficient reflects the maximum ability of the support structure to maintain stability under the action of external loads.

[0047] In this embodiment, when the three-dimensional model of the support structure is integrated with the tunnel excavation face model to generate a discrete element simulation model, and the distribution state set of the three-dimensional model of the support structure in the tunnel excavation face model is defined in a linear distribution manner using the discrete element method, the three-dimensional model of the support structure is combined with the excavation face model to ensure that the layout of the support structure and the geometric relationship of the excavation face can accurately reflect the on-site construction conditions, and the physical properties of the support structure and the excavation face model are integrated so that their mechanical properties can be accurately reflected in the simulation, such as the strength and elastic modulus of the stratum soil; the tunnel excavation face and the support structure model are discretized into a large number of particle units, each unit represents a part of the structure, and the discrete element method DEM is to decompose the continuous medium into particles for mechanical calculation, and define the interaction rules between the particles for the discrete element model, including contact force, friction force, adhesion force, etc., to accurately simulate the actual interaction between the stratum and the support structure, and the force distribution of the support structure is divided in a linear manner according to the stratum stress during tunnel excavation to ensure that the gradual change process of the support force distribution meets the actual engineering needs. For example, under different excavation stages and geological conditions, the distribution of support force changes gradually from front to back. The initial stress state is set for the discrete element simulation model to reflect the original stress distribution in the stratum before excavation. At the same time, boundary conditions are defined for the tunnel excavation model to ensure that the model does not produce unreasonable deformation or boundary effects during the simulation, for example, fixed boundaries or free boundaries are set to simulate geological constraints.

[0048] In this embodiment, the support force corresponding to the three-dimensional model of the support structure of each distribution form under the distribution state set is defined, and the discrete element method is applied after setting the initial stress state and boundary conditions. When simulating the excavation process, different sets of support structure distribution states can be established based on actual engineering needs. Each set represents a different support structure distribution form, such as linear change of support force from front to back, uniform distribution or nonlinear distribution, etc.; define parameters for each support form, such as the spacing of the support structure, the gradient change of the support force, the distribution of support stiffness, etc. According to the engineering design standards and formation conditions, the corresponding support force is calculated for each distribution form. Considering the weight of the soil, the groundwater pressure, and the changes in formation stress generated during tunnel excavation, the support force is allocated to each support structure unit in the three-dimensional model. These support forces can be uniform or distributed according to some predefined rules, such as linear, exponential, etc.; according to the geological survey data of the tunnel excavation area, the initial stress state in the stratum is set, including horizontal stress, vertical stress and stratum self-weight stress, etc., and boundary conditions are set for the tunnel excavation surface, usually including fixed boundaries (simulating the boundaries of fixed strata) and free boundaries (simulating the open environment outside the stratum) to ensure that the model does not produce unreasonable stress concentration or deformation in the simulation. In water-rich strata, it is necessary to simulate the flow and pressure of groundwater and set corresponding boundary conditions to reflect the influence of water pressure on tunnel stability; the discrete element method DEM is used to simulate the interaction between the support structure and the stratum during the advancement of the shield machine.

[0049] In this embodiment, when the particle displacement generated by the water-rich stratum after excavation is obtained by using the discrete element method, and the stratum movement and deformation of the excavation surface stratum are judged based on the particle displacement, the displacement data of all particles can be recorded after the simulation is completed. The displacement vector of each particle can be obtained by the discrete element method, including the size and direction of the displacement, and the particle displacement data can be used to generate a three-dimensional displacement field image to intuitively display the movement trend of the stratum particles after excavation. At the same time, the particle displacement around the excavation surface is analyzed to determine the area in the stratum where large displacement or settlement may occur. The area is usually the most unstable part of the stratum during the excavation process, and there may be a risk of stratum instability or settlement; the displacement gradient is determined by calculating the displacement difference between the particles; the overall settlement or uplift of the stratum during or after the excavation is judged according to the displacement data; the overall deformation mode of the stratum is analyzed according to the particle displacement field, including shear failure, local collapse or large-scale deformation, and the stability of the stratum during the excavation process can be judged by these deformation modes.

[0050] In this embodiment, when analyzing the deformation and settlement law of the soil corresponding to the excavation surface stratum based on the stratum movement and deformation conditions, and combining the curve analysis technology to judge the ultimate support force of the support structure and the stability of the excavation surface, the heuristic search algorithm can be used to read the characteristic information of the stratum movement and deformation conditions data, initialize the feature set after standardization of the characteristic information, and generate a number of clusters in the solution space; based on the cluster point traversal feature information, the overall probability and average probability of the distribution of the feature are judged, and the fitness value of the cluster point is calculated according to the overall probability and the average probability of the distribution, and the fitness value is used to analyze the intensity and amplitude; select the pair of characteristic information corresponding to the maximum value of intensity and amplitude as the change characteristic quantity, and use grey correlation analysis to calculate the correlation between the change characteristic quantity and soil deformation and soil settlement, and analyze the soil deformation and settlement law curve corresponding to the excavation surface stratum based on the correlation; use curve analysis technology to analyze the soil deformation and settlement law curve to obtain the motion trajectory points and deformation mode of water-rich stratum particles, and calculate the difference between the motion trajectory points and the standard motion trajectory points; calculate the ratio of the deformation mode to the standard deformation mode, and judge the ultimate support force of the support structure and the stability of the excavation surface based on the difference result and the ratio result.

[0051] Heuristic Search Algorithm, also known as Informed Search, is a search algorithm guided by heuristic information. It evaluates the heuristic value of the search node through the evaluation function, thereby guiding the search direction and finding the optimal solution or a solution close to the optimal solution in the search space. Grey correlation analysis refers to a method of quantitatively describing and comparing the development and change trend of a system. It determines whether the connection is close by determining the degree of similarity between the collection shapes of the reference data column and several comparison data columns, reflecting the degree of correlation between the curves. If the trends of the changes of the two factors are consistent, that is, the degree of synchronous change is high, then the degree of correlation between the two is high; otherwise, it is low.

[0052] It should be noted that the difference calculation is to make a difference between the actual monitored deformation mode and the standard mode, such as displacement difference, settlement difference, etc. The larger the difference, the more serious the actual deformation deviates from the expected standard, and the corresponding stability is poor. The ratio calculation is to calculate the ratio of the actual deformation to the standard. When the ratio is greater than the set threshold, such as 1.2, it means that the actual deformation is significantly greater than the design standard, indicating that the formation stability is poor and the ultimate support force may be insufficient; when the ratio is close to 1 or less than 1, it means that the support force is relatively stable.

[0053] Among them, when the fitness value is used to analyze the intensity and amplitude of feature changes, the calculation formula for the intensity of feature changes is:

[0054]

[0055] The formula for calculating the magnitude of feature change is:

[0056]

[0057] Where P i Indicates the intensity of feature change, Q i Indicates the magnitude of feature change, M min With M max They represent the maximum and minimum values ​​of the fitness value of the clustering point, β represents the correction value, T represents the number of clustering points, and f i (X i ) represents the fitness value of the i-th cluster point.

[0058] In order to facilitate understanding of the above technical solution of the present invention, the ultimate support force of the support structure and the stability analysis operation of the excavation surface in the actual process of the present invention are described in detail below.

[0059] Step 1: Parameter definition;

[0060] Support structure material: reinforced concrete, the specific parameters include: elastic modulus E is 30GPa, Poisson's ratio ν is 0.2, compressive strength σ comp is 40MPa, and the shear strength σ shear 3Mpa;

[0061] The thickness α of the support structure is 0.35m, the width w is 1.0m, the length l is 5.0m, the saturated water content of the water-rich stratum in the construction environment is 45%, and the groundwater level depth H gw =7m.

[0062] According to the material and geometric dimensions of the supporting structure, the deformation resistance of the supporting structure is calculated, and the stiffness coefficient K = EA / L, where the cross-sectional area of ​​the supporting structure A = wα = 0.35m 2 , the deformation L = 5m, so the stiffness coefficient is: K = 530 × 10 9 *0.35 / 5=2.1×109N / m; Assuming the allowable stress σ=20MPa, the bearing capacity F max =σA, then F max =20×10 6 *0.35=7.0×10 6 N.

[0063] Step 2: Establish a discrete element simulation model;

[0064] The water-rich strata are discretized into particle units with a particle diameter d p = 0.1m, the model contains 1200012000 particles, and the contact stiffness between particles is defined as the normal contact stiffness K n=1×107N / m, tangential contact stiffness K s =0.5×107N / m, based on the groundwater level being H gw =7m, where the calculated water pressure is P gw =ρ*g*H gw =1000*9.8*7=68.6kPa.

[0065] Step 3: Definition of support force distribution method;

[0066] The support force of the support structure follows a linear distribution from the front to the back of the excavation surface. Assuming that the front support force is F min =1.5×106N, the rear support force is F max =7.0×106N, the linear distribution function is: F(x)=F min +l(F max -F min )*x, where x is the position of the supporting structure from front to back and l=5m is the length of the supporting structure.

[0067] Step 4: Set the initial stress state and boundary conditions;

[0068] According to geological survey data, the initial stress includes horizontal stress h = 50kPa, vertical stress σ v =120kPa, self-weight stress P g =ρ*g*H soil =1800*9.8*10=176.4kPa (assuming the formation thickness H soil =10m), the fixed boundary conditions are used for the upper and lower boundaries of the stratum to prevent unreasonable boundary effects, and the water pressure boundary condition P gw =68.6kPa acts on the groundwater area.

[0069] Step 5: Simulate the excavation process and analyze particle displacement and settlement;

[0070] Excavation face advancement: gradually remove the soil on the excavation face, record the displacement and settlement of stratum particles during the excavation process, gradually apply the support force of the support structure during the excavation advancement process, and update the support force value in real time according to different distribution forms.

[0071] After the excavation is completed, the displacement of each particle is recorded to obtain the three-dimensional particle displacement field. Assuming that the maximum particle displacement of the stratum after excavation is u max =15cm, the average displacement is u avg =5cm.

[0072] By calculating the displacement gradient, the area with the largest displacement change in the stratum is determined. Assuming that the displacement gradient of the local area is Δu / Δx=2cm / m, this area is the area with the most serious settlement.

[0073] Step 6: Analysis of soil deformation and settlement rules;

[0074] Settlement law: The soil settlement curve is generated based on the particle displacement data. Assuming that the maximum settlement value is Smax = 20cm, the settlement curve can be approximately expressed as: S(x) = Smax*(1-e -kx ), where k is the decreasing coefficient of the sedimentation curve, and k = 0.5.

[0075] The overall deformation mode of the stratum is analyzed in the soil deformation mode. It is assumed that the deformation mode is shear failure, the deformation angle θ = 10°, and a local collapse risk area appears.

[0076] Step 7: Analysis of ultimate support force and excavation surface stability;

[0077] Using curve analysis technology, the difference between the soil deformation curve S(x) and the particle motion trajectory is analyzed. Assume that the standard trajectory point is u s =10cm, calculate the difference between the actual displacement trajectory and the standard trajectory point Δu=u max -u s =5cm, combined with grey correlation analysis, assuming that the correlation degree r = 0.85, a high correlation degree indicates that the current support force is close to the ultimate support force. According to the deformation and settlement analysis, the ultimate support force is assumed to be F lim =6.5×10 6 N, this value is close to the maximum support force F max =7.0×10 6 N indicates that the current support scheme is close to the safety limit, and it is recommended to strengthen the support or adjust the support force distribution.

[0078] The present invention simulates the effect of the support structure during tunnel excavation through precise three-dimensional modeling, discrete element method and curve analysis method, and facilitates the subsequent excavation face stability control work by analyzing the ultimate support force and the stability of the excavation face.

[0079] Step S3, selecting the optimal support force distribution form based on the ultimate support force and stability, predicting the mechanical properties of the shield machine according to the strength of the water-rich stratum, the principal stress of the water-rich stratum and the structural parameters of the shield machine, and generating the excavation face stability control value in combination with the optimal support force distribution form.

[0080] In this embodiment, the generalized Pareto distribution is used to fit the ultimate support force of the support structure and the stability of the excavation face to generate a test index, and the optimal support force distribution form is selected based on the test index; the water-rich stratum information containing shear strength and principal stress is obtained, and the shield machine structural parameters including the cutter head diameter, thrust range, torque range and propulsion speed are collected based on the shield machine production information; the numerical simulation technology is used to analyze the resistance of the shield machine during the excavation process based on the water-rich stratum information and the shield machine structural parameters, and the mechanical properties of the shield machine are predicted based on the resistance results; the support force control value of the shield machine at any excavation process point is determined according to the optimal support distribution form, and the shield machine control value is generated based on the mechanical properties of the shield machine, and the support force control value is combined with the shield machine control value as the control value for achieving excavation face stability.

[0081] The generalized Pareto distribution is a continuous probability distribution used to describe the conditional distribution of random variables exceeding a given threshold. In this embodiment, when applying the generalized Pareto distribution to fit the ultimate support force of the support structure and the stability of the excavation surface to generate a test index, and selecting the optimal support force distribution form based on the test index, the soil deformation value and settlement value corresponding to the ultimate support force of the support structure and the stability of the excavation surface can be arranged in order from large to small, and the Hill graph method can be used to determine the Hill statistic of the minimum soil deformation value and settlement value; the error between the Hill statistic and the preset Hill statistic is determined, and the Hill statistic that meets the error requirement is used as the water-rich formation location parameter; a support probability function is established based on the water-rich formation location parameter and the generalized Pareto distribution technology, and the support probability function is used to analyze the probability value of collapse under different support force distribution forms; the probability value is used as a test index to measure the ultimate support force of the support structure and the stability of the excavation surface, and the support force distribution form corresponding to the minimum test index is defined as the optimal selection result.

[0082] Among them, the expression of the support probability function is:

[0083]

[0084] In the formula, g t represents the probability of collapse under the tth support force distribution form, represents the soil deformation value under the t-th support force distribution form, τ t represents the soil settlement value under the t-th support force distribution form, γ represents the shape parameter, δ t represents the support force corresponding to the tth support force distribution form, and ξ represents the location parameter of the water-rich stratum.

[0085] In this embodiment, when numerical simulation technology is applied to analyze the resistance of the shield machine during the excavation process based on the water-rich stratum information and the structural parameters of the shield machine, and the mechanical properties of the shield machine are predicted based on the resistance results, the force range when the cutterhead contacts the water-rich stratum within the thrust range can be established based on the structural information of the shield machine, the corresponding torque value is analyzed according to the force range, and the shield machine model is simplified using the torque value and the force range using a medium simulation program; the water-rich stratum information is used as a macro parameter to characterize the water-rich stratum, and a two-dimensional numerical model is established in combination with the particle flow technology, and numerical tests at different excavation intervals are completed using the two-dimensional numerical model and the simplified shield machine; the resistance of the shield machine during the excavation process is judged based on the numerical test results, and the correlation between the macro parameters and the structural information of the shield machine is analyzed based on the resistance results; the correlations are ranked in importance, and a prediction model is constructed based on the importance ranking results and a bidirectional long short-term memory neural network to predict the mechanical properties of the shield machine.

[0086] In this embodiment, after the relevance is sorted by importance, the macro parameters and shield machine structure information are preprocessed and used as input data; the maximum number of iterations and the range of hyperparameter values ​​are set, and based on the setting results, they are applied to a bidirectional long short-term memory neural network to establish a prediction model, and then the fitness of the neural network nodes is calculated using an adaptive optimization algorithm; the linear reduction factor of the neural network is reconstructed based on the node fitness value results, and the Gaussian mixture mutation value is introduced according to the linear reduction factor to update the node position to calculate the fitness of the optimal node; it is determined whether the individual has reached the maximum number of iterations, and if so, the maximum value of the hyperparameter is assigned to the bidirectional long short-term memory neural network to obtain the final prediction model, and if not, the linear reduction factor is reconstructed; the input data is input into the final prediction model, the resistance result is output, and the resistance result is compared with the preset standard resistance value, and the comparison value is defined as the mechanical performance value of the shield machine.

[0087] It should be noted that the node fitness is calculated using the adaptive optimization algorithm. The "node" here refers to the "node" in the bidirectional long short-term memory neural network (BiLSTM), which represents the neurons in the network structure. "Fitness" refers to the adaptability of each neuron node under a specific task (such as support force prediction, formation state prediction), reflecting the degree of match between the node output and the expected output.

[0088] The linear reduction factor is reconstructed based on the node fitness value results. The "linear reduction factor" is a parameter adjustment factor of the neural network, which is used to gradually reduce the learning step size to ensure smoother convergence in the later stages of training. The "linear reduction factor" is used to optimize the hyperparameters of the neural network to make the fitness calculation process more refined. The Gaussian mixture mutation here optimizes the node position and obtains a better neuron layout by introducing Gaussian noise mutation, thereby improving the robustness of the prediction results.

[0089] “Judge whether the individual has reached the maximum number of iterations. If so, assign the maximum value of the hyperparameter to the bidirectional long short-term memory neural network to obtain the final prediction model. If not, reconstruct the linear reduction factor.” Here, “individual” refers to the combination of multiple weight parameters formed during the neural network training process, that is, the “individual” parameter set in the model training process. The training cycle of each model individual is limited by setting the maximum number of iterations. If the iteration limit is reached, the hyperparameter value is fixed. Otherwise, the reduction factor is reconstructed to ensure that the final model can meet the optimal support force prediction requirements.

[0090] In this embodiment, the bidirectional long short-term memory neural network (Bidirectional Long Short-Term Memory, BiLSTM) is a special recurrent neural network (Recurrent Neural Network, RNN) for processing sequence data, capable of learning long-term dependencies, and by introducing gate mechanisms (input gates, forget gates, and output gates) to overcome the gradient vanishing and gradient explosion problems of traditional RNNs when processing long sequences. The optimization algorithm aims to find an optimal or approximately optimal solution so that a certain objective function (such as a loss function) is minimized or maximized. In machine learning and deep learning, optimization algorithms are often used to train models to minimize prediction errors, and adaptive optimization algorithms can automatically adjust key parameters such as learning rates according to gradient information or other statistical data during training. This flexibility enables adaptive optimization algorithms to have better performance and convergence speed when training deep neural networks. The core principle of the adaptive optimization algorithm is to dynamically adjust the learning rate according to gradient information or other statistical data. Adaptive optimization algorithms such as AdaGrad (adaptive gradient algorithm), RMSprop (root mean square propagation algorithm), Adam (adaptive moment estimation algorithm), AdaDelta (adaptive learning rate adjustment algorithm), etc.

[0091] In this embodiment, generating a tunneling excavation face stability control value in combination with an optimal support force distribution form includes the following steps: setting a distribution form of the support structure within the tunneling range of the shield machine based on the optimal distribution form, and establishing a distribution formula for the support force according to the distribution form to determine the support force control value of the shield machine; determining the thrust value of the shield machine based on the excavation face area and the positive stress during the tunneling process of the shield machine, and analyzing the torque force and propulsion speed of the shield machine based on the resistance value and the strength of the water-rich stratum and the principal stress; comparing the thrust value, torque force and propulsion speed with the mechanical performance value of the shield machine to verify the rationality of the shield machine setting, and after the verification is completed, combining them with the support force control value as the control value to achieve excavation face stability.

[0092] The method of the present invention also includes real-time monitoring of stratum changes, analyzing real-time ultimate support force, and dynamically adjusting the distribution of support force to obtain the optimal support force distribution form under the current stratum state, and then generating a stable control value for the excavation face based on the real-time stratum changes.

[0093] In this embodiment, the distribution form of the supporting force is obtained by:

[0094] F(x)=F min +α1*(F max -F min )*X w

[0095] In the formula, F(x) represents the shield machine advancing to position X w The support force at min Indicates the front support force, F max represents the rear support force, α1 represents the formation support force threshold adjustment factor, X w Indicates the displacement of the shield machine.

[0096] The influence coefficient of stratum support force is obtained by analyzing the stratum parameters. The specific steps include:

[0097] Obtaining a set of stratum parameter standards preset in a database, wherein the set of stratum parameter standards includes: soil deformation displacement verification value, stratum settlement verification value, water-rich stratum seepage pressure verification value, shield machine propulsion displacement verification value, shield machine propulsion speed verification value, and shield machine thrust verification value;

[0098] The formation parameter standard set and formation parameters are analyzed to obtain the formation support force influence coefficient;

[0099] The formation support force influence coefficient is used to characterize the influence degree of formation parameters on the formation support force.

[0100] The influence coefficient of stratum support force is obtained by:

[0101]

[0102] In the formula, α2 represents the influence coefficient of stratum support force, U1 represents soil deformation displacement, ΔU1 represents the verification value of soil deformation displacement, S1 represents stratum settlement, ΔS1 represents the verification value of stratum settlement, P s Indicates the osmotic pressure of water-rich formations, ΔP s represents the verification value of the seepage pressure of the water-rich stratum, X1 represents the displacement of the shield machine, ΔX1 represents the verification value of the displacement of the shield machine, V t Indicates the shield machine advancement speed, ΔV t Indicates the verification value of shield machine advancement speed, F tIndicates the thrust of the shield machine, ΔF t Indicates the thrust verification value of the shield machine. represents the influence weight of soil deformation and displacement, Indicates the influence weight of stratum settlement, Indicates the osmotic pressure of water-rich formations, Indicates the weight of shield machine advancement speed, Represents the influence weight of shield machine thrust.

[0103] Sensors are used to collect formation parameters in real time, and the formation parameters include: soil deformation displacement, formation settlement, seepage pressure of water-rich formations, shield machine propulsion speed and shield machine thrust; the formation support force influence coefficient is obtained by analysis based on the formation parameters; the formation support force correction value corresponding to each formation support force influence coefficient interval preset in the database is obtained, and the formation support force influence coefficient is matched with each influence coefficient interval. If the formation support force influence coefficient belongs to a preset formation support force influence coefficient interval, the formation support force correction value corresponding to the influence coefficient interval is obtained as the formation support force threshold adjustment factor α1; the front support force, the rear support force and the shield machine propulsion displacement are obtained, and the support force distribution is obtained by analysis based on the formation support force threshold adjustment factor.

[0104] It should be noted that soil deformation displacement, stratum settlement, seepage pressure of water-rich strata, shield machine advancement displacement, shield machine advancement speed and shield machine thrust are all real-time measurement data, which can be obtained through real-time measurement when used, and can also be obtained through real-time simulation of the three-dimensional model of the tunnel excavation surface. For example: soil deformation displacement can be obtained through real-time simulation of the three-dimensional model of the tunnel excavation surface, and can also be obtained by measuring the displacement difference between adjacent time points before and after. The stratum settlement can be obtained by GPS measurement method by comparing elevation data at different times. The seepage pressure of water-rich strata can be obtained by measuring with an osmometer. The shield machine advancement displacement, shield machine advancement speed and shield machine thrust can be obtained by sensor measurement.

[0105] It should be noted that the shield machine displacement X w It indicates the distance from the initial position of the shield machine to the real-time advancement position. By evaluating the support force at the advancement position of the shield machine, the changes in the ground conditions and the advancement of the shield can be evaluated and monitored in real time. The support force can be dynamically allocated to cope with the complexity of different parts of the formation, making the support structure more adaptable to the support of the excavation surface. In a complex formation environment, real-time support force distribution can effectively prevent safety risks such as settlement and infiltration caused by unreasonable support force distribution, improve the overall stability of shield excavation, and automatically optimize the support force distribution according to the formation support force requirements, which not only ensures the safety of the structure, but also avoids the waste of support materials and improves the economic benefits of the support project.

[0106] By obtaining the threshold adjustment factor of the stratum support force and evaluating it, the real-time support force is obtained, that is, the distribution form of the support force is obtained. By calculating the ratio of the actual stratum state parameters (such as deformation displacement, settlement, seepage pressure, etc.) to the corresponding allowable value, the degree of demand for support force of the real-time stratum conditions is reflected. The role of the stratum support force threshold adjustment factor is to standardize the influence of the stratum support force and dynamically correct the distribution of the support force to ensure that the support of the support structure force distribution on the excavation face is consistent with the actual situation of the stratum. The stratum support force threshold adjustment factor can realize the adaptive adjustment of the support structure, effectively prevent the situation of insufficient or excessive support force during the excavation process, improve the accuracy of the support force distribution of the excavation face, ensure the safety and stability of shield excavation, adapt to the complex conditions of water-rich strata, optimize the mechanical distribution of the support structure, avoid the construction risks caused by soil deformation or seepage pressure changes, and avoid the waste of resources.

[0107] The support force distribution is a dynamic calculation result based on the stratum support force threshold adjustment factor. By feeding back the support force distribution result to the tunnel excavation face model in real time, the support force distribution is automatically adjusted according to the real-time stratum status during the advancement of the shield machine. The support force distribution in the support structure model can therefore be continuously optimized according to the stratum changes, ensuring that the support force can accurately respond to the support requirements of the stratum under complex stratum conditions, further improving the stability of the excavation face. The dynamic calculation of the support force threshold adjustment factor realizes the precise control and optimal distribution of the support force distribution in the support structure model, effectively avoiding the phenomenon of excessive or insufficient support force, and ensuring that the mechanical properties of the support structure are consistent with the actual requirements of the stratum. By dynamically optimizing the size of the support force, the support structure can adapt to the changes in the complex stratum to the greatest extent, so that the support resources can be reasonably allocated, the support cost during the construction process can be reduced, and the construction safety can be enhanced.

[0108] The influence coefficient of stratum support force is obtained by real-time collection of stratum parameters, including soil deformation displacement, stratum settlement, seepage pressure of water-rich stratum and shield machine advancement displacement, taking into account the mutual influence between these parameters. For example, when shield tunneling causes soil deformation displacement to increase, the support structure is difficult to fully support the stratum, thus causing stratum settlement. At the same time, excessive stratum settlement will further increase the stress transfer inside the stratum, leading to the aggravation of soil deformation displacement. When the seepage pressure increases, the deformation displacement of the soil will increase, causing the stability of the stratum to decrease. During the settlement process, the effect of the seepage pressure will aggravate the deformation between soil particles and accelerate the settlement trend. As the advancement displacement increases, the stratum settlement, seepage pressure and soil deformation displacement will increase. The faster the shield machine advances, the more violent the disturbance of the excavation surface, making it difficult to stabilize the internal structure of the soil, thereby increasing the deformation and displacement of the soil. A higher advancement speed will increase the disturbance of the soil, causing the internal stress of the stratum to be released rapidly, triggering the settlement of the upper stratum or the surface. When the advancement speed is faster, the frequency of stratum disturbance increases, and the flow of water pressure in the pores will intensify the fluctuation of the seepage pressure. The greater the thrust of the shield machine, the stronger the pressure on the excavation surface soil, which increases the settlement trend of the soil in the vertical direction. The increase in thrust will increase the stress of the soil and increase the deformation and displacement of the stratum.

[0109] It should also be noted that the soil deformation displacement influence weight represents the numerical value of the influence degree of soil deformation displacement on the influence coefficient of stratum support force. When used, the influence weight corresponding to the soil deformation displacement can be directly obtained from the database, and the corresponding relationship can be a pre-set mapping relationship. For example, the soil deformation displacement and the soil deformation displacement influence weight form a mapping set, and the real-time soil deformation displacement is input into the mapping set to obtain the influence weight corresponding to the soil deformation displacement, wherein the mapping relationship can be a one-to-one correspondence or a many-to-one relationship. The stratum settlement influence weight represents the numerical value of the influence degree of stratum settlement on the influence coefficient of stratum support force. When used, the influence weight corresponding to the stratum settlement can be directly obtained from the database, and the corresponding relationship can be a pre-set mapping relationship. For example, the stratum settlement and the stratum settlement influence weight form a mapping set, and the real-time stratum settlement is input into the mapping set to obtain the influence weight corresponding to the stratum settlement, wherein the mapping relationship can be a one-to-one correspondence or a many-to-one relationship. The permeability pressure influence weight of the water-rich stratum indicates the numerical value of the influence degree of the permeability pressure of the water-rich stratum on the influence coefficient of the stratum support force. When used, the influence weight corresponding to the permeability pressure of the water-rich stratum can be directly obtained from the database, and the corresponding relationship can be a pre-set mapping relationship. For example, the permeability pressure of the water-rich stratum and the permeability pressure influence weight of the water-rich stratum form a mapping set, and the real-time permeability pressure of the water-rich stratum is input into the mapping set to obtain the influence weight corresponding to the permeability pressure of the water-rich stratum, wherein the mapping relationship can be a one-to-one correspondence or a many-to-one relationship. The shield machine propulsion speed influence weight indicates the numerical value of the influence degree of the shield machine propulsion speed on the influence coefficient of the stratum support force. When used, the influence weight corresponding to the shield machine propulsion speed can be directly obtained from the database, and the corresponding relationship can be a pre-set mapping relationship. For example, the shield machine propulsion speed and the shield machine propulsion speed influence weight form a mapping set, and the real-time shield machine propulsion speed is input into the mapping set to obtain the influence weight corresponding to the shield machine propulsion speed, wherein the mapping relationship can be a one-to-one correspondence or a many-to-one relationship. The shield machine thrust influence weight represents the numerical value of the influence degree of the shield machine thrust on the stratum support force influence coefficient. When used, the influence weight corresponding to the shield machine thrust can be directly obtained from the database, and the corresponding relationship can be a pre-set mapping relationship. For example, the shield machine thrust and the shield machine thrust influence weight form a mapping set, and the real-time shield machine thrust is input into the mapping set to obtain the influence weight corresponding to the shield machine thrust. The mapping relationship can be a one-to-one correspondence or a many-to-one relationship.

[0110] It should be understood that the "real-time monitoring" function introduced in this step can continuously update the formation state parameters during the excavation process, and adjust the support force distribution according to the changes in the formation state parameters, so as to achieve the effect of maintaining stability in a dynamic environment, thereby making up for the shortcomings of the static support force distribution, and on the basis of the aforementioned neural network prediction, the real-time collection and dynamic feedback of the formation parameters are added, and the real-time data monitored is linked to the support force distribution, so that the support force distribution can adapt to the changes in the formation, thereby improving the real-time and accuracy of the support. The mechanism of feeding back the formation monitoring data to the excavation face support structure model, so that the support force can be adjusted according to the real-time state of the formation, effectively fills the limitations of static design, and enables the support structure model to have dynamic adjustment capabilities and better adapt to the dynamic changes of the formation.

[0111] In order to facilitate understanding of the above technical solution of the present invention, the method for obtaining the stability control value of the excavation face in the actual process of the present invention is described in detail below.

[0112] Step 1: Use the generalized Pareto distribution technology to select the optimal support force distribution form;

[0113] (1) Analysis of ultimate support force and excavation face stability:

[0114] Assume that the maximum deformation and settlement of soil under different support force distributions are calculated by discrete element method, which are U max =20cm and S max =15cm.

[0115] Hill statistic calculation: Arrange the soil deformation values ​​from large to small, assuming that the minimum deformation value U min =5cm, the Hill statistic is calculated by the Hill plot method, assuming that the calculated Hill statistic is H = 0.1, and the preset Hill statistic is H pre =0.12, the error is |HH pre ∣=0.02, meeting the error requirement.

[0116] By fitting the soil deformation and settlement values, the generalized Pareto distribution position parameter ξ=10cm, shape parameter γ=0.2, and the collapse probability P collapse =0.01, the minimum value of the inspection index corresponding to the optimal support force distribution is P min =0.005.

[0117] (2) Optimal support force distribution form:

[0118] Through the generalized Pareto distribution and collapse probability analysis, the optimal support force distribution form finally selected is linear distribution, and the front support force F min =2.0×106N, rear end support force F max=6.5×106N, the distribution function is: F=F min +L(F max -F min )*x.

[0119] Step 2: Prediction of mechanical properties of shield machine;

[0120] (1) Shield machine structural parameter collection:

[0121] Assume that the diameter of the shield machine cutter head is D = 6.5m and the thrust range is F push =15000~50000kN, torque range T=500~5000kN / cdotpm, propulsion speed v advance =40~80mm / min.

[0122] (2) Numerical simulation analysis of shield machine resistance:

[0123] The water-rich formation information includes shear strength c = 25 kPa, internal friction angle φ = 30°, maximum principal stress σ1 = 150 kPa, and minimum principal stress σ3 = 80 kPa.

[0124] A two-dimensional numerical model was established based on particle flow technology to set the force range A when the cutterhead contacts the water-rich formation. contact =20m 2 , assuming that according to the numerical simulation results, the maximum resistance of the shield machine during the advancement process is R max =25000kN, average resistance is R avg =18000kN.

[0125] (3) Mechanical properties prediction model:

[0126] Based on the resistance results of numerical tests and the structural parameters of the shield machine, a bidirectional long short-term memory neural network (BiLSTM) was used to predict the mechanical properties of the shield machine. The maximum number of iterations was set to 1000 and the hyperparameter learning rate was 0.001. It was assumed that the mechanical performance value of the shield machine predicted by the model was P performance =0.95, close to the standard value P std =1.0.

[0127] Step 3: Generate a stable control value for the excavation face;

[0128] (1) Determination of support force control value:

[0129] According to the optimal support distribution form, the support force control value of the shield machine at any excavation process point is calculated. Assuming that d = 3m at a certain process point, the support force at this point is F(d = 3) = 2.0 × 106 + 5(6.5 × 106 - 2.0 × 106) / 5*3 = 4.9 × 10 6 N.

[0130] (2) Determination of shield machine control value:

[0131] Based on the thrust value, torque and propulsion speed of the shield machine, the control value of the shield machine is determined. Assuming that the thrust F push =20000kN, torque T = 3000kN / cdotpm, propulsion speed v advance =60mm / min, by comparing the predicted values ​​of resistance and mechanical properties, the rationality of shield machine control is verified, assuming that the ratio of thrust and resistance is R / F push =0.9, which meets the standard requirements.

[0132] (3) Stability control value of excavation face:

[0133] The support force control value F support =4.9×10 6 After N is combined with the shield machine control value, the excavation face stability control value C is generated. stability , assuming that the calculated stable control value is C stability =0.92, meeting the safety standard.

[0134] Therefore, the optimal support force distribution form can be selected through the generalized Pareto distribution technology, and combined with the structural parameters and mechanical performance prediction of the shield machine, a reasonable excavation face stability control value can be generated during the shield machine excavation process, which helps to improve the safety of the excavation face and the working efficiency of the shield machine.

[0135] Step S4, adjusting the tunneling parameters of the shield machine according to the excavation face stability control value, applying the adjusted shield machine to real-time tunnel excavation work, judging the stress state of the excavation face after the excavation work is completed, and verifying the control ability of the excavation face stability control value based on the stress state.

[0136] In this embodiment, according to the calculated excavation face stability control value C stability If the value is close to or lower than the preset safety standard (such as C stability ≥0.9), the tunneling parameters of the shield machine need to be adjusted, mainly involving the following parameters:

[0137] Thrust: Increase or decrease the thrust of the shield machine. The adjustment range is usually 10,000 to 50,000 kN.

[0138] Cutter torque: Adjust the torque according to the resistance and formation strength. The common range is 500-5000 kN / cdotpm.

[0139] Advance speed: Adjust the advance speed to control the friction between the cutter head and the formation and the cutting effect, usually in the range of 40 to 80 mm / min.

[0140] Calculates thrust, torque and propulsion speed adjustments based on stable control values. For example:

[0141] Assume that the current excavation face stability control C stability , lower than standard C std =0.9, it is necessary to increase the thrust and cutter torque.

[0142] Assuming the current thrust is 20000 kN, it can be increased by 10% according to the adjustment principle, that is, the adjusted thrust is 20000×1.1=22000 kN;

[0143] At the same time, the cutter disc torque is adjusted to 3000 kN\cdotpm×1.1=3300 kN\cdotpm.

[0144] The calculated optimized thrust, torque and propulsion speed are input into the shield machine control system, and the operating parameters of the shield machine are adjusted to match the requirements of stable control values.

[0145] During the excavation process, the thrust, torque, propulsion speed and cutterhead speed are monitored in real time through the sensors and monitoring system of the shield machine to ensure that the adjusted parameters are effectively implemented in actual operations.

[0146] At the same time, the resistance of the cutter head in contact with the formation is monitored and the real-time resistance R is recorded. real And calculate the resistance R calc Comparison is made to ensure that the changes in resistance during excavation are within a controllable range. If abnormal changes in excavation surface resistance, settlement, etc. are monitored, the thrust, torque and other parameters of the shield machine are dynamically adjusted according to real-time data. For example, if the resistance exceeds the expected value, it may be necessary to further reduce the propulsion speed or increase the thrust.

[0147] After excavation is completed, the in-situ stress testing method is used to obtain the stress state around the excavation face after excavation is completed. The main stresses include the maximum principal stress σ1, which is usually the vertical stress; the minimum principal stress σ3, which is usually the horizontal stress. At the same time, settlement observation points are installed in the excavation face area to monitor the settlement and soil displacement in real time to ensure that the soil deformation meets the design requirements.

[0148] The stress state measured after excavation is completed is compared with the excavation face stability control value C generated during excavation. stability A comparison is made to verify the accuracy and control capability of the control value, and the difference between the actual stress state and the stable control value is calculated. For example, assuming that the measured maximum principal stress σ1 = 140 kPa and the minimum principal stress σ3 = 75 kPa are within 5% of the predicted stress, the control value is considered accurate.

[0149] Based on the analysis of settlement and stress state, the effectiveness of the excavation face stability control value is judged: if the soil settlement and principal stress are within the safe range, it means that the stability control value can effectively control the stability of the excavation face. If the control value deviates greatly, it may be necessary to adjust the parameters of the control model or correct the boundary conditions used in the numerical simulation.

[0150] Therefore, by adjusting the excavation parameters of the shield machine according to the excavation face stability control value during tunnel excavation, and verifying the effectiveness of the control value by monitoring the stress state after the excavation is completed, the excavation process of the shield machine is ensured to be safe and efficient.

[0151] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for controlling the stability of the excavation face of a cable shield tunnel in a water-rich stratum, characterized in that: The method includes: The virtual experiment technology is used to calibrate the simulation parameters of water-rich strata, and the simulation parameters of water-rich strata are combined with the shield machine to establish the tunnel excavation face model; The tunnel excavation face model and discrete element method are used to simulate the deformation and settlement of the soil in the excavation face under different support force distribution conditions, and the ultimate support force of the support structure and the stability of the excavation face are analyzed based on the deformation and settlement of the soil. The optimal support force distribution form is selected based on the ultimate support force and the stability of the excavation face. The mechanical properties of the shield machine are predicted according to the strength of the water-rich stratum, the principal stress of the water-rich stratum and the structural parameters of the shield machine. The stability control value of the excavation face is generated in combination with the optimal support force distribution form. The excavation parameters of the shield machine are adjusted according to the stability control value of the excavation face, and the adjusted shield machine is applied to real-time tunnel excavation work to judge the stress state of the excavation face after the excavation work is completed. The control ability of the excavation face stability control value is verified based on the stress state to achieve stable control of the shield tunnel excavation face.

2. The method for controlling the excavation face stability of a cable shield tunnel in a water-rich stratum according to claim 1 is characterized in that: The tunnel excavation face model and discrete element method are used to simulate the soil deformation and settlement law of the excavation face stratum under different support force distribution conditions of the support structure, and the ultimate support force of the support structure and the stability of the excavation face are analyzed according to the soil deformation and settlement law, including: A three-dimensional model of the support structure is established by using three-dimensional modeling technology, and a support stiffness coefficient and a bearing capacity coefficient are defined for the three-dimensional model of the support structure based on the support structure definition criteria; The three-dimensional model of the support structure is integrated with the tunnel excavation face model to generate a discrete element simulation model, and the distribution state set of the three-dimensional model of the support structure in the tunnel excavation face model is defined in a linear distribution manner using the discrete element method; Define the support force corresponding to the three-dimensional model of the support structure in each distribution form under the distribution state set, and use the discrete element method to simulate the excavation process after setting the initial stress state and boundary conditions; The discrete element method is used to obtain the particle displacement of the water-rich stratum after excavation is completed, and the stratum movement and deformation of the excavation surface stratum are determined based on the particle displacement; Based on the movement and deformation of the stratum, the deformation and settlement laws of the soil corresponding to the excavation surface stratum are analyzed, and the ultimate support force of the support structure and the stability of the excavation surface are determined in combination with curve analysis technology.

3. The method for controlling the excavation face stability of a cable shield tunnel in a water-rich stratum according to claim 2 is characterized in that: The analysis of soil deformation and settlement law corresponding to the excavation surface stratum based on stratum movement and deformation conditions, and the determination of the ultimate support force of the support structure and the stability of the excavation surface in combination with curve analysis technology include: The characteristic information of stratum movement and deformation data is read by using a heuristic search algorithm, the characteristic information is normalized and then the characteristic set is initialized, and several clusters are generated in the solution space; Based on the cluster point traversal feature information, the overall probability and average probability of feature distribution are determined, and the fitness value of the cluster point is calculated based on the overall probability and average probability of distribution, and the strength and amplitude of feature changes are analyzed using the fitness value; The characteristic information corresponding to the maximum value of intensity and amplitude is selected as the variation characteristic quantity, and the correlation between the variation characteristic quantity and soil deformation and soil settlement is calculated by using grey correlation analysis. The soil deformation and settlement law curve corresponding to the excavation surface stratum is analyzed based on the correlation degree. The curve analysis technology is used to analyze the deformation and settlement law curve of the soil to obtain the movement trajectory points and deformation modes of the water-rich stratum particles, and the difference between the movement trajectory points and the standard movement trajectory points is calculated; The ratio of the deformation mode to the standard deformation mode is calculated, and the ultimate support force of the support structure and the stability of the excavation surface are judged based on the difference results and the ratio results. The larger the difference and the ratio, the worse the stability.

4. The method for controlling the excavation face stability of a cable shield tunnel in a water-rich stratum according to claim 3 is characterized in that: The calculation formula of the intensity of the characteristic change is: The calculation formula for the amplitude of the characteristic change is: Where P i Indicates the intensity of feature change, Q i Indicates the magnitude of feature change, M min With M max They represent the maximum and minimum values ​​of the fitness value of the clustering point, β represents the correction value, T represents the number of clustering points, and f i (X i ) represents the fitness value of the i-th cluster point.

5. The method for controlling the excavation face stability of a cable shield tunnel in a water-rich stratum according to claim 1, characterized in that: The optimal support force distribution form is selected based on the ultimate support force and stability, the mechanical properties of the shield machine are predicted according to the strength of the water-rich stratum, the principal stress of the water-rich stratum and the structural parameters of the shield machine, and the stability control value of the excavation face is generated in combination with the optimal support force distribution form, including: The generalized Pareto distribution is used to fit the ultimate support force of the support structure and the stability of the excavation surface, and the test index is generated. The optimal support force distribution form is selected based on the test index. Obtain water-rich stratum information including shear strength and principal stress, and collect shield machine structural parameters including cutter head diameter, thrust range, torque range and propulsion speed based on shield machine production information; Based on the water-rich stratum information and the numerical application of shield machine structural parameters, the resistance of the shield machine during the excavation process is simulated and analyzed, and the mechanical performance of the shield machine is predicted based on the resistance results; The support force control value of the shield machine at any excavation process point is determined according to the optimal support distribution form, and the shield machine control value is generated based on the mechanical performance of the shield machine. The support force control value and the shield machine control value are combined as the excavation face stability control value.

6. The method for controlling the excavation face stability of a cable shield tunnel in a water-rich stratum according to claim 5, characterized in that: The application of generalized Pareto distribution to fit the ultimate support force of the support structure and the stability of the excavation surface to generate a test index, and selecting the optimal support force distribution form based on the test index includes: The ultimate support force of the support structure and the soil deformation and settlement values ​​corresponding to the stability of the excavation surface are arranged in descending order, and the Hill plot method is used to determine the Hill statistics of the minimum soil deformation and settlement values; Determine the error between the Hill statistic and the preset Hill statistic, and use the Hill statistic that meets the error requirement as a water-rich formation location parameter; Based on the location parameters of water-rich strata and generalized Pareto distribution, a support probability function is established, and the probability of collapse under different support force distribution forms is analyzed using the support probability function. The probability value is used as a test index to measure the ultimate support force of the support structure and the stability of the excavation surface, and the support force distribution form corresponding to the minimum test index is defined as the optimal selection result.

7. A method for controlling the stability of excavation face of cable shield tunneling in water-rich strata according to claim 6, characterized in that: The expression of the support probability function is: In the formula, g t represents the probability of collapse under the tth support force distribution form, represents the soil deformation value under the t-th support force distribution form, τ t represents the soil settlement value under the t-th support force distribution form, γ represents the shape parameter, δ t represents the support force corresponding to the tth support force distribution form, and ξ represents the location parameter of the water-rich stratum.

8. The method for controlling the excavation face stability of a cable shield tunnel in a water-rich stratum according to claim 5, characterized in that: The method of simulating and analyzing the resistance of the shield machine during the excavation process based on the water-rich stratum information and the application value of the shield machine structural parameters, and predicting the mechanical properties of the shield machine according to the resistance results includes: Based on the structural information of the shield machine, the force range when the cutterhead contacts the water-rich stratum within the thrust range is established, the corresponding torque value is analyzed according to the force range, and the shield machine model is simplified using the torque value and the force range using a medium simulation program. The water-rich stratum information is used as the macroscopic parameter to characterize the water-rich stratum, and a two-dimensional numerical model is established in combination with the particle flow technology. The two-dimensional numerical model and a simplified shield machine are used to complete numerical tests at different excavation intervals. The resistance of the shield machine during tunneling is determined based on the numerical test results, and the correlation between the macro parameters and the structural information of the shield machine is analyzed based on the resistance results; The correlations are ranked by importance, and a prediction model is constructed based on the importance ranking results and a bidirectional long short-term memory neural network to predict the mechanical performance of the shield machine.

9. The method for controlling the excavation face stability of a cable shield tunnel in a water-rich stratum according to claim 8, characterized in that: The relevance is sorted by importance, and a prediction model is constructed based on the importance sorting result and a bidirectional long short-term memory neural network to predict the mechanical properties of the shield machine, including: After ranking the relevance in terms of importance, the macro parameters and shield machine structure information are preprocessed and used as input data; Set the maximum number of iterations and the range of hyperparameter values ​​of the bidirectional long short-term memory neural network, establish a prediction model based on the setting results, and then use the adaptive optimization algorithm to calculate the fitness of the neural network nodes; Reconstruct the linear reduction factor of the neural network based on the node fitness value results, introduce Gaussian mixture mutation values ​​according to the linear reduction factor to update the node position, and calculate the fitness of the optimal node; Determine whether the maximum number of iterations has been reached. If so, assign the maximum value of the hyperparameter to the bidirectional long short-term memory neural network to obtain the final prediction model. If not, reconstruct the linear reduction factor. Input data into the final prediction model, output resistance results, and compare the resistance results with the preset standard resistance value. The comparison value is defined as the mechanical performance value of the shield machine.

10. The method for controlling the stability of excavation face of cable shield tunneling in water-rich strata according to claim 5, characterized in that: The method of determining the support force control value of the shield machine at any excavation process point according to the optimal support distribution form, generating the shield machine control value based on the mechanical performance of the shield machine, and combining the support force control value with the shield machine control value as the control value for achieving excavation face stability includes: The distribution form of the support structure is set within the excavation range of the shield machine based on the optimal distribution form, and the distribution formula of the support force is established according to the distribution form to determine the support force control value of the shield machine; The thrust value of the shield machine is determined based on the excavation surface area and normal stress during the tunneling process of the shield machine, and the torque force and propulsion speed of the shield machine are analyzed based on the resistance value, the strength of the water-rich stratum and the main stress; The thrust value, torque force and propulsion speed are compared with the mechanical performance value of the shield machine to verify the rationality of the shield machine settings. After the verification is completed, they are combined with the support force control value as the control value to achieve the stability of the excavation face.

11. The method for controlling the excavation face stability of a cable shield tunnel in a water-rich stratum according to claim 1, characterized in that: The method also includes monitoring the formation changes in real time, analyzing the real-time ultimate support force, and dynamically adjusting the distribution of the support force to obtain the optimal support force distribution form under the current formation state, including: Using sensors to collect formation parameters in real time, the formation parameters include: soil deformation displacement, formation settlement, seepage pressure of water-rich formations, shield machine propulsion speed and shield machine thrust; The influence coefficient of stratum support force is obtained by analyzing the stratum parameters; Obtain the formation support force correction value corresponding to each influence coefficient interval preset in the database, and match the formation support force influence coefficient with each influence coefficient interval. If the formation support force influence coefficient belongs to a certain preset formation support force influence coefficient interval, obtain the formation support force correction value corresponding to the influence coefficient interval as the formation support force threshold adjustment factor; The front support force, rear support force and shield machine displacement are obtained, and the support force distribution is obtained by analysis based on the stratum support force threshold adjustment factor.

Images