Quality assessment method for construction layout of prestressed anchor cables

Abstract

The present invention provides a method for evaluating the quality of prestressed anchor cable construction layout, which relates to the field of underground engineering and includes the following steps: S1, model construction: establishing a three-dimensional numerical model of different anchor arrangement schemes; S2, simulation calculation and result evaluation: performing excavation simulation calculations based on the three-dimensional numerical models of different anchor arrangement schemes, obtaining numerical simulation calculation results of surrounding rock deformation, stress distribution, damage and rupture zone distribution, and support structure stress; evaluating the numerical simulation calculation results of different anchor arrangement schemes, obtaining evaluation indicators, and determining the optimal scheme based on the evaluation indicators. The coordination between the above structures has the following beneficial effects: First, it can comprehensively improve the evaluation accuracy, by constructing a comprehensive evaluation system and applying data analysis models and algorithms to convert the data into intuitive and quantitative evaluation results; second, by establishing an error correction model, it can accurately calibrate the deviation between numerical simulation and on-site monitoring data.

Description

Technical Field

[0001] The present invention relates to the field of underground engineering, and in particular to a method for evaluating the construction layout quality of prestressed anchor cables. Background Art

[0002] my country will usher in a new round of engineering construction in many fields such as new energy, hydropower, urban construction, national defense and security. The development and utilization of underground space resources such as underground caverns in hydropower stations, underground carbon dioxide storage, underground nuclear waste storage, national defense hangars, and underground oil depots will become more and more common.

[0003] According to different construction needs, the geological environment in which underground caverns are located is complex, and they are gradually developing towards deep burial, large span, large scale, and high safety construction requirements. During the construction period and long-term operation of underground caverns, ensuring the dual safety of surrounding rock stability and support structure stress is one of the issues that must be focused on in the safety management of engineering buildings throughout their life cycle.

[0004] Currently, in the design of prestressed anchor cable support for the remaining excavation layer of the main powerhouse of Shuangjiangkou Hydropower Station, there is a need to optimize the spacing of prestressed anchor cables. However, there is currently a lack of an effective method to determine the optimal anchor cable arrangement scheme. Therefore, the key technical problem lies in how to effectively evaluate and determine the optimal spacing of prestressed anchor cables to ensure the support effect while reducing surrounding rock deformation and stress concentration, thereby improving construction efficiency and economic benefits. Summary of the Invention

[0005] In view of the deficiencies in the above-mentioned prior art, the technical problem to be solved by the present invention is to provide a method for evaluating the construction layout quality of prestressed anchor cables, which can effectively evaluate and determine the optimal spacing of prestressed anchor cables to ensure the support effect while reducing surrounding rock deformation and stress concentration.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is as follows: the present invention provides a method for evaluating the construction layout quality of prestressed anchor cables, comprising the following steps:

[0007] S1. Basic data preparation: Acquire and preprocess geological survey data of underground caverns, and establish a three-dimensional numerical model based on the preprocessed geological survey data;

[0008] S2. Model basic settings: Determine the surrounding rock types in different areas based on geological survey data, determine rock mass parameters based on the surrounding rock types, and select constitutive models based on the surrounding rock parameters, including elastic-plastic models and stress-crack strain evolution constitutive models;

[0009] S3. Model property research: Obtain the mechanical property parameters of the rock mass through specific test methods, establish a stress-crack strain evolution constitutive model, analyze the crack evolution law, determine the criteria for the surrounding rock relaxation zone, and conduct secondary development of the constitutive model;

[0010] S4. Construction simulation: simulate excavation sequence design based on the graded excavation scheme, analyze anchor support strength, determine support parameters and evaluate;

[0011] S5. Geostress field analysis: Using the initial geostress multivariate regression method and the geostress field inversion method of the generative adversarial network (GAN), the surrounding rock stress state before construction excavation is calculated and the mechanical parameters are determined;

[0012] S6. Scheme optimization decision: Through numerical simulation analysis of surrounding rock deformation, stress distribution and plastic failure zone distribution characteristics under different anchor cable layout schemes, the anchor cable layout scheme is optimized.

[0013] In a preferred solution, in step S1, the geological survey data includes physical and mechanical parameters of the rock mass, ground stress measurement data, and structural surface characteristics;

[0014] Among them, the physical and mechanical parameters of rock mass include natural density, deformation modulus, Poisson's ratio, and shear strength;

[0015] The preprocessing operation is to organize and screen the collected geological survey data, remove abnormal data and normalize them for model input and calculation;

[0016] A three-dimensional numerical model of the main chambers including the installation room, main and auxiliary powerhouse caverns, main transformer room, and tailwater surge tank was established based on the pre-processed geological survey data.

[0017] The three-dimensional numerical model uses three-dimensional 8-node hexahedron isoparametric solid elements and their degenerate elements to simulate rock mass and faults.

[0018] The overall model has 91,479 solid element nodes, 181,231 solid elements, 51,803 anchor element nodes, and 26,089 anchor elements.

[0019] In a preferred solution, in step S2, the physical and mechanical parameters of the rock mass are calculated according to empirical formulas to obtain rock mass parameters, and the calculated rock mass parameters are corrected in combination with engineering experience and on-site monitoring data;

[0020] Among them, the elastic-plastic model determines the cracking conditions of rock materials through macroscopic strength and determines that the rock mass enters the plastic state based on the Druker-Prager criterion;

[0021] When applying the constitutive model to the three-dimensional numerical model, the deformation, stress distribution and failure characteristics of the surrounding rock are analyzed by setting the parameters of the constitutive model and the boundary conditions of the three-dimensional numerical model and performing numerical simulation calculations using specific methods;

[0022] Among them, the parameter settings of the constitutive model include the yield stress and hardening parameters of the elastic-plastic model and the crack closure and expansion parameters of the stress-crack-strain evolution constitutive model.

[0023] In a preferred solution, in step S3, the mechanical characteristic parameters of the rock mass include peak stress, elastic modulus, and Poisson's ratio, and a triaxial compression test is selected as a specific test method. Through rock sample preparation and test loading, the stress-strain curve, threshold stress, elastic modulus and Poisson's ratio, stress-crack strain evolution constitutive model, crack evolution law, and surrounding rock relaxation zone criterion are analyzed, and the constitutive model code is compiled and verified in FLAC3D. The specific steps are as follows:

[0024] Collect rock samples and use a rock triaxial testing machine;

[0025] And let the peak stress be σ p , confining pressure σ3, the value range of confining pressure σ3 is: σ 3 ∈[1,40]MPa, and,σ 3 ∈1, 3, 5, 10, 20, 30, 40 MPa, the loading rate is 0.05 MPa / s for the confining pressure σ3, and the axial load increases at 0.5 MPa / s until the rock sample fails;

[0026] Peak stress σ p As the confining pressure σ3 increases, the peak stress σ p The relationship between it and the confining pressure σ3 is expressed by the following formula:

[0027] σ p =a+bσ3;

[0028] Assume that the threshold stresses are the crack closure stress σ cc , crack initiation stress σ ci , crack damage stress σ cd ;

[0029] The relationship between the ratio of threshold stress to peak strength Rcc d, Rci d, and Rcd d and the confining pressure can be fitted by the following formula:

[0030] R=c+dσ3;

[0031] Among them, a, b, c, and d are fitting parameters;

[0032] In the preferred solution, by fitting the rock sample test data under multiple confining pressures, the corresponding peak stress σ under different confining pressures σ3 is obtained. p They are 114.72MPa, 139.65MPa, 157.22MPa, 182.97MPa, 216.30MPa, 345.73MPa, and 374.63MPa respectively;

[0033] The crack closure stress σ under different confining pressure σ3 cc , crack initiation stress σ ci , crack damage stress σ cd The ratio of the peak intensity is different, and the values ​​of Rcc d, Rci d, and Rcd d are 0.22σ respectively. p , 0.41σ p and 0.78σ p , and gradually decreases with the increase of confining pressure σ3.

[0034] In a preferred solution, in step S3, a creep test is selected as a specific test method, and a multi-step loading method is adopted. By setting the rock sample parameters and the test method, the creep strain, the constitutive relationship between stress and instantaneous strain, the creep strain, the long-term strength and the creep strain rate are analyzed, and the Nishihara creep model is improved to verify the correctness of the initial parameters. The specific steps are as follows:

[0035] Assume that the confining pressure σ 3 The values ​​are ∈1, 3, 5, 10 MPa, and the axial stress is 0.55-1.00 times the peak strength;

[0036] The axial creep strain increases with time and is more obvious at high deviatoric stress levels. The total strain consists of instantaneous strain and creep strain. The creep strain includes viscoelastic creep strain and viscoplastic creep strain.

[0037] By analyzing the creep strain data under different confining pressures and deviatoric stresses, the instantaneous strain ε is obtained. m and the deviatoric stress-peak strength ratio R d The relationship can be expressed as:

[0038] ε m =k1×R d +b1;

[0039] Viscoelastic creep strain ε ve With R d The relationship can be expressed as:

[0040]

[0041] Viscoplastic creep strain ε vp The relationship with the confining pressure σ3 can be expressed as:

[0042] ε vp =k2×σ3+b2;

[0043] Among them, k1, b1, k2, and b2 are variables.

[0044] In a preferred solution, in step S4, the hierarchical excavation plan includes excavation sequence, anchor support strength analysis, support parameters and evaluation;

[0045] The excavation sequence was simulated according to the actual construction process, the anchor support strength analysis was based on indoor test results and field monitoring data, and the support parameters and evaluation were combined with theoretical analysis and numerical simulation results.

[0046] The support strength is determined by calculating the cohesion increment of the surrounding rock of the prestressed anchor cable support and comparing it with the value calculated by the empirical formula.

[0047] In a preferred solution, in step S6, the following steps are further included:

[0048] S61. Deformation characteristics of surrounding rock: Obtain the deformation of each part of the tunnel during each level of excavation, and analyze the changing trend of deformation at each level of excavation and the deformation differences at different parts;

[0049] S62. Stress distribution characteristics: Determine the stress concentration areas of the main caverns after each level of excavation, and compare the stress extreme value changes in the stress concentration areas of different schemes;

[0050] S63. Distribution characteristics of plastic failure zone: Obtain the depth of the plastic failure zone of the surrounding rock of each unit;

[0051] S64. Determine the three-dimensional numerical models under different anchor cable arrangement schemes according to the surrounding rock deformation characteristics, stress distribution characteristics, and plastic failure zone distribution characteristics, and perform excavation simulation calculations based on the three-dimensional numerical models under different anchor cable arrangement schemes to obtain the surrounding rock deformation results, stress distribution results, and plastic failure zone results.

[0052] In a preferred solution, in step S64, the surrounding rock stability under different anchor cable arrangement schemes is determined based on the surrounding rock deformation results, stress distribution results and plastic failure zone results, and the optimal solution is determined based on the surrounding rock stability under different anchor cable arrangement schemes.

[0053] In the preferred solution, the deformation, stress distribution and damage zone conditions of on-site monitoring are obtained;

[0054] The surrounding rock deformation results, stress distribution results and plastic failure zone results are compared with the deformation, stress distribution and failure zone conditions monitored on site, and statistical analysis methods are used for quantitative evaluation to obtain the comparison results;

[0055] Wherein, the statistical analysis method is one of the root mean square error or correlation coefficient;

[0056] An error correction model of the evaluation method is established based on the comparison results, which is used to pre-adjust the parameters under different anchor cable layout schemes.

[0057] The present invention provides a method for evaluating the construction layout quality of prestressed anchor cables. Through the coordination between the above structures, the following beneficial effects are achieved:

[0058] First, it can accurately simulate and analyze the construction process of underground caverns. By acquiring detailed geological survey data and establishing precise three-dimensional numerical models, combined with a variety of advanced constitutive models and experimental research results, it can accurately reflect the mechanical properties and geological conditions of the rock mass, provide a scientific basis for the formulation of construction plans, and effectively reduce uncertainties and risks during the construction process.

[0059] Second, optimize the anchor cable layout scheme and improve the support effect. By systematically analyzing the deformation, stress distribution and plastic failure zone characteristics of the surrounding rock under different anchor cable layout schemes, the optimal anchor cable layout parameters can be determined, so that the anchor cables can fully play their role in reinforcing the surrounding rock, effectively control the deformation of the surrounding rock, reduce the degree of stress concentration, and reduce the scope of the plastic failure zone, thereby enhancing the stability and safety of the underground cavern group and extending its service life.

[0060] Third, to improve the accuracy and reliability of engineering quality assessment, a variety of methods and data are used in combination throughout the assessment process, including field monitoring data, indoor test data, and numerical simulation results, which verify and supplement each other. The accuracy of the assessment results is ensured through strict logical steps and scientific calculation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The present invention will be further described below with reference to the accompanying drawings and examples:

[0062] Figure 1 It is the calculation range diagram of the underground powerhouse in the present invention;

[0063] Figure 2 This is a typical geological cross-section of the underground powerhouse of the present invention;

[0064] Figure 3 It is the overall three-dimensional finite element simulation model of the present invention;

[0065] Figure 4 This is a time-dependent variation trend diagram of C3 of the present invention;

[0066] Figure 5 This is a time-dependent displacement cloud diagram of the surrounding rock displacement at the 1# unit in the first-level excavation under the support working condition of the present invention;

[0067] Figure 6This is the time-dependent displacement cloud diagram of the surrounding rock displacement at the 4# unit in the first-level excavation under the support working condition of the present invention;

[0068] Figure 7 This is the contour map of the first-level excavation displacement Uz of the installation room of the present invention;

[0069] Figure 8 It is the contour map of the first-level excavation displacement Uy of the installation room of the present invention. DETAILED DESCRIPTION

[0070] In order to better understand the purpose, structure and function of the present invention, the embodiments and features in the embodiments of the present invention can be combined with each other without conflict. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0071] Example 1

[0072] like Figure 1 As shown in Figure 1-X, this embodiment provides a method for evaluating the construction layout quality of prestressed anchor cables, including the following steps:

[0073] S1. Basic data preparation: Acquire and preprocess geological survey data of underground caverns, and establish a three-dimensional numerical model based on the preprocessed geological survey data;

[0074] In a preferred solution, in step S1, the geological survey data includes physical and mechanical parameters of the rock mass, ground stress measurement data, and structural surface characteristics;

[0075] Among them, the physical and mechanical parameters of rock mass include natural density, deformation modulus, Poisson's ratio, and shear strength;

[0076] Specifically, the physical and mechanical parameters of the rock mass are determined by using a variety of geological survey methods, such as on-site sampling and testing, geophysical exploration, etc., to obtain key physical and mechanical parameters such as the natural density, deformation modulus, Poisson's ratio, and shear strength of the rock mass. These parameters reflect the basic mechanical properties of the rock mass. For example, in the granite rock mass of the Shuangjiangkou Hydropower Station, the natural density of the Class II surrounding rock mass is 2.6g / cm 3 , deformation modulus 32.5GPa, Poisson's ratio 0.25, etc. These data provide the basis for subsequent analysis;

[0077] Geostress measurement data uses specialized measurement methods such as stress relief and hydraulic fracturing to determine the distribution of geostress in the area where the underground caverns are located, including its magnitude, direction, and distribution pattern. Geostress is a significant factor affecting rock mass stability, and its measurement results are crucial for accurately simulating the mechanical response of rock masses during excavation.

[0078] Detailed investigation of the location, occurrence, spacing, and filling properties of structural surfaces such as faults, joints, and fissures within the rock mass. The presence of these structural surfaces can significantly weaken the strength and integrity of the rock mass. For example, in some areas, the presence of faults can cause sudden changes in the mechanical properties of the rock mass, so accurate recording of these relevant information is required during data acquisition.

[0079] The preprocessing operation is to organize and screen the collected geological survey data, remove abnormal data and normalize them for model input and calculation;

[0080] Specifically, the large amount of geological survey data collected is systematically organized, and based on professional geological knowledge and statistical analysis methods, obvious abnormal data are identified and removed. Furthermore, if a set of rock mass strength data differs greatly from the data in the surrounding area and does not conform to geological laws, it may be an outlier caused by testing errors or special geological conditions and needs to be eliminated to ensure the reliability and accuracy of the data;

[0081] Use an appropriate normalization method, i.e., linear normalization or standardization, to unify geological survey data of different magnitudes and units into a reasonable numerical range to facilitate the input and calculation of subsequent models. Furthermore, for stress data, the formula can be used: Perform normalization processing;

[0082] Among them, X n is the normalized parameter value, X is the original parameter value, and X min and X max are the minimum and maximum values ​​of the parameters, respectively, so that all stress data are analyzed and compared within the range of 0 to 1.

[0083] A three-dimensional numerical model of the main chambers including the installation room, main and auxiliary powerhouse caverns, main transformer room, and tailwater surge tank was established based on the pre-processed geological survey data.

[0084] The three-dimensional numerical model uses three-dimensional 8-node hexahedron isoparametric solid elements and their degenerate elements to simulate rock mass and faults.

[0085] Specifically, if Figure 1 、 2 As shown in the figure, the calculation range of the three-dimensional numerical model is determined according to the actual layout of the underground cavern group and engineering requirements;

[0086] Furthermore, in the x-axis direction, from unit 1 to unit 4, the intercept length is 275.7m (from 0-098.82 on the horizontal axis to 0+176.88 on the horizontal axis);

[0087] The y-axis is horizontally pointing from the main power building to the tail adjustment room, with a cut length of 300.0m;

[0088] The z-axis is vertically upward, and the bottom is taken to The top of the model extends to the surface of the mountain top, ensuring that the model can cover the key geological areas of the main caverns such as the installation room, main and auxiliary powerhouse caverns, main transformer room, tailwater surge tank, and their surrounding areas, and fully reflect the relationship between geological conditions and engineering structures, as shown in Table 1 below:

[0089] Table 1 Typical cross-section locations of Shuangjiangkou underground powerhouse

[0090] Location Pile number X(m) Installation room 0-052.320 Center line of unit 1# 0+000.00 Centerline of Unit 2# 0+030.02 Centerline of Unit 3# 0+060.04 Center line of unit 4# 0+90.06 Auxiliary factory 0+135.00

[0091] A three-dimensional 8-node hexahedron isoparametric solid element and its degenerate element are used to simulate rock masses and faults. This element type can better adapt to complex geological shapes and mechanical behaviors. Through reasonable meshing, a three-dimensional numerical model with 91,479 solid element nodes, 181,231 solid elements, 51,803 anchor element nodes and 26,089 anchor elements is constructed, providing an accurate geometric and physical model foundation for subsequent numerical simulations.

[0092] S2. Model basic settings: Determine the surrounding rock types in different areas based on geological survey data, determine rock mass parameters based on the surrounding rock types, and select constitutive models based on the surrounding rock parameters, including elastic-plastic models and stress-crack strain evolution constitutive models;

[0093] In a preferred solution, in step S2, the physical and mechanical parameters of the rock mass are calculated according to empirical formulas to obtain rock mass parameters, and the calculated rock mass parameters are corrected in combination with engineering experience and on-site monitoring data;

[0094] Specifically, first, based on the type of surrounding rock, relevant empirical formulas are used to calculate the physical and mechanical parameters of the rock mass. For example, for different types of granite, existing empirical formulas can be used to calculate their deformation modulus, shear strength and other parameters. However, the calculation results of these empirical formulas may have certain errors, so they need to be corrected in combination with engineering experience and on-site monitoring data. By analyzing the data on deformation and stress changes of the surrounding rock monitored on-site, and referring to the actual experience of similar projects, the calculated rock mass parameters are reasonably adjusted to make them more in line with the actual engineering situation. For example, if on-site monitoring finds that the deformation of the surrounding rock in a certain area exceeds expectations, it may be necessary to appropriately reduce the calculated rock mass strength parameters to more accurately reflect the actual mechanical properties of the rock mass in that area.

[0095] Among them, the elastoplastic model determines the cracking conditions of rock materials through macroscopic strength and determines the entry of rock into the plastic state based on the Druker-Prager criterion. The elastoplastic model can better describe the plastic deformation and yield behavior of rock masses under stress. The stress-crack strain evolution constitutive model focuses more on considering the influence of the generation, expansion and closure of cracks in the rock mass on the mechanical properties of the rock mass. It is particularly suitable for rock masses with developed cracks under high ground stress conditions.

[0096] When applying the constitutive model to the three-dimensional numerical model, the deformation, stress distribution and failure characteristics of the surrounding rock are analyzed by setting the parameters of the constitutive model and the boundary conditions of the three-dimensional numerical model and performing numerical simulation calculations using specific methods;

[0097] Specifically, if Figure 3 As shown in the figure, when applying the constitutive model to a three-dimensional numerical model, a series of model parameters need to be set, including the yield stress and hardening parameters of the elastic-plastic model, and the crack closure and expansion parameters of the stress-crack-strain evolution constitutive model. The values ​​of these parameters directly affect the accuracy of the model's simulation of the mechanical behavior of the rock mass. Furthermore, for the elastic-plastic model, based on the rock mass strength parameters determined by indoor tests and field monitoring data, combined with the Druker-Prager criterion, reasonable yield stress and hardening parameters are set to accurately simulate the conditions and process of the rock mass entering the plastic state. At the same time, by setting boundary conditions of the three-dimensional numerical model, such as displacement constraints and stress boundary conditions, the boundary stress conditions of the rock mass in actual engineering are simulated. Then, specific numerical calculation methods (such as the finite element method) are used to perform numerical simulation calculations to analyze the deformation, stress distribution, and failure characteristics of the surrounding rock during excavation and support processes, providing a theoretical basis for subsequent engineering construction simulation and scheme optimization.

[0098] Among them, the parameter settings of the constitutive model include the yield stress and hardening parameters of the elastic-plastic model and the crack closure and expansion parameters of the stress-crack-strain evolution constitutive model.

[0099] S3. Model property research: Obtain the mechanical property parameters of the rock mass through specific test methods, establish a stress-crack strain evolution constitutive model, analyze the crack evolution law, determine the criteria for the surrounding rock relaxation zone, and conduct secondary development of the constitutive model;

[0100] In a preferred solution, in step S3, the mechanical characteristic parameters of the rock mass include peak stress, elastic modulus, and Poisson's ratio, and a triaxial compression test is selected as a specific test method. Through rock sample preparation and test loading, the stress-strain curve, threshold stress, elastic modulus and Poisson's ratio, stress-crack strain evolution constitutive model, crack evolution law, and surrounding rock relaxation zone criterion are analyzed, and the constitutive model code is compiled and verified in FLAC3D. The specific steps are as follows:

[0101] Specifically, representative rock samples are collected from the project site and processed into φ50×100mm cylindrical standard specimens according to strict standards to ensure that the diameter, height-to-diameter ratio, end face parallelism and other parameters of the rock sample meet the test requirements. For example, the diameter needs to be controlled between 48-52mm, the height-to-diameter ratio should be controlled between 1.9-2.1, and the error of the parallelism of the two end faces should be less than 5×10 -2 mm to ensure the accuracy and comparability of the test results. The advanced MTS815FlexTestGT rock triaxial testing machine was used for the test. The instrument was precisely calibrated before the test to ensure that the load and deformation measurement accuracy met the test requirements.

[0102] And let the peak stress be σ p , confining pressure σ3, the value range of confining pressure σ3 is: σ 3 ∈[1,40]MPa, and,σ 3 ∈1, 3, 5, 10, 20, 30, 40 MPa, the loading rate is 0.05 MPa / s for the confining pressure σ3, and the axial load increases at 0.5 MPa / s until the rock sample fails;

[0103] During the test, the stress and strain data of the rock sample are collected in real time, including axial stress, radial stress, axial strain, radial strain, etc., and the mechanical response of the rock sample at different loading stages is recorded.

[0104] Peak stress σ p As the confining pressure σ3 increases, the peak stress σ p The relationship between it and the confining pressure σ3 is expressed by the following formula:

[0105] σ p =a+bσ3;

[0106] The experimental data of granite rock samples from Shuangjiangkou Hydropower Station were fitted, and the peak stress σ3 under different confining pressures was obtained. p They are 114.72MPa, 139.65MPa, 157.22MPa, 182.97MPa, 216.30MPa, 345.73MPa, and 374.63MPa respectively;

[0107] Assume that the threshold stresses are the crack closure stress σ cc , crack initiation stress σ ci , crack damage stress σ cd ;

[0108] The relationship between the ratio of threshold stress to peak strength Rcc d, Rci d, and Rcd d and the confining pressure can be fitted by the following formula:

[0109] R=c+dσ3;

[0110] Among them, a, b, c, and d are fitting parameters;

[0111] Through the crack closure stress σ cc , crack initiation stress σ ci , and crack damage stress σ cd at different confining pressures σ3, and the ratios of these stresses to the peak strength are different. The values of Rcc d, Rci d, and Rcd d are 0.22σ p , 0.41σ p , and 0.78σ p respectively, and they gradually decrease with the increase of the confining pressure σ3.

[0112] Elastic modulus and Poisson's ratio: The elastic modulus and Poisson's ratio have an exponential relationship with the confining pressure and tend to a constant value when the confining pressure increases.

[0113] Stress-crack strain evolution constitutive model: Use quasi-viscous and plastic elements to represent the relationship between axial and radial crack strains and the deviator stress-peak strength ratio Rd in the crack development stage, and determine the crack strain expressions and related parameters in each stage, such as the quasi-viscosity coefficient in the crack closure stage, stable growth stage, unstable growth stage, and post-peak stage.

[0114] Crack evolution law: When σ3 = 1 MPa, analyze the variation laws of crack closure degrees C1 and C3 under different R<s d , as well as the ratios of axial and radial crack strains to the total strain in the unstable growth stage and post-peak stage under high, medium, and low confining pressures, indicating that the radial crack expansion is the main cause of rock failure, which is specifically expressed as follows:

[0115] As Figure 4 shown, when σ3 = 1 Mpa, in the stable growth stage (Rcc d < R d < Rcd d), the crack evolution is slow, and the crack closure degree C3 gradually decreases to 0; while in the unstable growth stage (Rcd d < R d < 1), the crack expansion rate gradually increases, and the crack closure degree C3 rapidly decreases with the increase of R d . When R d = 1, C3 = -17.71, and the rock begins to fail; entering the softening section, R d decreases, and C3 continues to decrease to -28.0, and the rock completely disintegrates. Under the action of time effect, the tangential and radial stresses on the tunnel wall will gradually be released. The setting of anchor cables and bolts provides radial compressive stress for the surrounding rock, compensating for the radial stress released by the surrounding rock itself, making the difference between the tangential stress and the radial stress smaller than that in the case without support, and thus making the crack closure degree C3 of the rock mass in the non-destruction stage.

[0116] The criterion for the surrounding rock relaxation zone is shown in the following table:

[0117] Table 2 Relaxation and failure zones of surrounding rock

[0118]

[0119]

[0120] According to the crack closure degree C3, the surrounding rock unloading relaxation zone is divided into five zones: destruction zone (C3<-25), strong relaxation zone (-25≤C3<-17), weak relaxation zone (-17≤C3<-2), stress disturbance zone (-2≤C3<0) and original rock zone (C3≥0), and the corresponding wave velocity and apparent deformation modulus zoning indicators are given;

[0121] Furthermore, in step S3, the creep test is selected as the specific test method. The multi-step loading method is adopted. By setting the rock sample parameters and the test method, the creep strain, the constitutive relationship between stress and instantaneous strain, the creep strain, the long-term strength and the creep strain rate are analyzed. The Nishihara creep model is improved and the correctness of the initial parameters is verified. The specific steps are as follows:

[0122] Assume that the confining pressure σ 3 The values ​​are ∈1, 3, 5, 10 MPa, and the axial stress is 0.55-1.00 times the peak strength;

[0123] Analysis of the test data shows that the axial creep strain increases with time, especially at high deviatoric stress levels. The total strain consists of instantaneous strain and creep strain, which includes viscoelastic creep strain and viscoplastic creep strain. By fitting and analyzing the creep strain data under different confining pressures and deviatoric stresses, the instantaneous strain ε is obtained. m and the deviatoric stress-peak strength ratio R d The relationship can be expressed as:

[0124] ε m =k1×R d +b1;

[0125] Viscoelastic creep strain ε ve With R d The relationship can be expressed as:

[0126]

[0127] Viscoplastic creep strain ε vp The relationship with the confining pressure σ3 can be expressed as:

[0128] ε vp =k2×σ3+b2;

[0129] Among them, k1, b1, k2, and b2 are variables.

[0130] According to the relationship between viscoelastic creep strain and viscoplastic creep strain, the traditional Nishihara creep model is improved, and the confining pressure and R d Introduce into the model and determine the instantaneous elastic modulus E in the improved model M , Kelvin elastic modulus E K , viscosity coefficient η K and η M The isoparametric expression can more accurately describe the time-dependent behavior of rock mass, including the decay creep and steady-state creep stages.

[0131] Further, if Figure 5 、 6 As shown, initial parameter determination and verification: Determine the initial parameters of the improved Nishihara creep model: η M =25GPa.d, G K =19.98GPa, η K =196GPa.d). The correctness of the initial parameters was verified by comparing the numerical calculation of the time-dependent deformation after the first level of excavation in the supported excavation condition with the on-site monitoring data. Furthermore, the calculation results of the time-dependent displacement cloud diagram of the surrounding rock at Unit 1 of the main powerhouse were compared with the on-site monitoring data. The calculated values ​​were close to the monitored values ​​and had the same deformation trend, indicating that the determined initial parameters can reasonably reflect the creep characteristics of the rock mass and provide a reliable parameter basis for subsequent engineering analysis and prediction.

[0132] S4. Construction simulation: simulate excavation sequence design based on the graded excavation scheme, analyze anchor support strength, determine support parameters and evaluate;

[0133] In a preferred solution, in step S4, the hierarchical excavation plan includes excavation sequence, anchor support strength analysis, support parameters and evaluation;

[0134] The excavation sequence was simulated according to the actual construction process, the anchor support strength analysis was based on indoor test results and field monitoring data, and the support parameters and evaluation were combined with theoretical analysis and numerical simulation results.

[0135] The support strength is determined by calculating the cohesion increment of the surrounding rock of the prestressed anchor cable support and comparing it with the value calculated by the empirical formula.

[0136] Specifically, a detailed, graded excavation plan was developed based on the actual construction process and geological conditions of the project. For example, for the main powerhouse of the Shuangjiangkou Hydropower Station, excavation was carried out sequentially from Layer A to Layer H. Support was provided promptly after each layer was excavated, and the starting and ending elevations of each layer's excavation and support were clearly defined. For example, the excavation and support starting and ending elevations for Layer A of the main powerhouse were 2282.32 to 2273.32 meters. During the excavation process, full consideration was given to factors such as changes in geological conditions, the mutual influence between chambers, and construction safety. The excavation sequence and progress were rationally arranged to minimize disturbance to the surrounding rock and ensure the stability of the construction process, as shown in the following table:

[0137] Table 3 Layered excavation characteristics of the powerhouse on the left bank of Shuangjiangkou

[0138] Project Parts Start and end elevation (m) A layer excavation and support 2282.32～2273.32 B layer excavation and support 2273.32～2269.02 C layer excavation and support 2269.02～2261.42 D layer excavation and support 2261.42～2253.60 E layer excavation and support 2253.60～2246.60 F layer excavation and support 2246.60～2240.10 G layer excavation and support 2240.10～2231.60 H layer excavation and support 2231.60～2214.00 total 2282.32～2214.00

[0139] Analysis based on test and monitoring data: Anchor cable support strength analysis is based on laboratory test results and field monitoring data. By analyzing extensive data from indoor triaxial compression and creep tests, combined with field monitoring of surrounding rock deformation, stress changes, and anchor cable stress, we conduct in-depth research on the support effectiveness of anchor cables under different geological conditions and construction stages. For example, based on the rock mass mechanical parameters obtained from the tests and the anchor cable's anchoring performance, we analyze the stress distribution and deformation characteristics of the anchor cable under surrounding rock pressure to determine the maximum support force and effective support range that the anchor cable can provide.

[0140] Calculation and comparison of empirical formulas: Through statistical analysis of data on underground powerhouses of 20 domestic hydropower projects, empirical formulas for anchor support strength, strength-stress ratio, and excavation span were proposed. Furthermore, through analysis of data from 12 domestic large and medium-sized hydropower stations, empirical formulas for anchor support strength, strength-stress ratio, and powerhouse span were obtained. Furthermore, the relationship formula between the cohesion increment of the surrounding rock reinforced by anchor cables and the strength-stress ratio was developed. Based on the specific geological conditions and engineering parameters of the Shuangjiangkou Hydropower Station, the theoretical support strength of the anchor cables was calculated. At the same time, the cohesion increment of the surrounding rock of the actual designed anchor cable support was calculated and compared with the value calculated using the empirical formula to assess whether the anchor cable support meets the strength standards required by the project. By defining a dimensionless anchor cable support index, the anchor cable support strength was quantitatively evaluated to determine its effectiveness and reliability in the project:

[0141]

[0142] Support parameter determination: Based on theoretical analysis, numerical simulation results, and on-site monitoring data, combined with engineering experience, the detailed layout parameters of anchor cables and bolts in various locations, such as the main powerhouse, main transformer room, and tailgate room, were determined. These included anchor cable location, pile number, elevation, anchor cable internal force, length, spacing, angle, and other parameters, as well as anchor bolt type, length, and spacing. For example, the anchor cables at the upstream arch foot of the main and auxiliary powerhouses were T = 2000kN and L = 20m. Parameters such as spacing were rationally designed based on the mechanical properties of the surrounding rock and the structural form of the cavern, ensuring that the support structure could effectively restrain surrounding rock deformation and improve cavern stability.

[0143] Support effect evaluation: The support effect of the determined support parameters is evaluated. The deformation, stress distribution, and plastic failure zone of the surrounding rock under different support parameters are calculated through numerical simulation, and compared with the situation when no support is used or when other support parameters are used. For example, the surrounding rock deformation and stress concentration of the side walls of the main powerhouse of the Shuangjiangkou Hydropower Station are calculated under different anchor spacing and lengths to evaluate the improvement effect of the anchors on the surrounding rock stability. For prestressed anchor cables, their adjustment effect on the surrounding rock stress state and their suppression effect on the plastic failure zone under different prestress levels are analyzed. Based on the evaluation results, the support parameters are optimized and adjusted to achieve the best support effect. At the same time, factors such as construction cost and construction period are taken into consideration to achieve comprehensive optimization of the support scheme.

[0144] S5. Geostress field analysis: Using the initial geostress multivariate regression method and the geostress field inversion method of the generative adversarial network (GAN), the surrounding rock stress state before construction excavation is calculated and the mechanical parameters are determined;

[0145] Specifically, the geostress regression calculation adopts the fitting idea of ​​the combined action of self-weight stress and boundary load, and uses the multivariate linear regression method. The geostress regression calculation value is used as the dependent variable, and the stress calculation value under each working condition of the finite element calculation is used as the independent variable. The multivariate regression coefficient matrix is ​​determined by the principle of least squares method, and then the regression initial stress of any point is obtained. In the calculation process, the influence of geological structure, rock mechanical properties, topography and other factors on the geostress distribution is fully considered. Through statistical analysis of a large amount of on-site geostress measurement data and finite element calculation results, the mathematical relationship between geostress and various influencing factors is established to achieve accurate inversion of the initial geostress field.

[0146] The distribution of the geostress field calculated using multivariate regression was compared and verified with the measured geostress data on site. For example, the error between the measured and fitted geostress values ​​at each measuring point was calculated to evaluate the fitting effect. In the inversion of the geostress field at the Shuangjiangkou Hydropower Station, a geostress inversion analysis was performed based on the measured geostress, ignoring measuring points far from the calculation range. This yielded a fitted geostress distribution map before construction excavation. The maximum fitting error was calculated to occur at the second principal stress component σ2 at measuring point σ1SPD3-1, with a difference of 1.794 MPa. The fitted geostress field retained the characteristics of the measured geostress field, with a fitting complex correlation coefficient of 0.9024, indicating a good fitting effect. This provided reliable initial geostress conditions for subsequent construction simulation and scheme optimization.

[0147] A Generative Adversarial Network (GAN)-based inversion method for geostress fields: Lateral stress coefficient-based paleostress field inversion uses the lateral stress coefficient to invert the paleostress field. The relationship between the lateral stress coefficient and burial depth is determined using a related formula. An approximate lateral stress coefficient is obtained from measured geostress values, and an approximate paleostress field is constructed. This method, based on the fundamental principles of rock mechanics, considers the evolution of geostress over geological history. By analyzing the lateral stress coefficient at different burial depths, the magnitude and direction of paleostress are inferred, providing important evidence for the geomechanical history of the study area.

[0148] The GAN-based inversion of the present-day geostress field introduces a GAN to optimize the present-day geostress field. The GAN consists of a generator and a discriminator, which compete to reach a Nash equilibrium. The regression factor value range is determined based on uniform design experiments, and training samples are constructed. The optimized regression factor of the paleostress field is then fed into the GAN for training, resulting in the optimized paleostress field and present-day geostress field. In this process, the GAN effectively learns the complex nonlinear relationship between the geostress field and geological parameters. Through a large number of training samples, the model is continuously optimized, improving the accuracy of the geostress field inversion. Compared with the BP neural network, the GAN achieves smaller relative errors in the stress components of the measured points when inverting the geostress field. For example, the maximum relative error of the GAN measured points is 17.46%. The geostress field distribution is more reasonable, more accurately reflecting the actual geostress distribution, and providing more accurate geostress data support for engineering design and construction.

[0149] Equivalent mechanical parameters of faults and weak structural surfaces: The composite unit containing rock mass and faults is simplified into a layered, transversely isotropic equivalent unit. The deformation and stress relationship of the equivalent unit in different directions and the calculation formulas for parameters such as equivalent elastic modulus and Poisson's ratio are determined. This step takes into account the significant impact of adverse geological bodies such as faults and weak structural surfaces on the mechanical properties of the rock mass. Through the calculation of equivalent parameters, the mechanical behavior of these special geological structures can be more reasonably simulated in the numerical model, reducing the difficulty of numerical model modeling and calculation time, while improving the accuracy and reliability of the calculation results, making the ground stress field analysis results closer to the actual engineering situation.

[0150] S6. Scheme optimization decision: Through numerical simulation analysis of surrounding rock deformation, stress distribution and plastic failure zone distribution characteristics under different anchor cable layout schemes, the anchor cable layout scheme is optimized.

[0151] In a preferred solution, in step S6, the following steps are further included:

[0152] S61. Surrounding rock deformation characteristics: Using the professional numerical simulation software FLAC3D, based on the precise three-dimensional numerical model established in the early stage, the excavation process at all levels was simulated. Monitoring points were set at key locations in the model, including the top arch, bottom plate, and side walls, to accurately obtain deformation data for each location. Furthermore, under working condition 1 (excavation without support), at all levels of excavation:

[0153] like Figure 7 and Figure 8 As shown, in the first level of excavation, the top arch of the installation section sinks 16.66mm, and the formula To calculate the total subsidence, follow these steps:

[0154] The total number of design calculation steps n = 100, and the sinking increment of the top arch in each calculation step Δh i The average value is 0.1666mm, so the total sinking is 16.66mm.

[0155] Where Δh i is the sinking increment of the top arch in each calculation step, and n is the total number of calculation steps;

[0156] The bottom plate is lifted by 15.54 mm, and the displacement change is calculated according to the corresponding cumulative formula.

[0157] The main building arch of unit 1# section sank by 18.59 mm. Through statistical analysis of the displacement of each node in the model, it was found that the bottom plate rose by 18.29 mm, which was calculated based on the displacement vector of the corresponding node in the model.

[0158] The top arch of the main powerhouse of unit 2# sank by 21.26mm, which was obtained by accumulating the displacement data of each node of the top arch during the simulation process, and the bottom plate rose by 17.74mm, which was obtained by calculating the displacement of the bottom plate nodes.

[0159] The top arch of the main powerhouse of unit 3# sank by 18.23mm, which was calculated based on the displacement of the top arch node of the model, and the bottom plate rose by 19.52mm, which was obtained based on the statistical displacement changes of the bottom plate nodes.

[0160] The top arch of the main powerhouse of unit 4# sank 16.22mm. Through the analysis and calculation of the displacement of the top arch node, the bottom plate rose 22.05mm, which was calculated based on the displacement data of the bottom plate node.

[0161] As the excavation level increases, the deformation trend is analyzed in depth;

[0162] Assuming the excavation level is m and the deformation is D, the quadratic function relationship determined by fitting the deformation data under different excavation levels, that is, by the least squares method, is:

[0163] D=am 2 +bm+c;

[0164] Among them, multiple sets of arch subsidence data of different excavation levels were fitted, and the values ​​of parameters a = -0.005, b = 0.5, and c = 10 were determined by the least squares method, and their relationship with the excavation depth was studied;

[0165] Then: D = -0.005m 2 +0.5m+10c

[0166] Considering the influence of rock mass mechanical properties on deformation, based on the previously determined rock mass parameters, including elastic modulus E and Poisson's ratio ν, combined with the theory of elastic mechanics, the relationship between deformation and rock mass mechanical properties follows the formula:

[0167]

[0168] Where ΔL is the deformation, P is the external force, L is the length, and A is the cross-sectional area;

[0169] Furthermore, for a specific rock mass area, the external force P = 100 kN, the length L = 5 m, and the cross-sectional area A = 0.5 m are known. 2 And elastic modulus E = 20GPa, Poisson's ratio ν = 0.25, through the formula Calculate and analyze the deformation, then:

[0170]

[0171] Regarding the impact of ground stress release, the relationship between ground stress release and deformation is analyzed based on ground stress measurement data and stress changes in numerical simulations.

[0172] Assuming the ground stress release amount is Δσ, there is a linear relationship between the deformation amount and the ground stress release amount:

[0173] D = KΔσ + d;

[0174] Furthermore, by statistically analyzing the monitoring data of the stress release and corresponding deformation in different regions, the parameters k = 0.5 and d = 2 were determined, and the following results were obtained:

[0175] D = 0.5Δσ + 2;

[0176] In the preferred solution, the deformation differences of different parts are compared, for example, the deformation of some areas of the main powerhouse upstream is greater than that of the downstream;

[0177] Assume that the upstream deformation is D 上 =20mm, the downstream deformation is D 下 =15mm, calculate the deformation difference: ΔD=D 上 -D 下 =20-15=5mm;

[0178] By comparing and analyzing the displacement data of corresponding upstream and downstream nodes in the model, the areas with large deformation differences are determined;

[0179] The displacement of the downstream side wall of the auxiliary powerhouse is affected by the lamprophyre vein, and the elastic modulus of the vein is E 岩脉 =15GPa. Based on the displacement data of the side wall nodes in the numerical simulation, the influence of the dyke on the side wall deformation is analyzed. Due to the presence of the dyke, the side wall displacement increases by 5mm.

[0180] Furthermore, due to the low elastic modulus of the dyke, the deformation coordination conditions of the surrounding rock mass change, thereby increasing the displacement of the sidewall. By analyzing these deformation differences, we can gain a deeper understanding of the influence of rock mass heterogeneity and geological structure on surrounding rock deformation, and thus take targeted measures during the scheme optimization process, such as adjusting the layout and direction of the anchor cables and strengthening the support strength in areas with larger deformation to reduce deformation differences and improve the overall stability of the surrounding rock.

[0181] S62. Stress distribution characteristics: Determine the stress concentration areas of the main caverns after each level of excavation, and compare the stress extreme value changes in the stress concentration areas of different schemes;

[0182] In the preferred solution, the stress cloud map and data from the numerical simulation results are used to determine the stress concentration areas of the main caverns after each level of excavation by setting stress thresholds;

[0183] Furthermore, for the two arch ends of the main powerhouse, the arch end of the tail transfer room and the bottom corner of the cavern side wall, a stress value threshold σ is set. th =45MPa, these areas are identified as stress concentration areas. By screening and analyzing the stress data of each node in the model, the location and range of these stress concentration areas are clearly displayed, providing clear target areas for subsequent stress analysis and solution optimization.

[0184] Under the original anchor cable arrangement, statistical analysis of the numerical simulation results showed that the stress concentration at the upstream arch end of the installation room was greater than that at the downstream arch end, with the maximum value being approximately 50 to 55 MPa.

[0185] Assume that the maximum principal stress at the upstream arch end is σ 1上 =52MPa, the minimum principal stress is σ 3上 =48MPa, according to the calculation formula of stress concentration degree: Calculate the degree of stress concentration;

[0186] After optimizing the anchor cable arrangement, the numerical simulation calculation is re-performed by adjusting parameters such as anchor cable spacing, length, and prestress.

[0187] Assuming that the stress extreme value of the upstream arch end of the installation room is reduced to about 48MPa after optimization, the change of stress concentration degree is analyzed according to the above stress concentration degree calculation formula.

[0188] At the same time, by comparing the changes in the range of stress concentration areas, and through comparative analysis of stress extremes and concentration area ranges under different schemes, we can intuitively evaluate the improvement effects of different schemes on stress distribution, providing a key basis for selecting the optimal scheme.

[0189] S63. Distribution characteristics of plastic failure zone: Obtain the depth of the plastic failure zone of the surrounding rock of each unit;

[0190] Extract the depth data of the plastic failure zone of surrounding rock of each unit from the numerical simulation results;

[0191] Furthermore, the depth of the plastic failure zone of the top arch of the installation section is about 1.85m shallow plastic failure zone;

[0192] For the main building of Unit 2, which is affected by fault f1, the depth of the plastic failure zone at the top arch is approximately 2.29 m. Based on the location and mechanical properties of the fault and the distribution of plastic strain in the numerical simulation, the depth of the plastic failure zone near the fault was determined.

[0193] Similarly, the depth of the plastic failure zone at the top arch of the main powerhouse of Unit 3 is about 6.61m, which reflects the degree and scope of damage to the surrounding rock during the excavation process and is of great significance for evaluating the stability of the surrounding rock and the effectiveness of the support scheme.

[0194] Analyze the relationship between the range of the plastic failure zone and the stability of the surrounding rock. According to rock mechanics theory, the expansion of the plastic failure zone will lead to a decrease in the strength of the surrounding rock and a deterioration in its stability.

[0195] Assuming the range of the plastic failure zone is S and the surrounding rock stability index is K, through statistical analysis of the range of the plastic failure zone and the surrounding rock stability index under different schemes, a negative correlation is obtained: K = e -0.1S ;

[0196] Among them, the range data of plastic failure zone and the corresponding surrounding rock stability index data under multiple different schemes are obtained, and the parameter f=0.1 is determined by fitting analysis of the data;

[0197] Generally speaking, the smaller the plastic failure zone, the higher the stability of the surrounding rock. By comparing the changes in the plastic failure zone under different schemes, such as in the optimization scheme, by reasonably arranging the anchor cables, the depth and range of the plastic failure zone can be effectively controlled, thereby improving the stability of the surrounding rock and providing an important reference for determining the optimal scheme.

[0198] S64. Scheme determination and verification: Determine the three-dimensional numerical models for different anchor cable arrangement schemes based on the surrounding rock deformation characteristics, stress distribution characteristics, and plastic failure zone distribution characteristics. Perform excavation simulation calculations based on the three-dimensional numerical models for different anchor cable arrangement schemes to obtain surrounding rock deformation results, stress distribution results, and plastic failure zone results.

[0199] Taking all the above factors into consideration, an evaluation index system for surrounding rock stability is established;

[0200] Furthermore, the weighted average method is used, and the deformation weight is set as w D =0.4, stress weight is w σ =0.3, the weight of the plastic failure zone is w S =0.3;

[0201] Among them, K D It can be determined based on the ratio of deformation to allowable deformation, and the formula is as follows:

[0202]

[0203] In the anchor cable arrangement scheme, the actual deformation D is obtained by monitoring 实际 =18mm, the allowable deformation D is determined according to engineering specifications and experience 允许 =25mm, then

[0204] K σ It can be determined based on the ratio of stress concentration degree to allowable stress concentration degree. The formula is as follows:

[0205]

[0206] In a certain area, the actual stress concentration degree C 实际 =2.5MPa, the allowable stress concentration level C is determined based on rock mass strength theory and engineering experience 允许 =3.5MPa, then

[0207] K S It can be determined based on the ratio of the plastic failure zone range to the allowable plastic failure zone range. The formula is as follows:

[0208]

[0209] Assume that the actual plastic failure zone range S 实际 =4m, the allowable plastic failure zone range S is determined according to engineering specifications and experience 允许 =6m, then

[0210] Substitute the above calculation results into the surrounding rock stability index K 总 formula:

[0211] K 总 =w D K D +w σ K σ +w S K S ;

[0212] We can get: K 总 =0.4×0.28+0.3×0.29+0.3×0.33=0.302.

[0213] By comparing K under different schemes 总 Compare the values ​​and determine the optimal solution;

[0214] Specifically, different anchor cable arrangements are calculated to obtain their respective K 总 Value, further, K of Scheme 1 总 =0.28, K of Scheme 2 总 =0.32, K of Scheme 3 总 =0.35.

[0215] By comparison, select K 总 The solution with the largest value is the optimal solution, namely Solution 3. At the same time, the specific layout parameters of the anchor cables in the optimal solution are clarified, including the anchor cable spacing of 3m, length of 25m, angle of 25°, and layout position at the arch foot of the main powerhouse.

[0216] Verification and feedback: Obtain the deformation, stress distribution and damage zone conditions of on-site monitoring. By arranging monitoring instruments at the project site, including multi-point displacement meters, stress meters, and acoustic wave detectors, the actual deformation, stress and damage data of the surrounding rock can be obtained in real time.

[0217] Compare the surrounding rock deformation results, stress distribution results and plastic failure zone results with the deformation, stress distribution and failure zone conditions monitored on site;

[0218] The root mean square error (RMSE) formula is used to calculate the error between the two. Assume that the deformation data sequence obtained by simulation is D 模拟,i (i=1,2,…,n), where n is the number of data points, and the deformation data sequence obtained by monitoring is D 监测,i , the stress value data sequence is σ 模拟,j and σ 监测,j (j=1,2,…,m, where m is the number of stress data points and the depth data sequence of the plastic failure zone is h 模拟,k and h 监测,k (k=1,2,…,l), where l is the number of data points in the plastic failure zone.

[0219] Multi-point displacement meters, stress meters, acoustic wave detectors and other monitoring instruments are arranged at the project site to obtain the actual deformation, stress and damage data of the surrounding rock in real time. In a certain period of time, the deformation data sequence monitored by the multi-point displacement meter is D 监测,i ={17,20,19,18,21}(unit: mm), the stress value data sequence monitored by the strain gauge is σ 监测,j ={46,48,45,47,49}(unit: MPa), the data sequence of the depth of the plastic failure zone monitored by the acoustic wave detector is h 监测,k ={3.5,4.0,3.8,3.6,4.2}(unit: m).

[0220] Calculate the root mean square error of deformation:

[0221] The deformation data sequence obtained by numerical simulation is D 模拟,i ={16,22,18,19,20}, then:

[0222]

[0223] Calculate the root mean square error of stress:

[0224] The deformation data sequence obtained by numerical simulation is σ 模拟,j ={45,49,44,46,48}, then:

[0225]

[0226] Calculate the root mean square error of the plastic failure zone depth:

[0227] The deformation data sequence obtained by numerical simulation is h 模拟,k ={3.2,4.5,3.6,3.8,4.0}, then:

[0228]

[0229] Taking these three root mean square error values ​​into consideration, a total root mean square error RMSE can be obtained by weighted average method. 总 ;

[0230] Assume that the weights of deformation, stress, and plastic failure zone depth are w1 = 0.5, w2 = 0.3, and w3 = 0.2, respectively, and satisfy w1 + w2 + w3 = 1. Then, according to the root mean square error formula RMSE 总 =w1RMSE D +w2RMSE σ +w3RMSE h , we can get:

[0231] RMSE 总 =0.5×1.49+0.3×1.41+0.2×0.28≈1.27;

[0232] According to the comparison results, the error correction model of the evaluation method is established. If the RMSE 总 If the value is large, it means that there is a large deviation between the simulation results and the actual situation, and the parameters in the numerical model need to be adjusted.

[0233] Assume that the parameter to be adjusted is x. In order to determine g and h, collect the corresponding RMSE under multiple sets of different parameter values ​​x. 总 The data is shown in the following table:

[0234] Table 4

[0235] Parameter x value <![CDATA[RMSE 总 Value]]> <![CDATA[x1=2.5m]]> <![CDATA[RMSE 总1 =1.50]]> <![CDATA[x2=3.0m]]> <![CDATA[RMSE 总2 =1.20]]> <![CDATA[x3=3.5n]]> <![CDATA[RMSE 总3 =1.00]]> <![CDATA[x4=4.0m]]> <![CDATA[RMSE 总4 =0.85]]> <![CDATA[x5=4.5m]]> <![CDATA[RMSE 总5 =0.70]]>

[0236] The least squares method is used to fit and determine the values ​​of g and h, and according to the principle of least squares method, the equation group is established:

[0237]

[0238] Where q is the number of data sets. Substituting the data in Table 1, we get:

[0239]

[0240] in, We can get:

[0241] Calculation can be obtained:

[0242]

[0243] Substituting the above values ​​into the equations, we get:

[0244]

[0245] Solving the system of equations yields:

[0246] From the first equation in the system, we get h = 1.05 - 3.5g. Substituting this into the second equation in the system, we get: 18.2 = g × 61.5 + (1.05 - 3.5g) × 3.5.

[0247] 18.2 = 61.5 g + 3.675 - 12.25 g;

[0248] 18.2-3.675=49.25g;

[0249] 14.525 = 49.25 g;

[0250] g≈0.295;

[0251] Substituting g≈0.295 into h=1.05-3.5g, we can obtain: h≈1.05-3.5×0.295≈0.02;

[0252] When the RMSE obtained in a simulation 总 =1.3, according to the established linear relationship: RMSE 总 =gx+h, the correction value Δx of parameter x can be deduced inversely:

[0253]

[0254] Among them, x 当前 Get the value of the current parameter;

[0255] If the current parameter x 当前 =3m, according to the formula: We can get:

[0256]

[0257] Adjust the parameter x to: 调整后 =x 当前 +Δx=3.0+1.12=4.12m, and numerical simulation was performed again and compared with the field monitoring data;

[0258] Repeat the above process and continuously optimize the parameters until the error index of the root mean square error meets the accuracy range required by the project.

[0259] In addition, we can further analyze the influence of different parameters on the error, and further calculate the error change rate corresponding to each parameter value, that is: To determine the impact of parameter adjustment on the error. If the change of a certain parameter has a greater impact on the error, more attention should be paid to the adjustment of this parameter in the subsequent optimization process, so as to achieve continuous improvement and perfection of the evaluation method and anchor cable layout plan, improve the quality and safety of engineering construction, and ensure the long-term stable operation of the underground cavern group.

Claims

1. A method for evaluating the construction layout quality of prestressed anchor cables, characterized in that: The following steps are involved: S1. Basic data preparation: Acquire and preprocess geological survey data of underground caverns, and establish a three-dimensional numerical model based on the preprocessed geological survey data; S2. Model basic settings: Determine the surrounding rock types in different areas based on geological survey data, determine rock mass parameters based on the surrounding rock types, and select constitutive models based on the surrounding rock parameters, including elastic-plastic models and stress-crack strain evolution constitutive models; S3. Model property research: Obtain the mechanical property parameters of the rock mass through specific test methods, establish a stress-crack strain evolution constitutive model, analyze the crack evolution law, determine the criteria for the surrounding rock relaxation zone, and conduct secondary development of the constitutive model; In step S3, the creep test is selected as the specific test method. The multi-step loading method is used. By setting the rock sample parameters and the test method, the creep strain, stress and instantaneous strain, the constitutive relationship of creep strain, long-term strength and creep strain rate are analyzed. The Nishihara creep model is improved and the correctness of the initial parameters is verified. The specific steps are as follows: The axial creep strain increases with time and is more obvious at high deviatoric stress levels. The total strain consists of instantaneous strain and creep strain. The creep strain includes viscoelastic creep strain and viscoplastic creep strain. By analyzing the creep strain data under different confining pressures and deviatoric stresses, the instantaneous strain ε is obtained. m and the deviatoric stress-peak strength ratio R d The relationship can be expressed as: e m =k1×R d +b1; Viscoelastic creep strain ε ve With R d The relationship can be expressed as: Viscoplastic creep strain ε vp The relationship with the confining pressure σ3 can be expressed as: ε vp =k2×σ3+b2; Among them, k1, b1, k2, and b2 are variables; S4. Construction simulation: simulate excavation sequence design based on the graded excavation scheme, analyze anchor support strength, determine support parameters and evaluate; S5. Geostress field analysis: Using the initial geostress multivariate regression method and the geostress field inversion method of the generative adversarial network (GAN), the surrounding rock stress state before construction excavation is calculated and the mechanical parameters are determined; S6. Scheme optimization decision: Through numerical simulation analysis of surrounding rock deformation, stress distribution and plastic failure zone distribution characteristics under different anchor cable layout schemes, the anchor cable layout scheme is optimized.

2. The method for evaluating the construction layout quality of prestressed anchor cables according to claim 1, wherein: In step S1, the geological survey data includes the physical and mechanical parameters of the rock mass, ground stress measurement data, and structural surface characteristics; Among them, the physical and mechanical parameters of rock mass include natural density, deformation modulus, Poisson's ratio, and shear strength; The preprocessing operation is to organize and screen the collected geological survey data, remove abnormal data and normalize them for model input and calculation; A three-dimensional numerical model of the main chambers including the installation room, main and auxiliary powerhouse caverns, main transformer room, and tailwater surge tank was established based on the pre-processed geological survey data. The three-dimensional numerical model uses three-dimensional 8-node hexahedron isoparametric solid elements and their degenerate elements to simulate rock mass and faults. The overall model has 91,479 solid element nodes, 181,231 solid elements, 51,803 anchor element nodes, and 26,089 anchor elements.

3. The method for evaluating the construction layout quality of prestressed anchor cables according to claim 2, wherein: In step S2, the physical and mechanical parameters of the rock mass are calculated according to the empirical formula to obtain the rock mass parameters, and the calculated rock mass parameters are corrected in combination with engineering experience and field monitoring data; Among them, the elastic-plastic model determines the cracking conditions of rock materials through macroscopic strength and determines that the rock mass enters the plastic state based on the Druker-Prager criterion; When applying the constitutive model to the three-dimensional numerical model, the deformation, stress distribution and failure characteristics of the surrounding rock are analyzed by setting the parameters of the constitutive model and the boundary conditions of the three-dimensional numerical model and performing numerical simulation calculations using specific methods; Among them, the parameter settings of the constitutive model include the yield stress and hardening parameters of the elastic-plastic model and the crack closure and expansion parameters of the stress-crack-strain evolution constitutive model.

4. The method for evaluating the construction layout quality of prestressed anchor cables according to claim 1, wherein: In step S3, the mechanical property parameters of the rock mass include peak stress, elastic modulus, and Poisson's ratio. The triaxial compression test is selected as the specific test method. Through rock sample preparation and test loading, the stress-strain curve, threshold stress, elastic modulus and Poisson's ratio, stress-crack strain evolution constitutive model, crack evolution law, and surrounding rock relaxation zone criterion are analyzed. The constitutive model code is compiled and verified in FLAC3D. The specific steps are as follows: Collect rock samples and use a rock triaxial testing machine; And let the peak stress be σ p , confining pressure σ3, the value range of confining pressure σ3 is: σ 3 ∈[1,40]MPa, and,σ 3 ∈1, 3, 5, 10, 20, 30, 40 MPa, the loading rate is 0.05 MPa / s for the confining pressure σ3, and the axial load increases at 0.5 MPa / s until the rock sample fails; Peak stress σ p As the confining pressure σ3 increases, the peak stress σ p The relationship between it and the confining pressure σ3 is expressed by the following formula: s p =a+bσ3; Assume that the threshold stresses are the crack closure stress σ cc , crack initiation stress σ ci , crack damage stress σ cd ; The relationship between the ratio of threshold stress to peak strength Rcc d, Rci d, and Rcd d and the confining pressure can be fitted by the following formula: R=c+dσ3; Among them, a, b, c, and d are fitting parameters.

5. The method for evaluating the construction layout quality of prestressed anchor cables according to claim 4, wherein: By fitting the rock sample test data under multiple confining pressures, the corresponding peak stress σ under different confining pressures σ3 is obtained. p They are 114.72MPa, 139.65MPa, 157.22MPa, 182.97MPa, 216.30MPa, 345.73MPa, and 374.63MPa respectively; The crack closure stress σ under different confining pressure σ3 cc , crack initiation stress σ ci , crack damage stress σ cd The ratio of the peak intensity is different, and the values ​​of Rcc d, Rci d, and Rcd d are 0.22σ respectively. p , 0.41σ p and 0.78σ p , and gradually decreases with the increase of confining pressure σ3.

6. The method for evaluating the construction layout quality of prestressed anchor cables according to claim 5, characterized in that: Assume that the confining pressure σ 3 The values ​​are ∈1, 3, 5, 10MPa, and the axial stress is 0.55-1.00 times the peak strength.

7. The method for evaluating the construction layout quality of prestressed anchor cables according to claim 6, wherein: In step S4, the hierarchical excavation plan includes excavation sequence, anchor support strength analysis, support parameters and evaluation; The excavation sequence was simulated according to the actual construction process, the anchor support strength analysis was based on indoor test results and field monitoring data, and the support parameters and evaluation were combined with theoretical analysis and numerical simulation results. The support strength is determined by calculating the cohesion increment of the surrounding rock of the prestressed anchor cable support and comparing it with the value calculated by the empirical formula.

8. The method for evaluating the construction layout quality of prestressed anchor cables according to claim 7, wherein: In step S6, the following steps are also included: S61. Deformation characteristics of surrounding rock: Obtain the deformation of each part of the tunnel during each level of excavation, and analyze the changing trend of deformation at each level of excavation and the deformation differences at different parts; S62. Stress distribution characteristics: Determine the stress concentration areas of the main caverns after each level of excavation, and compare the stress extreme value changes in the stress concentration areas of different schemes; S63. Distribution characteristics of plastic failure zone: Obtain the depth of the plastic failure zone of the surrounding rock of each unit; S64. Determine the three-dimensional numerical models under different anchor cable arrangement schemes according to the surrounding rock deformation characteristics, stress distribution characteristics, and plastic failure zone distribution characteristics, and perform excavation simulation calculations based on the three-dimensional numerical models under different anchor cable arrangement schemes to obtain the surrounding rock deformation results, stress distribution results, and plastic failure zone results.

9. The method for evaluating the construction layout quality of prestressed anchor cables according to claim 8, wherein: In step S64, the surrounding rock stability under different anchor cable arrangement schemes is determined according to the surrounding rock deformation results, stress distribution results and plastic failure zone results, and the optimal scheme is determined according to the surrounding rock stability under different anchor cable arrangement schemes.

10. The method for evaluating the construction layout quality of prestressed anchor cables according to claim 9, wherein: Obtain deformation, stress distribution and damage zone conditions from on-site monitoring; The surrounding rock deformation results, stress distribution results and plastic failure zone results are compared with the deformation, stress distribution and failure zone conditions monitored on site, and statistical analysis methods are used for quantitative evaluation to obtain the comparison results; Wherein, the statistical analysis method is one of the root mean square error or correlation coefficient; An error correction model of the evaluation method is established based on the comparison results, which is used to pre-adjust the parameters under different anchor cable layout schemes.

Images