Hydropower station underground cavern group crustal stress field nonlinear inversion analysis method

Abstract

The invention provides a hydropower station underground cavern group crustal stress field nonlinear inversion analysis method, which comprises the following steps: S1, based on regional geologic structure conditions and actually measured crustal stress analysis, analyzing the geologic structure conditions of an engineering area where a hydropower station plant area is located, and carrying out field crustal stress measurement on a plurality of measurement points in the plant area, the local geological structure condition of the underground cavern group area is obtained; s2, performing crustal stress field nonlinear inversion on the hydropower station plant area based on an artificial neural network, and establishing a three-dimensional numerical model to obtain crustal stress distribution characteristics of the hydropower station plant area; s3, ground stress field characteristic analysis of the cavern area; and S4, performing field verification analysis on the ground stress of the cavern group. Based on the artificial neural network theory and in combination with the crustal stress field inversion method, the nonlinear inversion method of the crustal stress field is provided, the distribution characteristics of the crustal stress field in the underground multi-cavern group environment of the hydropower station can be calculated and analyzed more accurately, and the accuracy of crustal stress field inversion is greatly improved.

Description

Technical Field

[0001] This invention relates to the field of geostress field inversion technology, and in particular to a nonlinear inversion analysis method for geostress field of underground cavern groups in hydropower stations. Background Technology

[0002] The construction of hydropower stations requires analysis of the distribution characteristics of the geostress field. Since multiple underground chambers need to be constructed, the geostress field distribution needs further consideration. For most geotechnical engineering projects, using on-site measured data and considering topographic and geological characteristics to conduct theoretical or numerical analyses for inversion, regression, and simulation to infer geostress values ​​and provide initial geostress field models is an important and effective approach. A crucial component of this approach is inversion analysis, which mainly takes two forms: ① calculating the magnitude and direction of geostress based on on-site measured displacement data and referring to the mechanical properties of local rocks; ② inverting geostress field models based on measured stress values ​​at certain control points in the engineering area, considering topographic, geological, and lithological factors.

[0003] The main methods used for calculating and inverting the initial stress field of rock mass are as follows: (1) Heim's Hypothesis; (2) The initial stress field of rock mass is approximately equal to the stress field of rock mass self-weight; (3) Lateral pressure coefficient method; (4) Boundary load adjustment method; (5) Finite element mathematical model regression analysis method; (6) Stress function trend analysis method of stress field; (7) Stress function regression analysis method of stress field; (9) Intelligent inversion analysis method.

[0004] However, most of these methods suffer from drawbacks such as being susceptible to human factors and being highly arbitrary in the calculation process. Summary of the Invention

[0005] The present invention aims to at least partially solve one of the technical problems in the related art.

[0006] To achieve the above objectives, this invention proposes a nonlinear inversion analysis method for the geostress field of underground cavern groups in hydropower stations, comprising the following steps:

[0007] S1. Based on the regional geological structure and measured in-situ stress analysis, the geological structure of the engineering area where the hydropower station is located is analyzed, and in-situ in-situ stress measurements are carried out at multiple measuring points in the plant area to obtain the local geological structure of the underground cavern group area.

[0008] S2. Based on artificial neural networks, nonlinear inversion of the ground stress field in the hydropower plant area is performed, and a three-dimensional numerical model is established to obtain the ground stress distribution characteristics of the hydropower plant area.

[0009] S3. Analysis of the geostress field characteristics of the cavern area, including the geostress values ​​in the cavern group area, the geostress direction in the cavern group area, the influence of typical faults on the geostress in the plant area, and the changes in geostress in the plant area under reservoir impoundment.

[0010] S4. Conduct on-site verification and analysis of the geostress of the cavern group.

[0011] This invention provides a nonlinear inversion method for geostress fields based on artificial neural network theory and geostress field inversion method. It can more accurately calculate and analyze the geostress field distribution characteristics of underground multi-cavity groups in hydropower stations, while reducing the influence of human factors and arbitrariness, and greatly improving the accuracy of geostress field inversion.

[0012] Optionally, S1 specifically includes:

[0013] S11. Based on regional tectonics, the geological structure of the hydropower station site is analyzed according to the geostress distribution in the World Stress Map to obtain a preliminary judgment on the maximum horizontal direction of geostress in the hydropower station area.

[0014] S12. Conduct structural exploration of the hydropower station project area, determine the main underground faults at the dam site, and select multiple measuring points at different locations and depths in the site area to conduct in-situ stress tests.

[0015] Furthermore, in step S12, when determining the fault, it is necessary to determine the number, occurrence, fracture zone width, fault characteristics, and exposure location of the faults below the plant site.

[0016] Furthermore, in performing nonlinear inversion of geostress, to establish a nonlinear mapping between the model boundary conditions and the stress values ​​at the measured geostress points within the model, the nonlinear relationship between the boundary conditions to be inverted and the stress values ​​can be described using a set of neural networks:

[0017]

[0018] In the formula, P = (p1, p2, ... p n ) represents the input node representation of the neural network; D = (d1, d2, ... d n ) is the output node representation of a neural network; NN(n,h1,...h p (m) represents the established multi-layer neural network structure; where n is the input layer F. x The number of nodes, h1 to h p Corresponding to hidden layers F1 to F2 respectively p The number of nodes, m is the output layer F y The number of nodes.

[0019] Furthermore, in S2, the specific steps of the nonlinear inversion include:

[0020] S21. Establish a three-dimensional initial geostress field calculation model based on the geological and geomorphological characteristics of the hydropower station site to provide a stress distribution model basis for nonlinear inversion.

[0021] S22, Rock mass elastoplastic constitutive model and calculation parameters: Construct a parameter combination table for boundary load conditions according to the orthogonal or uniform design principle, and obtain the calculated geostress value information of the measurement location through numerical calculation software or other calculation tools, thereby establishing samples for neural network learning and training;

[0022] S23. Nonlinear regression calculation: The calculated ground stress value is used as the input vector of the network, and the boundary conditions are used as the output vector of the neural network to train the neural network.

[0023] S24. When a mature network structure and training times are obtained, the stress values ​​of the measured stress points are used as the input vector, and the output vector obtained by the neural network is used as the boundary conditions that can be adopted.

[0024] S25. Substitute the obtained boundary conditions into the numerical calculation software to perform a forward calculation to obtain the regional geostress field.

[0025] Furthermore, in S21, solid elements are used for each fault in the three-dimensional initial geostress field calculation model.

[0026] Furthermore, in step S22, the numerical calculation software used is ANSYS or FLAC numerical calculation software.

[0027] Furthermore, in S22, for situations where the rock mass in the area where the hydropower station is located has undergone geological processes of varying degrees, the rock mass will experience compression-shear and tension failures during geological changes, and an elastoplastic calculation method is used when calculating the rock mass parameters.

[0028] Furthermore, in step S23, the nonlinear regression calculation of the geostress field employs a comprehensive method based on uniform design, artificial neural network, and finite difference, specifically including:

[0029] S231. Based on uniform design, the sample construction is carried out. According to the measured geostress characteristics and engineering geological analysis, the uniform design method is adopted to make a sample horizontal uniform design table by combining the numerically calculated displacement boundary conditions and gravity correction coefficients, and multiple sets of learning and training samples are combined for neural network learning and training.

[0030] S232. Input the sample schemes of each group based on the boundary conditions obtained by uniform design into the finite difference software for calculation, obtain the stress calculation values ​​at multiple measurement points of each sample, use them as the learning and training input values ​​of the neural network, use the set boundary conditions as the corresponding output values, establish an evolutionary neural network model of the nonlinear mapping relationship between the stress components of the measurement points and the boundary conditions, and use a genetic algorithm to optimize the optimal network structure and connection weights of the neural network.

[0031] S233. Using the trained evolutionary neural network model, the measured stress component values ​​of each measuring point were input to obtain the displacement boundary conditions and gravity correction coefficient of the geostress field.

[0032] S234. The boundary conditions obtained by nonlinear regression inversion are calculated again using finite difference forward calculation, thereby obtaining the distribution characteristics of the geostress field in the underground engineering area of ​​the hydropower station.

[0033] S235. Conduct on-site testing of the stress values ​​at each measuring point and compare them with the calculation results to determine the rationality of the distribution characteristics of the geostress field.

[0034] Furthermore, in S3, when analyzing the stress characteristics of the underground cavern group area, it is necessary to analyze the characteristics, magnitude and distribution law of the stress field in the hydropower station project area based on the stress distribution characteristics of the hydropower station plant area obtained by nonlinear inversion of the stress field. The specific analysis parts of the project area include: the vertical cross-section of the axis on both the upstream and downstream sides of the underground cavern group, the horizontal cross-section of the top and bottom of the underground cavern group, and the horizontal cross-section of the middle part of the underground cavern group.

[0035] Furthermore, when analyzing the impact of typical faults on the ground stress in the plant area, numerical simulation was used to calculate and analyze the distribution of the ground stress field in the plant area under the condition of no fault, and to obtain the distribution of ground stress on the vertical cross-section of the axial direction on both the upstream and downstream sides of the underground cavern group under the condition of no fault.

[0036] By comparing the distribution of geostress on the vertical cross-sections of the upstream and downstream sides of the cavern group under fault conditions, the influence of typical faults on the geostress in the plant area can be obtained.

[0037] During engineering construction, the impact of underground cavern groups and dam construction on the in-situ stress of rock masses near faults should be monitored, and timely safety assessments of the surrounding rock of underground caverns should be conducted.

[0038] Furthermore, in analyzing the impact of dam abutment thrust on the stability of the powerhouse cavern group, a three-dimensional finite element model was established based on the characteristics of ground stress distribution in the hydropower station area, the axis of the underground powerhouse caverns, the cavern spacing, the excavation sequence, and the initial support scheme. Thrust was applied at the corresponding dam abutment positions, and the displacement and stress of the characteristic points of each unit section in the underground caverns were calculated after the application of the dam abutment arch thrust. The displacement and stress of the characteristic points of each unit section were compared between considering and not considering the dam abutment thrust to obtain the impact of dam abutment thrust on the stability of the underground caverns.

[0039] Furthermore, in analyzing the changes in ground stress in the power plant area under reservoir impoundment, a three-dimensional numerical simulation method was used to calculate and analyze the displacement and stress distribution of the power plant area under normal impoundment water level and low water level conditions. The influence of dam abutment thrust was also considered. It is necessary to analyze the changes in ground stress in the power plant area under reservoir impoundment under three operating conditions, including:

[0040] Operating condition 1: Dam abutment thrust + water pressure at normal water level (water surface elevation 2535m);

[0041] Condition 2: Dam abutment thrust + low water level (water surface elevation 2525m) water pressure;

[0042] Condition 3, Dam abutment thrust;

[0043] The parts that need to be calculated and analyzed include: the crown, upstream sidewall, downstream sidewall, and wall ends;

[0044] The parameters that need to be calculated for each part under each working condition include: displacement, maximum principal stress, minimum principal stress, anchor stress, and anchor cable load; for the wall end part, only the anchor stress needs to be calculated.

[0045] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0046] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0047] Figure 1 This is a schematic diagram illustrating the steps of a nonlinear inversion analysis method for the geostress field of an underground cavern group in a hydropower station according to the present invention.

[0048] Figure 2(a) is a schematic diagram of the three-dimensional numerical model structure of a nonlinear inversion analysis method for the geostress field of underground cavern group in a hydropower station according to the present invention, mainly showing the mountain structure;

[0049] Figure 2(b) is a schematic diagram of another three-dimensional numerical model structure of the nonlinear inversion analysis method of the underground cavern group in a hydropower station according to the present invention. It mainly shows the structure of the underground cavern group, where F115 is a typical fault.

[0050] Figure 3(a) is a schematic diagram of the distribution characteristics of the geostress field in the underground engineering area of ​​a hydropower station obtained by step S234 of the nonlinear inversion analysis method of the geostress field of the underground cavern group of a hydropower station according to the present invention, mainly showing the first stress.

[0051] Figure 3(b) is a schematic diagram of the distribution characteristics of the geostress field in the underground engineering area of ​​a hydropower station obtained by step S234 of the nonlinear inversion analysis method of the geostress field of the underground cavern group of a hydropower station according to the present invention, mainly showing the second stress.

[0052] Figure 3(c) is a schematic diagram of the distribution characteristics of the geostress field in the underground engineering area of ​​a hydropower station obtained by step S234 of the nonlinear inversion analysis method of the geostress field of underground cavern group of a hydropower station according to the present invention, mainly showing the third stress.

[0053] Figure 4(a) is a diagram of the stress distribution in the vertical profile of the powerhouse axis during the stress value characteristic analysis of the underground cavern group region in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention. It mainly shows the maximum principal stress σ1.

[0054] Figure 4(b) is a vertical profile of the powerhouse axis showing the stress distribution during the stress value characteristic analysis of the underground cavern group region in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention. It mainly shows the minimum principal stress σ3.

[0055] Figure 5(a) is a vertical profile of the stress distribution of the tail chamber axis during the stress value characteristic analysis of the underground cavern group in the cavern group area in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention. It mainly shows the maximum principal stress σ1, where F115 is a typical fault.

[0056] Figure 5(b) is a vertical profile of the stress distribution of the tail chamber axis during the stress value characteristic analysis of the underground cavern group in the cavern group region in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention. It mainly shows the minimum principal stress σ3, where F115 is a typical fault.

[0057] Figure 6(a) is a schematic diagram of the stress direction of a typical part of the underground cavern group in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention. It mainly shows the stress direction of the horizontal profile in the middle part of the underground cavern group.

[0058] Figure 6(b) is a schematic diagram of the stress direction of a typical part of the underground cavern group in step S3 of the nonlinear inversion analysis method of the underground cavern group of the present invention, mainly showing the stress direction of the vertical section of the powerhouse axis of the underground cavern group.

[0059] Figure 6(c) is a magnified view of a portion of Figure 6(b);

[0060] Figure 7(a) is a schematic diagram of the stress distribution of the vertical cross-section of the axial direction on both the upstream and downstream sides of the underground cavern group in the case of no fault, when performing the influence analysis of typical faults on the ground stress in the power plant area in step S3 of the nonlinear inversion analysis method of the ground stress field of the underground cavern group of the present invention. It mainly shows the case where the maximum principal stress σ1 of the typical fault is not considered.

[0061] Figure 7(b) is a schematic diagram of the stress distribution of the vertical cross-section of the axial direction on both the upstream and downstream sides of the underground cavern group in the absence of faults during the analysis of the influence of typical faults on the ground stress in the power plant area in step S3 of the nonlinear inversion analysis method of the ground stress field of the underground cavern group of the present invention. It mainly shows the case where the maximum principal stress σ3 of the typical fault is not considered.

[0062] Figure 8 This is a schematic diagram of the arch thrust model of the arch dam abutment in step S3 of a nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station according to the present invention.

[0063] Figures 9(a1) and (a2) are cloud maps of the vertical profile σ1 of the powerhouse axis when analyzing the influence of the dam abutment thrust on the stability of the surrounding rock of the powerhouse cavern group in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention.

[0064] Figure 9(b) is a cloud map of the distribution of σ1 on the horizontal profile of the arch crown in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention, when analyzing the influence of the dam abutment thrust on the stability of the surrounding rock of the powerhouse cavern group.

[0065] Figure 9(c) is a cloud map of the distribution of σ1 on the horizontal profile of the turbine in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention, when analyzing the influence of the dam abutment thrust on the stability of the surrounding rock of the powerhouse cavern group.

[0066] Figure 9(d) is a cloud map of the distribution of σ1 on the horizontal profile at the bottom of the powerhouse during step S3 of the nonlinear inversion analysis method for the geostress field of the underground cavern group of a hydropower station according to the present invention, when analyzing the influence of the dam abutment thrust on the stability of the surrounding rock of the powerhouse cavern group.

[0067] Figure 10This is a schematic diagram of the dam abutment thrust + normal water level pressure under working condition 1, during the analysis of the stress change in the power plant area under reservoir impoundment in step S3 of a nonlinear inversion analysis method for the geostress field of underground cavern group of a hydropower station according to the present invention.

[0068] Figure 11(a) is a schematic diagram of the displacement of the surrounding rock in the power plant area under working condition 1 during step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention, when analyzing the change of ground stress in the power plant area under reservoir impoundment. The unit is meters.

[0069] Figure 11(b) is a schematic diagram of the first principal stress σ1 in the power plant area under working condition 1 during step S3 of the nonlinear inversion analysis method for the ground stress field of underground cavern group of a hydropower station according to the present invention, when analyzing the changes in ground stress in the power plant area under reservoir impoundment. The unit is Pa.

[0070] Figure 11(c) is a schematic diagram of the third principal stress σ3 in the power plant area under working condition 1 during the analysis of the stress change in the power plant area under reservoir impoundment in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention. The unit is Pa.

[0071] Figure 12 In step S3 of the nonlinear inversion analysis method for the geostress field of underground cavern group of a hydropower station according to the present invention, when analyzing the change of geostress in the power plant area under reservoir impoundment, under working condition 2, a schematic diagram of dam shoulder thrust + low water level pressure is shown.

[0072] Figure 13(a) is a schematic diagram of the displacement of the surrounding rock in the power plant area under working condition 2 during step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention, when analyzing the change of ground stress in the power plant area under reservoir impoundment. The unit is meters.

[0073] Figure 13(b) is a schematic diagram of the first principal stress σ1 in the power plant area under working condition 2 during the analysis of the stress change in the power plant area under reservoir impoundment in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention. The unit is Pa.

[0074] Figure 13(c) is a schematic diagram of the third principal stress σ3 in the power plant area under working condition 2 during the analysis of the stress change in the power plant area under reservoir impoundment in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention. The unit is Pa.

[0075] Figure 14(a) is a schematic diagram of the displacement of the surrounding rock in the power plant area under working condition 3 in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention, when analyzing the change of ground stress in the power plant area under reservoir impoundment. The unit is meters.

[0076] Figure 14(b) is a schematic diagram of the first principal stress σ1 in the power plant area under working condition 3 during the analysis of the stress change in the power plant area under reservoir impoundment in step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention. The unit is Pa.

[0077] Figure 14(c) is a schematic diagram of the third principal stress σ3 in the power plant area under working condition 3 during step S3 of the nonlinear inversion analysis method of the underground cavern group of a hydropower station according to the present invention, when analyzing the change of ground stress in the power plant area under reservoir impoundment. The unit is Pa. Detailed Implementation

[0078] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0079] This invention provides a nonlinear inversion analysis method for the geostress field of underground cavern groups in hydropower stations, as described below. Figures 1 to 1 4. Provide a detailed explanation.

[0080] A nonlinear inversion analysis method for the in-situ stress field of underground cavern groups in a hydropower station includes the following steps:

[0081] S1. Based on the regional geological structure and measured in-situ stress analysis, the geological structure of the engineering area where the hydropower station is located is analyzed, and in-situ in-situ stress measurements are carried out at multiple measuring points in the plant area to obtain the local geological structure of the underground cavern group area.

[0082] S2. Based on artificial neural networks, nonlinear inversion of the ground stress field in the hydropower plant area is performed, and a three-dimensional numerical model is established to obtain the ground stress distribution characteristics of the hydropower plant area.

[0083] S3. Analysis of the geostress field characteristics of the cavern area, including the geostress values ​​in the cavern group area, the geostress direction in the cavern group area, the influence of typical faults on the geostress in the plant area, and the changes in geostress in the plant area under reservoir impoundment.

[0084] S4. Conduct on-site verification and analysis of the geostress of the cavern group.

[0085] This invention provides a nonlinear inversion method for geostress fields based on artificial neural network theory and geostress field inversion method. It can more accurately calculate and analyze the geostress field distribution characteristics of underground multi-cavity groups in hydropower stations, while reducing the influence of human factors and arbitrariness, and greatly improving the accuracy of geostress field inversion.

[0086] In some embodiments, S1 specifically includes:

[0087] S11. Based on regional tectonics, the geological structure of the hydropower station site is analyzed according to the geostress distribution in the World Stress Map to obtain a preliminary judgment on the maximum horizontal direction of geostress in the hydropower station area.

[0088] S12. Conduct structural exploration of the hydropower station project area, determine the main underground faults at the dam site, and select multiple measuring points at different locations and depths in the site area to conduct in-situ stress tests.

[0089] Furthermore, in S12, when determining the fault, it is necessary to determine the number, occurrence, fracture zone width, fault characteristics, and exposure location of the faults below the plant site.

[0090] Through geological surveys of the excavated adits and surface areas in the underground cavern complex area, the local geological structure of the underground cavern complex area was obtained. In order to determine the distribution of in-situ stress in the underground cavern complex area, multiple measuring points were selected at different locations and depths in the plant site area to conduct in-situ stress tests. The testing methods included stress relief method and hydraulic fracturing method.

[0091] By testing the distribution of ground stress at multiple points, the direction and mean of the maximum principal stress in the area where the underground cavern group powerhouse is located are analyzed, and a preliminary analysis of the ground stress situation in the hub area is conducted.

[0092] In some embodiments, during nonlinear inversion of geostress, to establish a nonlinear mapping between the model boundary conditions and the stress values ​​at measured geostress points within the model, the nonlinear relationship between the boundary conditions to be inverted and the stress values ​​can be described using a set of neural networks:

[0093]

[0094] In the formula, P = (p1, p2, ... p n ) represents the input node representation of the neural network; D = (d1, d2, ... d n ) is the output node representation of a neural network; NN(n,h1,...h p (m) represents the established multi-layer neural network structure; where n is the input layer F. x The number of nodes, h1 to h p Corresponding to hidden layers F1 to F2 respectively p The number of nodes, m is the output layer F y The number of nodes.

[0095] In some embodiments, the nonlinear inversion steps in S2 include:

[0096] S21. Establish a three-dimensional initial geostress field calculation model based on the geological and geomorphological characteristics of the hydropower station site to provide a stress distribution model basis for nonlinear inversion.

[0097] S22, Rock mass elastoplastic constitutive model and calculation parameters: Construct parameter combination tables for boundary load conditions according to orthogonal or uniform design principles, and obtain the calculated geostress value information of the measurement location by using numerical calculation software such as ANSYS and FLAC or other calculation tools, thereby establishing samples for neural network learning and training.

[0098] S23. Nonlinear regression calculation: The calculated ground stress value is used as the input vector of the network, and the boundary conditions are used as the output vector of the neural network to train the neural network.

[0099] S24. When a mature network structure and training times are obtained, the stress values ​​of the measured stress points are used as the input vector, and the output vector obtained by the neural network is used as the boundary conditions that can be adopted.

[0100] S25. Substitute the obtained boundary conditions into the numerical calculation software to perform a forward calculation to obtain the regional geostress field.

[0101] In some embodiments, in S21, considering that the main purpose of in-situ stress inversion is to obtain the characteristics of in-situ stress distribution in the plant area, solid elements are used for each fault in the three-dimensional initial in-situ stress field calculation model, such as... Figure 1 - As shown in Figure 2.

[0102] In some embodiments, during S22, the boundary of the hypothetical engineering region is gradually applied in the numerical calculation. In the initial stage of rock mass deformation, each rock mass element is in an elastic state, and the limit of this elastic state is called the yield condition. When the stress state of the rock mass element reaches this limit, further loading may cause irreversible plastic deformation. For rock masses in hydropower station areas that have undergone varying degrees of geological processes, this process is accompanied by varying degrees of elastoplastic deformation. These rock masses will experience compressive-shear and tensile failures during geological changes. Therefore, an elastoplastic calculation method is used to calculate the rock mass parameters. Furthermore, during the calculation, the corresponding mechanical parameters adopt the design recommendation values, as shown in the table below.

[0103] Table 1 Recommended values ​​for physical and mechanical parameters of surrounding rock in underground caverns

[0104]

[0105] In some embodiments, in S23, the nonlinear regression calculation of the geostress field adopts a comprehensive method based on uniform design, artificial neural network, and finite difference. This method fully utilizes the advantages of neural network nonlinear mapping and organically combines neural networks with finite difference numerical forward calculation methods. Specifically, it includes:

[0106] S231. Based on uniform design, the sample construction is carried out. According to the measured geostress characteristics and engineering geological analysis, the uniform design method is adopted to make a sample horizontal uniform design table by combining the numerically calculated displacement boundary conditions and gravity correction coefficients, and multiple sets of learning and training samples are combined for neural network learning and training.

[0107] In one embodiment, seven parameters are set for the displacement boundary conditions and gravity correction coefficient, each divided into 23 levels (as shown in Table 2), using U... 23 * (twenty three 7 A uniform design table was used to construct 23 sets of learning and training sample test combinations (as shown in Table 3), and 3 sets of samples were used to test the network prediction effect (as shown in Table 4).

[0108] Table 2 Sample Level Design Table

[0109]

[0110]

[0111] Table 3 Neural Network Learning and Training Samples

[0112]

[0113]

[0114] Table 4 Neural Network Test Samples

[0115]

[0116] S232. Input the sample schemes of each group based on the boundary conditions obtained by uniform design into the finite difference software for calculation, obtain the stress calculation values ​​at multiple measurement points of each sample, use them as the learning and training input values ​​of the neural network, use the set boundary conditions as the corresponding output values, establish an evolutionary neural network model of the nonlinear mapping relationship between the stress components of the measurement points and the boundary conditions, and use a genetic algorithm to optimize the optimal network structure and connection weights of the neural network.

[0117] S233. Using the trained evolutionary neural network model, the measured stress component values ​​of each measuring point were input to obtain the displacement boundary conditions and gravity correction coefficient of the geostress field.

[0118] In one embodiment, the sample schemes of boundary conditions obtained from the above uniform design are substituted into finite difference software for calculation to obtain the stress calculation values ​​at 6 measuring points of each sample. These values ​​are used as the learning and training input values ​​of the neural network, while the set boundary conditions are used as the corresponding output values. An evolutionary neural network model of the nonlinear mapping relationship between the stress components of the measuring points and the boundary conditions is established. In this process, a genetic algorithm is used to optimize the optimal network structure and connection weights of the neural network. During the calculation, the optimal network structure of the neural network is 156-52-26-7, that is, the input layer has 156 nodes, the intermediate hidden layers have two layers, the first layer has 52 nodes, the second layer has 26 nodes, and the final output layer has 7 nodes. Using this trained evolutionary neural network model, the measured stress component values ​​of each measuring point are input to obtain the displacement boundary conditions and gravity correction coefficients of the geostress field, as shown in Table 5.

[0119] Table 5. Displacement boundary conditions and gravity correction coefficients obtained from evolutionary neural networks.

[0120]

[0121] S234. The boundary conditions obtained by nonlinear regression inversion are calculated again using finite difference forward calculation, thereby obtaining the distribution characteristics of the geostress field in the underground engineering area of ​​the hydropower station. The specific calculation results are shown in Figures 3(a)-(c), which respectively show the distribution of the three principal stresses in the geostress field, namely the first principal stress in Figure 3(a), the second principal stress in Figure 3(b), and the third principal stress in Figure 3(c).

[0122] S235. Conduct on-site testing of the stress values ​​at each measuring point and compare them with the calculated results to determine the rationality of the geostress field distribution characteristics. If the nonlinear inversion results and the measured geostress values ​​agree well and maintain consistency in their patterns, it indicates that the geostress field distribution obtained through inversion is reasonable.

[0123] In some embodiments, during step S3, when analyzing the stress characteristics of the underground cavern group area, it is necessary to analyze the characteristics, magnitude, and distribution patterns of the stress field in the hydropower station engineering area based on the stress distribution characteristics obtained from the nonlinear inversion of the stress field. Specific analysis areas within the engineering area include: vertical cross-sectional views of the axes on both the upstream and downstream sides of the underground cavern group (vertical cross-sections of the powerhouse and tailrace chamber axes), horizontal cross-sections of the top and bottom of the underground cavern group, and horizontal cross-sections of the middle section of the underground cavern group. Numerical stress distribution maps of the powerhouse axis and the tailrace chamber axis are obtained respectively. In one embodiment, stress analysis diagrams are obtained as shown in Figures 4 and 5.

[0124] In some embodiments, when analyzing the geostress direction in a cavern complex area, the geostress direction distribution at typical locations within the cavern complex is obtained based on the inversion results of the initial geostress field in the cavern complex area. This is illustrated in Figure 6.

[0125] In some embodiments, when analyzing the impact of typical faults on the ground stress in the plant area, numerical simulation is used to calculate and analyze the distribution of the ground stress field in the plant area under the condition of no fault, and to obtain the distribution of ground stress on the vertical cross-section of the axis on both the upstream and downstream sides of the underground cavern group under the condition of no fault; as shown in Figure 7 in some embodiments.

[0126] By comparing the distribution of geostress on the vertical cross-section of the upstream and downstream sides of the cavern group under fault conditions (as shown in Figure 5), the influence of typical faults on the geostress in the plant area can be obtained.

[0127] During engineering construction, the impact of underground cavern groups and dam construction on the in-situ stress of rock masses near faults should be monitored, and timely safety assessments of the surrounding rock of underground caverns should be conducted.

[0128] In some embodiments, when analyzing the impact of abutment thrust on the stability of the surrounding rock of the powerhouse cavern group, a three-dimensional finite element model is established based on the characteristics of the geostress distribution in the hydropower station area, the axis of the underground powerhouse caverns, the cavern spacing, the excavation sequence, and the initial support scheme. Thrust is applied at the corresponding abutment positions of the arch dam to obtain a schematic diagram of the arch thrust at the abutment. In one embodiment, the schematic diagram of the arch thrust at the abutment is as follows: Figure 8 As shown in Figure 9, the displacement and stress at characteristic points of the cross-sections of each unit (unit refers to equipment in the underground cavern) in the underground cavern are calculated after the application of the dam abutment arch thrust. The displacement and stress at characteristic points of each unit's cross-section are compared between cases considering and not considering the dam abutment thrust to obtain the influence of the dam abutment thrust on the stability of the underground cavern. In some embodiments, the calculation results of the influence of the dam abutment thrust on the stability of the underground cavern are shown in Figure 9, and the numerical results are shown in the table below.

[0129] Table 6 Summary of Calculation Results for Typical Sections Before and After Applying Thrust to Dam Abutments

[0130]

[0131]

[0132]

[0133]

[0134] As shown in Table 6 and Figure 9, the compressive stress generated by the arch thrust decreases rapidly with increasing depth. The recommended influence value of the arch thrust compressive stress in the affected area of ​​the main powerhouse cavern at the recommended site is approximately 0.05 MPa, accounting for only 0.2% of the inverted ground stress (approximately 25 MPa at the arch crown). This proportion is very small and has minimal impact on the overall stability of the cavern surrounding rock. Table 6 shows that after the underground cavern excavation, considering the dam abutment thrust, the displacement of the surrounding rock of the underground cavern changes very little compared to the displacement without considering the dam abutment thrust, with a maximum change of less than 1 mm. For the cavern surrounding stress in both cases, the major principal stress increases slightly after considering the arch thrust, while the minor principal stress changes very little. Therefore, it can be considered that the dam abutment thrust has a very small impact on the stability of the surrounding rock of the underground cavern, and the distance between the underground cavern and the dam is appropriate.

[0135] In some embodiments, when analyzing the changes in ground stress in the power plant area under reservoir impoundment, a three-dimensional numerical simulation method is used to calculate and analyze the displacement and stress distribution of the power plant area under normal impoundment water level and low water level conditions, while simultaneously considering the influence of dam abutment thrust. It is necessary to analyze the changes in ground stress in the power plant area under reservoir impoundment under three operating conditions, including:

[0136] Operating condition 1: Dam abutment thrust + water pressure at normal water level (water surface elevation 2535m);

[0137] Condition 2: Dam abutment thrust + low water level (water surface elevation 2525m) water pressure;

[0138] Condition 3, Dam abutment thrust;

[0139] The parts that need to be calculated and analyzed include: the crown, upstream sidewall, downstream sidewall, and wall ends;

[0140] The parameters that need to be calculated for each part under each working condition include: displacement, maximum principal stress, minimum principal stress, anchor stress, and anchor cable load; for the wall end part, only the anchor stress needs to be calculated.

[0141] In one embodiment, the analysis under various operating conditions is as follows.

[0142] Operating condition 1: Dam abutment thrust + water pressure at normal water level (water surface elevation 2535m). A schematic diagram of the water pressure under this condition is shown below. Figure 10 As shown in Figure 11, the displacement and stress distribution of the surrounding rock in the plant area under this working condition were calculated. The maximum pressure in the reservoir area under normal water level is 2.63 MPa. Under this working condition, the displacement of the surrounding rock in the plant area ranges from 9.1 to 64.7 mm, the maximum principal stress ranges from -6.6 to -54.5 MPa, and the minimum principal stress ranges from -18.1 to 1.0 MPa.

[0143] Condition 2: Dam abutment thrust + low water level (water surface elevation 2525m) water pressure. A schematic diagram of the water pressure under this condition is shown below. Figure 12 As shown in Figure 13, the displacement and stress distribution of the surrounding rock in the plant area under this working condition were calculated. At the low water level (water surface elevation 2525m), the maximum water pressure in the reservoir area is 2.53MPa. Under this working condition, the displacement of the surrounding rock in the plant area ranges from 9.0 to 64.6 mm, the maximum principal stress ranges from -6.6 to -54.3 MPa, and the minimum principal stress ranges from -18.2 to 1.0 MPa.

[0144] Under working condition 3, with dam abutment thrust plus low water level (water surface elevation 2525m) water pressure, the displacement and stress distribution of the surrounding rock in the plant area were calculated. The stress distribution in the plant area is shown in Figure 14. Under this condition, the displacement of the surrounding rock in the plant area is 8.8–64.3 mm, the maximum principal stress is -6.5–-54.0 MPa, and the minimum principal stress is -18.3–0.96 MPa.

[0145] Table 7 Summary of Plant Area Calculation Results under Three Operating Conditions

[0146]

[0147] Table 7 presents the calculation results for the plant area under three operating conditions in this embodiment. As shown in the table, after the reservoir is filled, the stress on the surrounding rock in the plant area increases slightly with the rise in water level. Among the three operating conditions, the displacement and stress of the surrounding rock in the plant area are greatest under condition 1 (dam abutment thrust + water pressure at normal water level (water surface elevation 2535m)). The displacement of the surrounding rock is 9.1–64.7 mm, the maximum principal stress is -6.6–-54.5 MPa, and the minimum principal stress is -18.1–1.0 MPa. Compared to condition 3 (dam abutment thrust), the displacement of the surrounding rock in condition 1 increased by 0.1–0.4 mm, the maximum principal stress increased by 0.1–0.5 MPa, and the minimum principal stress increased by 0.0–0.2 MPa. In condition 2, the displacement of the surrounding rock increased by 0.1–0.3 mm, the maximum principal stress increased by 0.1–0.3 MPa, and the minimum principal stress increased by 0.0–0.1 MPa. Under condition 1, the increase in the maximum principal stress of the surrounding rock in the plant area is only 0.9% to 1.1% of that under condition 3, which is a very small proportion. It can be seen that the reservoir impoundment has little impact on the overall stability of the surrounding rock of the cavern.

[0148] In some embodiments, during S4, excavation verification is performed in the cavern group area, and the nonlinear inversion results of the underground cavern group are compared with the field excavation verification results to verify the effectiveness of the geostress inversion results.

[0149] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0150] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0151] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0152] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0153] In this invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0154] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station, characterized in that, Includes the following steps: S1. Based on the regional geological structure and measured in-situ stress analysis, the geological structure of the engineering area where the hydropower station is located is analyzed, and in-situ in-situ stress measurements are carried out at multiple measuring points in the plant area to obtain the local geological structure of the underground cavern group area. S2. Based on artificial neural networks, nonlinear inversion of the ground stress field in the hydropower plant area is performed, and a three-dimensional numerical model is established to obtain the ground stress distribution characteristics of the hydropower plant area. S3. Analysis of the geostress field characteristics of the cavern area, including the geostress values ​​of the cavern group area, the geostress direction of the cavern group area, the influence of typical faults on the geostress of the plant area, and the changes in geostress of the plant area under reservoir impoundment. S4. Conduct on-site verification and analysis of the geostress of the cavern group.

2. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 1, characterized in that, Specifically, S1 includes: S11. Based on regional tectonic analysis of geostress field, the geological structure of the hydropower station site is analyzed according to the geostress distribution in the World Stress Map, and a preliminary judgment is made on the maximum horizontal direction of geostress in the hydropower station area. S12. Conduct structural exploration of the hydropower station project area, determine the main underground faults at the dam site, and select multiple measuring points at different locations and depths in the site area to conduct in-situ stress tests.

3. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 2, characterized in that, In step S12, when determining faults, it is necessary to determine the number, occurrence, fracture zone width, fault characteristics, and exposure location of faults below the plant site.

4. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 1, characterized in that, In performing nonlinear inversion of geostress, to establish a nonlinear mapping between the model boundary conditions and the stress values ​​at measured geostress points within the model, the nonlinear relationship between the boundary conditions to be inverted and the stress values ​​can be described using a set of neural networks: In the formula, P = (p1, p2, ... p n ) represents the input node representation of the neural network; D = (d1, d2, ... d n ) is the output node representation of a neural network; NN(n,h1,...h p (m) represents the established multi-layer neural network structure; where n is the input layer F. x The number of nodes, h1 to h p Corresponding to hidden layers F1 to F2 respectively p The number of nodes, m is the output layer F y The number of nodes.

5. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 1, characterized in that, In step S2, the specific steps of the nonlinear inversion include: S21. Establish a three-dimensional initial geostress field calculation model based on the geological and geomorphological characteristics of the hydropower station site to provide a stress distribution model basis for nonlinear inversion. S22, Rock mass elastoplastic constitutive model and calculation parameters: Construct a parameter combination table for boundary load conditions according to the orthogonal or uniform design principle, and obtain the calculated geostress value information of the measurement location through numerical calculation software or other calculation tools, thereby establishing samples for neural network learning and training; S23. Nonlinear regression calculation: The calculated ground stress value is used as the input vector of the network, and the boundary conditions are used as the output vector of the neural network to train the neural network. S24. When a mature network structure and training times are obtained, the stress values ​​of the measured stress points are used as the input vector, and the output vector obtained by the neural network is used as the boundary conditions that can be adopted. S25. Substitute the obtained boundary conditions into the numerical calculation software to perform a forward calculation to obtain the regional geostress field.

6. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 5, characterized in that, In S21, solid elements are used for each fault in the three-dimensional initial geostress field calculation model.

7. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 5, characterized in that, In step S22, the numerical calculation software used is ANSYS or FLAC numerical calculation software.

8. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 5, characterized in that, In S22, for rock masses in the area where the hydropower station is located that have undergone geological processes of varying degrees, the rock mass will experience compression-shear and tension failures during geological changes. Therefore, an elastoplastic calculation method is used when calculating the rock mass parameters.

9. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 5, characterized in that, In S23, the nonlinear regression calculation of the geostress field adopts a comprehensive method based on uniform design, artificial neural network, and finite difference, specifically including: S231. Based on uniform design, the sample construction is carried out. According to the measured geostress characteristics and engineering geological analysis, the uniform design method is adopted to make a sample horizontal uniform design table by combining the numerically calculated displacement boundary conditions and gravity correction coefficients, and multiple sets of learning and training samples are combined for neural network learning and training. S232. Input the sample schemes of each group based on the boundary conditions obtained by uniform design into the finite difference software for calculation, obtain the stress calculation values ​​at multiple measurement points of each sample, use them as the learning and training input values ​​of the neural network, use the set boundary conditions as the corresponding output values, establish an evolutionary neural network model of the nonlinear mapping relationship between the stress components of the measurement points and the boundary conditions, and use a genetic algorithm to optimize the optimal network structure and connection weights of the neural network. S233. Using the trained evolutionary neural network model, the measured stress component values ​​of each measuring point were input to obtain the displacement boundary conditions and gravity correction coefficient of the geostress field. S234. The boundary conditions obtained by nonlinear regression inversion are calculated again using finite difference forward calculation, thereby obtaining the distribution characteristics of the geostress field in the underground engineering area of ​​the hydropower station. S235. Conduct on-site testing of the stress values ​​at each measuring point and compare them with the calculation results to determine the rationality of the distribution characteristics of the geostress field.

10. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 1, characterized in that, In S3, when analyzing the stress characteristics of the underground cavern group area, it is necessary to analyze the characteristics, magnitude and distribution law of the stress field in the hydropower station project area based on the stress distribution characteristics of the hydropower station plant area obtained by nonlinear inversion of the stress field. The specific analysis parts of the project area include: the vertical cross-section of the axis on both the upstream and downstream sides of the underground cavern group, the horizontal cross-section of the top and bottom of the underground cavern group, and the horizontal cross-section of the middle part of the underground cavern group.

11. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 10, characterized in that, When analyzing the impact of typical faults on the ground stress in the plant area, numerical simulation is used to calculate and analyze the distribution of the ground stress field in the plant area under the condition of no fault, and to obtain the distribution of ground stress on the vertical cross-section of the axial direction on both the upstream and downstream sides of the underground cavern group under the condition of no fault. By comparing the distribution of geostress on the vertical cross-sections of the upstream and downstream sides of the cavern group under fault conditions, the influence of typical faults on the geostress in the plant area can be obtained. During engineering construction, the impact of underground cavern groups and dam construction on the in-situ stress of rock masses near faults should be monitored, and timely safety assessments of the surrounding rock of underground caverns should be conducted.

12. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 10, characterized in that, When analyzing the impact of dam abutment thrust on the stability of the powerhouse cavern group, a three-dimensional finite element model was established based on the characteristics of the ground stress distribution in the hydropower station area, the axis of the underground powerhouse cavern, the cavern spacing, the excavation sequence, and the initial support scheme. Thrust was applied at the corresponding dam abutment positions, and the displacement and stress of the characteristic points of each unit section in the underground cavern were calculated after the dam abutment thrust was applied. The displacement and stress of the characteristic points of each unit section were compared between the cases considering dam abutment thrust and the cases not considering dam abutment thrust, so as to obtain the impact of dam abutment thrust on the stability of the underground cavern.

13. The nonlinear inversion analysis method for the geostress field of underground cavern groups in a hydropower station as described in claim 1, characterized in that, When analyzing the changes in ground stress in the power plant area under reservoir impoundment, a three-dimensional numerical simulation method was used to calculate and analyze the displacement and stress distribution of the power plant area under normal impoundment water level and low water level conditions. The influence of dam abutment thrust was also considered. The analysis of ground stress changes in the power plant area under reservoir impoundment under three operating conditions is required, including: Operating condition 1: Dam abutment thrust + water pressure at normal water level (water surface elevation 2535m); Condition 2: Dam abutment thrust + low water level (water surface elevation 2525m) water pressure; Condition 3, Dam abutment thrust; The parts that need to be calculated and analyzed include: the crown, upstream sidewall, downstream sidewall, and wall ends; The parameters that need to be calculated for each part under each working condition include: displacement, maximum principal stress, minimum principal stress, anchor stress, and anchor cable load; for the wall end part, only the anchor stress needs to be calculated.

Images