High ground stress tunnel surrounding rock stability analysis method and device

Abstract

The application provides a high ground stress tunnel surrounding rock stability analysis method and device, relates to the technical field of tunnel safety, and comprises the following steps: acquiring tunnel information, wherein the tunnel information comprises section information, historical monitoring data and real-time monitoring data; performing numerical simulation according to multi-dimensional characteristics in the tunnel information and extracting characteristic parameters; optimizing a random forest regression model based on a chaotic sparrow search algorithm, inputting the characteristic parameters into the optimized regression model for prediction, and obtaining surrounding rock prediction results; inputting parameters of the surrounding rock prediction results into a preset surrounding rock physical model to calculate a potential energy function, performing cusp catastrophe analysis on the potential energy function, and constructing a bifurcation set equation; performing state discrimination based on the bifurcation set equation and a preset surrounding rock instability constraint condition, and outputting a stability result. The safety problems of tunnel face instability convergence deformation, collapse and peeling caused by surrounding rock instability are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel safety, in particular to a high ground stress tunnel surrounding rock stability analysis method and device. BACKGROUND

[0002] In the field of tunnel safety technology, the construction process of high ground stress tunnel faces many complex challenges, which are mainly rooted in the extreme complexity of geological conditions, the significant looseness of rock structure, and the shortcomings in construction factors and support measures. Specifically, the complex complexity of geological structure and the instability of rock stratum, when the surrounding environment is disturbed by excavation operation, if the design or implementation of support measures is improper, combined with the selection of site construction parameters being limited by actual working conditions, and lacking sufficient theoretical support, the risk of surrounding rock stability will be greatly increased, leading to the safety problems of surrounding rock instability, tunnel face instability convergence deformation, collapse and spalling.

[0003] There is an urgent need for a high ground stress tunnel surrounding rock stability analysis method and device to solve the safety problems of surrounding rock instability, tunnel face instability convergence deformation, collapse and spalling. SUMMARY

[0004] The purpose of the present application is to provide a high ground stress tunnel surrounding rock stability analysis method and device to improve the above problems. In order to achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0005] In a first aspect, the present application provides a high ground stress tunnel surrounding rock stability analysis method, comprising:

[0006] Obtaining tunnel information, the tunnel information including cross-section information, historical monitoring data and real-time monitoring data;

[0007] Numerical simulation is carried out according to the multi-dimensional characteristics in the tunnel information, and characteristic parameters are extracted;

[0008] The random forest regression model is optimized based on the chaotic sparrow search algorithm, the characteristic parameters are input into the optimized regression model for prediction, and the surrounding rock prediction result is obtained;

[0009] The parameters of the surrounding rock prediction result are input into a preset surrounding rock physical model to calculate a potential energy function, a cusp catastrophe analysis is carried out on the potential energy function, and a bifurcation set equation is constructed;

[0010] State discrimination is carried out based on the bifurcation set equation and a preset surrounding rock instability constraint condition, and a stability result is output.

[0011] In a second aspect, the present application further provides a high ground stress tunnel surrounding rock stability analysis device, comprising:

[0012] The processor, the memory and the I / O interface;

[0013] The processor runs the acquisition module, the simulation module, the prediction module, the construction module and the judgment module in logic association;

[0014] The acquisition module is used for acquiring tunnel information and writing the tunnel information into the memory through the I / O interface, wherein the tunnel information includes section information, historical monitoring data and real-time monitoring data;

[0015] The simulation module is used for calling the tunnel information in the memory, performing numerical simulation according to multi-dimensional features in the tunnel information, extracting feature parameters and storing the feature parameters back to the memory;

[0016] The prediction module is used for calling the feature parameters in the memory, optimizing a random forest regression model based on a chaotic sparrow search algorithm, inputting the feature parameters into the optimized regression model for prediction, obtaining a surrounding rock prediction result and storing the surrounding rock prediction result back to the memory;

[0017] The construction module is used for calling the surrounding rock prediction result in the memory, inputting parameters of the surrounding rock prediction result into a preset surrounding rock physical model to calculate a potential energy function, performing cusp catastrophe analysis on the potential energy function, constructing a bifurcation set equation and storing the bifurcation set equation back to the memory;

[0018] The judgment module is used for calling the bifurcation set equation in the memory, performing state discrimination based on the bifurcation set equation and a preset surrounding rock instability constraint condition, feeding back a discrimination result to a terminal through the I / O interface and outputting a stability result.

[0019] The present application has the following advantages:

[0020] The present application has the following advantages:

[0021] Other features and advantages of the present application will be described in the following description, and some will become apparent from the description, or will be learned through practice of the present application. The objects and other advantages of the present application will be achieved and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the process for analyzing the stability of surrounding rock in high-stress tunnels as described in this embodiment of the invention.

[0024] Figure 2 This is a schematic diagram of the structure of the high-stress tunnel surrounding rock stability analysis device described in an embodiment of the present invention.

[0025] The markings in the diagram are: 800, High-stress tunnel surrounding rock stability analysis device; 801, Processor; 802, Memory; 803, Multimedia component; 804, I / O interface; 805, Communication component. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0027] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0028] Example 1:

[0029] This embodiment provides a method for analyzing the stability of surrounding rock in tunnels under high ground stress.

[0030] See Figure 1 The figure shows that the method includes steps S1 to S5, including:

[0031] S1: obtaining tunnel information, the tunnel information including section information, historical monitoring data and real-time monitoring data;

[0032] S2: performing numerical simulation according to multi-dimensional features in the tunnel information, and extracting feature parameters;

[0033] To make the specific acquisition method of the feature parameters, step S2 includes S21 to S23, specifically:

[0034] S21: constructing a model according to geological conditions and structural features in the tunnel information to obtain a numerical model;

[0035] In this step, the numerical model is used to simulate the mechanical behavior of the tunnel surrounding rock, reflecting the actual mechanical behavior of the tunnel surrounding rock;

[0036] The geological conditions include lithology, ground stress, underground water, rock type, geological structure, physical and mechanical properties of rock mass, etc., and the structural features include tunnel section shape, size, lining structure, support structure, etc.

[0037] S22: preprocessing the tunnel information, inputting the preprocessed tunnel information into the numerical model to set boundary conditions and initial conditions for numerical simulation, and obtaining numerical simulation results;

[0038] In this step, the numerical simulation operation is to calculate through a numerical simulation software to simulate the mechanical behavior of the tunnel under various conditions, and finally obtain numerical simulation results. The boundary conditions and the initial conditions can make the numerical simulation more consistent with the actual engineering situation, and thus obtain reliable simulation results.

[0039] The preprocessing includes data cleaning, standardization, removal of outliers, etc., the boundary conditions include displacement boundary conditions, stress boundary conditions, etc., and the initial conditions include initial ground stress field, etc.

[0040] S23: extracting multi-dimensional features from the numerical simulation results, processing the multi-dimensional features through dimension reduction, combining the reduced features with the numerical simulation results to extract surrounding rock stability feature parameters, and obtaining feature parameters.

[0041] In this step, the reduced features and the numerical simulation results are further analyzed and extracted to obtain feature parameters closely related to the stability of the surrounding rock, such as the maximum principal stress of the surrounding rock, the maximum displacement value, the displacement change rate, etc. The feature parameters directly reflect the stability state of the surrounding rock, and the multi-dimensional features reflect the mechanical state of the tunnel surrounding rock from different angles.

[0042] Preferably, the multiple dimensional features include stress, displacement, strain and the like of different positions of the surrounding rock over time; and the dimension reduction processing includes principal component analysis (PCA) or linear discriminant analysis.

[0043] S3: optimizing the random forest regression model based on the chaotic sparrow search algorithm, and predicting the surrounding rock prediction result by inputting the feature parameters into the optimized regression model;

[0044] S31: optimizing the random forest regression model based on the chaotic sparrow search algorithm to obtain an optimized regression model;

[0045] In the optimization process, the chaotic sparrow search algorithm initializes the population through chaotic mapping, thereby enhancing the global search capability. The chaotic sequence of the chaotic sparrow search algorithm is used to generate initial candidate solutions of hyperparameters, and then the node splitting feature number and tree depth and other key parameters of the random forest are optimized. After optimization, the root mean square error of the prediction result of the random forest regression model is reduced by 23%, and the optimization efficiency is improved by 40%. The improvement effectively solves the problem of insufficient exploration of traditional optimization algorithms in complex geological parameter space.

[0046] To clarify the specific way of obtaining the optimized regression model, steps S311 to S314 are included in step S31, specifically:

[0047] S311: generating an initial population based on chaotic mapping in the chaotic sparrow search algorithm, and calculating the initial population through the prediction performance indicators of the random forest regression model to obtain the fitness function value;

[0048] In this step, the chaotic mapping has the characteristics of nonlinearity, ergodicity and randomness, which is used to expand the search range and increase the diversity of the population. By generating an initial population through chaotic mapping, compared with the traditional random generation method, the problem of poor quality of the initial population is avoided.

[0049] S312: updating the position of the initial population according to the fitness function value to obtain the optimized population position;

[0050] In this step, the position of the individual in the initial population is updated according to the size of the fitness function value. The position updating mode of the individual with a high fitness function value will make it closer to the global optimal solution; the individual with a low fitness function value will be adjusted through certain strategies to improve its fitness.

[0051] S313: performing global search optimization based on the optimized population position and the sine-cosine algorithm to obtain the global optimal solution;

[0052] In this step, in the global search optimization process, the sine-cosine algorithm will guide the individuals in the population to approach the global optimal solution according to the current optimization population position, and finally obtain the global optimal solution. The periodicity and amplitude variation of the sine and cosine functions are used to update the position of individuals in the population, so as to realize global search optimization.

[0053] S314: updating the population position according to the global optimal solution to obtain a new population position, retraining the random forest regression model through the new population position, and judging the prediction performance index of the retrained model against the preset constraint condition, and outputting the optimized regression model when the prediction performance index of the retrained model meets the preset constraint condition.

[0054] In this step, the global optimal solution is used as a reference to update the position of individuals in the population to obtain a new population position. This process makes the individuals in the population closer to the global optimal solution, thereby improving the fitness of the entire population. The new population position is used as input to retrain the random forest regression model, and the prediction performance index of the retrained model is compared with the preset constraint condition to determine whether it meets the requirements. If the preset constraint condition is met, the optimized regression model is output; otherwise, the optimization process continues.

[0055] S32: inputting the feature parameters into the optimized regression model for prediction, comparing the prediction result with the preset actual observation value, and calculating an evaluation index to obtain an evaluation result;

[0056] In this step, the evaluation index is calculated to measure the difference between the prediction result and the actual observation value, which is used to judge the performance of the initial regression model.

[0057] The calculation of the evaluation index includes methods such as mean square error (MSE), root mean square error (RMSE), and mean absolute error (MAE).

[0058] S33: randomly generating multiple sets of hyperparameter combinations according to the evaluation result, and optimizing and adjusting the multiple sets of hyperparameter combinations through the self-adaptive weight clustering verification algorithm to obtain optimized hyperparameters.

[0059] To clarify the specific acquisition method of the optimized hyperparameters, steps S33 include S331 to S333, which are as follows:

[0060] S331: determining the value range of the hyperparameters of the optimized regression model according to the evaluation result, and randomly sampling multiple sets of hyperparameters within the determined value range through a random search method.

[0061] In this step, the performance of the optimized regression model under different hyperparameters is analyzed based on the evaluation results to determine a reasonable range of values ​​for each hyperparameter. Then, within the determined range of hyperparameter values, multiple sets of different hyperparameter combinations are generated through multiple random samplings. Determining the range of hyperparameter values ​​can narrow the search space and improve search efficiency.

[0062] S332: Improve cross-validation based on adaptive weighted clustering validation algorithm by using multiple sets of hyperparameters to train the optimized regression model, and evaluate the performance of each set of hyperparameters through improved cross-validation to obtain performance evaluation results;

[0063] In this step, the adaptive weighted clustering verification algorithm adopts a static weight allocation mode, using the local density of the verification set samples as the weight benchmark, assigning a higher penalty coefficient to the prediction error of areas with significant rock mass deformation, dynamically adjusting the error contribution weight of each verification set, and focusing on the prediction accuracy of geological anomaly intervals.

[0064] Among them, adaptive error weights are assigned to the validation set samples through a weight calculation expression;

[0065] The weight calculation expression is as follows:

[0066] (1);

[0067] In the above formula (1), Indicates the first The weights of each sample, Indicates the first The actual value of each sample Sensitivity coefficient This represents the standard deviation of the error.

[0068] S333: Select the optimal hyperparameter combination based on the performance evaluation results to obtain the optimized hyperparameters.

[0069] In this step, the predictive performance of the regression model is improved by selecting the optimal combination of hyperparameters.

[0070] S34: Input the optimized hyperparameters into the regression model for optimization training, and predict the feature parameters through the optimized regression model to obtain the surrounding rock prediction result.

[0071] In this step, the optimized hyperparameters are input into the regression model, the regression model is retrained, and during the training process, the regression model adjusts its parameters according to the optimized hyperparameters, so that the model can better fit the data. After training, the previous feature parameters are input into the optimized regression model again, and the model will calculate according to the new parameters to output the prediction result of the surrounding rock state. Training and optimizing the regression model using the optimized hyperparameters can improve the prediction accuracy and stability of the regression model, and obtain more reliable surrounding rock prediction results.

[0072] In this step, the problems of insufficient prediction accuracy and unreasonable selection of hyperparameters in the traditional regression model are effectively solved.

[0073] S4: input the parameters of the surrounding rock prediction result into a preset surrounding rock physical model to calculate a potential energy function, and construct a bifurcation set equation by analyzing the cusp catastrophe of the potential energy function;

[0074] To clarify the specific acquisition method of the bifurcation set equation, step S4 includes S41 to S44, specifically:

[0075] S41: input the parameters of the surrounding rock prediction result into the surrounding rock physical model, simplify the non-circular tunnel section by the equivalent radius method, and obtain the section equivalent radius;

[0076] In this step, the parameters of the surrounding rock prediction result are input into the surrounding rock physical model, and the equivalent radius method is used to simplify the non-circular tunnel section, and the geometric characteristics of the non-circular section are converted into an equivalent circular section radius;

[0077] The section equivalent radius is:

[0078] (2);

[0079] In the above formula (2), represents the equivalent circle radius, represents the section shape correction coefficient, represents the tunnel section area.

[0080] S42: based on the mechanical equilibrium condition, solve the surrounding rock mechanical model through the section equivalent radius to obtain the mechanical properties of the surrounding rock;

[0081] To clarify the specific acquisition method of the mechanical properties of the surrounding rock, step S42 includes S421 to S423, specifically:

[0082] S421: based on the mechanical equilibrium condition of the rock mass, analyze the balance and geometric relationship of the surrounding rock mechanical model through the tunnel section equivalent radius to obtain the macro stress distribution;

[0083] In this step, the macro stress distribution is:

[0084] (3);

[0085] In the above formula (3), dR represents a small increment, represents a radial stress, represents a radial stress, represents a change in radial distance, represents a hoop stress, represents the current radius;

[0086] (4);

[0087] In the above formula (4), dR represents a small increment, represents a radial strain, represents a change in radial displacement, represents a change in radial distance;

[0088] (5);

[0089] In the above formula (5), dR represents a small increment, represents a hoop strain, represents a radial displacement, represents the current radius;

[0090] The macro stress distribution considers factors such as the geometry, size, and boundary conditions of the surrounding rock, and the force balance and geometric relationship in the analysis model.

[0091] S422: According to the macro stress distribution, analyze the stress change of the surrounding rock from a micro perspective, and construct a micro mechanical distribution;

[0092] In this step, the micro mechanical distribution is:

[0093] (6);

[0094] In the above formula (6), dR represents a small increment, represents the elastic modulus of the surrounding rock, represents the strain corresponding to the yield point, represents the stress of the material, represents the stress corresponding to the yield point, represents the deformation degree of the material, and represent undetermined constants;

[0095] The micro mechanical analysis considers features such as micro cracks and structural planes within the surrounding rock, which affect the transmission and distribution of stress.

[0096] S423: constructing based on the macro stress distribution and the micro mechanical distribution to obtain the surrounding rock mechanical properties.

[0097] In this step, the surrounding rock mechanical properties are used to describe the surrounding rock mechanical properties of the surrounding rock mechanical properties.

[0098] S43: constructing a deformation zone mechanical model according to the surrounding rock mechanical properties and the preset boundary conditions and calculating to obtain a total potential energy function;

[0099] To clarify the specific acquisition method of the total potential energy function, steps S431 to S433 are included in step S43, specifically:

[0100] S431: establishing according to the surrounding rock mechanical properties and the preset boundary conditions to obtain a stress-strain relationship model of the deformation zone, the deformation zone including a plastic zone and an elastic zone;

[0101] In this step, under the condition that the plastic zone is incompressible, the plastic zone satisfies:

[0102] (7);

[0103] In the above formula (7), represents the hoop strain, represents the radial strain, represents the strain in the third direction perpendicular to the radial and hoop directions, represents the internal strain, represents the initial displacement or the displacement amount under a certain specific condition, represents the current radius, represents the boundary radius of the plastic zone, represents the internal strain, represents the hoop stress, represents the radial stress; the plastic zone radial stress calculation formula is:

[0104] (8);

[0105] In the above formula (8), represents the plastic zone radial stress, represents the stress corresponding to the yield point, represents the initial displacement or the displacement amount under a certain specific condition, represents the boundary radius of the plastic zone, represents the current radius, and represents an undetermined constant ;

[0106] The plastic zone hoop stress calculation formula is:

[0107] (9);

[0108] The formula (9) above, represents the hoop stress in the plastic zone, represents the radial stress in the plastic zone, and represents an undetermined constant , represents the stress corresponding to the yield point, represents the initial displacement or the displacement under a certain condition, represents the boundary radius of the plastic zone, represents the current radius;

[0109] According to the plastic mechanics analysis, the radius of the plastic zone of the surrounding rock is obtained:

[0110] (10);

[0111] The formula (10) above, represents the radius of the plastic zone of the surrounding rock, represents the boundary radius of the plastic zone, represents the internal friction angle, represents the pressure of the surrounding rock, represents the cohesion;

[0112] The stress-strain relationship of the elastic zone is:

[0113] (11);

[0114] The formula (11) above, represents the radial strain of the elastic zone, represents the initial displacement or the displacement under a certain condition, represents the boundary radius of the plastic zone, represents the current radius, represents the hoop strain of the elastic zone, represents the radial stress of the elastic zone; represents the pressure of the surrounding rock, represents the elastic modulus of the surrounding rock, represents the Poisson's ratio, represents the hoop stress of the elastic zone;

[0115] The stress-strain relationship model of the deformation zone is established to accurately describe the mechanical behavior of the surrounding rock in different zones.

[0116] S432: Simplify the stress-strain relationship model of the deformation zone based on the deformation zone mechanics theory, and obtain the potential energy function of the plastic zone and the elastic zone by integrating the simplified relationship model;

[0117] In this step, the plastic zone potential energy function is:

[0118] (12);

[0119] In the above formula (12), represents the potential energy function of the plastic zone, represents the radius of the surrounding rock plastic zone, represents the radial stress of the plastic zone, represents the radial strain of the plastic zone, represents the hoop stress of the plastic zone, represents the hoop strain of the plastic zone, represents the current radius, represents a small increment, represents an angular variable;

[0120] the elastic zone potential energy function;

[0121] (13);

[0122] In the above formula (13), represents the potential energy function of the elastic zone, represents the radius of the surrounding rock plastic zone, represents the radial stress of the elastic zone, represents the radial strain of the elastic zone, represents the hoop stress of the elastic zone, represents the hoop strain of the elastic zone, represents the current radius, represents a small increment, represents the surrounding rock elastic modulus, represents the boundary radius of the plastic zone, represents the Poisson's ratio, represents the initial displacement or the displacement under a certain condition.

[0123] S433: Integrating based on the plastic zone potential energy function and the elastic zone potential energy function to obtain a total potential energy function.

[0124] In this step, the total potential energy is:

[0125] (14);

[0126] In the above formula (14), represents the total potential energy, represents the potential energy function of the plastic zone, represents the potential energy function of the elastic zone.

[0127] S44: Deriving the total potential energy function based on the cusp catastrophe theory to obtain a bifurcation set equation.

[0128] To make clear the specific obtaining method of the bifurcation equation, steps S44 to S445 are included in step S44, specifically:

[0129] S441: Perform Tschirnhaus transformation on the total potential energy function based on the cusp catastrophe theory to obtain a cusp catastrophe standard equation;

[0130] In this step, the Tschirnhaus transformation is used to convert the time domain to the complex frequency domain; preferably, the Tschirnhaus transformation is a Tschirnhaus transformation.

[0131] S442: Perform linear transformation processing on the displacement variable in the cusp catastrophe standard equation to obtain a new dimensionless displacement variable;

[0132] S443: Substitute the new dimensionless displacement variable into the total potential energy function to reconstruct the total potential energy function to obtain a new total potential energy function;

[0133] In this step, the new dimensionless displacement variable is substituted into the original total potential energy function. Since the displacement variable in the total potential energy function is replaced by the new dimensionless displacement variable, the form of the function will change accordingly. Through reorganization and calculation, a new total potential energy function is obtained.

[0134] S444: Perform first-order derivative on the new total potential energy function to obtain an instability catastrophe model;

[0135] In this step, the new total potential energy function is differentiated with respect to the new dimensionless displacement variable to obtain a first-order derivative. According to the energy principle, the equilibrium state corresponds to the extreme point of the total potential energy function, i.e., the point where the first-order derivative of the total potential energy function is zero. The equation obtained after the first-order derivative represents the condition under which the system is in equilibrium. When the state of the system changes, the equilibrium condition is destroyed, and instability catastrophe occurs.

[0136] The instability catastrophe model is:

[0137] (15);

[0138] In the above formula (15), represents a variable, , , and represent the influence of different parameters on the cusp catastrophe equilibrium surface equation

[0139] The instability catastrophe model describes the critical condition of the system from a stable state to an unstable state, providing a basis for judging whether the surrounding rock will experience instability catastrophe.

[0140] S445: constructing based on the destabilization mutation model and the new total potential function function to obtain a bifurcation set equation.

[0141] In this step, the bifurcation set equation is:

[0142] (16);

[0143] In the above formula (16), represents the bifurcation set, represents the variable, , , and represent the influence of different parameters on the cusp mutation equilibrium surface equation;

[0144] S5: state discrimination based on the bifurcation set equation and the preset surrounding rock destabilization constraint condition, and output of the stability result.

[0145] Based on the bifurcation set equation and the surrounding rock destabilization constraint condition, the stability of the tunnel surrounding rock is discriminated, and a stable equilibrium state result is obtained. Subsequently, the surrounding rock mechanical stress-strain test curve is simplified according to the stable equilibrium state result, the stability of the tunnel surrounding rock is analyzed through the simplified test curve, and finally the stability result is obtained.

[0146] In this step, the discriminant of the high ground stress tunnel in different states is:

[0147] If Δ>0, the surrounding rock of the tunnel is in a stable equilibrium state;

[0148] If Δ=0, the surrounding rock of the tunnel is in a stable equilibrium critical state;

[0149] If Δ<0, the surrounding rock of the tunnel is in a non-stable equilibrium state.

[0150] The bifurcation set of Δ is simplified, and according to the necessary condition of surrounding rock destabilization Δ<0, that is, .

[0151] Let , , according to substitute into , obtain , let , ;

[0152] represents the relationship formula about the ratio of post-peak strain to peak strain, represents the elastic modulus , the modulus reduction value , the peak stress , the Poisson's ratio and ground stress related parameters;

[0153] Thus, the discriminant of the simplified high ground stress tunnel in different states is:

[0154] If , the surrounding rock of the tunnel is in a stable equilibrium state;

[0155] If , the surrounding rock of the tunnel is in a stable equilibrium critical state;

[0156] If , the surrounding rock of the tunnel is in an unstable equilibrium state.

[0157] Embodiment 2:

[0158] The embodiment provides a high ground stress tunnel surrounding rock stability analysis device, and the high ground stress tunnel surrounding rock stability analysis device described below can be correspondingly referred to the high ground stress tunnel surrounding rock stability analysis method described above, comprising:

[0159] Figure 2 A block diagram of a high ground stress tunnel surrounding rock stability analysis device 800 is shown according to an exemplary embodiment. As shown in Figure 2 , the high ground stress tunnel surrounding rock stability analysis device 800 includes a processor 801, a memory 802, a multimedia component 803, an I / O interface 804, and a communication component 805.

[0160] The processor 801 runs the overall operation of the logically associated acquisition module, simulation module, prediction module, construction module, and judgment module inside;

[0161] The acquisition module is configured to acquire tunnel information and write the tunnel information into the memory 802 through the I / O interface 804, wherein the tunnel information includes cross-section information, historical monitoring data, and real-time monitoring data;

[0162] The simulation module is configured to call the tunnel information in the memory 802, perform numerical simulation according to the multi-dimensional features in the tunnel information, extract feature parameters, and store the feature parameters back to the memory 802;

[0163] The prediction module is configured to call the feature parameters in the memory 802, optimize a random forest regression model based on a chaotic sparrow search algorithm, input the feature parameters into the optimized regression model for prediction, obtain surrounding rock prediction results, and store the surrounding rock prediction results back to the memory 802;

[0164] To clearly show the specific acquisition method of the prediction module, the specific acquisition method includes:

[0165] An optimization unit is configured to write the chaotic sparrow search algorithm and the random forest regression model into the storage 802 through the I / O interface 804, optimize the random forest regression model based on the chaotic sparrow search algorithm, obtain an optimized regression model, and store the optimized regression model in the storage 802.

[0166] To determine the specific acquisition method of the optimization unit, the following specific methods are provided.

[0167] A calculation subunit is configured to write the chaotic sparrow search algorithm into the storage 802 through the I / O interface 804, generate an initial population based on the chaotic mapping in the chaotic sparrow search algorithm, calculate the initial population based on the prediction performance index of the random forest regression model, obtain a fitness function value, and store the fitness function value in the storage 802.

[0168] An update subunit is configured to call the fitness function value in the storage 802, update the position of the initial population based on the fitness function value, obtain an optimized population position, and store the optimized population position in the storage 802.

[0169] An optimization subunit is configured to call the optimized population position in the storage 802, perform global search optimization based on the optimized population position and the sine-cosine algorithm, obtain a global optimal solution, and store the global optimal solution in the storage 802.

[0170] A training subunit is configured to call the global optimal solution in the storage 802, update the population position based on the global optimal solution, obtain a new population position, retrain the random forest regression model based on the new population position, judge the prediction performance index of the retrained model against a preset constraint condition, and output an optimized regression model and store the optimized regression model in the storage 802 when the prediction performance index of the retrained model meets the preset constraint condition.

[0171] An evaluation unit is configured to call the optimized regression model in the storage 802, input the feature parameters into the optimized regression model to predict a state, compare the prediction result with a preset actual observation value, calculate an evaluation index, obtain an evaluation result, and store the evaluation result in the storage 802.

[0172] A generation unit is configured to call the evaluation result in the storage 802, randomly generate a plurality of groups of hyperparameter combinations based on the evaluation result, optimize and adjust the plurality of groups of hyperparameter combinations based on the adaptive weight clustering verification algorithm, obtain optimized hyperparameters, and store the optimized hyperparameters in the storage 802.

[0173] A training unit is configured to call the optimized hyperparameters in the storage 802, input the optimized hyperparameters into the regression model for optimization training, predict the feature parameters based on the optimized regression model, obtain a surrounding rock prediction result, and store the surrounding rock prediction result in the storage 802.

[0174] The construction module is configured to call the surrounding rock prediction result in the memory 802, input parameters of the surrounding rock prediction result into a preset surrounding rock physical model to calculate a potential energy function, perform cusp catastrophe analysis on the potential energy function, construct a bifurcation set equation, and store the bifurcation set equation in the memory 802.

[0175] To clearly show the specific acquisition mode of the construction module, the specific acquisition mode includes the following.

[0176] The simplification unit is configured to call the surrounding rock prediction result in the memory 802, input parameters of the surrounding rock prediction result into the surrounding rock physical model, simplify a non-circular tunnel section by using an equivalent radius method, obtain a section equivalent radius, and store the section equivalent radius in the memory 802.

[0177] The solving unit is configured to call the section equivalent radius in the memory 802, solve a surrounding rock mechanical model based on a mechanical equilibrium condition, obtain surrounding rock mechanical characteristics by using the section equivalent radius, and store the surrounding rock mechanical characteristics in the memory 802.

[0178] The calculation unit is configured to call the surrounding rock mechanical characteristics in the memory 802, construct a deformation zone mechanical model according to the surrounding rock mechanical characteristics and a preset boundary condition, and calculate a total potential energy function, and store the total potential energy function in the memory 802.

[0179] The derivation unit is configured to call the total potential energy function in the memory 802, derive the total potential energy function based on a cusp catastrophe theory, obtain a bifurcation set equation, and store the bifurcation set equation in the memory 802.

[0180] The judgment module is configured to call the bifurcation set equation in the memory 802, perform state discrimination on the bifurcation set equation and a preset surrounding rock instability constraint condition, feed back a discrimination result to a terminal through the I / O interface 804, and output a stability result.

[0181] It should be noted that the processor 801 is configured to control overall operation of the high ground stress tunnel surrounding rock stability analysis device 800 to achieve all or part of the steps in the high ground stress tunnel surrounding rock stability analysis method described above. The memory 802 is configured to store various types of data to support the operation of the high ground stress tunnel surrounding rock stability analysis device 800, which can include, for example, instructions for any application or method operating on the high ground stress tunnel surrounding rock stability analysis device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, and the like. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk. The multimedia component 803 can include a screen and an audio component. The screen can be, for example, a touch screen, and the audio component is configured to output and / or input audio signals. For example, the audio component can include a microphone configured to receive external audio signals. The received audio signals can be further stored in the memory 802 or transmitted through the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, which can be a keyboard, a mouse, a button, and the like. These buttons can be virtual buttons or physical buttons. The communication component 805 is configured to enable wired or wireless communication between the high ground stress tunnel surrounding rock stability analysis device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G or 4G, or one or more of them or a combination of them, so the corresponding communication component 805 can include a Wi-Fi module, a Bluetooth module, an NFC module.

[0182] In an exemplary embodiment, the high stress tunnel surrounding rock stability analysis device 800 can be implemented by one or more Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor or other electronic elements for executing the high stress tunnel surrounding rock stability analysis method described above.

[0183] It should be noted that, as for the device in the above-mentioned embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be described in detail here.

[0184] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0185] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0185] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A high ground stress tunnel surrounding rock stability analysis method, characterized in that, The method comprises the following steps: acquiring tunnel information, wherein the tunnel information comprises section information, historical monitoring data and real-time monitoring data; performing numerical simulation according to multi-dimensional features in the tunnel information, and extracting feature parameters; optimizing a random forest regression model based on a chaotic sparrow search algorithm, inputting the feature parameters into the optimized regression model for prediction, and obtaining surrounding rock prediction results; wherein the specific acquisition method of the surrounding rock prediction results comprises: optimizing the random forest regression model based on the chaotic sparrow search algorithm to obtain an optimized regression model; inputting the feature parameters into the optimized regression model for prediction, comparing the prediction results with preset actual observation values, calculating evaluation indexes, and obtaining evaluation results; randomly generating a plurality of groups of hyperparameter combinations according to the evaluation results, optimizing and adjusting the plurality of groups of hyperparameter combinations through a self-adaptive weight clustering verification algorithm, and obtaining optimized hyperparameters; inputting the optimized hyperparameters into the regression model for optimization training, predicting the feature parameters through the optimized regression model, and obtaining surrounding rock prediction results; inputting parameters of the surrounding rock prediction results into a preset surrounding rock physical model to calculate a potential energy function, performing cusp catastrophe analysis on the potential energy function, and constructing a bifurcation set equation; performing state discrimination based on the bifurcation set equation and a preset surrounding rock instability constraint condition, and outputting a stability result.

2. The high ground stress tunnel surrounding rock stability analysis method according to claim 1, characterized in that, performing numerical simulation according to multi-dimensional features in the tunnel information, and extracting feature parameters, comprising: constructing a model according to geological conditions and structural features in the tunnel information, and obtaining a numerical model; preprocessing the tunnel information, inputting the preprocessed tunnel information into the numerical model to set boundary conditions and initial conditions for numerical simulation, and obtaining numerical simulation results; extracting multi-dimensional features from the numerical simulation results, performing dimension reduction processing on the multi-dimensional features, combining the dimension-reduced features with the numerical simulation results to extract surrounding rock stability feature parameters, and obtaining feature parameters.

3. The high ground stress tunnel surrounding rock stability analysis method according to claim 1, characterized in that, optimizing the random forest regression model based on the chaotic sparrow search algorithm to obtain an optimized regression model, comprising: generating an initial population based on chaotic mapping in the chaotic sparrow search algorithm, calculating the initial population through a prediction performance index of the random forest regression model to obtain a fitness function value; updating the position of the initial population according to the fitness function value to obtain an optimized population position; performing global search optimization based on the optimized population position and a sine-cosine algorithm to obtain a global optimal solution; updating the population position according to the global optimal solution to obtain a new population position, retraining a random forest regression model through the new population position, judging a prediction performance index of the retrained model against a preset constraint condition, and outputting the optimized regression model when the prediction performance index of the retrained model meets the preset constraint condition.

4. The high ground stress tunnel surrounding rock stability analysis method according to claim 1, characterized in that, inputting parameters of the surrounding rock prediction results into a preset surrounding rock physical model to calculate a potential energy function, performing cusp catastrophe analysis on the potential energy function, and constructing a bifurcation set equation, comprising: Inputting the parameters of the surrounding rock prediction result into the surrounding rock physical model, simplifying the non-circular tunnel section by the equivalent radius method to obtain the section equivalent radius; Solving the surrounding rock mechanical model based on the mechanical equilibrium condition through the section equivalent radius to obtain the mechanical properties of the surrounding rock; According to the mechanical properties of the surrounding rock and the preset boundary conditions, a deformation zone mechanical model is constructed and calculated to obtain a total potential energy function; Deriving the total potential energy function based on the cusp catastrophe theory to obtain a bifurcation set equation.

5. The high ground stress tunnel surrounding rock stability analysis method according to claim 4, characterized in that, According to the mechanical properties of the surrounding rock and the preset boundary conditions, a deformation zone mechanical model is constructed and calculated to obtain a total potential energy function, including: According to the mechanical properties of the surrounding rock and the preset boundary conditions, a deformation zone stress-strain relationship model is constructed, and the deformation zone includes a plastic zone and an elastic zone; Based on the deformation zone mechanical theory, the deformation zone stress-strain relationship model is simplified, and the plastic zone and the elastic zone potential energy functions are obtained by integrating the simplified relationship model; Based on the plastic zone potential energy function and the elastic zone potential energy function, the total potential energy function is obtained.

6. A high ground stress tunnel surrounding rock stability analysis device, characterized in that, It includes: a processor (801), a memory (802) and an I / O interface (804); the processor (801) runs the logically associated acquisition module, simulation module, prediction module, construction module and judgment module; wherein the acquisition module is used to acquire tunnel information and write it into the memory (802) through the I / O interface (804), the tunnel information including section information, historical monitoring data and real-time monitoring data; the simulation module is used to call the tunnel information in the memory (802), perform numerical simulation according to the multi-dimensional characteristics in the tunnel information, extract feature parameters and store them back to the memory (802); the prediction module is used to call the feature parameters in the memory (802), optimize the random forest regression model based on the chaotic sparrow search algorithm, input the feature parameters into the optimized regression model for prediction, obtain the surrounding rock prediction result and store it back to the memory (802); wherein the prediction module includes: the optimization unit is used to write the chaotic sparrow search algorithm and the random forest regression model into the memory (802) through the I / O interface (804), optimize the random forest regression model based on the chaotic sparrow search algorithm, obtain the optimized regression model and store it back to the memory (802); the evaluation unit is used to call the optimized regression model in the memory (802), input the feature parameters into the optimized regression model for prediction, compare the prediction result with the preset actual observation value, calculate the evaluation index, obtain the evaluation result and store it back to the memory (802); the generation unit is used to call the evaluation result in the memory (802), randomly generate a plurality of groups of hyperparameter combinations according to the evaluation result, optimize and adjust the plurality of groups of hyperparameter combinations through the adaptive weight clustering verification algorithm, obtain the optimized hyperparameters and store them back to the memory (802); The training unit is configured to call the optimized hyperparameters in the memory (802), input the optimized hyperparameters into the regression model for optimization training, predict the characteristic parameters by the optimized regression model, obtain a surrounding rock prediction result, and store the surrounding rock prediction result in the memory (802) again; The construction module is configured to call the surrounding rock prediction result in the memory (802), input the parameters of the surrounding rock prediction result into a preset surrounding rock physical model to calculate a potential energy function, analyze the potential energy function by a cusp catastrophe analysis, construct a bifurcation set equation, and store the bifurcation set equation in the memory (802) again; The judgment module is configured to call the bifurcation set equation in the memory (802), perform state discrimination based on the bifurcation set equation and a preset surrounding rock instability constraint condition, and feed back a discrimination result to a terminal through the I / O interface (804) to output a stability result.

7. The high ground stress tunnel surrounding rock stability analysis device according to claim 6, characterized in that, The optimization unit comprises: The calculation sub-unit is configured to write the chaotic sparrow search algorithm into the memory (802) through the I / O interface (804), generate an initial population based on a chaotic mapping in the chaotic sparrow search algorithm, calculate the initial population by a prediction performance index of the random forest regression model, obtain a fitness function value, and store the fitness function value in the memory (802) again; The update sub-unit is configured to call the fitness function value in the memory (802), update a position of the initial population according to the fitness function value, obtain an optimized population position, and store the optimized population position in the memory (802) again; The optimization sub-unit is configured to call the optimized population position in the memory (802), perform global search optimization based on the optimized population position and a sine-cosine algorithm, obtain a global optimal solution, and store the global optimal solution in the memory (802) again; The training sub-unit is configured to call the global optimal solution in the memory (802), update a population position according to the global optimal solution, obtain a new population position, retrain the random forest regression model by the new population position, judge a model prediction performance index after retraining and a preset constraint condition, and output an optimized regression model and store the optimized regression model in the memory (802) again when the model prediction performance index after retraining meets the preset constraint condition.

8. The high ground stress tunnel surrounding rock stability analysis device according to claim 6, characterized in that, The construction module comprises: The simplification unit is configured to call the surrounding rock prediction result in the memory (802), input parameters of the surrounding rock prediction result into the surrounding rock physical model, simplify a non-circular tunnel section by an equivalent radius method, obtain a section equivalent radius, and store the section equivalent radius in the memory (802) again; The solving unit is configured to call the section equivalent radius in the memory (802), solve a surrounding rock mechanical model by the section equivalent radius based on a mechanical equilibrium condition, obtain surrounding rock mechanical characteristics, and store the surrounding rock mechanical characteristics in the memory (802) again; The calculation unit is configured to call the surrounding rock mechanical characteristics in the memory (802), construct a deformation zone mechanical model according to the surrounding rock mechanical characteristics and a preset boundary condition, and calculate a total potential energy function, and store the total potential energy function in the memory (802) again; The derivation unit is configured to call the total potential energy function in the memory (802), derive the total potential energy function based on a cusp catastrophe theory, obtain a bifurcation set equation, and store the bifurcation set equation in the memory (802) again.

Images