A tunnel surrounding rock natural frequency determination method and system based on numerical samples

Abstract

The application provides a tunnel surrounding rock inherent frequency determination method and system based on numerical sample, relates to the technical field of tunnel engineering, and comprises the following steps: obtaining geological survey data of a target deep buried joint rock mass tunnel engineering; according to the geological survey data, combining a three-dimensional numerical inversion method to establish a parameter model and set parameters, obtaining a parameterized model configuration; according to the parameterized model configuration, carrying out numerical model construction processing to generate an inherent frequency numerical sample data set; according to the inherent frequency numerical sample data set, carrying out sensitivity analysis to obtain a main control factor screening result; according to the main control factor screening result, establishing a prediction model to obtain an inherent frequency prediction model; inputting real-time main control factor data obtained through field monitoring into the inherent frequency prediction model for prediction processing to obtain inherent frequency time-frequency characteristics and main frequency information of the target deep buried joint rock mass tunnel. The application significantly improves construction safety and engineering decision efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel engineering, in particular to a tunnel surrounding rock natural frequency determination method and system based on numerical samples. BACKGROUND

[0002] With the continuous expansion of tunnel engineering construction in China to the deep part, the stress environment of deep-buried tunnel surrounding rock presents high complexity and dynamic change characteristics under the influence of multiple factors such as geological structure and excavation unloading, which poses a severe challenge to the safety and stability of engineering construction. Among them, the unfavorable combination of structural planes represented by jointed fissure zones becomes the key control factor of hard rock surrounding rock stability. Under the combined action of high ground stress and construction disturbance, deep-buried jointed rock mass tunnel faces many engineering geological disaster risks. Various external dynamic disturbance sources (such as TBM (Tunnel Boring Machine) vibration, blasting vibration, seismic wave, train vibration, etc.) in the construction process are transmitted into the surrounding rock, and if the disturbance frequency is close to the natural frequency of the surrounding rock, significant resonance effect will be caused. This resonance phenomenon will accelerate the initiation and propagation of cracks in jointed rock mass, and then induce rock burst disaster, causing personnel casualties and equipment damage. Therefore, it is urgent to establish a method for quickly obtaining the natural frequency of deep-buried jointed rock mass tunnel, to lay a foundation for preventing and controlling rock burst triggered by dynamic disturbance, and to ensure the safety of on-site personnel construction. The existing natural frequency acquisition methods mainly rely on field tests or numerical simulation calculations. Although field tests are direct and reliable, they are long in period, high in cost and greatly affected by environmental factors; while numerical simulation calculations are relatively accurate, but require the construction of a fine geometric model and the setting of material parameters, and the calculation process is complex and time-consuming.

[0003] Based on the above-mentioned shortcomings of the prior art, there is an urgent need for a tunnel surrounding rock natural frequency determination method and system based on numerical samples. SUMMARY

[0004] The purpose of the present application is to provide a tunnel surrounding rock natural frequency determination method and system based on numerical samples, which aims to overcome the above-mentioned shortcomings of the prior art and provide a layout method and system with high modeling efficiency, accurate positioning, self-consistent logic and strong adaptability. In order to achieve the above-mentioned purpose, the technical solutions adopted by the present application are as follows:

[0005] In a first aspect, the present application provides a tunnel surrounding rock natural frequency determination method based on numerical samples, comprising:

[0006] Obtaining geological survey data of a target deep-buried jointed rock mass tunnel engineering;

[0007] According to the geological survey data, a parameter model is established and parameters are set by combining a three-dimensional numerical inversion method, and a parameterized model configuration is obtained by setting excavation rate, tunnel burial depth, lateral pressure coefficient and surrounding rock grade as core influencing factors;

[0008] performing numerical model construction processing according to the parameterized model configuration, simulating joint distribution and mechanical response under different surrounding rock qualities, and performing dynamic mechanical analysis in combination with working condition cross combination to generate an inherent frequency numerical sample data set;

[0009] performing sensitivity analysis according to the inherent frequency numerical sample data set, quantifying the correlation degree of each influence factor and the inherent frequency based on a preset grey correlation theory model, and obtaining a main control factor screening result;

[0010] performing prediction model establishment according to the main control factor screening result, extracting features from the numerical sample and constructing a mapping relationship between each main control factor parameter and a target deep jointed rock mass tunnel inherent frequency main frequency based on a preset machine learning model, and obtaining an inherent frequency prediction model;

[0011] inputting real-time main control factor data obtained through field monitoring into the inherent frequency prediction model for prediction processing, and obtaining inherent frequency time-frequency characteristics and main frequency information of the target deep jointed rock mass tunnel.

[0012] In a second aspect, the present application also provides a tunnel surrounding rock inherent frequency determination system based on numerical samples, comprising:

[0013] an acquisition module configured to acquire geological survey data of a target deep jointed rock mass tunnel project;

[0014] a configuration module configured to establish and set a parameter model according to the geological survey data in combination with a three-dimensional numerical inversion method, set an excavation rate, a tunnel burial depth, a lateral pressure coefficient, and a surrounding rock grade as core influence factors, and obtain a parameterized model configuration;

[0015] a construction module configured to perform numerical model construction processing according to the parameterized model configuration, simulate joint distribution and mechanical response under different surrounding rock qualities, and perform dynamic mechanical analysis in combination with working condition cross combination to generate an inherent frequency numerical sample data set;

[0016] an analysis module configured to perform sensitivity analysis according to the inherent frequency numerical sample data set, quantify the correlation degree of each influence factor and the inherent frequency based on a preset grey correlation theory model, and obtain a main control factor screening result;

[0017] a modeling module configured to perform prediction model establishment according to the main control factor screening result, extract features from the numerical sample and construct a mapping relationship between each main control factor parameter and a target deep jointed rock mass tunnel inherent frequency main frequency based on a preset machine learning model, and obtain an inherent frequency prediction model;

[0018] A prediction module is configured to input real-time master factor data acquired by field monitoring into the inherent frequency prediction model for prediction processing, so as to obtain inherent frequency time-frequency characteristics and master frequency information of the target deep-buried jointed rock mass tunnel.

[0019] The present application has the following beneficial effects:

[0020] The present application integrates the parameterized model configuration driven by geological survey data, the inherent frequency sample data set generated by numerical simulation, the master factor screening dominated by the grey correlation theory, and the inherent frequency prediction model constructed by machine learning, so as to realize the rapid and high-precision acquisition of the inherent frequency of the deep-buried jointed rock mass tunnel, provide a scientific basis for the active prevention and control of the rock burst triggered by dynamic disturbance in the high ground stress environment, and significantly improve the construction safety and engineering decision efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0022] Figure 1 A flowchart of a tunnel surrounding rock inherent frequency determination method based on numerical samples according to an embodiment of the present application;

[0023] Figure 2 A structural schematic diagram of a tunnel surrounding rock inherent frequency determination system based on numerical samples according to an embodiment of the present application;

[0024] Figure 3 A schematic diagram of deep-buried jointed rock mass tunnel models of different surrounding rock grades;

[0025] Figure 4 A schematic diagram of the influence of GSI changes on the inherent frequency of deep-buried jointed tunnels.

[0026] In the figure, 901 is an acquisition module, 902 is a configuration module, 903 is a construction module, 904 is an analysis module, 905 is a modeling module, and 906 is a prediction module. DETAILED DESCRIPTION

[0027] 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.

[0028] 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.

[0029] Example 1:

[0030] This embodiment provides a method for determining the natural frequency of tunnel surrounding rock based on numerical samples.

[0031] See Figure 1 The figure shows that the method includes steps S100 to S600.

[0032] Step S100: Obtain geological survey data for the target deep-buried jointed rock mass tunnel project;

[0033] To address the complex dynamic response mechanism of deeply buried jointed rock tunnels, this step focuses on obtaining the complete engineering geological background characteristics. By extracting basic parameters such as tunnel geometry, surrounding rock mechanical strength, and geostress distribution, it provides a physical basis for constructing a numerical model under real high geostress conditions, thereby avoiding the limitations of idealized geological parameters in traditional empirical formulas.

[0034] Step S200: Based on geological survey data, a parameter model is established and parameters are set using a three-dimensional numerical inversion method. The core influencing factors are set as excavation rate, tunnel depth, lateral pressure coefficient, and surrounding rock grade to obtain the parameterized model configuration.

[0035] It can be understood that, in order to solve the complexity of quantitative characterization of multi-factor coupling effect, this step is based on the combination of geological survey data and three-dimensional numerical inversion method, sets three core variables of excavation disturbance (rate), ground pressure environment (burial depth, lateral pressure coefficient) and surrounding rock integrity (level), establishes a physical model of dynamic correlation of construction process-geological conditions, and the parameter configuration directly determines the reduction degree of nonlinear response of jointed rock mass in subsequent numerical simulation.

[0036] Step S300, numerical model construction processing is performed according to the parameterized model configuration, the joint distribution and mechanical response under different surrounding rock quality are simulated, and dynamic mechanics analysis is performed in combination with working condition cross combination to generate a numerical sample data set of natural frequency;

[0037] It should be noted that, in view of the training data requirement of machine learning, the rock mass structure characteristics of different surrounding rock quality (GSI) are restored through the discrete joint network, the dynamic mechanics calculation of high ground stress working condition combination is combined to generate a natural frequency sample library covering the actual engineering variability of surrounding rock conditions, and the constraints of field measured data scarcity and sample deviation on model training are eliminated.

[0038] Step S400, sensitivity analysis is performed according to the numerical sample data set of natural frequency, the correlation degree of each influencing factor and natural frequency is quantified based on the preset gray correlation theory model, and the main control factor screening result is obtained;

[0039] In view of the problem that the influence weight of multiple factors is not clear, this step uses the gray correlation theory to analyze the action path of excavation parameters, ground stress and surrounding rock characteristics on natural frequency in the numerical sample, quantifies the contribution degree of each factor to dynamic response, excludes secondary variables to focus on key rock burst causes, and improves the effectiveness and interpretability of subsequent model input.

[0040] Step S500, a prediction model is established according to the main control factor screening result, and based on the preset machine learning model, the mapping relationship between each main control factor parameter and the natural frequency of the target deep buried jointed rock mass tunnel is constructed, and the natural frequency prediction model is obtained;

[0041] This step is based on the structured data set of main control factors, extracts the jointed rock mass stiffness attenuation law and vibration modal characteristics specific to high ground stress environment through machine learning, and constructs the nonlinear mapping relationship between the ground stress boundary, the surrounding rock integrity and the natural frequency.

[0042] Step S600, real-time main control factor data obtained by field monitoring is input into the natural frequency prediction model for prediction processing, and the natural frequency time-frequency characteristics and main frequency information of the target deep buried jointed rock mass tunnel are obtained.

[0043] Finally, by combining the vibration monitoring data stream during construction, the main control parameters of the surrounding rock monitored in real time are input into the prediction model, and the natural frequency time-frequency characteristics and main frequency value based on the current geological conditions are directly output. This enables online prediction of dynamic characteristics from numerical simulation to engineering scenarios, providing real-time decision parameters for proactive prevention and control of dynamic disturbance risks in deep-buried tunnels.

[0044] Further, step S200 includes steps S210 to S230.

[0045] Step S210: Combining geological survey data with three-dimensional numerical inversion methods, basic parameter extraction processing is performed. By identifying tunnel geometric dimensions, surrounding rock mechanical parameters and geostress characteristics, a set of basic parameters is obtained.

[0046] Step S220: Based on the basic parameter set, the core influencing factors are set and processed. By selecting the excavation rate, tunnel depth, lateral pressure coefficient and surrounding rock grade as dynamic variables, the surrounding rock grade is quantitatively classified based on the geological strength index to obtain the definition of the core factors.

[0047] Step S230: Based on the definition of core factors, perform parameter integration processing, construct the coupling relationship of influencing factors by associating geological strength and geostress characteristics, and obtain the parameterized model configuration.

[0048] Preferably, in the scenario of a deeply buried jointed rock tunnel, step S210 first combines geological survey data with three-dimensional numerical inversion methods to identify tunnel geometric dimensions (such as tunnel diameter), surrounding rock mechanical parameters (strength indices such as cohesion and internal friction angle), and geostress characteristics (vertical stress to lateral pressure ratio), forming a set of basic parameters characterizing the intrinsic properties of a high geostress environment; step S220 selects excavation rate, tunnel depth, lateral pressure coefficient, and surrounding rock grade as dynamic variables based on this parameter set, wherein the surrounding rock grade is quantitatively classified through geological strength indices, which will... The structural features such as the number of joint groups and spacing are converted into scale values ​​in the range of 50-80 (for example, low values ​​correspond to multiple dense joint groups, and high values ​​correspond to intact rock masses), realizing the quantifiable classification of complex jointed rock masses; step S230 constructs a coupled relationship model of construction disturbance-rock mass integrity-geostress environment by associating geological strength indicators with geostress characteristics (such as high ground pressure inhibiting joint opening and deep burial conditions strengthening surrounding rock constraints), generating a parameterized configuration that takes into account geological heterogeneity and construction dynamics, providing a physical basis for the subsequent accurate simulation of the dynamic response of jointed rock masses.

[0049] Further, step S300 includes steps S310 to S330.

[0050] Step S310: Based on the parameterized model configuration, the jointed rock mass model is constructed. A three-dimensional discrete joint network is established by mapping the number of joint groups, spacing and spatial distribution characteristics through geological strength index, and a parameterized surrounding rock quality model is obtained.

[0051] Step S320, multi-working condition dynamic response analysis processing is performed according to the parameterized surrounding rock quality model, vibration characteristics of surrounding rock in a high ground stress environment are simulated through cross combination of tunnel burial depth, lateral pressure coefficient and excavation rate, and dynamic response data are obtained;

[0052] Step S330, inherent frequency sample generation processing is performed according to the dynamic response data, frequency value is quantified by extracting the fundamental frequency vibration mode under different surrounding rock quality, and an inherent frequency numerical sample data set is obtained.

[0053] Specifically, considering that rock burst disasters generally occur in grade II-III surrounding rock, in order to truly reflect the gradual process of the surrounding rock environment from bad to good, the surrounding rock quality is quantitatively divided into GSI=50-80 in combination with the GSI (Geological Strength Index) scale. The selection of this range is mainly based on the surrounding rock characteristics of deep-buried tunnels: deep-buried tunnels are mostly in a high ground stress environment, and the rock mass classification corresponding to GSI=50-80 is grade II-IV, covering typical surrounding rock types from “relatively stable” (grade II-III, GSI=60-80) to “easily unstable” (grade IV, GSI=50-55). At the same time, the GSI value is closely related to the joint environment: typically, the more the number of joint sets, the lower the GSI value. For example, GSI=80 corresponds to 2 joint sets (relatively complete rock mass), GSI=60-75 corresponds to 3 joint sets (moderately developed), and GSI=50-55 corresponds to 4 joint sets (relatively developed), indicating that the more developed the joint, the worse the rock mass integrity, and the GSI decreases; joint spacing: the larger the joint spacing (increasing from 2.0m to 5.5m) and the shorter the length (decreasing from 60m to 30m), the higher the GSI value, reflecting the weakening of the cutting effect of joints on the rock mass and the improvement of the integrity. Based on the above characteristics, discontinuous surfaces with different spacings, inclinations and friction coefficients are introduced in the research area to simulate the rock mass quality of grade II-IV surrounding rock, and deep-buried jointed rock mass tunnel models of different surrounding rock grades are established, as shown in FIG. 1. Figure 3 The corresponding relationship table between the engineering rock mass classification and GSI is as follows:

[0054] Table 1 Corresponding relationship between engineering rock mass classification and GSI

[0055]

[0056] Step S310 maps the structural characteristics of the jointed rock mass of the deep-buried tunnel by the geological strength index (GSI), converts the index value in the interval of 50-80 into specific joint group number (2-4 groups), interval (2.0-5.5 m), and spatial distribution (different inclination / angle combinations), establishes a three-dimensional discrete joint network model, accurately restores the gradual change characteristics of joint development in a high ground stress environment (such as rock mass fragmentation caused by joint densification corresponding to a decrease in the index value), overcomes the defects of the traditional method of simplified modeling of complex rock mass structure, and outputs a parameterized surrounding rock quality model;

[0057] Step S320, based on the quality model, simulates the vibration propagation constraints of surrounding rock in a high ground stress environment (such as deep overburden pressure inhibiting joint opening and rapid excavation inducing unloading dynamic waves) by cross-combining three typical variables of deep-buried tunnel, such as tunnel burial depth (500-2500 m), lateral pressure coefficient, and excavation rate (0.1-20 m / day) in the discrete element platform. The core innovation lies in quantifying the dynamic coupling effect of construction disturbance-ground stress boundary-joint network, and generating dynamic response data reflecting the variability of actual working conditions;

[0058] Step S330 extracts the fundamental frequency vibration mode (rock mass dominant resonant harmonic component) from the dynamic response data, constructs a sample set by quantifying the fundamental frequency value under different surrounding rock quality, reveals the core law that the increase of the geological strength index leads to the increase of rock mass integrity, the increase of surrounding rock stiffness, and the monotonic increase of natural frequency, and compares the research results. In a high-stress deep-buried joint hard rock tunnel, as the GSI scale increases from 50 to 80, the natural frequency of the deep-buried tunnel increases from 10 Hz to 26.15 Hz. The more complete the surrounding rock is (Ⅳ level→Ⅱ level), the higher the natural frequency of the deep-buried joint rock mass tunnel is. The larger the internal crack volume of the rock mass structure is, the smaller the overall stiffness of the surrounding rock is, and the greater the natural frequency is. The obtained conclusion is consistent with the laboratory test results, as shown in Figure 4 The process realizes the digital representation of the stiffness attenuation characteristics of the deep-buried joint rock mass, and provides a high-fidelity training sample library covering Ⅱ-Ⅳ surrounding rock for machine learning.

[0059] Further, step S400 includes step S410 to step S430.

[0060] Step S410, according to the numerical sample data set of the natural frequency, an evaluation sequence construction process is performed, the excavation rate, the tunnel burial depth, the lateral pressure coefficient, and the surrounding rock level are defined as the evaluation index sequence, the natural frequency is defined as the reference sequence, and an initial evaluation matrix is obtained;

[0061] Step S420, data standardization processing is performed according to the initial evaluation matrix, the mean value is used to eliminate the dimension difference of each factor, and a non-dimensional data sequence matrix is generated;

[0062] Step S430: Perform correlation quantification processing on the dimensionless data sequence matrix, calculate the correlation coefficient and mean correlation degree, output the correlation ranking of each influencing factor and the inherent frequency, and obtain the screening results of the main control factors.

[0063] This process, based on numerical samples, conducts a sensitivity analysis of the natural frequencies of deeply buried jointed rock tunnels. A grey relational analysis model is used to calculate the correlation coefficients of each influencing factor. The specific process is as follows:

[0064] A matrix of evaluation indexes is established to determine the influencing factors of the natural frequency of the surrounding rock in deeply buried jointed rock tunnels. (The matrix is ​​provided in the original text.) The evaluation sequence matrix is ​​composed of data from the evaluation indicators and reference indicators:

[0065] ;

[0066] In the formula, This represents the total number of evaluation index sequences and reference index sequences; This indicates the total number of evaluation indicators and reference indicators; Represents the evaluation sequence matrix; to This indicates specific evaluation indicators and reference indicators.

[0067] Next, determine the evaluation indicator list. Compared with the reference indicator series :

[0068] Evaluation index sequence : ;

[0069] Reference indicator sequence: : ;

[0070] Then, the indicator data is made dimensionless.

[0071] Considering the different physical meanings of the factors, the dimensions of the data may not be the same. This paper uses mean normalization to make the reference index series dimensionless:

[0072] ;

[0073] ;

[0074] In the formula, Indicates the first The first evaluation indicator Dimensionless data after mean normalization; Indicates the first The first of the evaluation indicators (such as excavation rate, tunnel depth, etc.) Raw data, i.e. original observation values without standardization processing; The serial number of a sample or data point.

[0075] The dimensionless data sequence matrix is as follows:

[0076] ;

[0077] In the formula, represents the dimensionless data sequence matrix, to represents the specific element value contained in the dimensionless data sequence matrix.

[0078] Then, the absolute value of the difference between each evaluation index sequence and the reference index sequence is calculated:

[0079]

[0080] In the formula, represents the absolute value of the difference between the reference sequence (intrinsic frequency) and the evaluation index sequence (excavation rate, tunnel depth, lateral pressure coefficient, surrounding rock grade) after dimensionless processing in the sample; represents the sample data after mean dimensionless processing of the reference sequence (intrinsic frequency); represents the evaluation index sequence (influencing factor) after mean dimensionless processing of the sample data.

[0081] The maximum and minimum values in the difference between the evaluation and reference index sequences are determined as and :

[0082] ;

[0083] ;

[0084] Then, the correlation coefficient of each evaluation and reference index sequence is calculated:

[0085] ;

[0086] In the formula, represents the correlation coefficient of the evaluation index sequence (influencing factor) and the reference sequence (intrinsic frequency) at the sample point; represents the resolution coefficient, which is a parameter for adjusting the sensitivity of the correlation coefficient, and the value range is usually .

[0087] For each evaluation indicator, the mean of the correlation coefficient between the individual indicator and the corresponding elements of the reference sequence is calculated to reflect the correlation between each evaluation indicator and the reference indicator sequence, i.e., the correlation degree.

[0088] ;

[0089] In the formula, Indicates the first The correlation between a series of evaluation indicators and a reference series (inherent frequency).

[0090] Based on the correlation between the evaluation indicators of the natural frequency of the deep-buried jointed rock mass tunnel determined in the above steps and the reference indicators, the evaluation indicators are ranked.

[0091] Input optimization: Key control factors with high correlation are selected as input parameters. Data preprocessing is performed on the input parameters, and the Pearson correlation coefficient is used to assess the correlation between the key control factors. Highly correlated redundant parameters are eliminated to improve the generalization ability of the subsequent prediction model. The Pearson correlation coefficient reflects the degree of linear correlation between two variables: when the linear relationship between two variables strengthens, the correlation coefficient tends to 1 or -1; when one variable increases, the other also increases, indicating a positive correlation with a correlation coefficient greater than 0; conversely, a negative correlation indicates a negative correlation with a correlation coefficient less than 0; if the correlation coefficient equals 0, it indicates that there is no linear correlation between them. The formula is:

[0092] ;

[0093] ;

[0094] In the formula, Representing variables and The Pearson correlation coefficient between two variables is used to quantify the degree of linear correlation between the two variables. Representing variables and covariance; and Representing variables respectively and Standard deviation; Representing variables The One observation value; Representing variables The average of all observations; Representing variables The One observation value; Representing variables the average of all observations of the variable.

[0095] Further, the step S500 includes steps S510 to S530.

[0096] Step S510, according to the training data set construction processing of the master control factor screening result, the surrounding rock grade, tunnel depth and lateral pressure coefficient parameter in the frequency label are obtained by associating the frequency value sample, and the master control factor training set is obtained;

[0097] Step S520, according to the feature space mapping processing of the master control factor training set, the stiffness attenuation and vibration modal characteristics of jointed rock mass under high stress environment are extracted by nonlinear transformation, and the high-dimensional feature vector is obtained.

[0098] Step S530, according to the high-dimensional feature vector, the prediction model training processing is carried out, the mapping weight of master control factor parameter and natural frequency is optimized by error back propagation, and the natural frequency prediction model is obtained.

[0099] Specifically, the above process adopts random sampling method to divide the screened master control factor data set into training set and test set according to the proportion of 8:2, and establishes the natural frequency prediction model of deep buried jointed rock mass tunnel based on SVM algorithm. The model is trained combined with the training data set, the input data is each master control factor parameter, and the output data is the natural frequency time-frequency characteristics and main frequency information of deep buried jointed rock mass tunnel. Based on the control variable and cross validation analysis, the change of penalty factor and kernel function coefficient in SVM model in natural frequency prediction process is determined to determine the optimal value of hyperparameter. After completing the model construction, the performance of the model is also evaluated: the prediction accuracy of SVM model is verified by using test set, and the root mean square error RMSE, mean absolute percentage error MAPE and goodness of fit R 2 are used as prediction model evaluation indexes. Under the same data set and parameter setting, the prediction effects of other machine learning models (random forest, neural network, etc.) are established and compared to verify the superiority of SVM model.

[0100] Further, the step S600 includes steps S610 to S630.

[0101] Step S610, according to the real-time acquisition and processing of vibration signal in the field construction environment, the surrounding rock vibration waveform time domain data in each direction is obtained by synchronously acquiring the vibration monitor, and the original vibration signal set is obtained.

[0102] Step S620, according to the original vibration signal set, the master control factor feature extraction processing is carried out, the jointed rock mass fundamental frequency harmonic component under high stress environment is separated by filtering and denoising and time-frequency transformation, and the standardized master control factor input vector is obtained.

[0103] Step S630, according to the standardized master factor input vector, an inherent frequency prediction process is performed, a base frequency response value dominated by the stiffness attenuation of surrounding rock is calculated through forward propagation, and inherent frequency time-frequency characteristics and dominant frequency information of the target deep buried jointed rock mass tunnel are obtained.

[0104] Step S610, multi-directional surrounding rock vibration time domain data (X / Y / Z three-axis waveforms) are synchronously collected by a vibration monitor, a mixed signal set containing construction disturbance and real response of the rock mass is obtained, this multi-directional synchronous collection strategy is specially designed for the spatial anisotropy problem of the dynamic response of the deep buried tunnel jointed rock mass, and can separate key components such as axial excavation disturbance, radial stress wave and vertical ground pressure wave, thereby providing undistorted original vibration signals for subsequent feature extraction;

[0105] It should be noted that step S620 performs master factor feature extraction based on the signal set, filters and denoises to eliminate high-frequency noise interference such as TBM tunneling, and separates base frequency harmonic components specific to the jointed rock mass under high ground stress environment, the core of the processing is to quantify the joint stiffness attenuation characteristics and the ground stress boundary response, and finally output a dimensionally standardized master factor input vector, which is strictly consistent in parameter structure and numerical sample; step S630 inputs the standardized input vector into the pre-trained machine learning model, and calculates the base frequency response value dominated by the stiffness attenuation of surrounding rock through forward propagation, wherein the internal learning mechanism of the model strictly follows the mapping rule of the geological strength index and the inherent frequency, and automatically corrects the boundary effect of the high ground stress environment, and finally generates inherent frequency time-frequency characteristics and dominant frequency information which can be directly used for rock burst resonance risk warning.

[0106] Embodiment 2:

[0107] As shown in Figure 2 The embodiment provides a tunnel surrounding rock inherent frequency determination system based on numerical samples, and the system comprises:

[0108] An acquisition module 901 is configured to acquire geological survey data of a target deep buried jointed rock mass tunnel project;

[0109] A configuration module 902 is configured to establish a parameter model and set parameters according to the geological survey data in combination with a three-dimensional numerical inversion method, to obtain a parameterized model configuration by setting an excavation rate, a tunnel burial depth, a lateral pressure coefficient and a surrounding rock grade as core influencing factors.

[0110] A construction module 903 is configured to perform numerical model construction processing according to the parameterized model configuration, to simulate joint distribution and mechanical response under different surrounding rock qualities, and to perform dynamic mechanical analysis in combination with working condition cross combination, thereby generating an inherent frequency numerical sample data set.

[0111] The analysis module 904 is configured to perform sensitivity analysis based on the inherent frequency numerical sample data set, quantize the correlation degree between each influence factor and the inherent frequency based on a preset grey correlation theory model, and obtain a main control factor screening result.

[0112] The modeling module 905 is configured to perform prediction model establishment based on the main control factor screening result, extract features from the numerical sample based on a preset machine learning model, and construct a mapping relationship between each main control factor parameter and the inherent frequency main frequency of the target deep buried jointed rock mass tunnel, and obtain an inherent frequency prediction model.

[0113] The prediction module 906 is configured to input real-time main control factor data obtained through field monitoring into the inherent frequency prediction model for prediction processing, and obtain inherent frequency time-frequency characteristics and main frequency information of the target deep buried jointed rock mass tunnel.

[0114] In one specific embodiment of the present application, the configuration module 902 includes:

[0115] The first configuration unit is configured to perform basic parameter extraction processing in combination with geological survey data and a three-dimensional numerical inversion method, obtain a basic parameter set by identifying tunnel geometric dimensions, surrounding rock mechanical parameters and ground stress characteristics, and the like.

[0116] The second configuration unit is configured to perform core influence factor setting processing based on the basic parameter set, select excavation rate, tunnel depth, lateral pressure coefficient and surrounding rock grade as dynamic variables, wherein the surrounding rock grade is quantitatively divided based on the geological strength index, obtain a core factor definition, and the like.

[0117] The third configuration unit is configured to perform parameter integration processing based on the core factor definition, construct an influence factor coupling relationship by associating the geological strength and the ground stress characteristics, and obtain a parameterized model configuration.

[0118] In one specific embodiment of the present application, the construction module 903 includes:

[0119] The first construction unit is configured to perform jointed rock mass model construction processing based on the parameterized model configuration, establish a three-dimensional discrete joint network by mapping the joint group number, spacing and spatial distribution characteristics of the geological strength index, and obtain a parameterized surrounding rock quality model.

[0120] The second construction unit is configured to perform multi-working-condition dynamic response analysis processing based on the parameterized surrounding rock quality model, simulate the surrounding rock vibration characteristics in a high ground stress environment by cross-combining the tunnel depth, lateral pressure coefficient and excavation rate, and obtain dynamic response data.

[0121] The third construction unit is configured to perform inherent frequency sample generation processing based on the dynamic response data, extract the fundamental frequency vibration mode under different surrounding rock qualities and quantize the frequency value, and obtain an inherent frequency numerical sample data set.

[0122] In an embodiment of the present application, the analysis module 904 comprises:

[0123] a first analysis unit configured to perform evaluation sequence construction processing according to the inherent frequency numerical sample data set, to define the excavation rate, the tunnel depth, the lateral pressure coefficient and the surrounding rock grade as the evaluation index sequence, to define the inherent frequency as the reference sequence, and to obtain an initial evaluation matrix;

[0124] a second analysis unit configured to perform data standardization processing according to the initial evaluation matrix, to eliminate the dimensional differences of each factor by mean value, and to generate a dimensionless data sequence matrix;

[0125] a third analysis unit configured to perform correlation quantification processing according to the dimensionless data sequence matrix, to calculate the correlation coefficient and the mean correlation degree, to output the correlation ranking of each influencing factor and the inherent frequency, and to obtain the main control factor screening result.

[0126] In an embodiment of the present application, the modeling module 905 comprises:

[0127] a first modeling unit configured to perform training data set construction processing according to the main control factor screening result, to correlate the surrounding rock grade, the tunnel depth and the lateral pressure coefficient parameters in the inherent frequency numerical sample with the frequency label, and to obtain a main control factor training set;

[0128] a second modeling unit configured to perform feature space mapping processing according to the main control factor training set, to extract the stiffness attenuation and vibration modal characteristics of jointed rock mass under high stress environment by nonlinear transformation, and to obtain a high-dimensional feature vector;

[0129] a third modeling unit configured to perform prediction model training processing according to the high-dimensional feature vector, to optimize the mapping weight of the main control factor parameters and the inherent frequency by error back propagation, and to obtain an inherent frequency prediction model.

[0130] In an embodiment of the present application, the prediction module 906 comprises:

[0131] a first prediction unit configured to perform real-time vibration signal acquisition processing according to the field construction environment, to synchronously obtain the surrounding rock vibration waveform time domain data in each direction by a vibration monitor, and to obtain an original vibration signal set;

[0132] a second prediction unit configured to perform main control factor feature extraction processing according to the original vibration signal set, to separate the jointed rock mass fundamental frequency harmonic component under high stress environment by filtering and noise reduction and time-frequency transformation, and to obtain a standardized main control factor input vector;

[0133] The third prediction unit is configured to perform inherent frequency prediction processing according to the standardized master factor input vector, to calculate an output of a base frequency response value dominated by stiffness attenuation of surrounding rock through forward propagation, and to obtain inherent frequency time-frequency characteristics and dominant frequency information of the target deep-buried joint rock mass tunnel.

[0134] The above merely provides a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be encompassed in the protection scope of the present application.

Claims

1. A method for determining the natural frequency of tunnel surrounding rock based on numerical samples, characterized in that, The method comprises the following steps: obtaining geological survey data of a target deep-buried jointed rock mass tunnel engineering; establishing a parameter model and setting parameters according to the geological survey data in combination with a three-dimensional numerical inversion method, obtaining a parameterized model configuration by setting an excavation rate, a tunnel burial depth, a lateral pressure coefficient and a surrounding rock grade as core influencing factors; performing numerical model construction processing according to the parameterized model configuration, simulating joint distribution and mechanical response under different surrounding rock qualities, and performing dynamic mechanical analysis in combination with working condition cross combination to generate an inherent frequency numerical sample data set; performing sensitivity analysis according to the inherent frequency numerical sample data set, quantifying the correlation degree of each influencing factor and the inherent frequency based on a preset gray correlation theory model to obtain a main control factor screening result; establishing a prediction model according to the main control factor screening result, extracting features from the numerical sample based on a preset machine learning model to construct a mapping relationship between each main control factor parameter and a target deep-buried jointed rock mass tunnel inherent frequency main frequency, and obtaining an inherent frequency prediction model; inputting real-time main control factor data obtained by field monitoring into the inherent frequency prediction model for prediction processing to obtain inherent frequency time-frequency characteristics and main frequency information of the target deep-buried jointed rock mass tunnel; wherein the numerical model construction processing according to the parameterized model configuration, the simulation of joint distribution and mechanical response under different surrounding rock qualities, and the dynamic mechanical analysis in combination with working condition cross combination to generate an inherent frequency numerical sample data set comprise: performing jointed rock mass model construction processing according to the parameterized model configuration, establishing a three-dimensional discrete joint network by mapping joint group number, interval and spatial distribution characteristics based on a geological strength index to obtain a parameterized surrounding rock quality model; performing multi-working condition dynamic response analysis processing according to the parameterized surrounding rock quality model, simulating surrounding rock vibration characteristics in a high ground stress environment by cross combination of tunnel burial depth, lateral pressure coefficient and excavation rate to obtain dynamic response data; performing inherent frequency sample generation processing according to the dynamic response data, extracting fundamental frequency vibration modes under different surrounding rock qualities and quantifying frequency values to obtain an inherent frequency numerical sample data set.

2. The method according to claim 1, wherein, The method comprises the following steps: obtaining geological survey data of a target deep-buried jointed rock mass tunnel engineering; establishing a parameter model and setting parameters according to the geological survey data in combination with a three-dimensional numerical inversion method, obtaining a parameterized model configuration by setting an excavation rate, a tunnel burial depth, a lateral pressure coefficient and a surrounding rock grade as core influencing factors; performing numerical model construction processing according to the parameterized model configuration, simulating joint distribution and mechanical response under different surrounding rock qualities, and performing dynamic mechanical analysis in combination with working condition cross combination to generate an inherent frequency numerical sample data set; performing sensitivity analysis according to the inherent frequency numerical sample data set, quantifying the correlation degree of each influencing factor and the inherent frequency based on a preset gray correlation theory model to obtain a main control factor screening result; establishing a prediction model according to the main control factor screening result, extracting features from the numerical sample based on a preset machine learning model to construct a mapping relationship between each main control factor parameter and a target deep-buried jointed rock mass tunnel inherent frequency main frequency, and obtaining an inherent frequency prediction model; inputting real-time main control factor data obtained by field monitoring into the inherent frequency prediction model for prediction processing to obtain inherent frequency time-frequency characteristics and main frequency information of the target deep-buried jointed rock mass tunnel; wherein the numerical model construction processing according to the parameterized model configuration, the simulation of joint distribution and mechanical response under different surrounding rock qualities, and the dynamic mechanical analysis in combination with working condition cross combination to generate an inherent frequency numerical sample data set comprise: performing jointed rock mass model construction processing according to the parameterized model configuration, establishing a three-dimensional discrete joint network by mapping joint group number, interval and spatial distribution characteristics based on a geological strength index to obtain a parameterized surrounding rock quality model; performing multi-working condition dynamic response analysis processing according to the parameterized surrounding rock quality model, simulating surrounding rock vibration characteristics in a high ground stress environment by cross combination of tunnel burial depth, lateral pressure coefficient and excavation rate to obtain dynamic response data; performing inherent frequency sample generation processing according to the dynamic response data, extracting fundamental frequency vibration modes under different surrounding rock qualities and quantifying frequency values to obtain an inherent frequency numerical sample data set. The method comprises the following steps: obtaining geological survey data of a target deep-buried jointed rock mass tunnel engineering; establishing a parameter model and setting parameters according to the geological survey data in combination with a three-dimensional numerical inversion method, obtaining a parameterized model configuration by setting an excavation rate, a tunnel burial depth, a lateral pressure coefficient and a surrounding rock grade as core influencing factors; performing numerical model construction processing according to the parameterized model configuration, simulating joint distribution and mechanical response under different surrounding rock qualities, and performing dynamic mechanical analysis in combination with working condition cross combination to generate an inherent frequency numerical sample data set; performing sensitivity analysis according to the inherent frequency numerical sample data set, quantifying the correlation degree of each influencing factor and the inherent frequency based on a preset gray correlation theory model to obtain a main control factor screening result; establishing a prediction model according to the main control factor screening result, extracting features from the numerical sample based on a preset machine learning model to construct a mapping relationship between each main control factor parameter and a target deep-buried jointed rock mass tunnel inherent frequency main frequency, and obtaining an inherent frequency prediction model; inputting real-time main control factor data obtained by field monitoring into the inherent frequency prediction model for prediction processing to obtain inherent frequency time-frequency characteristics and main frequency information of the target deep-buried jointed rock mass tunnel; wherein the numerical model construction processing according to the parameterized model configuration, the simulation of joint distribution and mechanical response under different surrounding rock qualities, and the dynamic mechanical analysis in combination with working condition cross combination to generate an inherent frequency numerical sample data set comprise: performing jointed rock mass model construction processing according to the parameterized model configuration, establishing a three-dimensional discrete joint network by mapping joint group number, interval and spatial distribution characteristics based on a geological strength index to obtain a parameterized surrounding rock quality model; performing multi-working condition dynamic response analysis processing according to the parameterized surrounding rock quality model, simulating surrounding rock vibration characteristics in a high ground stress environment by cross combination of tunnel burial depth, lateral pressure coefficient and excavation rate to obtain dynamic response data; performing inherent frequency sample generation processing according to the dynamic response data, extracting fundamental frequency vibration modes under different surrounding rock qualities and quantifying frequency values to obtain an inherent frequency numerical sample data set.

3. The method according to claim 1, wherein, According to the natural frequency numerical sample data set, sensitivity analysis is performed, the correlation degree of each influencing factor and the natural frequency is quantified based on a preset grey correlation theory model, and a main control factor screening result is obtained, including: According to the natural frequency numerical sample data set, an evaluation sequence construction process is performed, the excavation rate, tunnel depth, lateral pressure coefficient and surrounding rock grade are defined as the evaluation index sequence, and the natural frequency is defined as the reference sequence, so as to obtain an initial evaluation matrix; According to the initial evaluation matrix, data standardization processing is performed, the dimension difference of each factor is eliminated through mean value, and a dimensionless data sequence matrix is generated; According to the dimensionless data sequence matrix, correlation quantification processing is performed, the correlation coefficient and the mean correlation degree are calculated, the correlation ranking of each influencing factor and the natural frequency is output, and the main control factor screening result is obtained.

4. The method of claim 1, wherein, According to the main control factor screening result, a prediction model is established, features are extracted from the numerical sample based on a preset machine learning model, and a mapping relationship between each main control factor parameter and the target deep buried jointed rock mass tunnel natural frequency main frequency is constructed, so as to obtain a natural frequency prediction model, including: According to the main control factor screening result, a training data set construction process is performed, the surrounding rock grade, tunnel depth and lateral pressure coefficient parameters in the natural frequency numerical sample are associated with the frequency label, and a main control factor training set is obtained; According to the main control factor training set, feature space mapping processing is performed, the stiffness attenuation and vibration modal characteristics of the jointed rock mass in the high ground stress environment are extracted through nonlinear transformation, and a high-dimensional feature vector is obtained; According to the high-dimensional feature vector, prediction model training processing is performed, the mapping weight of the main control factor parameter and the natural frequency is optimized through error back propagation, and the natural frequency prediction model is obtained.

5. A system for determining the natural frequency of tunnel surrounding rock based on numerical samples, characterized in that, It includes: An acquisition module is configured to acquire geological survey data of a target deep buried jointed rock mass tunnel project; A configuration module is configured to establish a parameter model and set parameters based on the geological survey data and a three-dimensional numerical inversion method, set the excavation rate, tunnel depth, lateral pressure coefficient and surrounding rock grade as core influencing factors, and obtain a parameterized model configuration; A construction module is configured to construct a numerical model based on the parameterized model configuration, simulate joint distribution and mechanical response under different surrounding rock qualities, and perform dynamic mechanical analysis by combining working condition cross combination to generate a natural frequency numerical sample data set; An analysis module is configured to perform sensitivity analysis based on the natural frequency numerical sample data set, quantify the correlation degree of each influencing factor and the natural frequency based on a preset grey correlation theory model, and obtain a main control factor screening result; A modeling module is configured to establish a prediction model based on the main control factor screening result, extract features from the numerical sample based on a preset machine learning model, and construct a mapping relationship between each main control factor parameter and the target deep buried jointed rock mass tunnel natural frequency main frequency, to obtain a natural frequency prediction model; A prediction module is configured to input real-time main control factor data obtained by field monitoring into the natural frequency prediction model for prediction processing to obtain natural frequency time-frequency characteristics and main frequency information of the target deep buried jointed rock mass tunnel. The construction module includes: The first construction unit is configured to perform a jointed rock mass model construction process according to the parameterized model configuration, to establish a three-dimensional discrete joint network by mapping the number of joint sets, spacing and spatial distribution characteristics of the geological strength index, and to obtain a parameterized surrounding rock quality model; The second construction unit is configured to perform a multi-working condition dynamic response analysis process according to the parameterized surrounding rock quality model, to simulate the vibration characteristics of surrounding rock in a high ground stress environment by cross-combining the tunnel depth, lateral pressure coefficient and excavation rate, and to obtain dynamic response data; The third construction unit is configured to perform an inherent frequency sample generation process according to the dynamic response data, to extract the fundamental frequency vibration mode under different surrounding rock qualities and to quantify the frequency value, and to obtain an inherent frequency numerical sample data set.

6. The system for determining the natural frequency of the tunnel surrounding rock based on numerical samples according to claim 5, wherein, The configuration module comprises: The first configuration unit is configured to perform a basic parameter extraction process according to the geological survey data in combination with a three-dimensional numerical inversion method, to obtain a basic parameter set by identifying the tunnel geometric size, surrounding rock mechanical parameters and ground stress characteristics; The second configuration unit is configured to perform a core influencing factor setting process according to the basic parameter set, to select the excavation rate, tunnel depth, lateral pressure coefficient and surrounding rock grade as dynamic variables, wherein the surrounding rock grade is quantitatively divided based on the geological strength index, and to obtain a core factor definition; The third configuration unit is configured to perform a parameter integration process according to the core factor definition, to construct the influencing factor coupling relationship by associating the geological strength and ground stress characteristics, and to obtain a parameterized model configuration.

7. The system for determining the natural frequency of the tunnel surrounding rock based on numerical samples according to claim 5, characterized in that, The analysis module comprises: The first analysis unit is configured to perform an evaluation sequence construction process according to the inherent frequency numerical sample data set, to define the excavation rate, tunnel depth, lateral pressure coefficient and surrounding rock grade as an evaluation index sequence, and the inherent frequency as a reference sequence, and to obtain an initial evaluation matrix; The second analysis unit is configured to perform a data standardization process according to the initial evaluation matrix, to eliminate the dimensional differences of each factor by mean value, and to generate a dimensionless data sequence matrix; The third analysis unit is configured to perform a correlation quantification process according to the dimensionless data sequence matrix, to calculate the correlation coefficient and the mean correlation degree, to output the correlation ranking of each influencing factor and the inherent frequency, and to obtain a main control factor screening result.

8. The system for determining the natural frequency of the tunnel surrounding rock based on numerical samples according to claim 5, wherein, The modeling module comprises: The first modeling unit is configured to perform a training data set construction process according to the main control factor screening result, to associate the surrounding rock grade, tunnel depth and lateral pressure coefficient parameters in the inherent frequency numerical sample with the frequency label, and to obtain a main control factor training set; The second modeling unit is configured to perform a feature space mapping process according to the main control factor training set, to extract the stiffness attenuation and vibration mode characteristics of the jointed rock mass under high ground stress environment by nonlinear transformation, and to obtain a high-dimensional feature vector; The third modeling unit is configured to perform a prediction model training process according to the high-dimensional feature vector, to optimize the mapping weight of the main control factor parameters and the inherent frequency by error back propagation, and to obtain an inherent frequency prediction model.

Images