Multi-dimensional comprehensive prediction and characterization method and system for water gushing in water-rich sandy cobble stratum tunnel

Abstract

The application discloses a kind of water-rich sand cobble stratum tunnel water gushing multidimensional comprehensive prediction and characterization method and system, it is related to tunnel engineering technical field, including: to water-rich sand cobble stratum tunnel construction area is carried out geological survey, obtains geological data, and lays out multiple monitoring sensors, real-time collection hydrodynamics and hydrochemical data;Physical model of tunnel area is built, including three-dimensional geological model, hydrodynamics model and hydrochemical model;Based on hydrodynamics model and hydrochemical model, according to the real-time collection hydrodynamics data and hydrochemical data are dynamically water gushing prediction, output the next time step prediction water gushing intensity, spatial distribution and time variation trend;Predictive data is mapped to three-dimensional geological model, generates flow field vector diagram and concentration distribution heat map, and the prediction result is compared with the monitoring data of next time step, to dynamically adjust and update model parameters.The application can realize high-precision dynamic water gushing prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel engineering, and particularly relates to a method and system for multi-dimensional comprehensive prediction and characterization of water gushing in a water-rich sandy pebble stratum. BACKGROUND

[0002] The statements in this section merely provide background information related to the present application and do not necessarily constitute the prior art.

[0003] In the process of tunnel construction, water-rich sandy pebble strata are often areas with a high incidence of water gushing. Water gushing not only affects the progress and safety of tunnel construction, but also can cause geological disasters, equipment damage, and casualties of construction personnel, etc. Therefore, accurate prediction and prevention of water gushing are important tasks in tunnel engineering. Traditional water gushing prediction methods can be divided into four categories: prediction methods based on geological exploration data, prediction methods based on hydrogeological modeling, single analysis methods based on hydrodynamic parameters, and analysis methods based on water chemical characteristics. However, water gushing is the result of the combined action of multiple factors, and multi-factor data includes the dynamic characteristics of water flow, the source of groundwater, and its chemical composition, etc. The above traditional methods usually rely on a single type of data such as geological data or hydrological data, while ignoring the multi-dimensional characteristics of water flow and the mutual influence between multi-source data, resulting in difficulty in providing comprehensive prediction results and low accuracy of water gushing prediction. Moreover, traditional methods are usually based on historical data or static analysis, and are difficult to reflect the dynamic changes of water gushing in real time, and cannot cope with the rapid changes of water gushing in complex geological environments.

[0004] With the continuous development of tunnel construction technology, especially in the face of complex water-rich sandy pebble strata, traditional prediction methods have been unable to meet the demand for high-precision prediction. Therefore, more and more research has begun to focus on multi-dimensional data fusion technology, combining hydrodynamic parameters, geological background information, and water chemical characteristics to conduct more comprehensive analysis and prediction. However, although existing methods consider the dynamics of water flow and water chemical characteristics, these methods still have many shortcomings, including:

[0005] (1) Existing monitoring methods rely on a small number of sensors, and the monitoring points are usually laid out and selected in the tunnel, which cannot comprehensively cover all potential water gushing risk areas, resulting in the inability to timely predict and discover water gushing risks in some key areas;

[0006] (2) Existing prediction methods often rely on fixed historical data, and cannot dynamically track changes in water gushing areas, making it impossible to achieve real-time and accurate prediction of the water gushing process;

[0007] (3) Existing prediction models are usually based on empirical formulas, which lack flexibility and cannot adapt to changes in different tunnel conditions, resulting in poor final prediction accuracy. SUMMARY

[0008] To solve the above problems of the prior art, the application provides a water-rich sand and gravel stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization method and system, which comprehensively considers spatial distribution and water gushing risk, rationally arranges a dense sensor network at multiple key positions such as a tunnel and a geological layer around the tunnel and a potential water gushing area, comprehensively considers real-time monitoring data of water flow, water pressure and water chemical composition, combines geological background information, constructs a comprehensive and dynamic water gushing prediction model through multi-dimensional data fusion, and realizes high-precision and dynamic water gushing prediction, thereby providing accurate water gushing prediction for water-rich sand and gravel stratum tunnel construction.

[0009] In a first aspect, the application provides a water-rich sand and gravel stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization method.

[0010] A water-rich sand and gravel stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization method comprises the following steps.

[0011] Geological investigation is performed on a water-rich sand and gravel stratum tunnel construction area to obtain geological data of the tunnel area, and various monitoring sensors are arranged to collect water dynamics data and water chemistry data in real time.

[0012] A physical model of the tunnel area is constructed, including a three-dimensional geological model, a water dynamics model and a water chemistry model.

[0013] Based on the water dynamics model and the water chemistry model, dynamic water gushing prediction is performed according to the real-time collected water dynamics data and water chemistry data, and the predicted water gushing intensity, spatial distribution and time variation trend of the next time step are output.

[0014] The predicted water gushing intensity and spatial distribution of the next time step are mapped into the three-dimensional geological model to generate a flow field vector diagram and a concentration distribution heat map, and the prediction result is compared with the monitoring data of the next time step, so as to dynamically adjust and update the model parameters.

[0015] In a further technical solution, the geological data comprises underground stratum structure, hydrogeological condition and geological feature data, wherein the underground stratum structure comprises the thickness and permeability coefficient of the sand and gravel layer, the hydrogeological condition comprises underground water level, hydraulic gradient and aquifer distribution, and the geological feature comprises a fault zone and a karst potential water gushing channel.

[0016] The water dynamics data comprises water level, water flow, flow rate and water pressure parameter data, and the water chemistry data comprises ion concentration, pH value, oxidation-reduction potential and dissolved solid parameter data in the water body.

[0017] In a further technical solution, the construction of the three-dimensional geological model comprises the following steps.

[0018] A three-dimensional geological model of the tunnel area is established by using a three-dimensional geological modeling tool;

[0019] The obtained geological data is integrated as an input layer to generate a three-dimensional geological model of the three-dimensional sandy pebble stratum, fault zone, aquifer and karst distribution.

[0020] Further technical solutions, the construction of the hydrodynamic model and the water chemical model, comprising:

[0021] According to the principle of seepage mechanics, a mathematical model describing the flow of groundwater is established, and a partial differential equation set describing the flow of water is established by combining the fluid mechanics equation, as the mathematical basis of the model; wherein the fluid mechanics equation includes Darcy's law and continuity equation;

[0022] According to the convection-diffusion-reaction equation, the mathematical basis equation of the water chemical model is established, which is used to simulate the migration and reaction of chemical substances in groundwater;

[0023] Coupling the data basis equation of the hydrodynamic model and the water chemical model, a coupled model is constructed for analyzing the influence of chemical reaction on permeability and the feedback of flow rate on water chemical change;

[0024] The coupled model is solved by using the finite element method to obtain the hydrodynamic equation and the water chemical equation, thereby constructing the hydrodynamic model and the water chemical model as the dynamic water inflow prediction model.

[0025] Further technical solutions, the coupled model is solved by using the finite element method to obtain the hydrodynamic equation and the water chemical equation, comprising:

[0026] Taking the coupled model equation set as the control equation, the tunnel area is defined according to the three-dimensional geological structure model, and the boundary conditions and initial conditions of the area are defined;

[0027] The tunnel area is discretized, and the weighted residual method is used to solve the control equation to obtain the weak form, that is, the hydrodynamic equation and the water chemical equation.

[0028] Further technical solutions, the dynamic water inflow prediction is:

[0029] The hydrodynamic equation and the water chemical equation are discretized to form discrete equations, and the stiffness matrix K e and the load vector F e of the water dynamics and water chemistry of the calculation area unit are calculated.

[0030] The stiffness matrix K e and the load vector F eAssembled into a global matrix K and a load vector F, combined with real-time collected hydrodynamic data and hydrochemical data, time discretization and solving are performed to obtain a prediction equation of water head in hydrodynamics and a prediction equation of chemical component concentration in hydrochemistry;

[0031] Based on the prediction equation, for each time step, a global linear equation group is solved to predict the water head and chemical component concentration of the next time step and to update.

[0032] In a second aspect, the present application provides a water-rich sand and gravel stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization system.

[0033] A water-rich sand and gravel stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization system comprises:

[0034] A data acquisition module is configured to conduct geological investigation on a water-rich sand and gravel stratum tunnel construction area, acquire geological data of the tunnel area, and arrange a plurality of monitoring sensors to collect real-time hydrodynamic data and hydrochemical data.

[0035] A physical modeling module is configured to construct a physical model of the tunnel area, including a three-dimensional geological model, a hydrodynamic model and a hydrochemical model.

[0036] A dynamic water gushing prediction module is configured to perform dynamic water gushing prediction based on the hydrodynamic model and the hydrochemical model according to the real-time collected hydrodynamic data and hydrochemical data, and output the predicted water gushing intensity, spatial distribution and time variation trend of the next time step.

[0037] A water gushing characterization module is configured to map the predicted water gushing intensity and spatial distribution of the next time step to the three-dimensional geological model to generate a flow field vector diagram and a concentration distribution heat map, and compare the prediction result with the monitoring data of the next time step to dynamically adjust and update the model parameters.

[0038] In a third aspect, the present application further provides an electronic device comprising a memory for storing executable instructions and a processor for executing the executable instructions stored in the memory to implement the water-rich sand and gravel stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization method.

[0039] In a fourth aspect, the present application further provides a computer readable storage medium storing executable instructions for causing a processor to execute the executable instructions to implement the water-rich sand and gravel stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization method.

[0040] In a fifth aspect, the present application also provides a computer program product, which comprises executable instructions stored in a computer readable storage medium; wherein the processor of the electronic device reads the executable instructions from the computer readable storage medium and executes the executable instructions to implement the above-mentioned method for multi-dimensional comprehensive prediction and characterization of water inrush in a water-rich sandy cobble stratum tunnel.

[0041] The above one or more technical solutions have the following beneficial effects:

[0042] 1. The present application provides a method and system for multi-dimensional comprehensive prediction and characterization of water inrush in a water-rich sandy cobble stratum tunnel, which comprehensively considers spatial distribution and water inrush risk by reasonably arranging a dense sensor network at multiple key positions such as the tunnel and surrounding geological layers, potential water inrush areas, etc., to comprehensively consider real-time monitoring data of water flow, water pressure, and water chemical composition, combine with geological background information, and construct a comprehensive and dynamic water inrush prediction model through multi-dimensional data fusion. This model not only considers the physical properties of water flow but also combines the chemical properties of water, enabling more comprehensive and high-precision dynamic water inrush prediction, comprehensive and accurate prediction of water inrush intensity, time, and spatial distribution, and significantly improving the accuracy and reliability of the prediction, providing accurate water inrush prediction for water-rich sandy cobble stratum tunnel construction.

[0043] 2. Compared with the traditional monitoring and prediction method relying on a small number of single sensors, the present application arranges a dense sensor network in the tunnel construction area and the surrounding potential water inrush risk area, i.e., the arrangement is not limited to the tunnel itself but also includes geological layers, potential water inrush areas, underground water levels, underground water flow directions, and multiple key positions. By comprehensively considering spatial distribution and water inrush risk, reasonably arranging sensors, and the cooperative work of multiple sensors, real-time monitoring of all potential water inrush sources and their paths is ensured, and this dense sensor arrangement helps to fill the monitoring blind area in traditional methods and achieves comprehensive coverage of water inrush risk areas.

[0044] 3. The present application dynamically optimizes and feedback adjusts the prediction model in combination with real-time monitoring data to dynamically update the prediction results according to the actual situation and water inrush changes, and optimizes risk assessment and intervention measures, making the prediction and risk management more flexible, real-time, and accurate.

[0045] 4. The present application uses water chemical fingerprint identification technology to accurately identify the specific source of water inrush, such as groundwater, surface water leakage, karst water, etc., by analyzing ion concentration, pH value, isotope, and other chemical components in the water sample. This method not only improves the accuracy of water source identification but also provides more accurate input data for subsequent water inrush path prediction.

[0046] 5、The application can accurately predict the water gushing path by combining the water dynamics and water chemistry models, and can not only identify the trend of the water gushing channel, but also can real-time draw the spatial distribution map of the water gushing, intuitively display the high-risk area, and provide scientific risk assessment and construction decision basis for the tunnel construction personnel.

[0047] Advantages of the additional aspects of the application will be partly given in the following description, partly will become obvious from the following description, or will be understood by the practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0048] The drawings accompanying the specification of this application form a part thereof, serve to provide further understanding of the application, and together with the description of the exemplary embodiments of the application given below, explain the application, and do not limit the application in an inappropriate manner.

[0049] Figure 1 The overall flowchart of the water-rich sandy pebble stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization method described in the embodiments of the application. DETAILED DESCRIPTION

[0050] It should be noted that the following detailed description is exemplary only, is only for the purpose of describing specific embodiments, and is intended to provide further explanation of the application, and is not intended to limit the exemplary embodiments according to the application. Unless otherwise indicated, all technical and scientific terms used herein have the same meaning as generally understood by those skilled in the art to which the application belongs. In addition, it should be understood that when the terms "comprise" and / or "include" are used in the specification, they refer to the presence of a feature, step, operation, device, component and / or their combination.

[0051] Embodiment one

[0052] The embodiment provides a water-rich sandy pebble stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization method, as shown in the figure, specifically comprising the following steps: Figure 1 As shown in the figure, specifically comprising the following steps:

[0053] Step S1, carrying out geological investigation on the water-rich sandy pebble stratum tunnel construction area, obtaining the geological data of the tunnel area, and laying multiple monitoring sensors to collect water dynamics data and water chemistry data in real time.

[0054] Specifically, detailed underground geological investigation is carried out on the water-rich sandy pebble stratum tunnel construction area, such as through geological exploration (such as drilling, geological profile, etc.), resistivity detection, seismic wave detection, etc., to obtain the geological data of the area, which includes underground stratum structure (thickness of sandy pebble layer, permeability coefficient, etc.), hydrogeological conditions (such as underground water level, hydraulic gradient, aquifer distribution, etc.), geological features (such as fault zone, karst, etc. potential water gushing channel) data, etc.

[0055] Further, according to the geological survey results, various hydrodynamic sensors are arranged in the tunnel construction area, such as water level sensors, flow meters, flow meters, pressure sensors, etc., to detect water level, water flow, flow rate, water pressure and other hydrodynamic parameter data in real time; Similarly, various water chemical monitoring equipment is arranged, and through the installation of chemical sensors such as ion concentration meter, pH meter, etc., ion concentration, pH value, oxidation-reduction potential, dissolved solids and other water chemical parameter data in the water body are collected and obtained in real time.

[0056] Preferably, a real-time data transmission system is established to transmit the monitoring data to a remote central processing unit for subsequent data analysis and processing. Among them, various sensors measure the physical changes (such as flow rate and pressure) and chemical composition (such as ion concentration) of the water flow in real time, and the sampling frequency of each sensor can be set according to specific needs, for example, once a minute or once a second. In this way, various sensors continuously collect hydrodynamic and hydrochemical parameter data in the tunnel area and send the data to the sensor network gateway; the sensor network gateway receives and preliminarily processes the raw data, and after noise filtering and format conversion, etc., it is transmitted to the central processing unit through wired or wireless network, and reliable data transmission protocol (such as HTTP protocol) can be used in the transmission process to ensure the accuracy and real-time of data transmission.

[0057] Step S2, a physical model of the tunnel area is constructed, including a three-dimensional geological model, a hydrodynamic model and a hydrochemical model.

[0058] Step S2.1, first, a three-dimensional geological model of the tunnel area is established using a three-dimensional geological modeling tool (such as RockWorks, etc.), integrating the geological, drilling, geophysical data obtained in the previous step as the input layer, generating a three-dimensional geological model of the three-dimensional sand and gravel stratum, fault zone, aquifer and karst distribution, and outputting as the output layer.

[0059] Step S2.2, secondly, modeling of hydrodynamics and hydrochemistry is performed.

[0060] Step S2.2.1, according to the principles of seepage mechanics and fluid mechanics equations, the mathematical basis equations of the hydrodynamic model are established to describe the flow of groundwater. Specifically, according to the principles of seepage mechanics, a mathematical model is established to describe the flow of groundwater, and combined with the fluid mechanics equations, a partial differential equation set is established to describe the flow of water as the mathematical basis of the model. Among them, the fluid mechanics equations include Darcy's law and continuity equation.

[0061] Among them, Darcy's law is:

[0062] ;

[0063] In the above formula, qis the seepage flux, k is the hydraulic conductivity, h is the hydraulic head.

[0064] The continuity equation is:

[0065] ;

[0066] In the above equation, is the storage coefficient, denotes the rate of change of hydraulic head with respect to time.

[0067] Thus, the basic partial differential equations describing three-dimensional groundwater flow are established as:

[0068] ;

[0069] In the above equation, is the hydraulic head, which is an unknown quantity; k x 、k y 、k z is the hydraulic conductivity in different directions (anisotropy is considered here); S s is the storage coefficient, which describes the storage of aquifer per unit volume caused by the change of hydraulic head; W is the source / sink term, which represents the rate of external water flow into (such as surface infiltration) or out of (such as pumping) the system; t is the time variable.

[0070] Step S2.2.2. Establish the mathematical foundation equation of the water chemistry model according to the convection-diffusion-reaction equation, which is used to simulate the migration and reaction of chemical substances in groundwater. Specifically, the control equation of chemical substance migration and reaction is shown in the following formula, which describes the influence of chemical components in the aquifer by fluid convection ( ), diffusion ( ) and reaction ( ), which is:

[0071] ;

[0072] In the above equation, is the concentration of chemical components (such as ion concentration), which is an unknown quantity; , which represents the flow velocity vector of groundwater (provided by the hydrodynamic model); K is the influence on the hydraulic conductivity; D is the diffusion coefficient of chemical substances (which considers porosity and formation characteristics); is the chemical reaction term, which represents the contribution of concentration change caused by chemical reaction.

[0073] Step S2.2.3. Coupling the data base equations of the hydrodynamic model and the hydrochemical model to construct a coupled model for analyzing the influence of chemical reactions on permeability and the feedback of flow velocity on hydrochemical changes.

[0074] Specifically, in tunnel water inflow prediction, the hydrodynamic model and the hydrochemical model need to be solved simultaneously. To this end, the above equations are coupled to establish a coupled model. The main equation groups of the coupled model are Darcy's law, the basic partial differential equation, and the convection-diffusion-reaction equation. There are two coupling points: one is that water flow velocity affects chemical migration, and the flow velocity q is provided by the hydrodynamic model, directly affecting the convection term in the hydrochemical model ); the other is that chemical changes feed back hydraulic properties. Chemical reactions may cause precipitation or dissolution, such as carbonate precipitation, and this property includes the influence on the permeability K and the diffusion coefficient D .

[0075] Step S2.2.4. Solving the coupled model using the finite element method (FEM) to obtain the hydrodynamic equation and the hydrochemical equation, thereby constructing the hydrodynamic model and the hydrochemical model as a dynamic water inflow prediction model.

[0076] Specifically, first, problem description and equation establishment are performed, including: determining the control equation (i.e., the above coupled equation group), defining the region (i.e., the three-dimensional geological structure model Ω) and the boundary conditions, and the initial conditions, wherein the boundary conditions include: for hydrodynamics, the given boundary water pressure h or the permeable boundary flux ; for hydrochemistry, the given boundary concentration C or the chemical flux ; the initial conditions include: the initial water head distribution , and the initial concentration distribution .

[0077] Then, finite element discretization is performed, and the tunnel region Ω is discretized (i.e., meshing, divided into multiple unit regions). In the discretization process, the piecewise linear function is used as the trial function, where is the shape function, is the i th unit region. i

[0078] After that, the weighted residual method is used to solve the control equation to obtain the weak form of the hydrodynamic equation and the hydrochemical equation, which are represented as:

[0079] ;

[0080] ;​

[0081] wherein, is the weight function, the above integral term is calculated in the element.

[0082] Step S3, based on the hydrodynamic model and the hydrochemical model, according to the real-time collected hydrodynamic data and hydrochemical data, dynamic water inflow is predicted, and the predicted water inflow intensity, spatial distribution and time variation trend of the next time step are output.

[0083] Specifically, the hydrodynamic equation and the hydrochemical equation are discretized to form discrete equations, and the stiffness matrix K e and the load vector F e of the calculation unit are calculated. e Among them, the stiffness matrix K e and the load vector F e of the hydrodynamics are:

[0084] ;

[0085] ;

[0086] The stiffness matrix K e and the load vector F e of the hydrochemistry are:

[0087] ;

[0088] .

[0089] Then, the stiffness matrix K e and the load vector F h of all elements are assembled into the global matrix K and the load vector F, according to , , combined with the real-time collected hydrodynamic data and hydrochemical data, the nodes with known water head or concentration are directly assigned and the corresponding matrix rows and columns are adjusted. Among them, H and C are the global vectors of water head and concentration; K c and K h are the stiffness matrixes, which correspond to the system stiffness of the hydrodynamic (h) and hydrochemical (C) models respectively, K c is the hydrodynamic stiffness matrix, which describes the spatial variation relationship of water flow (water head), related to the gradient of water flow, permeability coefficient, boundary condition and other factors, while K h is the hydrochemical stiffness matrix, which describes the spatial distribution relationship of chemical components (such as ion concentration), considering chemical reaction, diffusion, convection and other factors; F c and F hFw is the water dynamic load vector, which represents the source term of external water flow influence, such as external water source, seepage water flow, etc. c Fch is the water chemical load vector, which represents the external chemical source term, such as external dissolved matter, reaction source, etc.

[0090] Then, time discretization and solving are performed, i.e., the time term is discretized using the backward Euler method to obtain the prediction equation of water head in water dynamics and the prediction equation of chemical component concentration in water chemistry, as follows:

[0091]

[0092]

[0093] Based on the above equations, for each time step, the global linear equation group K h H n+1 =F h and K C C n+1 =F C are solved to predict the water head H n+1 and the chemical component concentration C n+1 of the next time step and update, where n represents the current time or the current time step, n +1 represents the next time or the next time step.

[0094] As another implementation, the present embodiment can also identify the gushing source, and the analysis of identifying the gushing source is an important part of the prediction model optimization and parameter updating as an optimization step of the gushing prediction result.

[0095] In the present embodiment, identifying the gushing source includes water chemical fingerprint identification, water dynamic gradient analysis and mixing analysis, wherein:

[0096] (1) Water chemical fingerprint identification: the type of water source is distinguished by analyzing the ion concentration, pH value, isotope, etc. in the water sample. Specifically, by analyzing the water chemical data, mainly the ion concentration (such as Na + , Ca 2+ , HCO3 - , Cl - , etc.), water pH value, oxidation-reduction potential, water temperature and dissolved gas concentration (such as CO2, O2, etc.), the water chemical characteristics (such as ion composition characteristics) are analyzed, and compared with known water samples (such as surface water, groundwater, karst water, etc.), to identify the source of the water body and its relationship with different geological units.

[0097] ​​(2) Hydrodynamic gradient analysis: By analyzing the water level gradient and flow velocity field, the flow direction and flow source of the water flow are determined. Specifically, by collecting hydrodynamic data (such as water pressure, flow velocity, gradient), the flow characteristics and gradient distribution of the groundwater around the tunnel are obtained, and the dynamic source of the water gushing is determined.

[0098] (3) Mixed analysis: Combined with multi-source water chemical data, the contribution proportion of each water source is estimated by mass balance method, and the mass balance method formula is: wherein, is the contribution proportion of source type i is the chemical concentration of source type i

[0099] In addition, through geological background data such as tunnel geological profile, aquifer distribution, fault and karst channel position, boundary conditions are provided for fluid flow; the water level contour of the area around the tunnel is calculated, the water flow direction and the possible position of the water gushing source are analyzed; with the help of groundwater flow model (i.e. partial differential equation set), the relationship between the water gushing point and the water source area is predicted.

[0100] Through the above water chemical fingerprint identification, hydrodynamic gradient analysis and mixed analysis and other technical means, the source of water gushing (such as groundwater, surface water, karst water, etc.) can be accurately determined, and the strength, flow path, etc. of the water gushing are optimized and adjusted according to the identification results, and then the prediction model can be optimized and the parameters are updated according to the data after the optimization and adjustment, so as to improve the accuracy of water gushing prediction, and provide more accurate input data for subsequent model updating.

[0101] In addition, according to the different properties and sources of the water gushing source, targeted prevention and control measures can be taken to improve the scientificity and effectiveness of water gushing risk management.

[0102] Further, in the embodiment, considering that the water gushing is only predicted according to the constructed hydrodynamic and hydrochemical model, such prediction is based on physical equations (such as Darcy's law, continuity equation, etc.) and hydrogeological data, and the physical model has certain parameter dependence and lacks flexibility, which leads to the fact that the prediction accuracy is not optimal, wherein the parameter dependence is that many parameters in the physical model such as the permeability coefficient, the water head distribution, etc. may have great uncertainty, and the measurement and estimation of these parameters usually need a large amount of field data or assumptions; lack of flexibility: the physical model generally operates under the assumption of fixed geological conditions and flow patterns, and it is difficult to handle particularly complex and dynamically changing actual engineering environments. On the basis of the above, the embodiment introduces machine learning and statistical modeling techniques to overcome the limitations of the physical model, and further improves the prediction accuracy and dynamic adaptability in a data-driven manner.

[0103] ​​Therefore, preferably, while predicting the water inrush by using the constructed hydrodynamic and hydrochemical model (i.e., the physical model), the embodiment also constructs a dynamic water inrush prediction model, i.e., a machine learning or statistical modeling technique is used to construct the dynamic water inrush prediction model. Specifically, in the model, real-time collected hydrodynamic, hydrochemical and geological information data are input, and the predicted water inrush intensity, spatial distribution and time variation trend are output. The model is verified and optimized by historical data, and the model parameters are dynamically adjusted according to the actual monitoring results to improve the prediction accuracy.

[0104] The physical model and the machine learning model work together, the physical model provides a preliminary framework for prediction, describes the basic physical mechanism and process of water inrush, and the machine learning model optimizes the parameters of the physical model and corrects the water inrush prediction results based on historical data and real-time data. That is, based on the machine learning model (i.e., the dynamic water inrush prediction model), the predicted water inrush intensity, spatial distribution and time variation trend at the next time step are predicted according to real-time hydrodynamic, hydrochemical and geological information data; the prediction results output by the model are used to correct the water inrush prediction results generated by the physical model (i.e., the hydrodynamic model and the hydrochemical model) to obtain more accurate water inrush prediction results. The correction can be realized by weighted average and the like, which will not be introduced one by one here.

[0105] By introducing the machine learning model, the prediction accuracy and adaptability can be improved: (1) model nonlinearization: machine learning can automatically learn complex nonlinear relationships from data by processing a large amount of historical data, which is particularly important for complex geological environments and dynamically changing water inrush phenomena. The machine learning model can identify details that the physical model fails to capture, such as nonlinear relationships between water flow and water chemical characteristics under certain conditions; (2) dynamic adaptability: the machine learning model has self-adaptive ability and can dynamically adjust model parameters according to real-time collected hydrodynamic data, hydrochemical data and geological data to improve prediction accuracy. For example, as new real-time data is input, the machine learning model can timely correct the predicted value of water inrush intensity and spatial distribution.

[0106] By introducing the machine learning model, the physical model can be optimized and verified: (1) historical data verification: although the physical model provides a theoretical-based prediction result, its accuracy is often affected by factors such as site conditions and parameter estimation. Therefore, by inputting historical data, the machine learning model can verify and optimize the physical model according to actual monitoring data, compensate for errors in the physical model, and the machine learning model can adjust itself within a certain range to adapt to new data patterns and changes; (2) self-adjustment and optimization: the machine learning model can adjust the weight and optimize the model structure in real time during operation, so that the difference between the output of the physical model and the actual observation data is minimized, thereby enhancing the prediction ability of the model.

[0107] Through the introduction of machine learning models, data fusion and complex pattern recognition can be achieved: (1) Data fusion: Machine learning can process and fuse data from multiple sources (such as hydrodynamics, hydrochemistry, geological data, etc.), extract the most meaningful features, and further refine the water inrush prediction; (2) Complex pattern recognition: Through a large amount of historical data and real-time monitoring data, machine learning algorithms can identify the complex patterns of water inrush intensity changes, spatial distribution, and their relationship with environmental factors, which are difficult to capture by traditional physical models.

[0108] Through the above combination of physical models and data-driven models, the uncertainty in complex geological environments can be better addressed, resulting in more accurate and flexible water inrush prediction.

[0109] Step S4, map the predicted water inrush intensity and spatial distribution of the next time step to the three-dimensional geological model, generate flow field vector maps and concentration distribution heat maps, and compare the prediction results with the monitoring data of the next time step to dynamically adjust and update the model parameters.

[0110] Specifically, combined with the prediction results, the formation causes, evolution process, and possible expansion trend of the water inrush channel are analyzed for three-dimensional spatial representation.

[0111] First, using known hydrodynamic data, the flow velocity field, flow direction field, and flow path are drawn through the output of the hydrodynamic model, and potential water inrush channels are identified; based on the analysis of changes in known hydrochemical data, the flow path and flow characteristics of the water flow are inferred, such as high , which may flow through karst areas; high , which may pass through sulfur mineral layers; high , , which may be surface water leakage or deep water rising, and reverse tracing analysis is performed, i.e., the path and residence time of the water flow are inferred based on flow velocity and direction. Among them, the possible water inrush path is identified based on the geological model, including: rock fissure and fault, karst channel, sand and gravel layer with high permeability, etc.

[0112] Second, the water head and concentration distribution results obtained by solving are mapped back to the three-dimensional geological model to generate flow field vector maps, concentration distribution heat maps, etc., and the numerical results are compared with the monitoring data to adjust the model parameters such as permeability coefficient and diffusion coefficient, and then combined with the updated monitoring data, the model is dynamically adjusted to predict the spatial and temporal distribution of tunnel water inrush intensity and chemical characteristics. In addition, based on the water inrush intensity and distribution prediction, the water inrush risk level of different regions is evaluated, and a risk assessment report is generated.

[0113] Through the above method, the spatial distribution of water flow path and water inrush channel is visualized in the three-dimensional geological model, helping engineers to intuitively understand the water inrush area.

[0114] Embodiment Two

[0115] The embodiment provides a water-rich sand and gravel stratum tunnel water gushing multi-dimensional comprehensive prediction and characterization system, which comprises the following.

[0116] The data acquisition module is used for geological investigation on a water-rich sand and gravel stratum tunnel construction area, acquires geological data of the tunnel area, and arranges a plurality of monitoring sensors to collect water dynamics data and water chemistry data in real time.

[0117] The physical modeling module is used for constructing a physical model of the tunnel area, including a three-dimensional geological model, a water dynamics model and a water chemistry model.

[0118] The dynamic water gushing prediction module is used for dynamically predicting water gushing based on the water dynamics model and the water chemistry model according to the water dynamics data and the water chemistry data collected in real time, and outputting predicted water gushing intensity, spatial distribution and time variation trend of a next time step.

[0119] The water gushing characterization module is used for mapping the predicted water gushing intensity and spatial distribution of the next time step to the three-dimensional geological model, generating a flow field vector diagram and a concentration distribution heat map, and comparing the prediction result with monitoring data of the next time step, so as to dynamically adjust and update model parameters.

[0120] Embodiment Three

[0121] The embodiment provides an electronic device, comprising a memory for storing executable instructions, and a processor for executing the executable instructions stored in the memory to implement the above method provided by the embodiment.

[0122] Embodiment Four

[0123] The embodiment also provides a computer readable storage medium storing executable instructions, which, when executed by a processor, cause the processor to execute the above method provided by the embodiment.

[0124] Embodiment Five

[0125] The embodiment provides a computer program product, which comprises executable instructions, and the executable instructions are computer instructions; the executable instructions are stored in a computer readable storage medium. When the processor of the electronic device reads the executable instructions from the computer readable storage medium, the processor executes the executable instructions, so that the electronic device executes the above method provided by the embodiment.

[0126] The steps involved in the above embodiments two to five correspond to the method of embodiment one, and the specific implementation can refer to the relevant description of embodiment one. The term "computer readable storage medium" should be understood to include a single medium or multiple media of one or more instruction sets; it should also be understood to include any medium capable of storing, encoding, or carrying the instruction sets for execution by a processor and causing the processor to perform any of the methods in the present application.

[0127] Those skilled in the art should understand that each module or step of the present application described above can be realized by a general computer device, alternatively, they can be realized by program codes executable by a computing device, so that they can be stored in a storage device for execution by a computing device, or they can be respectively made into each integrated circuit module, or a plurality of modules or steps among them can be made into a single integrated circuit module to realize. The present application is not limited to any specific combination of hardware and software.

[0128] The above description is only the preferred embodiments of the present application, although the specific embodiments of the present application are described in conjunction with the drawings, but it is not a limitation on the scope of protection of the present application, those skilled in the art should understand that various modifications or changes made on the basis of the technical solutions of the present application without the need for creative labor are still within the scope of protection of the present application.

Claims

1. A multi-dimensional comprehensive prediction and characterization method for water inflow in a water-rich sandy cobble stratum tunnel, characterized in that, The method comprises the following steps: carrying out geological investigation on the tunnel construction area of the water-rich sandy cobble stratum, obtaining geological data of the tunnel area, and laying multiple monitoring sensors to collect water dynamics data and water chemistry data in real time; constructing a physical model of the tunnel area, including a three-dimensional geological model, a water dynamics model, and a water chemistry model; based on the water dynamics model and the water chemistry model, dynamically predicting water inflow according to the real-time collected water dynamics data and water chemistry data, and outputting the predicted water inflow intensity, spatial distribution, and time variation trend of the next time step; mapping the predicted water inflow intensity and spatial distribution of the next time step to the three-dimensional geological model to generate a flow field vector diagram and a concentration distribution heat map, and comparing the prediction results with the monitoring data of the next time step to dynamically adjust and update the model parameters; the construction of the water dynamics model and the water chemistry model comprises: establishing a mathematical model for describing the flow of underground water according to the principle of seepage mechanics, and combining with the fluid mechanics equation to establish a partial differential equation set for describing the flow of water as the mathematical basis of the model; wherein the fluid mechanics equation includes Darcy's law and the continuity equation; establishing the mathematical basis equation of the water chemistry model according to the convection-diffusion-reaction equation, which is used to simulate the migration and reaction of chemical substances in underground water; coupling the data basis equations of the water dynamics model and the water chemistry model to construct a coupled model, which is used to analyze the influence of chemical reaction on permeability and the feedback of flow rate on water chemistry change; solving the coupled model by using the finite element method to obtain the water dynamics equation and the water chemistry equation, thereby constructing the water dynamics model and the water chemistry model as the dynamic water inflow prediction model.

2. The method according to claim 1, wherein, The geological data includes: underground stratum structure, hydrogeological condition, geological feature data; wherein the underground stratum structure includes the thickness and permeability coefficient of the sandy cobble layer; the hydrogeological condition includes the underground water level, hydraulic gradient, and aquifer distribution; and the geological feature includes the fracture zone and karst potential water inflow channel. The water dynamics data includes: water level, water flow, flow rate, and water pressure parameter data; and the water chemistry data includes: ion concentration, pH value, oxidation-reduction potential, and dissolved solid parameter data in water.

3. The method of claim 1, wherein the method is characterized by, The construction of the three-dimensional geological model comprises: using a three-dimensional geological modeling tool to establish a three-dimensional geological model of the tunnel area; integrating the obtained geological data as an input layer to generate a three-dimensional geological model of the three-dimensional sandy cobble stratum, fracture zone, aquifer, and karst distribution.

4. The method of claim 1, wherein the method is characterized by, Solving the coupled model by using the finite element method to obtain the water dynamics equation and the water chemistry equation comprises: using the coupled model equation set as the control equation, defining the tunnel area according to the three-dimensional geological structure model, and defining the boundary conditions and initial conditions of the area; discretizing the tunnel area, and solving the control equation by using the weighted residual method to obtain the weak form, i.e., the water dynamics equation and the water chemistry equation.

5. The method for multi-dimensional comprehensive prediction and characterization of water inflow in a water-rich sandy cobble stratum tunnel according to claim 4, characterized in that, The dynamic water inflow prediction comprises: Discretize the hydrodynamic equation and the water chemical equation to form discrete equations, and calculate the stiffness matrix K of the hydrodynamic and water chemical of the region unit e and the load vector F e ; Assemble the stiffness matrix K e and the load vector F e of all units into global matrix K and load vector F, combine real-time collected hydrodynamic data and water chemical data, carry out time discretization and solving, and obtain the prediction equation of water head in hydrodynamics and the prediction equation of chemical component concentration in water chemistry; based on the prediction equation, solving the global linear equation set for each time step to predict the water head and chemical component concentration of the next time step and update it.

6. A multi-dimensional comprehensive prediction and characterization system for water inflow in a water-rich sand and gravel stratum tunnel, characterized in that, ​ The data acquisition module is configured to perform geological investigation on a tunnel construction area in a water-rich sandy cobble stratum, to obtain geological data of the tunnel area, and to arrange a plurality of monitoring sensors to collect water dynamics data and water chemistry data in real time. The physical modeling module is configured to construct a physical model of the tunnel area, including a three-dimensional geological model, a water dynamics model, and a water chemistry model. The dynamic water gushing prediction module is configured to perform dynamic water gushing prediction based on the water dynamics model and the water chemistry model according to the water dynamics data and the water chemistry data collected in real time, and to output predicted water gushing intensity, spatial distribution, and time variation trend at a next time step. The water gushing characterization module is configured to map the predicted water gushing intensity and spatial distribution at the next time step to the three-dimensional geological model, to generate a flow field vector diagram and a concentration distribution heat map, and to compare the prediction result with monitoring data at the next time step, so as to dynamically adjust and update model parameters. The construction of the water dynamics model and the water chemistry model includes: According to the principle of seepage mechanics, a mathematical model describing the flow of groundwater is established, and a partial differential equation set describing the flow of water is established based on the fluid mechanics equation, serving as the mathematical basis of the model; wherein the fluid mechanics equation includes Darcy's law and the continuity equation; According to the convection-diffusion-reaction equation, the mathematical basis equation of the water chemistry model is established, which is used to simulate the migration and reaction of chemical substances in groundwater; The data basis equations of the water dynamics model and the water chemistry model are coupled to construct a coupled model, which is used to analyze the influence of chemical reaction on permeability and the feedback of flow rate on water chemistry change; The coupled model is solved by using the finite element method to obtain the water dynamics equation and the water chemistry equation, thereby constructing the water dynamics model and the water chemistry model as the dynamic water gushing prediction model.

7. An electronic device, comprising: It includes: A memory for storing executable instructions; A processor for executing the executable instructions stored in the memory to implement the method of any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The executable instructions are stored in the memory for causing the processor to execute the executable instructions to implement the method of any one of claims 1-5.

9. A computer program product, characterised in that, The computer program product includes executable instructions stored in a computer-readable storage medium; When the processor of the electronic device reads the executable instructions from the computer-readable storage medium and executes the executable instructions, the method of any one of claims 1-5 is implemented.

Images