A self-developed and controllable three-dimensional tunnel water inrush analysis method and system

Abstract

This invention relates to the field of tunnel engineering safety monitoring technology, specifically to a method and system for analyzing tunnel water inrush based on an autonomous and controllable three-dimensional model. The method includes: collecting multi-source heterogeneous geological, hydrological, and construction data; fusing the multi-source heterogeneous data; constructing a dynamically updated three-dimensional geological-construction coupled digital twin model after alignment and encoding; performing rolling prediction of water inrush evolution through multi-scale physical-AI hybrid modeling and correcting real-time data, and outputting the predicted data; constructing a three-dimensional visualization platform and initiating multi-level linkage early warning based on risk thresholds. The system includes: a data fusion module, a rolling prediction module for water inrush evolution, and a multi-level linkage early warning module. By fusing geological data, hydrological monitoring data, and construction parameters, combined with three-dimensional visualization rendering and deep learning fluid simulation technology, the system meets the needs for rapid early warning and timely decision-making regarding water inrush disasters during tunnel construction, providing intelligent support for engineering disaster prevention decisions.

Description

Technical Field

[0001] This invention relates to the field of tunnel engineering safety monitoring technology, and in particular to a method and system for analyzing tunnel water inflow based on autonomous and controllable three-dimensional methods. Background Technology

[0002] In the field of tunnel construction, water inrush hazards have long been a critical challenge hindering construction safety and project progress. With the continuous advancement of infrastructure construction in transportation, energy, and other sectors, tunnel projects are increasingly developing towards greater depth, length, and more complex geological conditions, significantly increasing the frequency and severity of water inrush hazards. Once a water inrush occurs, it can not only flood equipment inside the tunnel and interrupt construction, causing huge economic losses, but also lead to serious consequences such as casualties and irreversible damage to the surrounding ecological environment. Therefore, accurate prediction and effective prevention and control of tunnel water inrush hazards are of paramount importance for ensuring the safe construction and long-term stable operation of tunnel projects.

[0003] Currently, in existing tunnel water inrush analysis, the time for a single simulation can be reduced to within half an hour. However, under actual complex geological conditions, especially when large-scale geological models and detailed hydrological simulations are involved, the time for a single simulation remains long. For example, for some large tunnel projects, the geological model contains numerous geological units and complex geological structures, and the hydrological model needs to consider complex water flow equations and boundary conditions. A complete water inrush simulation may take several hours or even days. This severely affects the real-time performance of tunnel water inrush analysis and cannot meet the needs for rapid early warning and timely decision-making regarding water inrush hazards during tunnel construction. Summary of the Invention

[0004] The purpose of this invention is to provide an autonomous and controllable three-dimensional tunnel water inrush analysis method and system, which aims to meet the needs of rapid early warning and timely decision-making for water inrush disasters during tunnel construction.

[0005] To achieve the above objectives, this invention employs a self-controllable three-dimensional tunnel water inflow analysis method, comprising the following steps:

[0006] Collect multi-source heterogeneous geological, hydrological and construction data, fuse the multi-source heterogeneous data, and construct a dynamically updated three-dimensional geological-construction coupled digital twin model after alignment and encoding;

[0007] By using multi-scale physical-AI hybrid modeling and correcting real-time data, rolling predictions of water inrush evolution are made, and the predicted data is output.

[0008] Build a 3D visualization platform and initiate multi-level linkage early warning based on risk thresholds.

[0009] Among the steps involved are: collecting multi-source heterogeneous geological, hydrological, and construction data; fusing the multi-source heterogeneous data; and constructing a dynamically updated three-dimensional geological-construction coupled digital twin model after alignment and encoding.

[0010] Integrate geological exploration data, real-time hydrological monitoring data, and BIM parameters from the construction management system to obtain multi-source heterogeneous data;

[0011] A unified coordinate reference system is established by processing multi-source heterogeneous data through a spatiotemporal alignment algorithm.

[0012] Construct a three-dimensional geological-construction coupled digital twin model with permeability attribute labels.

[0013] Among the steps involved in integrating geological exploration data, real-time hydrological monitoring data, and BIM parameters from the construction management system to obtain multi-source heterogeneous data:

[0014] The multi-source heterogeneous geology includes millimeter-precision point cloud data obtained from 3D laser scanning, ground-penetrating radar detection results, and rock mechanics parameters obtained from borehole exploration; real-time hydrological monitoring data includes water pressure, flow rate, and temperature parameters collected by a distributed fiber optic sensor network; and construction management system BIM parameters include excavation progress and support strength information.

[0015] After processing multi-source heterogeneous data using a spatiotemporal alignment algorithm and establishing a unified coordinate reference system:

[0016] The fracture network topology is transformed into graph structure data containing multidimensional features, where node features include porosity and permeability coefficient.

[0017] Among them, the steps of performing rolling prediction of water inrush evolution through multi-scale physical-AI hybrid modeling, correcting real-time data, and outputting prediction data are as follows:

[0018] A hybrid modeling approach combining multi-scale physics-AI and graph neural networks was employed to capture the seepage correlation characteristics between fractures.

[0019] It receives and processes on-site monitoring data in real time, extracts spatiotemporal features, and dynamically corrects model parameters.

[0020] The process includes receiving and processing on-site monitoring data in real time, extracting spatiotemporal features, and dynamically correcting model parameters.

[0021] The system predicts the evolution of water inrush within a rolling time window and outputs data on the path of water inrush at different times.

[0022] Among them, after the step of outputting the path direction data of the water inrush at different times during the evolution of the water inrush within the rolling prediction time window:

[0023] Extract key risk indicators and perform quantification processing for each key indicator, and output the comprehensive risk value.

[0024] Among them, in the steps of building a three-dimensional visualization platform and initiating multi-level linkage early warning based on risk thresholds:

[0025] Build a three-dimensional visualization platform based on an autonomous and controllable graphics engine, and use progressive rendering to display the path trend data and comprehensive risk value;

[0026] Set a first threshold and a second threshold, compare the comprehensive risk value, the first threshold and the second threshold, and perform multi-level linkage early warning according to the comparison results.

[0027] Among them, in the steps of setting a first threshold and a second threshold, comparing the comprehensive risk value, the first threshold and the second threshold, and performing multi-level linkage early warning according to the comparison results:

[0028] When the comprehensive risk value is less than the first threshold, a third-level early warning is triggered;

[0029] When the comprehensive risk value is greater than the first threshold and less than the second threshold, a second-level early warning is triggered;

[0030] When the comprehensive risk value is greater than the second threshold, a first-level early warning is triggered.

[0031] The present invention also provides a three-dimensional tunnel water inrush analysis system based on autonomy and control, including a data fusion module, a water inrush evolution rolling prediction module, and a multi-level linkage early warning module; wherein:

[0032] The data fusion module is used to collect multi-source heterogeneous geological, hydrological and construction data, fuse and obtain multi-source heterogeneous data, and construct a dynamically updated three-dimensional geological-construction coupling digital twin model after alignment coding;

[0033] The water inrush evolution rolling prediction module is used to perform water inrush evolution rolling prediction through multi-scale physical-AI hybrid modeling and correct real-time data, and output prediction data;

[0034] The multi-level linkage early warning module is used to build a three-dimensional visualization platform and initiate multi-level linkage early warning based on risk thresholds.

[0035] This invention discloses a method and system for analyzing water inrush in tunnels based on autonomous and controllable three-dimensional technology. The system employs a data fusion module, a rolling prediction module for water inrush evolution, and a multi-level linkage early warning module, performing the following steps: collecting multi-source heterogeneous geological, hydrological, and construction data; fusing the multi-source heterogeneous data; constructing a dynamically updated three-dimensional geological-construction coupled digital twin model after alignment and encoding; performing rolling prediction of water inrush evolution through multi-scale physical-AI hybrid modeling and correcting real-time data, and outputting the predicted data; constructing a three-dimensional visualization platform and initiating multi-level linkage early warning based on risk thresholds; and by fusing geological data, hydrological monitoring data, and construction parameters, combined with three-dimensional visualization rendering and deep learning fluid simulation technology, achieving the goal of meeting the needs for rapid early warning and timely decision-making regarding water inrush disasters during tunnel construction, providing intelligent support for engineering disaster prevention decisions. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart of the steps of the three-dimensional tunnel water inrush analysis method based on autonomous and controllable technology of the present invention.

[0038] Figure 2 This is a flowchart of steps S100 of the present invention.

[0039] Figure 3 This is a flowchart of steps S200 of the present invention.

[0040] Figure 4 This is a flowchart of steps S300 of the present invention.

[0041] Figure 5 This is a schematic diagram of the structural principle of the autonomous and controllable three-dimensional tunnel water inrush analysis system of the present invention.

[0042] Figure 6 This is a schematic diagram of the electronic device of the present invention.

[0043] 401 - Data fusion module, 402 - Rolling prediction module for water inrush evolution, 403 - Multi-level linkage early warning module. Detailed Implementation

[0044] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0045] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0046] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0047] Please see Figures 1-4 This invention provides a method for analyzing water inflow in three-dimensional tunnels based on autonomous and controllable technology, comprising the following steps:

[0048] S100: Collect multi-source heterogeneous geological, hydrological and construction data, fuse the multi-source heterogeneous data, and construct a dynamically updated three-dimensional geological-construction coupled digital twin model after alignment and encoding.

[0049] In this embodiment, multi-source heterogeneous geological, hydrological, and construction data are collected, fused to obtain the multi-source heterogeneous data, and after alignment and encoding, a dynamically updated three-dimensional geological-construction coupled digital twin model is constructed. The specific process is as follows:

[0050] S101: Integrate geological exploration data, real-time hydrological monitoring data, and BIM parameters from the construction management system to obtain multi-source heterogeneous data. The multi-source heterogeneous geological data includes millimeter-precision point cloud data obtained from 3D laser scanning, ground-penetrating radar detection results, and rock mechanics parameters obtained from borehole exploration. The real-time hydrological monitoring data includes water pressure, flow rate, and temperature parameters collected by a distributed fiber optic sensor network. The BIM parameters from the construction management system include excavation progress and support strength information.

[0051] S102: Process multi-source heterogeneous data through a spatiotemporal alignment algorithm to establish a unified coordinate reference system;

[0052] S103: Transform the fracture network topology into graph structure data containing multi-dimensional features, where node features include porosity and permeability coefficient;

[0053] S104: Construct a three-dimensional geological-construction coupled digital twin model with permeability attribute labels.

[0054] In the above process, geological exploration data is integrated as follows: three-dimensional laser scanning technology is used to obtain point cloud data with millimeter-level precision, which can accurately present the morphological characteristics of the geological surface around the tunnel; ground-penetrating radar detection technology is used to obtain information on geological structures at different depths underground and identify possible geological anomalies such as faults and karst caves; rock mechanical parameters, such as the compressive strength and tensile strength of rocks, are obtained through borehole exploration, providing basic data for analyzing the impact of geological structures on water inflow.

[0055] Real-time hydrological monitoring data integration: Deploy a distributed fiber optic sensor network to collect hydrological parameters such as water pressure, flow rate, and temperature in real time. These sensors can accurately sense the dynamic changes in the water body around the tunnel and promptly reflect the impact of changes in hydrological conditions on the risk of water inrush.

[0056] Construction Management System BIM Parameter Integration: Information such as excavation progress and support strength is obtained from the construction management system. Excavation progress reflects the speed and location of tunnel excavation, while support strength reflects the tunnel's ability to support the surrounding rock. This information is crucial for assessing the impact of the construction process on water inflow.

[0057] A spatiotemporal alignment algorithm is employed to process the integrated multi-source heterogeneous data. After completing temporal and spatial alignment, all data are represented in the same coordinate system, establishing a unified coordinate reference system. Since the acquisition time and spatial location of different data sources may differ, the spatiotemporal alignment algorithm uses methods such as interpolation, timestamp matching, and coordinate transformation to adjust all data to a unified time base and spatial coordinate system. For example, geological exploration data, hydrological monitoring data, and construction BIM data are unified into the project's global coordinate system through coordinate transformation. Simultaneously, timestamp matching ensures the temporal consistency of all data for subsequent time-series analysis.

[0058] The fracture network topology is transformed into graph structure data containing multi-dimensional features. In this process, the nodes and edges in the graph structure are determined. Node features include porosity, permeability coefficient, and other parameters that reflect the physical properties of the fracture network; edge features characterize the connectivity of seepage channels and describe the flow path of water in the fracture network.

[0059] Based on the processed data, a three-dimensional geological-construction coupled digital twin model with permeability attribute labels is constructed. The permeability attribute labels assign specific numerical values ​​or descriptions related to the permeability of each geological element in the model, such as porosity and permeability coefficient. This model organically combines data and information from multiple aspects, including geological structure, hydrological conditions, and construction status, to form a unified whole model. It can realistically reflect the geological structure, hydrological conditions, and construction status around the tunnel, and supports dynamic incremental updates during construction, ensuring real-time synchronization between the model and actual on-site conditions, providing an accurate model foundation for subsequent water inflow analysis and prediction. Specifically, the geological-hydrological coupling involves fusing geological exploration data and hydrological monitoring data to analyze the impact of geological structures on water flow. For example, ground-penetrating radar detection results identify potential geological anomalies such as faults and karst caves, and the impact of these anomalies on water flow paths and inflow volumes is analyzed in conjunction with hydrological monitoring data. The geological-construction coupling involves fusing geological exploration data and BIM parameters from the construction management system to analyze the impact of the construction process on the geological structure. For example, by analyzing excavation progress and support strength information, the disturbance of construction activities to the geological structure surrounding the tunnel, and the potential risk of water inrush caused by such disturbance, can be determined. Hydrology-construction coupling: Hydrological monitoring data and BIM parameters from the construction management system are integrated to analyze the impact of the construction process on hydrological conditions. For example, by using real-time monitored water pressure and flow data, combined with excavation progress and support strength information, the impact of construction activities on the dynamic changes of water bodies surrounding the tunnel can be analyzed.

[0060] S200: Through multi-scale physical-AI hybrid modeling and correction of real-time data, it performs rolling predictions of water inrush evolution and outputs predicted data.

[0061] In this embodiment, multi-scale physical-AI hybrid modeling is used to correct real-time data, enabling rolling prediction of water inrush evolution, and the predicted data is output. The specific process is as follows:

[0062] S201: Employs a hybrid modeling approach combining multi-scale physics-AI and graph neural networks to capture the seepage correlation characteristics between fractures;

[0063] S202: Receives and processes on-site monitoring data in real time, extracts spatiotemporal features, and dynamically corrects model parameters;

[0064] S203: Rolling prediction of the evolution of water inrush within a time window, outputting the path direction data of water inrush at different times;

[0065] S204: Extract key risk indicators, quantify each key indicator, and output a comprehensive risk value.

[0066] In the above process, a multi-scale physical-AI hybrid modeling method is adopted, and a graph neural network (GNN) is combined to capture the seepage correlation characteristics between fractures. At the macroscopic scale, a hydrodynamic model is constructed based on the improved Navier-Stokes equation to describe the flow law of water in a large range; at the microscopic scale, a graph neural network is used to model the complex seepage behavior in the fracture network. Through a specially designed message passing mechanism, the non-Darcy seepage characteristics in the geological fracture network are learned, so as to more accurately simulate the flow of water in the fractures.

[0067] Receive and process on-site monitoring data in real time, and use a spatio-temporal convolutional neural network (ST-CNN) to extract the spatio-temporal characteristics of the data. The extracted features are fused with the model to dynamically correct the model parameters, so that the model can timely reflect the changes in the actual on-site situation and improve the accuracy and adaptability of the model.

[0068] Based on the corrected model, roll-predict the water inrush evolution process within a time window (such as the next 2 hours). By continuously updating the model parameters and inputting real-time data, continuously predict the path direction of water inrush at different times, and output the corresponding path direction data to provide dynamic information on the development of water inrush for engineering personnel.

[0069] Extract key risk indicators from the water inrush evolution prediction results, such as the water inrush flow rate, water pressure, and influence range in different regions. Quantify each key indicator, and set corresponding risk weights for each indicator according to the actual engineering situation and experience. Through a weighted calculation method, obtain the comprehensive risk value of each region to evaluate the magnitude of the water inrush risk. For example:

[0070] The water inrush flow rate may account for 40% of the risk assessment weight, the water pressure accounts for 30% of the weight, and the influence range accounts for 30% of the weight. Through a weighted calculation method, the quantified risk indicators are combined to obtain the comprehensive risk value of each region. For example, if the quantified value of the water inrush flow rate in a certain region is 80, the quantified value of the water pressure is 70, and the quantified value of the influence range is 60, according to the above weights, the comprehensive risk value of this region = 80×0.​​​​​​​​​​​S302: Set the first threshold and the second threshold, compare the comprehensive risk value with the first threshold and the second threshold, and conduct multi-level linkage warnings according to the comparison results.

[0075] Further, in the steps of setting the first threshold and the second threshold, comparing the comprehensive risk value with the first threshold and the second threshold, and conducting multi-level linkage warnings according to the comparison results:

[0076] When the comprehensive risk value is less than the first threshold, a level-three warning is triggered;

[0077] When the comprehensive risk value is greater than the first threshold and less than the second threshold, a level-two warning is triggered;

[0078] When the comprehensive risk value is greater than the second threshold, a level-one warning is triggered.

[0079] In the above process, a three-dimensional visualization platform is constructed based on an autonomous and controllable graphics engine, and progressive rendering technology is used to display the path trend data and the comprehensive risk value. This platform can intuitively present the geological structure around the tunnel, the water inrush path, and the risk distribution, support multi-scale dynamic rendering of the risk heat map, including both the overall risk distribution overview and the fine display of local high-risk areas, providing clear and intuitive visualization effects for engineering personnel.

[0080] Set risk thresholds: Set the first threshold and the second threshold to divide different risk levels. The setting of the thresholds can be adjusted according to the actual engineering situation, historical data, and relevant specification standards. For example:

[0081] The first threshold is 50, and the second threshold is 80.

[0082] Compare the thresholds with the comprehensive risk value: Compare the calculated comprehensive risk value with the first threshold and the second threshold.

[0083] Conduct warnings according to the comparison results:

[0084] When the comprehensive risk value is less than the first threshold, a level-three warning is triggered. At this time, mark the risk area in the three-dimensional model, generate disposal suggestions and push them to relevant terminals, reminding engineering personnel to pay attention to potential risks, but no emergency measures are taken temporarily.

[0085] When the comprehensive risk value is greater than the first threshold and less than the second threshold, a level-two warning is triggered. On the basis of the level-three warning, increase the monitoring frequency, closely monitor the risk changes, and make emergency preparations.

[0086] When the comprehensive risk value is greater than the second threshold, a level-one warning is triggered. Execute preset emergency response measures, such as stopping construction, starting drainage equipment, etc., and at the same time notify relevant personnel to arrive at the scene immediately for handling to ensure project safety. For example:

[0087] A risk heatmap is used, with different colors representing areas of different risk levels. Red indicates high-risk areas (overall risk value above 80), orange indicates high-risk areas (overall risk value 50-80), and yellow indicates medium-risk areas (overall risk value below 50), thus visually presenting the risk distribution of the entire tunnel area. In a certain area, the quantified value of water inflow is 80, the quantified value of water pressure is 70, and the quantified value of the affected area is 60. After weighted calculation, the overall risk value of this area = 80×0.4 + 70×0.3 + 60×0.3 = 71. Based on the calculated overall risk of 71, this area is a high-risk area, displayed in orange, and a level-two warning is triggered.

[0088] Corresponding to the aforementioned embodiments of the autonomous and controllable three-dimensional tunnel water inrush analysis method, this application also provides embodiments of the autonomous and controllable three-dimensional tunnel water inrush analysis system.

[0089] Figure 5 This is a block diagram illustrating an autonomous and controllable three-dimensional tunnel water inrush analysis system according to an exemplary embodiment. (Refer to...) Figure 5 The system may include: a data fusion module 401, a rolling prediction module for water inrush evolution 402, and a multi-level linkage early warning module 403; wherein:

[0090] The data fusion module 401 is used to collect multi-source heterogeneous geological, hydrological and construction data, fuse the multi-source heterogeneous data, and construct a dynamically updated three-dimensional geological-construction coupled digital twin model after alignment and encoding.

[0091] The water inrush evolution rolling prediction module 402 is used to perform water inrush evolution rolling prediction by multi-scale physical-AI hybrid modeling and correcting real-time data, and output prediction data.

[0092] The multi-level linkage early warning module 403 is used to construct a three-dimensional visualization platform and initiate multi-level linkage early warning based on risk thresholds.

[0093] In this embodiment, the data fusion module 401 collects multi-source heterogeneous geological, hydrological, and construction data, fuses the multi-source heterogeneous data, and constructs a dynamically updated three-dimensional geological-construction coupled digital twin model after alignment and encoding. The water inrush evolution rolling prediction module 402 performs rolling prediction of water inrush evolution through multi-scale physical-AI hybrid modeling and corrects real-time data, and outputs the predicted data. The multi-level linkage early warning module 403 constructs a three-dimensional visualization platform and initiates multi-level linkage early warning based on risk thresholds. By fusing geological data, hydrological monitoring data, and construction parameters, and combining three-dimensional visualization rendering and deep learning fluid simulation technology, the system meets the needs for rapid early warning and timely decision-making regarding water inrush disasters during tunnel construction, providing intelligent support for engineering disaster prevention decisions.

[0094] Regarding the system in the above embodiments, the specific ways in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0095] For the system embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0096] Accordingly, this application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; and when the one or more programs are executed by the one or more processors, causing the one or more processors to implement the above-described autonomous and controllable three-dimensional tunnel water inrush analysis method. Figure 6 The diagram shown is a hardware structure diagram of any device with data processing capabilities, based on an autonomous and controllable three-dimensional tunnel water inrush analysis system provided in an embodiment of the present invention. (Except for...) Figure 6 In addition to the processor, memory, and network interface shown, any data processing device in the embodiment may also include other hardware depending on the actual function of the data processing device, which will not be described in detail here.

[0097] Accordingly, this application also provides a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the aforementioned autonomous and controllable three-dimensional tunnel water inrush analysis method. The computer-readable storage medium can be an internal storage unit of any data processing device as described in any of the foregoing embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc., equipped on the device. Furthermore, the computer-readable storage medium can include both internal storage units of any data processing device and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the data processing device, and can also be used to temporarily store data that has been output or will be output.

[0098] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0099] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for analyzing water inflow in tunnels based on autonomous and controllable three-dimensional methods, characterized in that, The steps include: Collect multi-source heterogeneous geological, hydrological and construction data, fuse to obtain multi-source heterogeneous data, and construct a dynamically updated three-dimensional geological-construction coupling digital twin model after alignment coding; Through multi-scale physical-AI hybrid modeling, correct real-time data, conduct rolling prediction of water inrush evolution, and output prediction data; Construct a three-dimensional visualization platform and initiate multi-level linkage warning based on risk thresholds; In the step of through multi-scale physical-AI hybrid modeling, correcting real-time data, conducting rolling prediction of water inrush evolution, and outputting prediction data: Convert the fracture network topology structure into graph structure data containing multi-dimensional features, determine the nodes and edges in the graph structure, where the node features include porosity and permeability coefficient; The edge feature characterizes the connectivity of the seepage channel; Use multi-scale physical-AI and graph neural network hybrid modeling to capture the seepage correlation characteristics between fractures. On the macroscopic scale, construct a hydrodynamic model based on the Navier-Stokes equation to describe the flow law of water flow in a large range; on the microscopic scale, use graph neural network to model the complex seepage behavior in the fracture network and learn the non-Darcy seepage characteristics in the geological fracture network; Receive and process on-site monitoring data in real time, use spatio-temporal convolutional neural network to extract spatio-temporal features of the data, and dynamically correct model parameters; Rolling predict the water inrush evolution process within the time window, and output the path direction data of water inrush at different times; Extract risk key indicators from the water inrush evolution prediction results, quantify each key indicator, and set corresponding risk weights for each indicator according to the actual engineering situation and experience. Through weighted calculation, obtain the comprehensive risk value of each region; In the step of constructing a three-dimensional visualization platform and initiating multi-level linkage warning based on risk thresholds: Construct a three-dimensional visualization platform based on an autonomous and controllable graphics engine, and use progressive rendering to display the path direction data and comprehensive risk value; Set the first threshold and the second threshold, compare the comprehensive risk value, the first threshold and the second threshold, and conduct multi-level linkage warning according to the comparison results; When the comprehensive risk value is less than the first threshold, trigger a level-three warning; When the comprehensive risk value is greater than the first threshold and less than the second threshold, trigger a level-two warning; When the comprehensive risk value is greater than the second threshold, trigger a level-one warning.

2. The method for analyzing water inflow in a three-dimensional tunnel based on autonomous and controllable technology as described in claim 1, characterized in that, In the step of collecting multi-source heterogeneous geological, hydrological and construction data, fusing to obtain multi-source heterogeneous data, and constructing a dynamically updated three-dimensional geological-construction coupling digital twin model after alignment coding: Integrate geological exploration data, real-time hydrological monitoring data, and BIM parameters of the construction management system to obtain multi-source heterogeneous data; Process multi-source heterogeneous data through a spatio-temporal alignment algorithm to establish a unified coordinate reference system; Construct a three-dimensional geological-construction coupling digital twin model with permeability attribute labels.

3. The method for analyzing water inflow in a three-dimensional tunnel based on autonomous and controllable technology as described in claim 2, characterized in that, In the step of integrating geological exploration data, real-time hydrological monitoring data, and BIM parameters of the construction management system to obtain multi-source heterogeneous data: The multi-source heterogeneous geology includes millimeter-precision point cloud data obtained from 3D laser scanning, ground-penetrating radar detection results, and rock mechanics parameters obtained from borehole exploration; real-time hydrological monitoring data includes water pressure, flow rate, and temperature parameters collected by a distributed fiber optic sensor network; and construction management system BIM parameters include excavation progress and support strength information.

4. The method for analyzing water inflow in a three-dimensional tunnel based on autonomous and controllable technology as described in claim 3, characterized in that, After processing multi-source heterogeneous data using a spatiotemporal alignment algorithm and establishing a unified coordinate reference system: The fracture network topology is transformed into graph structure data containing multidimensional features, where node features include porosity and permeability coefficient.

5. A three-dimensional tunnel water inflow analysis system based on autonomous controllability, applied to the three-dimensional tunnel water inflow analysis method based on autonomous controllability as described in claim 1, characterized in that, It includes a data fusion module, a rolling prediction module for water inrush evolution, and a multi-level linkage early warning module; among which: The data fusion module is used to collect multi-source heterogeneous geological, hydrological and construction data, fuse the multi-source heterogeneous data, and construct a dynamically updated three-dimensional geological-construction coupled digital twin model after alignment and encoding. The rolling prediction module for water inrush evolution is used to perform rolling prediction of water inrush evolution by multi-scale physical-AI hybrid modeling and correcting real-time data, and output the prediction data. The multi-level linkage early warning module is used to build a three-dimensional visualization platform and initiate multi-level linkage early warning based on risk thresholds.

Images