Tunnel pipe gallery disaster monitoring method and system based on multi-physics field coupling

Abstract

The invention belongs to the technical field of underground disaster monitoring, and provides a tunnel pipe gallery disaster monitoring method and system based on multi-physics field coupling, which uses multi-physics field data for monitoring, can capture various subtle changes of soil bodies inside and around an underground tunnel / pipe gallery, and improves the monitoring accuracy of the underground tunnel / pipe gallery. Accurate data are provided for building of a tunnel / pipe gallery project disaster model and disaster prediction, the method is more reasonable compared with traditional single physical quantity monitoring and analysis, the situation that potential safety hazards are difficult to find due to incomplete data is greatly reduced, and the safety detection efficiency of underground engineering is remarkably improved; on this basis, time step alignment is carried out on the data, discrete multi-physical field data is mapped into continuous physical field distribution based on Kriging spatial interpolation, a seepage-mechanics-thermodynamics coupling equation is established to calculate and predict the settlement amount, and a reduced-order model algorithm is utilized to optimize the calculation time of more than 10 h of traditional data coupling calculation to be within 5 min.

Description

Technical Field

[0001] The present invention belongs to the technical field of underground disaster monitoring, and in particular relates to a tunnel corridor disaster monitoring method and system based on multi-physical field coupling. Background Art

[0002] With the continued development of economic development, a large number of integrated tunnels and pipeline corridors are now located underground in some newly developed construction areas. These tunnels and pipeline corridors essentially create underground tunnel spaces that integrate various engineering pipelines such as electricity, communications, gas, heating, and water supply and drainage, ensuring the normal operation of the city. In some plain areas, the main geological hazards are slow-moving disasters such as subsidence and ground fissures. These hazards do not cause significant damage in the short term. However, when heavy rainfall occurs in nearby areas, the combined effects of precipitation and ground fissure settlement can cause severe collapse and groundwater seepage, severely impacting the lifeline of tunnels and pipeline corridors in the area. Therefore, the development of multi-physics intelligent detection and early warning for subsidence in integrated underground tunnels and pipeline corridors in plain areas is extremely necessary and urgent.

[0003] Currently, conventional tunnel / pipeline corridor disaster prediction mainly relies on data transmitted by a single sensor, such as stress sensors and temperature sensors, and presents the data through two-dimensional charts. The level of visualization is not very advanced, and the accuracy of such methods is insufficient. First, the disaster-causing mode of tunnel / pipeline corridor disasters is usually the coupling of multiple factors, such as a complex process in the form of force, water, and heat coupling. Current detection modes mostly use a single type of sensor. Existing technologies lack the ability to model multi-field coupling mechanisms, resulting in insufficient accuracy of early warning models. Second, tunnel / pipeline corridor disasters are mostly immediate responses. Current disaster predictions are mostly based on offline data deduction, which is not timely and only has a relatively good effect in post-disaster simulation and review. It is unable to combine dynamic data to update the early warning model in real time, making it difficult to timely reflect the evolution trend of the disaster. Third, the visualization effect is limited. The display effect of two-dimensional charts and two-dimensional models is far less comprehensive than that of three-dimensional models. The data is separated from the model. At the same time, only static model display is currently available, which cannot intuitively display the changes in the spatiotemporal characteristics of multiple physical fields. It lacks interactive disaster model deduction and cannot be combined with disaster coupling monitoring. Summary of the Invention

[0004] In order to solve the above problems, the present invention proposes a tunnel and pipeline corridor disaster monitoring method and system based on multi-physics field coupling. The present invention uses multi-physics field data for monitoring, which can capture various subtle changes in the soil inside and around underground tunnels / pipelines, including but not limited to settlement displacement, seepage volume, temperature and humidity, and stress and strain, providing accurate data for the establishment of tunnel / pipeline corridor engineering disaster models and disaster prediction. Compared with traditional single physical quantity monitoring and analysis, it is more reasonable, greatly reduces the difficulty in discovering safety hazards caused by incomplete data, and significantly improves the efficiency of underground engineering safety detection; on this basis, the data is time-step aligned, and the discrete multi-physics field data is mapped to a continuous physical field distribution based on Kriging spatial interpolation, and a seepage-mechanics-thermodynamics coupling equation is established to calculate and predict settlement. The reduced-order model algorithm is used to optimize the traditional data coupling calculation time of more than 10 hours to within 5 minutes; through multi-physics field data fusion, dynamic coupling modeling and three-dimensional model technology, real-time deduction and precise prevention and control of tunnel / pipeline corridor disasters are realized, and the real-time performance and decision-making efficiency of tunnel / pipeline corridor safety management are improved.

[0005] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0006] In a first aspect, the present invention provides a tunnel corridor disaster monitoring method based on multi-physics field coupling, comprising:

[0007] Acquiring multi-physical field data of an underground monitoring object; wherein the multi-physical field data includes settlement displacement, stress and strain information, seepage information, temperature information, and gas concentration information;

[0008] Align low-frequency data and high-frequency data in multi-physics field data to a unified time base through interpolation algorithms; map discrete multi-physics field data to continuous physical field distribution based on Kriging spatial interpolation;

[0009] Based on the aligned and mapped multi-physics field data, a seepage-stress-thermomechanical coupling mathematical model is constructed. The order of the model is reduced through an order reduction algorithm, and the seepage-stress-thermomechanical coupling mathematical model is solved to obtain the prediction results.

[0010] The prediction results are displayed in a preset three-dimensional model, and graded warnings are issued based on the prediction results.

[0011] Furthermore, consistency processing is performed on the settlement displacement, the stress-strain information, the seepage information, the temperature information and the gas concentration information, and an isolation forest algorithm is used to monitor and eliminate abnormal data points.

[0012] Furthermore, Kalman filtering is used to align the timestamps of multi-physics field data and suppress noise; deep neural networks are used to learn the nonlinear coupling relationship between multi-physics fields. The network input is settlement displacement, stress and strain information, seepage information, temperature information and gas concentration information, and the output is the predicted value of settlement amount.

[0013] Furthermore, the seepage-stress-thermomechanical coupling mathematical model includes a seepage field equation, a stress field equation, and a temperature field equation; the seepage field equation is:

[0014]

[0015] Where k(T) is the temperature-dependent permeability coefficient; μ(T) is the fluid dynamic viscosity; φ is the porosity affected by strain; α is the Biot coefficient (characterizing the contribution of pore pressure to stress); β is the volume change coefficient caused by thermal expansion; is the pore water pressure; ρ is the fluid density; ∈ v is the volume strain;

[0016] The stress field equation is:

[0017]

[0018] Where C is the elastic matrix; u is the fluid dynamic viscosity; β T is the coefficient of thermal expansion; is temperature;

[0019] The temperature field equation is:

[0020]

[0021] Where ρ is the density; C p is the specific heat capacity at constant pressure; k T is the thermal conductivity coefficient; ρ f is the liquid density; C p , f is the specific heat capacity of the liquid at constant pressure; ν is the Poisson's ratio.

[0022] Furthermore, dynamic particles are used to simulate the seepage path, and the particle density is positively correlated with the seepage velocity; the structural deformation is calculated in real time through the vertex shader, and the crack extension direction is consistent with the direction of the maximum principal stress; the temperature field distribution is mapped using thermal map colors.

[0023] Furthermore, graded warnings include yellow warning, orange warning and red warning; a yellow warning is issued when a single physical field exceeds the threshold; an orange warning is issued when two physical fields exceed the threshold at the same time; a red warning is issued when three physical fields exceed the threshold at the same time and the temperature gradient exceeds the preset value.

[0024] In a second aspect, the present invention further provides a tunnel corridor disaster monitoring system based on multi-physics field coupling, comprising:

[0025] The data acquisition module is configured to: acquire multi-physical field data of the underground monitoring object; wherein the multi-physical field data includes settlement displacement, stress and strain information, seepage information, temperature information and gas concentration information;

[0026] The processing module is configured to: align low-frequency data and high-frequency data in the multi-physics field data to a unified time base through an interpolation algorithm; and map the discrete multi-physics field data into a continuous physical field distribution based on Kriging spatial interpolation;

[0027] The prediction module is configured to: construct a seepage-stress-thermomechanical coupling mathematical model based on the aligned and mapped multi-physics field data, reduce the order of the model through an order reduction algorithm, and solve the seepage-stress-thermomechanical coupling mathematical model to obtain a prediction result;

[0028] The early warning module is configured to: display the prediction results in a preset three-dimensional model and issue graded warnings based on the prediction results.

[0029] In a third aspect, the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the tunnel corridor disaster monitoring method based on multi-physical field coupling described in the first aspect.

[0030] In a fourth aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, the steps of the tunnel corridor disaster monitoring method based on multi-physical field coupling described in the first aspect are implemented.

[0031] In a fifth aspect, the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the tunnel corridor disaster monitoring method based on multi-physical field coupling described in the first aspect.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] Acquiring multi-physical field data of an underground monitoring object; wherein the multi-physical field data includes settlement displacement, stress and strain information, seepage information, temperature information, and gas concentration information;

[0034] Align low-frequency data and high-frequency data in multi-physics field data to a unified time base through interpolation algorithms; map discrete multi-physics field data to continuous physical field distribution based on Kriging spatial interpolation;

[0035] Based on the aligned and mapped multi-physics field data, a seepage-stress-thermomechanical coupling mathematical model is constructed. The order of the model is reduced through an order reduction algorithm, and the seepage-stress-thermomechanical coupling mathematical model is solved to obtain the prediction results.

[0036] The prediction results are displayed in a preset three-dimensional model, and graded warnings are issued based on the prediction results.

[0037] 1. The present invention uses multi-physics field data for monitoring, which can capture various subtle changes in the soil inside and around underground tunnels / pipelines, including but not limited to settlement displacement, seepage volume, temperature and humidity, and stress and strain, providing accurate data for the establishment of tunnel / pipeline corridor engineering disaster models and disaster prediction. Compared with traditional single physical quantity monitoring and analysis, it is more reasonable, greatly reduces the difficulty in discovering safety hazards caused by incomplete data, and significantly improves the efficiency of underground engineering safety detection; on this basis, the data is time-step aligned, and the discrete multi-physics field data is mapped to a continuous physical field distribution based on Kriging spatial interpolation, and a seepage-mechanics-thermodynamics coupling equation is established to calculate and predict settlement. The reduced-order model algorithm is used to optimize the traditional data coupling calculation time of more than 10 hours to within 5 minutes; through multi-physics field data fusion, dynamic coupling modeling and three-dimensional model technology, real-time deduction and precise prevention and control of tunnel / pipeline corridor disasters are realized, improving the real-time performance and decision-making efficiency of tunnel / pipeline corridor safety management.

[0038] 2. The data processing method of the present invention utilizes extended Kalman filtering (EKF) or lossless Kalman filtering (UKF) to process nonlinear prediction models, integrates heterogeneous data to make up for the limitations of single data, realizes spatiotemporal alignment and fusion of stress-seepage-temperature data, removes outliers and performs wavelet denoising on fiber optic sensor, GNSS displacement meter and piezometer data, constructs a state vector containing settlement, rate and acceleration, fuses multi-sensor observations to update the state estimate, and uses the LSTM network to use the previous 5 minutes of fused data as input to predict the settlement trend in the next 30 minutes, outputs a confidence interval, and effectively ensures the accuracy of settlement prediction.

[0039] 3. By combining virtual reality technology with multi-field coupling calculation theory, the present invention can display key data such as stress and strain, settlement, seepage, temperature and humidity at various structures in the tunnel / pipeline corridor in real time from a detailed and objective three-dimensional model perspective. At the same time, when the data changes or exceeds the specified threshold, color display changes and alarms are generated. Particle flow rendering technology is used to simulate the process of gas diffusion or seepage exceeding the limit, dynamically reflecting the process of engineering disasters, and providing virtual human-computer interaction to realize the simulated disaster prevention and control process of staff. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings constituting a part of the specification of this embodiment are used to provide a further understanding of this embodiment. The schematic embodiments and descriptions of this embodiment are used to explain this embodiment and do not constitute an improper limitation on this embodiment.

[0041] Figure 1 This is a block diagram of the method of embodiment 1 of the present invention. DETAILED DESCRIPTION

[0042] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0043] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0044] Example 1:

[0045] This embodiment provides a tunnel and pipeline corridor disaster monitoring method based on multi-physical field coupling, which is used to assist in predicting disasters in underground tunnels / pipelines; through multi-physical field data fusion, dynamic coupling modeling and VR virtual-reality synchronization technology, real-time deduction and precise prevention and control of tunnel / pipeline corridor disasters are achieved, and the human-computer interaction mode of virtual imaging is used to improve the real-time and decision-making efficiency of tunnel / pipeline corridor safety management. In this embodiment, the method mainly includes five main modules and steps, namely, the data acquisition layer module performs data acquisition, the data processing layer module performs data processing, the data coupling model layer module performs data coupling, the VR imaging and human-computer interaction layer module performs visual imaging, and the early warning model decision layer module performs the early warning model. Specifically:

[0046] Optionally, the data acquisition layer module uses stress and strain sensors, seepage sensors, temperature and humidity sensors, and gas concentration sensors to collect data, and performs remote data transmission in accordance with the 5G data transmission protocol. The sensors are respectively arranged at sensitive points inside the tunnel / pipeline corridor, such as the thermal pipeline relay connection port, the water supply pipeline relay connection port, and the natural gas supply pipeline relay connection port, to monitor the changes in various parameters in real time and transmit the parameters to the main control computer in real time.

[0047] Specifically, key data such as deformation, stress and strain monitoring, temperature and humidity changes, seepage field data changes, and gas concentration changes can be monitored using highly sensitive instruments such as fiber Bragg grating sensors, GNSS displacement meters, osmometers, inclinometers, and gas concentration sensors. The monitoring process involves long-term, regular observation of sensitive monitoring points. For example, horizontal baseline detection methods or three-dimensional laser scanning technology can be used to monitor settlement within tunnels and pipeline corridors in real time.

[0048] Optionally, the data processing layer module performs consistency processing on the acquired multi-source heterogeneous data such as stress and strain information, seepage information, temperature information and gas concentration information, uses the isolation forest algorithm to monitor and eliminate abnormal data points, and uses cubic spline interpolation or linear interpolation to align the time steps of low-frequency data, establishes a high-precision settlement model to perform multimodal fusion of data of different properties, uses extended Kalman filtering (EKF) or lossless Kalman filtering (UKF) to process nonlinear prediction models, and integrates heterogeneous data to make up for the limitations of single data.

[0049] Specifically, time step alignment refers to the processing of unifying the time base and mapping the monitoring data of different data sources (such as displacement monitors, temperature and humidity sensors, gas concentration sensors, osmometers, etc.) into the full-field settlement distribution, solving the data differences caused by differences in sensor sampling frequencies, and capturing the transient response (such as the sudden increase stage) and long-term trend of settlement. Unit standardization aims to eliminate the analytical obstacles caused by inconsistent units between different monitoring data and parameters, ensure that all data are processed with the same dimension or scale during the coupling process, and improve the consistency and comparability of the data. Abnormal data removal, including the identification and removal of outliers, is based on statistical methods (3σ criterion) or isolation forest algorithm to detect and remove abnormal data points; the removal of outliers depends on the specific situation, based on the actual situation of the data and the purpose of analysis.

[0050] Optionally, the data coupling layer module establishes the seepage-stress-thermodynamic coupling equation, reducing the calculation time from 10 hours to 5 minutes through the reduced-order model (ROM); optionally, the seepage-stress-thermodynamic coupling equation is established using Cosmol Multiphysics or other tools. Optionally, multi-physics field data coupling, the calculation process is as follows:

[0051] To focus on the core coupling effects, the following assumptions are made: porous media: the soil surrounding the tunnel / pipeline corridor is an isotropic saturated porous medium; small deformations: geometric nonlinearity is ignored, and a linear elastic constitutive relation is adopted; steady-state seepage: the seepage velocity is low, and inertia terms are ignored (Darcy's law applies); thermomechanical coupling: temperature changes cause thermal expansion of the soil and changes in fluid viscosity; the coupling relationship is: the seepage field (water pressure p) affects the stress field (effective stress); the stress field (strain ε) changes the porosity and affects the seepage permeability coefficient; the temperature field (T) affects the seepage viscosity, and the thermal expansion of the soil changes the seepage field and stress fields. A fully coupled seepage-stress-thermomechanical governing equation is established to calculate and predict settlement within the tunnel / pipeline corridor area.

[0052] The governing equations for each physical field are as follows (taking all calculation factors into account):

[0053] To simplify the tunnel / pipeline corridor settlement calculation process, the transient terms of the seepage and temperature fields can be ignored. If the temperature change is small (△T < 50°C), the effect of temperature on the permeability coefficient k(T) and viscosity μ(T) can be ignored, retaining only the thermal expansion effect. Given the symmetrical structure of the tunnel / pipeline corridor, a two-dimensional model in a cylindrical coordinate system is used to reduce the amount of calculation. After simplification, the seepage-stress-thermal coupling equation for the soil surrounding the tunnel / pipeline corridor is established as follows:

[0054] Seepage field equation:

[0055]

[0056] The permeability coefficient k is affected by temperature and reflects the thermal sensitivity of the clay in the preset area, satisfying:

[0057]

[0058] Stress field equation:

[0059]

[0060] The elastic matrix C is determined by the soil elastic modulus E = 50 MPa and Poisson's ratio ν = 0.3.

[0061] Temperature field equation:

[0062]

[0063] Initial temperature T0 = 20℃, thermal conductivity coefficient.

[0064] Optionally, the VR imaging and human-computer interaction layer module uses a 3D scanner to pre-scan the tunnel / pipeline corridor section, and constructs an initial three-dimensional geometric model of the tunnel / pipeline corridor based on the BIM design file (Revit model) or laser point cloud scanning data. The BIM model is exported in FBX / OBJ format and adapted to the Unity3D engine. The physical field data is associated with the model vertices / meshes, and the layers are divided according to the physical field type (such as seepage field / stress field), supporting independent display and superposition. Construct near / far perspective switching, the near perspective (<5m) displays crack micro-textures and bolt details (number of facets>100,000); the far perspective (≥5m) simplifies the model to simplified polygons (number of facets <10,000) to improve the real-time rendering efficiency of the seepage path based on the particle system and ensure instant changes to the model.

[0065] Specifically, the human-computer interaction in the VR imaging and human-computer interaction layer module relies on an external head-mounted display device and a control handle for human-computer interaction operations, allowing free movement, rotation, and zooming of the viewing angle. Preset disaster scenarios (such as pipe seepage, pipeline explosion, and tunnel / pipeline corridor roof collapse) can be selected to automatically trigger multi-physics field coupling calculation and deduction. Groundwater pressure, temperature, and other boundary conditions can be modified in real time to observe the model response. The handle is used for human-computer interaction to complete operations such as "grouting reinforcement" and "drainage and pressure reduction" and to rehearse treatment measures. The VR imaging and human-computer interaction layer module supports multi-terminal access (VR headsets, PCs, and mobile terminals), sharing the same virtual scene and marking risk points. Data can be synchronized to the cloud. The VR imaging and human-computer interaction layer module establishes a tunnel / pipeline corridor model, uses virtual reality equipment for human-computer interaction, and conducts real-time calculations or rehearsals of treatment plans for multiple scenarios. The coupling layer data is transmitted to the VR imaging layer module for color grading and real-time rendering of particle systems for different disaster modules. Optionally, use Unity3D to build a tunnel / pipeline gallery model, or use other tools to build a tunnel / pipeline gallery model, such as 3DmaxDENG.

[0066] Optionally, the warning model decision layer module establishes a three-level warning mechanism to color-code disasters in different situations.

[0067] Preferably, the early warning model decision layer module is combined with the VR imaging layer module. The early warning model decision layer module adopts a three-level (yellow, orange, and red) early warning level. The trigger condition is determined by combining the settlement amount, settlement rate, seepage field, and temperature field (the trigger condition is determined by the geological survey report data of each section). With reference to relevant standards and survey reports, the tunnel / pipeline corridor settlement warning value is recommended to be 10-30 mm, and the settlement rate warning value is 2 mm / day:

[0068] Table 1 Warning levels

[0069]

[0070]

[0071] The early warning model's decision-making layer module calculates normal fluctuations based on historical regional monitoring data (e.g., a one-year period). It then uses an LSTM neural network machine learning algorithm to aid in predicting subsidence trends, dynamically adjust thresholds, and perform real-time parameter optimization, providing specialized treatment for different regions. The early warning model's decision-making layer utilizes multi-terminal intelligent push notifications and emergency command coordination technology. Multi-terminal intelligent push notifications connect PCs, VR headsets, and mobile phones to improve emergency response speed and rescue efficiency.

[0072] Optional: Seepage Field: Particle system simulates water flow paths, with flow rate positively correlated with seepage pressure (dynamic particle density changes). Stress Field: Vertex displacement animation shows crack expansion, with color gradient (blue → red) indicating stress level. Temperature Field: Thermal map overlays infrared pseudo-color, with dynamic diffusion effect.

[0073] Based on this, subsequent data analysis and model construction simulation work can be carried out more accurately, and the development trends of tunnels / pipeline corridors during the construction process or operation and maintenance stage, such as changes in structural stability, can be more accurately analyzed and predicted, achieving the quality of virtual reality imaging of underground tunnel / pipeline corridor engineering disaster models and promoting the sustainable development of the project.

[0074] The five modules in this embodiment cooperate with each other to progressively complete the virtual reality imaging of the tunnel / pipeline corridor engineering disaster model based on multi-physical field monitoring coupling, while satisfying human-computer interaction.

[0075] The implementation process of the method in this embodiment may include:

[0076] S101. Activate the data acquisition layer module, deploy sensors at each sensitive monitoring point in the tunnel / pipeline corridor, use a 3D laser scanner to photograph the underground tunnel / pipeline corridor structure, and import the collected data into the main console or cloud via 5G transmission or wired transmission. The main data to be exported are the tunnel / pipeline corridor structure composition data and key indicators such as stress and strain, settlement, seepage, temperature and humidity, and gas concentration. The real-time monitoring system starts working, and the recovered data is transmitted to the data processing layer module and the VR visualization and human-computer interaction layer module respectively.

[0077] S201. Start the data processing layer module to check and identify the data of stress and strain, settlement, seepage, temperature and humidity, gas concentration and other indicators. Perform cubic spline interpolation or linear interpolation on the indicators used in the coupled calculation (stress and strain, temperature, seepage) to synchronize the low-frequency data in time and space. At the same time, map the data to the full-field distribution according to the plane coordinates of the tunnel / pipeline corridor. Screen and eliminate abnormal data based on statistical algorithms (3σ criterion) or isolation forest algorithms to improve data consistency. Establish a high-precision settlement model and perform multimodal fusion on the data of unified dimensions.

[0078] S301. Start the data coupling layer module. After the monitoring data is fused and calculated, the seepage-stress-thermal coupling control equation is established based on Biot theory and thermoelasticity theory. The settlement inside the tunnel / pipeline corridor is calculated and predicted. The data is then transmitted to the VR visualization and human-computer interaction layer module.

[0079] S401. Start the VR visualization and human-computer interaction layer module. The BIM model of the underground tunnel / pipeline corridor structure data captured by the 3D laser scanner is exported to the FBX / OBJ format and adapted to the Unity3D engine to perform 3D modeling of the underground tunnel / pipeline corridor tunnel. Relying on the external head-mounted display device and the control handle for human-computer interaction, the operator can freely move, rotate, and zoom the viewing angle. The boundary conditions such as groundwater pressure and temperature can be modified in real time to observe the model response. The handle is used for human-computer interaction to complete operations such as "grouting reinforcement" and "drainage and pressure reduction" to conduct a preview of treatment measures.

[0080] S501. Activate the early warning model decision-making layer module. Tunnel / pipeline corridor model sections displayed in the VR visualization environment will be displayed in primary, yellow, orange, and red colors. As described above, the three-level hazard classification system assigns different color alert levels to different sections in real time. Simultaneously, each tunnel / pipeline corridor model section in the VR visualization environment will issue different alarms and notifications according to the three-level early warning mechanism. Data from the data acquisition layer module will be uploaded to the VR visualization environment in real time. When monitoring data exceeds a given threshold, the early warning model decision-making layer module will evaluate and analyze the abnormal data. The trained prediction model will then update the data in real time to generate a comprehensive disaster warning report.

[0081] S601. Data synchronization and alarming are carried out based on mobile terminals, PC terminals, and VR head-mounted displays. The location, cause, and potential impact of possible disasters are explained and displayed in a VR visualization environment. Reference response suggestions are given according to the various treatment measures of the decision-making level.

[0082] This embodiment uses data recovered by multi-physics field detectors to predict and prevent disaster risks in tunnel / pipeline corridor structures imaged in virtual reality, providing industry personnel with three-dimensional visual action interaction, solving the aforementioned problem of incomplete presentation of two-dimensional imaging commonly used in the industry. At the same time, the coupling of multi-physics field monitoring breaks the previous subjective evaluation method of using a single physical quantity as a disaster assessment standard.

[0083] Example 2:

[0084] To further supplement the method in Example 1, this embodiment provides a tunnel and pipe gallery disaster monitoring method based on multi-physical field coupling, including a data acquisition layer module. By deploying high-precision sensors such as three-dimensional laser scanners, osmometers, GNSS displacement meters, temperature and humidity sensors, and gas concentration sensors, the module acquires data on the internal structure of the tunnel / pipe gallery and the surrounding physical fields, detects data in real time, and monitors abnormal data.

[0085] The data processing layer module uses the time-space step synchronization algorithm and the abnormal data elimination algorithm to normalize data of different dimensions, and uses the multimodal data fusion system to perform nonlinear data processing and calculation;

[0086] The data coupling layer module receives the data collected by the data processing layer module and the data acquisition layer module, and uses the three-physics field coupling equation to predict and calculate the settlement amount;

[0087] The VR imaging and human-computer interaction layer module provides VR headsets and controllers, generates virtual reality scenes, receives and displays data from the data coupling layer module and the data processing layer module, and displays it in real time, identifying, evaluating, warning, and providing feedback on the real-time situation of tunnels / pipelines.

[0088] The early warning model and decision-making layer module provides a three-level disaster classification system, combined with VR imaging and human-computer interaction layer modules to push real-time alerts and generate virtual disaster animations, and export analysis results in real time to mobile devices, PCs and VR terminals.

[0089] The data acquisition layer module includes five main detection and monitoring instruments. The three-dimensional laser scanner can obtain three-dimensional spatial information such as underground engineering geological structure and engineering structure with high precision, which is used for subsequent underground tunnel / pipeline corridor VR model building, quality inspection and deformation monitoring; the piezometer is a measuring instrument used to measure the seepage (void) water pressure inside the structure. The vibrating wire piezometer is used to measure the seepage water level inside the tunnel, which serves as the main data source for judging whether there are hidden dangers of disasters such as large-flow seepage and pipe bursts in the underground project; the GNSS displacement meter monitors surface displacement and building deformation caused by disasters. The data can be uploaded in real time via 4G signals to obtain more accurate deformation data; the temperature and humidity sensor monitors and uploads the temperature and humidity changes inside the tunnel / pipeline corridor in real time, and the recovered data is used for subsequent data coupling layer calculations; the gas concentration sensor is mainly deployed at the connection of the domestic gas pipeline in the underground tunnel / pipeline corridor to monitor the content of dangerous flammable and explosive gases in real time, which serves as an important data source for judging gas leaks in underground pipelines. 50 fiber optic sensor measurement points (40m apart) and 10 piezometer measurement points (distributed longitudinally along the tunnel / pipeline corridor) were deployed. The 3D model was exported in FBX format using Autodesk Revit, retaining structural details (including the tunnel / pipeline corridor roof, side walls, pipeline supports, etc.). The model was imported into Unity3D, and material mapping (concrete texture, metal reflective material) was set, optimizing the number of facets to less than 500,000.

[0090] The data processing layer module includes a data time-space synchronization system, an abnormal data elimination system, and a multimodal data fusion system; the data time-space synchronization system includes data coordinate mapping space alignment processing and cubic spline interpolation time alignment processing; the data coordinate mapping space alignment processing is based on the inverse distance weight (IDW) algorithm to map discrete point data to the same network, and use coordinate transformation to unify the sensor data to the tunnel / pipeline corridor global coordinate system (such as taking the tunnel / pipeline corridor entrance as the coordinate origin) and map the strain data of 50 measuring points (sampling rate 1kHz) to the model grid through vertex attributes; the cubic spline interpolation time alignment processing is to perform low-frequency data (such as GNSS 1Hz) on the data. Linear interpolation or cubic spline interpolation is performed to match the high-frequency data timestamp (such as 1kHz), and the high-frequency data is downsampled to match the low-frequency data, so as to unify the multi-source data into the same spatiotemporal benchmark; the Kriging interpolation algorithm (the variation function model is Gaussian) is used to generate the full-field seepage pressure distribution from the seepage pressure data (1Hz) of 10 measuring points; the abnormal data removal system includes the isolation forest algorithm and the statistical 3σ criterion for data processing; the multimodal data fusion system includes nonlinear observation model processing, which is refined into three commonly used algorithms, such as the improved Kalman filter algorithm, the lossless Kalman filter algorithm, and the extended Kalman filter algorithm, to predict future settlement based on the current monitored data.

[0091] The data coupling layer module uses three coupled physical field equations and utilizes seepage-temperature-stress coupling to predict and calculate the settlement amount, and uses the ROM model to reduce the coupling calculation time from 10 hours to within 5 minutes.

[0092] The VR imaging and human-computer interaction layer module includes a VR head display device and a handle, and a virtual reality scene construction; the VR head display device and the handle provide display and interaction of the virtual reality scene, and the staff can use the VR head display device to switch the perspective, and use the handle to zoom in and out of the perspective. The data processed by the data processing layer and the data coupling layer module will be synchronized in real time to the tunnel / pipeline corridor model of the VR perspective for display, and the user can query and modify the simulation value operation, conduct simulated disaster process drills and try out treatment measures. At the same time, the VR perspective provides a sound alarm system and a particle flow model to simulate the flow rate changes of gas and fluid; using the The data acquired by the 3D laser scanner in the data acquisition layer module is used to build a 3D model of the underground tunnel / pipeline gallery based on the Unity3D engine. The tunnel / pipeline gallery model is constructed with near / far perspective switching. The near perspective (<5m) shows the crack micro-texture and bolt details (number of faces>100,000); the far perspective (≥5m) simplifies the model to simplified polygons (number of faces<10,000) to improve the real-time rendering efficiency of the seepage path based on the particle system and ensure the instant change of the model. During the construction of the model, a data display module is constructed, and data display can be performed in real time in VR visualization. Data modification (stress, seepage flow, temperature) can be performed to simulate different disaster situations.

[0093] The early warning model and decision-making layer module include the setting of a three-level early warning mechanism and the setting of processing standards. The use of the early warning model and decision-making layer module is based on the VR imaging and human-computer interaction layer module. After the above monitoring data is recovered, it is transmitted to the data processing and data coupling module for data analysis. The analyzed data is synchronously transmitted to the tunnel / pipeline corridor VR model. The three-level grading system of the early warning model and decision-making layer module assigns colors to different sections (primary color: safe, yellow: low risk, orange: medium risk, red: high risk), and different treatment measures are implemented for sections assigned different colors. When the monitoring data exceeds the threshold, an early warning signal is issued and early warning information is generated based on the data information. It is uploaded and synchronized to the PC, mobile and VR terminals in real time. When an abnormality occurs, the early warning information is received first and a disaster early warning report is generated in real time, explaining the location, cause and potential impact of the possible disaster and giving comprehensive response suggestions.

[0094] The specific steps of the method in this embodiment can be:

[0095] S201. Start the data acquisition layer module. Use a 3D laser scanner to capture the underground tunnel / pipeline corridor's topography, engineering structures, and other data. Export data based on BIM design files (such as Revit models) or laser point cloud scanning to construct an initial 3D geometric model of the tunnel / pipeline corridor. Simultaneously, piezometers deployed around the tunnel / pipeline corridor monitor groundwater levels and seepage rates. GNSS displacement meters monitor soil displacement and tunnel / pipeline corridor deformation. Temperature and humidity sensors monitor internal temperature and humidity. Gas concentration sensors monitor sensitive and hazardous gas concentrations in real time. This data is exported to the data processing layer module via wired or wireless transmission. Use Autodesk Revit to export the 3D model in FBX format, preserving structural details (including the tunnel / pipeline corridor roof, sidewalls, pipeline supports, etc.). Import the model into Unity3D, set material mapping (concrete texture, metal reflective material), and optimize the number of facets to less than 500,000.

[0096] S202. Start the data processing layer module, perform data coordinate mapping and spatial alignment processing on the collected data, associate the sensor data (stress, seepage, temperature, etc.) with the model vertices / grids to bind the physical field data, and use Kriging interpolation to map the discrete point data to the full-line grid of the tunnel / pipeline corridor (resolution 0.5m); divide the layers according to the physical field type (such as "stress field" and "seepage field"), support independent display and overlay, and use cubic spline interpolation to perform time alignment, match the low-frequency data with the high-frequency data timestamp, downsample the high-frequency data to match the low-frequency data to complete the normalization of data of different quality levels, perform cubic spline interpolation on the GNSS and piezometer data, and align them to the 1kHz timestamp.

[0097] S203. The abnormal data elimination system operates to check, screen, and eliminate the data. Specifically, wavelet transform is used for denoising high-frequency sensors (such as optical fibers); sliding average filtering is used for low-frequency sensors (such as GNSS); and abnormal data points are detected and eliminated based on statistical methods (3σ criterion) or isolation forest algorithm. Specifically, the data dimension normalization operation formula is as follows:

[0098]

[0099] Where X is the collected data, X max With X min These are the maximum and minimum values of the collected data. The abnormal data processing system will mark them and issue an alarm when they exceed the specified threshold.

[0100] S204. The multimodal data fusion system processes the nonlinear observation model. Taking the improved Kalman filter (KF) as an example, specifically, the formula is as follows:

[0101] The amount of sedimentation s kis the state variable, considering the sedimentation rate v k and acceleration a k , the state equations are as follows (2-2), (2-3), and (2-4).

[0102]

[0103] x k+1 =Fx k +w k

[0104]

[0105] The observation equation is represented by different sensors z k (i) Mapping with status:

[0106]

[0107] For example, the strain ∈ observed by the optical fiber sensor is converted into the settlement z by Hooke's law: (1) =E∈·L (E is the elastic modulus, L is the length of the tunnel / pipeline corridor); GNSS directly observes the displacement z (2) =s. Extended Kalman filter (EKF) or undestructive Kalman filter (UKF) is used to process nonlinear observation models and fuse multi-sensor data to update state estimates. In terms of time series prediction, ARIMA, LSTM or Transformer models are used, with fused multimodal data as input to predict future settlement trends. Monte Carlo Dropout or Bayesian neural network is used to output the confidence interval of the predicted value to complete uncertainty quantification. The data panel displays the fused multi-physics field heat map (such as the red warning area of the stress field), and the model vertices in the VR scene deform slightly with the strain data.

[0108] S205, the data coupling layer module establishes the seepage-stress-thermal coupling equation based on the data collected by the data acquisition layer, and reveals the comprehensive impact of multi-physical field interaction on soil stability through mathematical models. Input and output of settlement prediction, input parameters:

[0109] Seepage field: pore water pressure p, permeability coefficient k(T) (temperature related)

[0110] Stress field: soil elastic modulus E, Poisson's ratio ν

[0111] Temperature field: initial temperature T0, thermal expansion coefficient β T

[0112] The corresponding output is:

[0113] Sedimentation s(x, y, z, t): displacement field distributed in time and space

[0114] Stress concentration area: potential crack initiation location

[0115] Disaster evolution trends: subsidence rate, risk level

[0116] The calculation equations are as follows:

[0117] Seepage field equation (Darcy's law correction), considering the effect of temperature on fluid viscosity:

[0118]

[0119] k(T) is the temperature-dependent permeability coefficient, μ(T) is the fluid dynamic viscosity (e.g., μ = μ0 / (1 + kT)), φ is the porosity affected by strain, α is the Biot coefficient (characterizing the contribution of pore pressure to pressure), and β is the volume change coefficient caused by thermal expansion.

[0120] Stress field equation (thermoelastic correction):

[0121] Based on the effective stress principle and thermal expansion effect:

[0122]

[0123] Constitutive relation (linear thermoelasticity):

[0124] σ′=C:∈-β T (T-T0)I

[0125] Geometric equations (small deformation):

[0126]

[0127] Where σ′ is the effective stress tensor, C is the elastic stiffness matrix, β T is the thermal expansion coefficient, u is the displacement vector

[0128] Temperature field equation (convection-conduction coupling)

[0129] Considering seepage, convection and thermoelastic work:

[0130]

[0131] where k T is the heat transfer coefficient, v is the seepage velocity, Q mech is the heat source caused by stress-seepage coupling.

[0132] The above equations are combined to obtain the fully coupled seepage-stress-thermal control equation:

[0133]

[0134] If the tunnel / pipeline corridor disaster evolution time scale is long, the transient terms of the seepage field and temperature field can be ignored. If the temperature change is small (△T < 50℃), the influence of temperature on the permeability coefficient k(T) and viscosity μ(T) can be ignored, and only the thermal expansion effect is retained. For the axisymmetric structure of the tunnel / pipeline corridor, a two-dimensional model in the cylindrical coordinate system is used to reduce the amount of calculation.

[0135] S206, VR imaging and human-computer interaction layer module uses virtual reality technology to visually display the disaster evolution process and supports interactive simulation. The data processed by the data coupling layer and the data processed by the data processing layer module will be jointly imported into the VR imaging and human-computer interaction layer module. The disaster scene triggering form is dynamically displayed through the VR scene:

[0136] Seepage field: The particle system simulates the water flow path, and the flow rate is positively correlated with the seepage pressure (particle density changes dynamically)

[0137] Stress field: Vertex displacement animation shows crack expansion, and the color gradient (primary color → yellow → orange → red) indicates the stress level

[0138] Temperature field: thermal map superimposed with infrared pseudo-color, dynamic diffusion effect

[0139] Presentation effect: In the VR scene, seepage particles flow into the cracks in the tunnel / pipeline gallery side walls, roof settlement animation (assuming a displacement of 8.7 mm), and a heat map showing the spread of high-temperature areas.

[0140] Users use the controller to interactively verify treatment measures and generate intelligent warning and emergency plans. Using the controller, users can grab virtual tools (such as grouting equipment) and target cracked areas to trigger repair operations. Environmental conditions (such as drainage rate) can be modified in real time to observe the model's response. Simultaneously, the system triggers three levels of warning (red, orange, and yellow) based on the coupled model results. High-risk areas in the VR scene flash, and a treatment report (PDF format) is automatically generated, including operation records, pressure curves, and recommended measures.

[0141] Example of presentation effect (for example): After the user triggers "Grouting Reinforcement", the density of seepage particles decreases by 50%, the cracks stop expanding, and the report panel pops up "Treatment successful: seepage pressure drops to 0.18 MPa."

[0142] S207. System shutdown and data archiving, safe preservation of simulation data, support for historical case review and analysis, export of VR simulation video, sensor raw data and fusion results to the cloud database, and automatic generation of a summary of this simulation (key event timestamps, maximum displacement value, warning records).

[0143] In some embodiments, a dynamic particle system is optionally used to simulate the seepage path, with particle density positively correlated with the seepage velocity; structural deformation is calculated in real time through a vertex shader, with the crack propagation direction consistent with the direction of the maximum principal stress; and thermal map color maps the temperature field distribution, supporting dynamic LOD (level of detail) adjustment when the user switches perspectives. The trigger conditions for graded warnings are: Yellow warning: a single physical field exceeds the threshold (osmotic pressure increases by 10% compared to the baseline or daily sedimentation ≥ 5mm); Orange warning: two physical fields simultaneously exceed the threshold (osmotic pressure ≥ 0.25MPa or sedimentation rate ≥ 2mm / day or sensitive gas concentration > 50% LEL); Red warning: three physical fields simultaneously exceed the threshold and the temperature gradient ≥ 10℃ / m.

[0144] The data acquisition layer module of the present invention utilizes a combination of high-precision sensors and a 3D geological scanner. These instruments collaborate to sense subtle changes in underground tunnels / pipeline corridors, including stratum displacement, temperature and humidity variations, gas concentration changes, and underground seepage field variations. This provides accurate data for VR imaging of underground tunnels / pipeline corridors and 3D modeling, as well as project status assessments. Construction and subsequent maintenance teams can monitor data changes, identify potential risks in real time, and effectively optimize subsequent construction or maintenance measures. This significantly reduces structural hazards caused by changes in multi-physics field data and significantly improves the maintenance and operational efficiency of underground tunnel / pipeline corridor projects.

[0145] The data processing layer module of this embodiment cleverly combines multi-source heterogeneous data, aligns the data time steps by mathematical numerical analysis methods, and maps the multi-source physical field data to the tunnel / pipeline corridor coordinates using data space coordinate mapping alignment. At the same time, the high and low frequency data are normalized using the Kalman filter algorithm, and abnormal data points are detected and eliminated based on statistical methods (3σ criterion) or isolation forest algorithms. The multimodal data fusion processing in the data processing layer module uses ARIMA, LSTM or Transformer models in time series prediction, and uses the fused multimodal data as input to predict future settlement trends. Monte Carlo Dropout or Bayesian neural network is used to output the confidence interval of the predicted value to complete uncertainty quantification. The data of different dimensions and properties inside the underground tunnel / pipeline corridor project are processed with the same evaluation criteria, effectively avoiding the one-sided analysis caused by single physical field data in the traditional monitoring and early warning mechanism, and then formulating a more scientific and reasonable early warning judgment mechanism, reducing the risk of engineering economic losses while enhancing monitoring sensitivity.

[0146] The data coupling layer module in this embodiment leverages Biot theory and thermoelasticity to establish a seepage-stress-thermodynamic coupling equation. This mathematical model reveals the combined impact of multi-physics field interactions on soil stability. Before or when an underground engineering disaster is imminent, settlement predictions are made based on multi-physics field monitoring and coupled calculations, leveraging sensitive changes in seepage, soil stress, and geothermal heat.

[0147] The VR imaging and human-computer interaction layer module of this embodiment uses the Unity3D engine to develop and construct a tunnel / pipeline corridor visualization model and a human-computer interaction handle to monitor the tunnel / pipeline corridor in real time. The monitoring data includes but is not limited to real-time soil stress, temperature and humidity, seepage rate, and sensitive gas concentration. At the same time, the dynamic rendering engine is used to express the process of gas and liquid seepage movement in the form of dynamic particle flow. The three-level color separation warning mechanism in the early warning model and the decision-making layer module is combined to divide the tunnel / pipeline corridor model section into colors according to the danger level. The alarm operation is performed using sound and light changes. The early warning information can be transmitted to the PC, mobile terminal, and VR terminal through wireless transmission. An operating handle is provided for data query, perspective switching and sliding, data modification, and pre-processing scheme implementation simulation. Users can perform virtual processing operations such as grouting reinforcement and emergency closure of gas and water transmission channels in the virtual scene (real-time development and function addition can be performed as needed). The early warning plan and construction treatment plan are effectively designed through the three-dimensional multi-source data model display and human-computer interaction operation, so that operators can clearly identify the disaster problem while maintaining a sense of participation and accurately handle the problem in a timely manner, breaking the limitations of the previous two-dimensional single data display model.

[0148] The early warning model and decision-making layer module of this embodiment uses reference correlation, geological survey reports, tunnel / pipeline gallery engineering technical specifications, and construction foundation pit engineering monitoring technical standards to design a hierarchical mechanism for tunnel / pipeline gallery engineering disaster model virtual reality imaging early warning. This can more accurately represent the development trends of underground tunnel / pipeline gallery projects at different stages, such as changes in structural stability or sensitive data. It can then provide early warning for any safety hazards in the tunnel / pipeline gallery, effectively implement tunnel / pipeline gallery engineering safety early warnings, and promote the sustainable development of tunnel / pipeline gallery projects.

[0149] The method in this embodiment also includes other technical features of the tunnel corridor disaster monitoring method based on multi-physical field coupling in Example 1, which will not be repeated here.

[0150] Example 3:

[0151] This embodiment provides a tunnel and pipe gallery disaster monitoring system based on multi-physics field coupling, including:

[0152] The data acquisition module is configured to: acquire multi-physical field data of the underground monitoring object; wherein the multi-physical field data includes settlement displacement, stress and strain information, seepage information, temperature information and gas concentration information;

[0153] The processing module is configured to: align low-frequency data and high-frequency data in the multi-physics field data to a unified time base through an interpolation algorithm; and map the discrete multi-physics field data into a continuous physical field distribution based on Kriging spatial interpolation;

[0154] The prediction module is configured to: construct a seepage-stress-thermomechanical coupling mathematical model based on the aligned and mapped multi-physics field data, reduce the order of the model through an order reduction algorithm, and solve the seepage-stress-thermomechanical coupling mathematical model to obtain a prediction result;

[0155] The early warning module is configured to: display the prediction results in a preset three-dimensional model and issue graded warnings based on the prediction results.

[0156] The working method of the system is the same as the tunnel corridor disaster monitoring method based on multi-physical field coupling in Example 1 and / or Example 2, and will not be repeated here.

[0157] Example 4:

[0158] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps of the tunnel corridor disaster monitoring method based on multi-physical field coupling described in Example 1 are implemented.

[0159] Example 5:

[0160] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, the steps of the tunnel corridor disaster monitoring method based on multi-physical field coupling described in Example 1 are implemented.

[0161] Example 6:

[0162] This embodiment provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the tunnel corridor disaster monitoring method based on multi-physical field coupling described in Example 1 are implemented.

[0163] The above description is merely a preferred embodiment of this embodiment and is not intended to limit this embodiment. Those skilled in the art will readily appreciate that this embodiment may be modified and varied in various ways. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this embodiment shall be within the scope of protection of this embodiment.

Claims

1. A tunnel and pipe gallery disaster monitoring method based on multi-physics field coupling is characterized by: include: Acquiring multi-physical field data of an underground monitoring object; wherein the multi-physical field data includes settlement displacement, stress and strain information, seepage information, temperature information, and gas concentration information; Align low-frequency data and high-frequency data in multi-physics field data to a unified time base through interpolation algorithms; map discrete multi-physics field data to continuous physical field distribution based on Kriging spatial interpolation; Based on the aligned and mapped multi-physics field data, a seepage-stress-thermomechanical coupling mathematical model is constructed. The order of the model is reduced through an order reduction algorithm, and the seepage-stress-thermomechanical coupling mathematical model is solved to obtain the prediction results. The prediction results are displayed in a preset three-dimensional model, and graded warnings are issued based on the prediction results.

2. The tunnel and pipe gallery disaster monitoring method based on multi-physics field coupling according to claim 1, characterized in that: The monitoring object is a tunnel or a pipe gallery; the settlement displacement, the stress-strain information, the seepage information, the temperature information and the gas concentration information are processed for consistency, and an isolation forest algorithm is used to monitor and eliminate abnormal data points.

3. The tunnel and pipe gallery disaster monitoring method based on multi-physics field coupling according to claim 1, characterized in that: Kalman filtering is used to align the timestamps of multi-physics field data and suppress noise; deep neural networks are used to learn the nonlinear coupling relationship between multi-physics fields. The network input is settlement displacement, stress and strain information, seepage information, temperature information and gas concentration information, and the output is the predicted value of settlement amount.

4. The tunnel and pipe gallery disaster monitoring method based on multi-physics field coupling according to claim 1 is characterized in that: The seepage-stress-thermomechanical coupling mathematical model includes the seepage field equation, the stress field equation, and the temperature field equation; the seepage field equation is: Where k(T) is the temperature-dependent permeability coefficient; μ(T) is the fluid dynamic viscosity; φ is the porosity affected by strain; α is the Biot coefficient (characterizing the contribution of pore pressure to stress); β is the volume change coefficient caused by thermal expansion; is the pore water pressure; ρ is the liquid density; ∈ v is the volume strain; The stress field equation is: Where C is the elastic matrix; u is the fluid dynamic viscosity; β T is the coefficient of thermal expansion; is temperature; The temperature field equation is: Where ρ is the density; C p is the specific heat capacity at constant pressure; k T is the thermal conductivity coefficient; ρ f is the liquid density; C p , f is the specific heat capacity of the liquid at constant pressure; ν is the Poisson's ratio.

5. The tunnel and pipe gallery disaster monitoring method based on multi-physics field coupling according to claim 1, characterized in that: Dynamic particles are used to simulate the seepage path, and the particle density is positively correlated with the seepage velocity; the structural deformation is calculated in real time through the vertex shader, and the crack extension direction is consistent with the direction of the maximum principal stress; the temperature field distribution is mapped using thermal map colors.

6. The tunnel and pipe gallery disaster monitoring method based on multi-physics field coupling according to claim 1, characterized in that: The graded warnings include yellow warning, orange warning and red warning; a yellow warning is issued when a single physical field exceeds the threshold; an orange warning is issued when two physical fields exceed the threshold at the same time; a red warning is issued when three physical fields exceed the threshold at the same time and the temperature gradient exceeds the preset value.

7. The tunnel and pipeline corridor disaster monitoring system based on multi-physics field coupling is characterized by: include: The data acquisition module is configured to: acquire multi-physical field data of the underground monitoring object; wherein the multi-physical field data includes settlement displacement, stress and strain information, seepage information, temperature information and gas concentration information; The processing module is configured to: align low-frequency data and high-frequency data in the multi-physics field data to a unified time base through an interpolation algorithm; and map the discrete multi-physics field data into a continuous physical field distribution based on Kriging spatial interpolation; The prediction module is configured to: construct a seepage-stress-thermomechanical coupling mathematical model based on the aligned and mapped multi-physics field data, reduce the order of the model through an order reduction algorithm, and solve the seepage-stress-thermomechanical coupling mathematical model to obtain a prediction result; The early warning module is configured to: display the prediction results in a preset three-dimensional model and issue graded warnings based on the prediction results.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the tunnel and pipe gallery disaster monitoring method based on multi-physical field coupling as described in any one of claims 1 to 6 are implemented.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: When the processor executes the program, the steps of the tunnel corridor disaster monitoring method based on multi-physical field coupling as described in any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the tunnel corridor disaster monitoring method based on multi-physical field coupling as described in any one of claims 1 to 6 are implemented.

Images