A modeling method and system for tunnel model

Abstract

The present invention relates to the field of data modeling technology, and in particular to a modeling method and system for tunnel models. The method comprises the following steps: deploying a multi-source sensor array at the tunnel engineering site to collect surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data, and performing data preprocessing to construct a multi-source original data set for the tunnel; performing spatiotemporal fusion processing on the multi-source original data set for the tunnel to generate a synchronized tunnel data set; using the synchronized tunnel data set to construct a three-dimensional geomechanical feature space and generate a three-dimensional tunnel feature tensor; therefore, the present invention solves the problems of data fragmentation and insufficient model accuracy in traditional tunnel support design through multi-source data fusion and digital-physical collaborative modeling, thereby improving the prediction accuracy and construction safety level of the tunnel support scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of data modeling, and in particular to a modeling method and system applied to a tunnel model. Background Art

[0002] Existing methods applied to tunnel modeling generally have significant technical shortcomings in terms of multi-source data fusion, model accuracy control, parameter optimization and engineering adaptability. Most methods still rely on single physical quantity information in the data acquisition link. For example, model construction is based only on surrounding rock stress or surface deformation, ignoring the comprehensive utilization of multimodal data such as anchor strain, support structure stress state and tunnel point cloud geometry changes, resulting in the model being one-sided when reflecting the actual force and deformation mechanism inside the tunnel. At the same time, due to the lack of high-precision time synchronization and spatial alignment mechanisms, there is a problem of inconsistent temporal and spatial scales between multi-source data, which seriously restricts the integrity and accuracy of the subsequent three-dimensional geomechanical feature construction. At the modeling verification level, existing methods mostly use pure numerical simulation to construct the response model of the surrounding rock-support system. There is a lack of a closed-loop verification mechanism between the actual physical model test, and it is impossible to achieve dynamic correction of model parameters and error compensation, resulting in insufficient reflection of the digital model on the actual engineering working conditions. Furthermore, existing methods often rely on empirically defined parameter combinations during support optimization design, making it difficult to construct a high-dimensional support solution space covering complex working conditions. They also lack an evaluation mechanism that integrates multiple deformation criteria and structural constraints, making it difficult to meet the multi-objective, multi-constrained, and refined prediction requirements of actual tunnel design. More importantly, most current models are not lightweight after construction, resulting in excessively large parameters and complex structures, which limits their real-time deployment and dynamic updating capabilities within field embedded systems or edge computing platforms. Summary of the Invention

[0003] Based on this, it is necessary to provide a modeling method and system for a tunnel model to solve at least one of the above technical problems.

[0004] To achieve the above object, a modeling method for a roadway model is provided, the method comprising the following steps:

[0005] Step S1: deploying a multi-source sensing device array at the tunnel engineering site to collect surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data, and performing data preprocessing to construct a multi-source original data set for the tunnel;

[0006] Step S2: performing spatiotemporal fusion processing on the multi-source original data sets of the roadway to generate a synchronized roadway data set; constructing a three-dimensional geomechanical feature space using the synchronized roadway data set and generating a three-dimensional roadway feature tensor; and generating an initial digital model of the roadway based on the three-dimensional roadway feature tensor;

[0007] Step S3: Using the initial digital model of the tunnel to simulate the repair support scheme, and conducting physical model tests to obtain physical test data; using the physical test data to perform digital-physical error analysis to obtain model compensation parameters; based on the model compensation parameters, modifying the parameters of the initial digital model of the tunnel to obtain an optimized digital model of the tunnel;

[0008] Step S4: Predict the repair effect based on the tunnel optimization digital model and generate support optimization prediction data; perform full-process modeling evaluation based on the support optimization prediction data, and generate a report to obtain a full-process model construction report for the tunnel model.

[0009] The beneficial effect of the present invention is that, by deploying a multi-source sensor array at the tunnel engineering site, surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data are collected in real time, forming a multi-dimensional, multi-spatiotemporal resolution data foundation, providing detailed and comprehensive raw data support for subsequent data fusion and model construction. The multi-source raw data sets are synchronized using a spatiotemporal fusion algorithm to eliminate spatiotemporal asynchrony and dimensional differences between the data, ensuring that the spatial positions and timestamps of different types of data are highly consistent, thereby constructing an accurate synchronized tunnel data set. On this basis, combined with the principles of geomechanics, the fused data is mapped to a three-dimensional feature space, and a high-dimensional characterization of the multi-physical quantity characteristics of the complex tunnel structure is achieved through tensor expression, effectively capturing the coupling relationship between the surrounding rock and the support structure, and then generating an initial digital model that accurately reflects the mechanical state of the tunnel. By combining digital models with physical model tests, the digital and physical test data are compared using error analysis methods, and model compensation parameters are extracted to correct the systematic deviations in the digital model, thereby achieving optimization and precision of the digital model. Based on the optimized digital model, the effect prediction of the repair support scheme is carried out. Relying on the structured support optimization prediction data, a full-process modeling evaluation is carried out to systematically quantify the feasibility and safety of the repair scheme, and generate a detailed evaluation report to form a closed-loop data-driven design and verification process. This process not only realizes the dynamic and refined monitoring of the tunnel structure status, but also improves the accuracy and reliability of the model through data-driven methods, and promotes the scientific and intelligent support design. Therefore, the present invention solves the problems of data fragmentation and insufficient model accuracy in traditional tunnel support design through multi-source data fusion and digital-physical collaborative modeling, and improves the prediction accuracy and construction safety level of the tunnel support scheme.

[0010] Preferably, step S1 includes the following steps:

[0011] Step S11: deploying fiber Bragg grating sensor arrays on the roadway roof and both sides with a sampling frequency of 10 Hz, a measurement range of 0-50 MPa, and an accuracy of ±0.1 MPa to collect surrounding rock deformation data;

[0012] Step S12: Arrange a laser scanner on the tunnel surface with a scanning accuracy of 0.1 mm and collect deformation point cloud data of the tunnel surface at intervals of 30 minutes;

[0013] Step S13: embedding strain sensors inside the anchor rods and lining structure with a sampling frequency of 5 Hz to collect support structure status data;

[0014] Step S14: using wavelet transform to perform noise reduction processing on the surrounding rock deformation data, the roadway surface deformation point cloud data, and the support structure status data to obtain a multi-source roadway noise reduction data set;

[0015] Step S15: performing data cleaning on the multi-source laneway denoising dataset to generate a laneway multi-source original dataset.

[0016] This method deploys high-precision fiber Bragg grating (FBG) sensor arrays on the tunnel roof and sides to achieve continuous, high-frequency monitoring of surrounding rock deformation. With a sampling frequency of 10 Hz, the system covers a measurement range of 0-50 MPa and boasts a high accuracy of ±0.1 MPa, enabling detailed capture of stress changes in the surrounding rock during different construction phases. Simultaneously, a laser scanner collects high-precision (0.1 mm) point cloud data of the tunnel surface, performing periodic scans at 30-minute intervals to ensure both spatial resolution and temporal continuity of the tunnel surface deformation, thereby obtaining rich three-dimensional deformation information. Strain sensors embedded within the anchor bolts and lining structures utilize a 5 Hz sampling frequency, a range of ±5000 με, and an accuracy of ±5 με. These sensors provide real-time information on the stress state and deformation of the support structure, generating a multi-dimensional, multi-physical sensor data set. Wavelet transform technology is used to perform multi-scale noise reduction on different types of sensor data, effectively removing measurement noise and environmental interference, enhancing the signal-to-noise ratio, and ensuring data accuracy and reliability. Subsequently, a systematic data cleaning process was performed on the denoised multi-source dataset to remove outliers and missing data, fill data gaps, unify data formats and standards, and form a structured multi-source original dataset for the tunnel. This dataset not only covers multidimensional information on the surrounding rock, surface, and support structure, but also achieves high precision and temporal and spatial consistency of the data, laying a solid data foundation for subsequent fusion analysis and digital model construction. Overall, this invention ensures the integrity and accuracy of tunnel structure monitoring data from the source through multi-sensor fusion acquisition and advanced data processing technology, greatly improving data quality and the effectiveness of subsequent analysis.

[0017] Preferably, in step S2, constructing a three-dimensional geomechanical feature space using the synchronized roadway dataset and generating a roadway feature tensor includes:

[0018] Extract stress distribution based on synchronized roadway data sets, mark rock stress maximums, and generate roadway rock stress characteristic data;

[0019] Based on the synchronized roadway data set, the roadway surface point displacement convergence analysis is performed, and the surface curvature change is measured to generate roadway point cloud feature data;

[0020] Based on the synchronized roadway data set, the support structure axis deformation analysis is performed, and the internal shear deformation of the structure is calculated to generate the internal structural characteristic data of the roadway;

[0021] The tunnel rock stress characteristic data, tunnel point cloud characteristic data and tunnel internal structure characteristic data are mapped to a three-dimensional spatial grid according to stress × displacement × strain to construct a three-dimensional geomechanical characteristic space and generate a three-dimensional tunnel characteristic tensor.

[0022] The present invention systematically extracts and comprehensively expresses the multidimensional mechanical characteristics of tunnels based on synchronized tunnel datasets. First, stress distribution is extracted from the collected synchronized data, and the stress intensity and its maximum and minimum values ​​at different locations of the surrounding rock are accurately calculated. Local stress concentration phenomena are effectively captured by rock stress maximum markers, forming a high-resolution tunnel rock stress characteristic dataset. Second, the displacement of key points on the tunnel surface is converged based on point cloud data, quantifying the displacement change trend and amplitude of the tunnel surface. Combined with the curvature change measurement of the surface points, point cloud feature data reflecting the overall and local deformation morphology of the tunnel are generated. Furthermore, by dynamically tracking the axial deformation of the support structure axis and defense line, quantitative calculation of the internal shear deformation of the structure is carried out, and the structural strain distribution and its variation law are obtained, thereby generating internal structural feature data of the tunnel. The above three types of feature data are mapped and fused in high dimensions according to the three dimensions of stress, displacement, and strain to construct a unified three-dimensional spatial grid system, achieving a three-dimensional geomechanical feature spatial characterization of the complex mechanical behavior of the tunnel, and ultimately generating a three-dimensional tunnel feature tensor containing multi-physical quantity coupling information. This tensor not only reflects the spatial distribution characteristics of the surrounding rock stress state, surface deformation and support structure response, but also realizes the effective fusion and high-dimensional expression of multi-source heterogeneous data, providing an accurate data basis and theoretical support for subsequent mechanical analysis, numerical simulation and optimization design, and significantly improving the scientificity and accuracy of tunnel structure health monitoring and safety assessment.

[0023] Preferably, generating the initial digital model of the tunnel based on the three-dimensional tunnel characteristic tensor in step S2 includes:

[0024] Based on the three-dimensional lane feature tensor, the spatial pattern is learned through a preset convolution layer to obtain the lane spatial feature data;

[0025] The 3D tunnel characteristic tensor is used to reconstruct the surrounding rock mass geometry to obtain a 3D tunnel structure model;

[0026] The geometric strain state attributes of the concrete lining are reconstructed from the 3D tunnel characteristic tensor to obtain a 3D support structure sub-model.

[0027] The spatial characteristic data of the tunnel are used to simulate the contact surface force transmission mechanism of the three-dimensional tunnel structure model and the three-dimensional support structure model to obtain the initial digital model of the tunnel.

[0028] The present invention realizes deep learning and spatial pattern extraction of multi-dimensional spatial mechanical characteristics in the tunnel by applying a multi-layer convolutional neural network with a preset convolution layer structure based on the three-dimensional tunnel characteristic tensor, systematically captures the complex correlation and potential spatial laws between the surrounding rock mass and the support structure, generates high-dimensional tunnel spatial characteristic data, and provides accurate feature expression for subsequent structural modeling. At the data level, the stress, displacement and strain information in the three-dimensional tunnel characteristic tensor are used to reconstruct the structural sub-model of the three-dimensional surrounding rock mass through a geometric reconstruction algorithm, accurately restore the spatial form and mechanical state of the tunnel surrounding rock mass, and realize a unified description of geometric deformation and mechanical response. At the same time, based on the strain state attributes in the tensor, the three-dimensional geometric model of the concrete lining is reconstructed, the strain distribution and local mechanical behavior inside the support structure are refined, and a three-dimensional support structure sub-model is obtained. Combined with the spatial characteristic data of the tunnel, the force transmission mechanism of the contact surface between the surrounding rock mass model and the support structure model was simulated. Through numerical calculation and mechanical coupling analysis, the interface mechanical interaction between the surrounding rock and the support structure was accurately characterized, reflecting the stress transmission path and deformation coordination. Ultimately, an initial digital model of the tunnel with high precision and multi-physics coupling characteristics was generated. This model fully integrates the structural information and mechanical characteristics of multi-source spatial data, improving the realism and detailed representation of the digital model. It provides a solid data foundation and theoretical support for subsequent simulation analysis, structural optimization, and safety assessment, significantly enhancing the scientific and practical nature of the digital design and management of tunnel engineering.

[0029] Preferably, step S3 includes the following steps:

[0030] Step S31: simulating the repair and support scheme using the initial digital model of the tunnel to obtain tunnel design repair and support simulation data;

[0031] Step S32: generating physical abrasive tool parameters based on the tunnel design repair and support simulation data, and performing a physical model test to obtain physical test data;

[0032] Step S33: Perform digital-physical error analysis using the physical test data and the tunnel design repair and support simulation data to obtain model compensation parameters; perform parameter correction on the initial tunnel digital model based on the model compensation parameters to obtain an optimized tunnel digital model.

[0033] The present invention generates detailed tunnel design and repair support simulation data by performing numerical simulation of the repair support scheme based on the initial digital model of the tunnel, thus achieving preliminary evaluation and effect prediction of different support schemes in a digital environment. At the data level, these simulation data are used to construct physical mold parameters to guide the implementation of physical model tests, collect multi-dimensional physical test data including stress, strain and displacement, and truly reflect the mechanical response and deformation behavior of the tunnel structure under repair support conditions. Subsequently, by performing quantitative digital-physical error analysis on the differences between the physical test data and the digital simulation data, the error distribution characteristics are extracted and the model compensation parameters are calculated, thereby correcting the deviations in the digital model. This process effectively utilizes the error feedback mechanism, so that the model compensation parameters can reflect the nonlinear mechanical behavior and complex interactions in the real physical environment, significantly improving the accuracy and credibility of the digital model. Based on these compensation parameters, the parameters of the initial digital model are corrected, and the generated tunnel optimization digital model more accurately reflects the actual mechanical state and structural response of the tunnel, enhances the digital model's predictive ability and adaptability to the repair support effect, and reduces the deviations caused by model simplification or assumptions. This method realizes the closed-loop integration of digital simulation and physical experiments at the data level. Through error-driven model optimization, it effectively improves the scientificity and practicality of the digital design of tunnel structures, provides reliable data support and theoretical basis for the optimization of subsequent support schemes, and significantly promotes the deepening application of digital twin technology in mine tunnel engineering.

[0034] Preferably, step S32 includes the following steps:

[0035] Step S321: performing scaling model size processing based on the tunnel design repair support simulation data to obtain tunnel scaling model parameters, wherein the geometric similarity ratio of the scaling model size processing is 1:50 and the stress similarity ratio is 1:75;

[0036] Step S322: extract anchor position points from the initial digital model of the tunnel to obtain model anchor position data; reserve anchor points based on the model anchor position data to obtain hole density parameter data;

[0037] Step S323: Perform an equivalent boundary load physical model experiment using the tunnel scaling model parameters and the hole density parameter data to obtain physical test data.

[0038] The present invention effectively achieves the scale matching and mechanical similarity conversion between the digital model and the physical test model by scaling the model size based on the tunnel design repair and support simulation data. Specifically, the tunnel design data is scaled using a geometric similarity ratio of 1:50 and a stress similarity ratio of 1:75 to ensure that the physical model is highly consistent with the actual tunnel in terms of spatial size and stress distribution, laying the foundation for subsequent physical experiments. At the data level, by accurately extracting the anchor position points in the initial digital model of the tunnel, a model anchor position data set is constructed. The data set contains the spatial coordinates and distribution density information of the anchor, and then the anchor point is reserved based on this data to generate hole density parameter data, which reflects the density of the anchor arrangement and its spatial distribution characteristics. By using the above-mentioned scaled model parameters and hole density parameter data, equivalent boundary load conditions are constructed in the physical experiment stage to achieve accurate simulation of the physical model in boundary mechanical loading, ensuring a high degree of consistency between the experimental environment and the actual tunnel mechanical state. The test data collected through physical model experiments covers multi-dimensional mechanical response information, including stress field distribution, deformation patterns, and local anchor response, forming a real physical test data set. This data set provides a key measured basis for subsequent digital-physical error analysis and effectively supports the optimization and correction of digital model parameters. Overall, this method achieves a closed-loop mapping from digital simulation to physical experiment at the data level, promotes the accurate maintenance of geometric and mechanical similarity during scale conversion, improves the representativeness and reliability of physical experimental results, and provides a solid data foundation and experimental guarantee for the scientific design and optimization of tunnel support structures.

[0039] Preferably, step S33 includes the following steps:

[0040] Step S331: using the physical test data and the tunnel design repair and support simulation data to calculate the displacement relative error, and obtain the tunnel displacement relative error data; using the physical test data and the tunnel design repair and support simulation data to calculate the strain relative error, and obtain the tunnel strain relative error data;

[0041] Step S332: When the relative error data of the roadway displacement is greater than 10% and the relative error data of the roadway strain is less than 0.85, a genetic algorithm is used to optimize the parameters of the surrounding rock constitutive model to obtain model compensation parameters;

[0042] Step S333: modifying the parameters of the initial digital model of the tunnel based on the model compensation parameters to obtain an optimized digital model of the tunnel.

[0043] The present invention achieves dynamic correction between the digital model and the actual physical state by performing detailed error analysis on physical test data and tunnel design repair and support simulation data. Specifically, the displacement data of the physical test and digital simulation results are first compared and calculated at the data level to obtain the relative error data of the tunnel displacement. This data reflects the degree of deviation between the tunnel deformation predicted by the digital model and the actual deformation of the physical model. At the same time, the strain data is subjected to the same comparative processing to obtain the relative error data of the tunnel strain. This data reflects the accuracy of the simulation of the strain state of the material inside the support structure. Based on the above error data, a judgment threshold mechanism is set. When the relative error of displacement exceeds 10% and the relative error of strain is less than 0.85, the genetic algorithm optimization process is initiated. As a global optimization method based on evolutionary mechanisms, the genetic algorithm encodes the parameters of the surrounding rock constitutive model as chromosomes, combines selection, crossover and mutation operations, and performs iterative optimization with the goal of minimizing the error to search for a set of optimal or near-optimal model compensation parameters. These compensation parameters are intended to correct the deviations of the mechanical properties, boundary conditions or other key parameters of the material in the digital model, so that the model is more consistent with the actual performance of the physical test. After obtaining the model compensation parameters, a data-driven parameter correction method was used to systematically adjust the initial digital model of the roadway, generating an optimized digital model with higher fit and prediction accuracy. This process achieved closed-loop feedback between the physical and digital models at the data level, enhancing the model's adaptability and reliability. This significantly improved the accuracy and practical value of the digital model in actual engineering applications, promoting the scientific and refined development of engineering support structure design and optimization.

[0044] Preferably, step S4 includes the following steps:

[0045] Step S41: Predicting the repair effect based on the tunnel optimization digital model and generating support optimization prediction data;

[0046] Step S42: Perform full-process modeling evaluation based on the support optimization prediction data, and update the mold design parameters to obtain full-process roadway modeling data;

[0047] Step S43: constructing a full-process model construction report of the lane model based on the full-process lane model modeling data.

[0048] This invention achieves forward-looking data analysis and evaluation of support structure performance by predicting repair effects based on an optimized digital model of the tunnel. Specifically, at the data level, support optimization prediction data is first generated using a simulation algorithm by inputting multidimensional parameters into the optimized digital model. This data includes quantitative prediction results for design variables (such as anchor spacing and lining thickness) and constraints (such as displacement limits), reflecting the potential performance of the repair solution in an actual engineering environment. Subsequently, a full-process modeling evaluation is conducted based on the support optimization prediction data. This process includes dynamic adjustment and feedback of parameters at each stage of the model, and real-time updating of mold design parameters based on the predicted data. This effectively ensures data consistency and accuracy between the physical model and the digital model, ensuring the continuity and closed-loop control of the modeling process. Through this multi-level and multi-angle parameter adjustment and evaluation, full-process tunnel model modeling data covering design, simulation, testing, and feedback is formed, fully demonstrating the organic integration from data acquisition, model simulation, to physical testing. On this basis, the full-process roadway modeling data was systematically summarized and analyzed to generate a full-process roadway model construction report. This report includes the evolution of key model parameters, error correction status, and quantitative indicators of optimization effects, providing data support and scientific basis for subsequent engineering decisions. This process ensures the accuracy and adaptability of the model through data-driven iterative optimization, while promoting the intelligent and standardized design of roadway support, significantly improving the systematicness and transparency of the entire modeling process.

[0049] Preferably, step S41 includes:

[0050] Based on the preset judgment criteria, the repair effect prediction of the tunnel optimization digital model is carried out to generate support optimization prediction data, where the repair effect prediction includes design variable prediction and constraint condition prediction;

[0051] When predicting design variables, the preset judgment criteria include anchor bolt horizontal spacing of 0.5-1.5m and lining thickness of 0.1-0.3m. When the support optimization prediction data is not, the remaining data will be marked as scrapped data;

[0052] When predicting the constraint conditions, the preset judgment criteria include top plate displacement less than 50mm, side displacement less than 30mm, and when the support optimization prediction data is the rest, the data will be marked as scrapped.

[0053] The present invention predicts the repair effect of the tunnel optimization digital model based on preset judgment criteria, realizes systematic data verification and screening of the support scheme design variables and constraints, and ensures the validity and reliability of the prediction data. At the data level, the optimized digital model is first used to generate support optimization prediction data. The data contains specific numerical information of design variables (such as horizontal anchor spacing and lining thickness) and constraints (such as roof displacement and side displacement), reflecting the performance characteristics of the repair scheme under actual working conditions. Subsequently, the design variable prediction data is screened according to the preset judgment criteria. The horizontal anchor spacing is limited to the range of 0.5 to 1.5 meters, and the lining thickness is limited to the range of 0.1 to 0.3 meters. All prediction data outside this range are marked as scrapped data, thereby eliminating abnormal data that does not conform to the actual project and design specifications, ensuring the accuracy and rationality of the data basis for subsequent analysis and application. At the same time, in the constraint prediction, threshold limits of less than 50 mm and 30 mm were set for the displacement of the roof and side respectively. Prediction data exceeding the threshold was also marked as scrapped data to exclude potential structural instability risk data and improve the model's prediction accuracy for structural safety. This dual judgment mechanism based on a strict threshold system effectively realizes dynamic verification and data quality control between design variables and constraint conditions, enhancing data availability and the reliability of model prediction. Through strict filtering and classification of multi-dimensional data, the overall process makes the support optimization prediction data not only engineering applicable, but also statistically valid, providing solid data support for subsequent support scheme optimization and risk assessment, thereby promoting the development of tunnel support design towards intelligence and refinement.

[0054] In this specification, a modeling system for a lane model is provided, which is used to execute the above-mentioned modeling method for a lane model. The modeling system for a lane model includes:

[0055] The multi-source data acquisition module is used to deploy a multi-source sensor array on-site in the tunnel project to collect surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data, and perform data preprocessing to construct a multi-source original data set for the tunnel;

[0056] The spatiotemporal fusion and 3D modeling module is used to perform spatiotemporal fusion processing on the multi-source original data sets of the roadway to generate a synchronized roadway data set; construct a 3D geomechanical feature space using the synchronized roadway data set and generate a 3D roadway feature tensor; and generate an initial digital model of the roadway based on the 3D roadway feature tensor;

[0057] The physical feedback and model correction module is used to simulate the repair and support scheme using the initial digital model of the tunnel, conduct physical model tests, and obtain physical test data; use the physical test data to perform digital-physical error analysis to obtain model compensation parameters; and based on the model compensation parameters, perform parameter correction on the initial digital model of the tunnel to obtain the optimized digital model of the tunnel;

[0058] The modeling evaluation and report generation module is used to predict the repair effect based on the tunnel optimization digital model and generate support optimization prediction data; perform full-process modeling evaluation based on the support optimization prediction data, and generate a report to obtain a full-process model construction report for the tunnel model.

[0059] The beneficial effect of the present invention is that, by deploying a multi-source sensor array at the tunnel engineering site, surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data are collected in real time, forming a multi-dimensional, multi-spatiotemporal resolution data foundation, providing detailed and comprehensive raw data support for subsequent data fusion and model construction. The multi-source raw data sets are synchronized using a spatiotemporal fusion algorithm to eliminate spatiotemporal asynchrony and dimensional differences between the data, ensuring that the spatial positions and timestamps of different types of data are highly consistent, thereby constructing an accurate synchronized tunnel data set. On this basis, combined with the principles of geomechanics, the fused data is mapped to a three-dimensional feature space, and a high-dimensional characterization of the multi-physical quantity characteristics of the complex tunnel structure is achieved through tensor expression, effectively capturing the coupling relationship between the surrounding rock and the support structure, and then generating an initial digital model that accurately reflects the mechanical state of the tunnel. By combining digital models with physical model tests, the digital and physical test data are compared using error analysis methods, and model compensation parameters are extracted to correct the systematic deviations in the digital model, thereby achieving optimization and precision of the digital model. Based on the optimized digital model, the effectiveness of the repair support scheme is predicted. Relying on structured support optimization prediction data, a full-process modeling evaluation is conducted. The feasibility and safety of the repair scheme are systematically quantified, and a detailed evaluation report is generated, forming a closed-loop data-driven design and verification process. This process not only enables dynamic and detailed monitoring of the tunnel structure status, but also improves the accuracy and reliability of the model through data-driven methods, promoting scientific and intelligent support design. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic flow chart of the steps of a modeling method applied to a roadway model;

[0061] Figure 2 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0062] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0063] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative work are within the scope of protection of the present invention.

[0064] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0065] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0066] To achieve this, please refer to Figures 1 to 2 A modeling method for a roadway model, comprising the following steps:

[0067] Step S1: deploying a multi-source sensing device array at the tunnel engineering site to collect surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data, and performing data preprocessing to construct a multi-source original data set for the tunnel;

[0068] Step S2: performing spatiotemporal fusion processing on the multi-source original data sets of the roadway to generate a synchronized roadway data set; constructing a three-dimensional geomechanical feature space using the synchronized roadway data set and generating a three-dimensional roadway feature tensor; and generating an initial digital model of the roadway based on the three-dimensional roadway feature tensor;

[0069] Step S3: Using the initial digital model of the tunnel to simulate the repair support scheme, and conducting physical model tests to obtain physical test data; using the physical test data to perform digital-physical error analysis to obtain model compensation parameters; based on the model compensation parameters, modifying the parameters of the initial digital model of the tunnel to obtain an optimized digital model of the tunnel;

[0070] Step S4: Predict the repair effect based on the tunnel optimization digital model and generate support optimization prediction data; perform full-process modeling evaluation based on the support optimization prediction data, and generate a report to obtain a full-process model construction report for the tunnel model.

[0071] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a flow chart of steps of a modeling method for a laneway model according to the present invention. In this example, the modeling method for a laneway model includes the following steps:

[0072] Step S1: deploying a multi-source sensing device array at the tunnel engineering site to collect surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data, and performing data preprocessing to construct a multi-source original data set for the tunnel;

[0073] In an embodiment of the present invention, during the deployment of a multi-source sensing array at a tunnel construction site, multiple types of sensors are deployed to monitor the deformation behavior of the surrounding rock and support structure with high temporal and spatial resolution, while simultaneously collecting point cloud data of the tunnel surface deformation. Specifically, surrounding rock deformation data is continuously acquired using fiber Bragg grating strain gauges, displacement meters, or distributed fiber optic sensing systems to capture the detailed evolution of micro-deformation within the surrounding rock. Point cloud data of the tunnel surface deformation is periodically scanned using laser scanners (such as 3D lidar) and structured light systems, forming a multi-viewpoint point cloud dataset with spatial density characteristics. Support structure status data is typically collected using a combination of sensors, including strain gauges, crack meters, and inclinometers, to reflect the stress and deformation state of anchor bolts, steel arches, or shotcrete layers. After acquisition, these raw data must first be time-stamped and coordinate-system-aligned. A time synchronization module is used to align asynchronous data sources, and a spatial projection algorithm is used to standardize data from different coordinate spaces. Furthermore, the data characteristics generated by different sensor devices require separate operations such as format conversion, noise filtering, missing value interpolation, and outlier identification. Point cloud data also requires further registration, filtering, and sparse reconstruction to improve the accuracy of subsequent feature extraction. Ultimately, all processed data is fused and organized into a structured, multi-source raw data set for the roadway. This data set preserves the original measurement dimensions and spatiotemporal distribution characteristics of each sensor source, providing a unified and reusable data foundation for subsequent modeling.

[0074] Step S2: performing spatiotemporal fusion processing on the multi-source original data sets of the roadway to generate a synchronized roadway data set; constructing a three-dimensional geomechanical feature space using the synchronized roadway data set and generating a three-dimensional roadway feature tensor; and generating an initial digital model of the roadway based on the three-dimensional roadway feature tensor;

[0075] In an embodiment of the present invention, after constructing a multi-source raw data set for a roadway, it is first necessary to perform spatiotemporal fusion processing on the various time series and spatial data in the data set. This process primarily involves two core steps: time alignment and spatial reconstruction. In terms of time alignment, the asynchronously acquired data are coordinated on a unified time scale by standardizing the timestamps of the multi-source sensor data and combining them with linear interpolation or a nonlinear alignment algorithm based on dynamic time warping (DTW). In terms of spatial reconstruction, a multi-source point cloud registration algorithm, such as a combined method based on iterative closest point (ICP) and voxel grid filtering, is employed to transform spatial measurement results from different sources into a global geological coordinate system. Multi-scale spatial interpolation techniques are then used to fill in measurement blind spots and sparse areas, thereby generating a continuous three-dimensional spatial field. Subsequently, based on the fused data, a three-dimensional geomechanical feature space is constructed. This space expresses the coupling relationship between surrounding rock stress, displacement field, structural response, and the state of the retaining system in the form of a tensor field. High-dimensional feature extraction techniques (such as tensor decomposition, principal component analysis, or three-dimensional convolution kernel embedding) are used to encode the spatiotemporal data and map it into a three-dimensional feature tensor of a fixed structure. This feature tensor serves as the basic input for constructing the initial digital model of the tunnel. Further data-driven three-dimensional modeling processes, such as graph neural networks or spatial mapping reconstruction algorithms, are used to reconstruct the tunnel geometry and simulate its mechanical response. This generates an initial digital model of the tunnel with structural integrity, spatial consistency, and physical logical rationality, providing fundamental numerical support for subsequent repair plan simulations and multi-field coupling calculations.

[0076] Step S3: Using the initial digital model of the tunnel to simulate the repair support scheme, and conducting physical model tests to obtain physical test data; using the physical test data to perform digital-physical error analysis to obtain model compensation parameters; based on the model compensation parameters, modifying the parameters of the initial digital model of the tunnel to obtain an optimized digital model of the tunnel;

[0077] In this embodiment of the present invention, after the initial digital model of the tunnel is constructed, a simulation of the repair and support scheme is conducted based on the model to assess its responsiveness to actual working conditions and the adaptability of the support strategy. This process relies on numerical simulation methods such as the finite element method (FEM), discrete element method (DEM), or coupled field intensity method. By applying different types of loading boundary conditions to the model, the deformation, crack propagation, and stress evolution of the tunnel structure under repair and support are simulated, generating a multi-period mechanical response data series. Simultaneously, physical model tests are conducted to compare the repair and support response under real material and structural conditions, obtaining physical observation data such as experimental stress-strain curves, displacement field evolution maps, and localized failure morphology images. To achieve consistency analysis between the two, the experimental data is quantified and feature extracted, and then compared with the spatiotemporal response data obtained during the simulation. Error mapping analysis, residual distribution evaluation, and multidimensional regression fitting are used to identify the sources of deviation in the digital model and the mismatch intervals of key parameters. Based on the error analysis results, a set of model compensation parameters was extracted, including correction factors for material mechanical parameters, support structure stiffness adjustment coefficients, and fine-tuning items for contact boundary conditions. A compensation parameter mapping table was then constructed. Finally, based on this parameter table, the relevant modules in the initial digital model were revised and updated. Parameter replacement and submodule reconstruction methods were used for local reconstruction and overall consistency verification, resulting in a revised optimized digital model of the roadway. This optimized model not only maintains a characteristic distribution consistent with the site at the geometric and structural levels but also achieves higher approximate accuracy in response behavior, providing a more realistic model foundation for the next stage of predictive analysis.

[0078] Step S4: Predict the repair effect based on the tunnel optimization digital model and generate support optimization prediction data; perform full-process modeling evaluation based on the support optimization prediction data, and generate a report to obtain a full-process model construction report for the tunnel model.

[0079] In an embodiment of the present invention, after completing the construction of the optimized digital model of the tunnel, a predictive analysis of the repair and support effects needs to be carried out based on the model. The core of this analysis is to use the numerical evolution capability of the structure-geology-support coupling to simulate virtual loading for different repair strategies, thereby forming a complete time-series response data set. The specific technical approach includes grouping and arranging the various support materials and structural parameters (such as anchor length, anchor angle, steel arch stiffness, shotcrete thickness, etc.) input into the optimization model to construct a multi-condition simulation sample set; for each parameter combination, the structural response is predicted using a three-dimensional finite element method (FEM) or a multi-field coupling simulation method, and the output includes multi-dimensional spatiotemporal response indicators such as surrounding rock stress redistribution, support structure bearing capacity attenuation trend, and crack expansion morphology. On this basis, the prediction results are normalized, and a set of response data indicators under typical support conditions is extracted to form support optimization prediction data. Subsequently, the predicted data was compared with the initial support condition response results, and the full-process optimization effect was quantitatively evaluated by constructing an evaluation function system (such as the support safety factor improvement rate, residual deformation reduction rate, plastic zone expansion rate, etc.). Error backtracking and sensitivity analysis were also performed to determine the influence weight of different parameters on the repair effect. After the full-process modeling evaluation was completed, the tunnel modeling analysis report was generated through an automated template generation system, combining simulation parameters, response curves, evaluation indicators and model evolution record data. This report presents the model construction process, key input parameters, optimized simulation path, predicted response data and evaluation indicator maps in a structured manner, and is accompanied by model stability analysis and numerical accuracy evaluation results, forming a systematic tunnel full-process model construction and evaluation document dataset.

[0080] Preferably, step S1 includes the following steps:

[0081] Step S11: deploying fiber Bragg grating sensor arrays on the roadway roof and both sides with a sampling frequency of 10 Hz, a measurement range of 0-50 MPa, and an accuracy of ±0.1 MPa to collect surrounding rock deformation data;

[0082] Step S12: Arrange a laser scanner on the tunnel surface with a scanning accuracy of 0.1 mm and collect deformation point cloud data of the tunnel surface at intervals of 30 minutes;

[0083] Step S13: embedding strain sensors inside the anchor rods and lining structure with a sampling frequency of 5 Hz to collect support structure status data;

[0084] Step S14: using wavelet transform to perform noise reduction processing on the surrounding rock deformation data, the roadway surface deformation point cloud data, and the support structure status data to obtain a multi-source roadway noise reduction data set;

[0085] Step S15: performing data cleaning on the multi-source laneway denoising dataset to generate a laneway multi-source original dataset.

[0086] In an embodiment of the present invention, during the acquisition and processing of tunnel monitoring data, a fiber grating sensor array is first deployed on the tunnel roof and both sides to collect surrounding rock stress change data in real time through the wavelength modulation principle. The operating frequency of the sensor is set to 10 Hz, which can cover a measurement range of 0 to 50 MPa, with a measurement accuracy of ±0.1 MPa, and can dynamically track the stress state of the surrounding rock. At the same time, a high-precision laser scanner is arranged on the tunnel surface, and a planar array scanning method is used to perform periodic sampling at intervals of 30 minutes. The scanning accuracy is controlled at 0.1 mm, and the output data is a dense spatial point cloud collection, which records the geometric characteristics of the deformation evolution of the tunnel surface. In addition, resistive strain sensors are embedded inside support structures such as anchor rods and concrete linings. The sampling frequency is set to 5 Hz, the range is ±5000 microstrain, and the accuracy is controlled at ±5 microstrain, which is used to obtain subtle stress change information inside the support structure. After the data of these three types of sensors are collected, they need to enter the signal preprocessing stage separately. Wavelet transform is used as the main noise reduction algorithm. Its local analysis ability in the time and frequency domain is used to perform multi-scale decomposition and reconstruction of high-frequency interference in non-stationary signals, thereby removing high-frequency noise in the surrounding rock deformation time series, outliers and ranging errors in point cloud data, and random disturbances in strain data. After the noise reduction process is completed, the multi-source data set is cleaned, mainly including missing value identification and interpolation, data format standardization, timestamp synchronization and sensor drift correction, so that all data are structurally unified. Finally, the surrounding rock stress data, surface point cloud deformation data and structural strain data that have undergone wavelet denoising and cleaning are integrated into a structured multi-source original data set of the tunnel, providing a clear, stable and high-precision data foundation for subsequent spatiotemporal fusion modeling.

[0087] Of particular importance is that step S2 of performing spatiotemporal fusion processing based on the multi-source original data sets of the roadway includes:

[0088] The multi-source original data sets were time-aligned with a time resolution of 0.1s using the cubic spline interpolation method to obtain multi-source time-aligned data.

[0089] Perform timestamp synchronization based on multi-source time alignment data to obtain multi-source timestamp synchronization data;

[0090] Spatial alignment is performed based on multi-source time-stamped synchronized data to obtain a synchronized laneway dataset.

[0091] In an embodiment of the present invention, the multi-source raw data sets collected on-site in the tunnel, including fiber Bragg grating strain data, laser point cloud deformation data, and support structure status data collected by embedded sensors, suffer from inconsistent sampling frequencies, different time acquisition starting points, and differences in data frame lengths. Therefore, unified temporal and spatial fusion processing is necessary. To achieve consistency in the time domain, a cubic spline interpolation method is used to perform equal time step interpolation resampling operations on various data channels. Specifically, the time resolution is set to 0.1s, so that all types of data have the same sampling point structure on a unified time axis, and missing or asynchronous segments are smoothly supplemented, thereby obtaining multi-source time-aligned data. Subsequently, to address the inherent acquisition time offset problem of different data channels, timestamp synchronization processing is further performed on the multi-source time-aligned data. Specifically, the local acquisition timestamps attached to each type of data are offset from the global reference timestamps. A time difference minimization algorithm is used to align the start times and update frequencies of multiple channels, thereby generating multi-source timestamp synchronized data under a unified reference frame structure. After completing time synchronization, to achieve spatial consistency, the original spatial position annotations must be transformed and aligned based on the spatial reference system of each data type and its mapping relationship in the roadway coordinate system. Through rigid transformation matrices and spatial rotation corrections, unified spatial matching of various data types in 3D coordinates is achieved. Ultimately, the multi-source data, which has undergone time alignment, timestamp synchronization, and spatial registration, is integrated into a synchronized roadway dataset with consistent structure and spatiotemporal continuity, providing a stable data foundation for subsequent 3D feature tensor construction and roadway digital model reconstruction.

[0092] Preferably, in step S2, constructing a three-dimensional geomechanical feature space using the synchronized roadway dataset and generating a roadway feature tensor includes:

[0093] Extract stress distribution based on synchronized roadway data sets, mark rock stress maximums, and generate roadway rock stress characteristic data;

[0094] Based on the synchronized roadway data set, the roadway surface point displacement convergence analysis is performed, and the surface curvature change is measured to generate roadway point cloud feature data;

[0095] Based on the synchronized roadway data set, the support structure axis deformation analysis is performed, and the internal shear deformation of the structure is calculated to generate the internal structural characteristic data of the roadway;

[0096] The tunnel rock stress characteristic data, tunnel point cloud characteristic data and tunnel internal structure characteristic data are mapped to a three-dimensional spatial grid according to stress × displacement × strain to construct a three-dimensional geomechanical characteristic space and generate a three-dimensional tunnel characteristic tensor.

[0097] In this embodiment of the present invention, based on a synchronized roadway dataset, the surrounding rock stress distribution is first extracted with high precision. This process, based on time-series stress data at each measuring point, constructs a continuous stress distribution field in a three-dimensional coordinate system using spatial interpolation algorithms (such as the inverse distance weighted method or Kriging interpolation). This process then combines maximum principal stress direction extraction with local gradient calculation to identify stress concentration areas and weak zones. This process then completes the marking of rock stress extremes, annotating the specific locations, directions, and amplitudes of the maximum and minimum stress points within the region, generating roadway rock stress characteristic data. Simultaneously, a point displacement differential analysis method is employed for the laser scanning point cloud data of the roadway surface. By calculating the coordinate changes of the same point at different time points, the deformation of each observation point is quantified. Furthermore, methods based on Bezier surface fitting or polynomial curvature analysis are used to fit and measure the roadway surface contour changes, obtaining the distribution of surface curvature changes per unit area and outputting the roadway point cloud characteristic data. At the structural level, combining the internal strain data of the anchor and lining structures, a structural deformation path model based on the axis is constructed. The axis displacement and rotation angle changes are quantified through Fourier difference analysis or generalized least squares method to complete the axis offset (anti-axis change) analysis. The shear strain distribution on the structural cross section is further calculated based on the strain gradient to form the internal structural characteristic data of the tunnel. Finally, the above three types of characteristic data are mapped to a unified three-dimensional spatial grid system, and a tensor mapping relationship of stress, displacement, and strain ternary data is established, where each grid node contains a multi-field coupling characteristic vector expressed in tensor form. Through tensor splicing and spatial normalization processing, a three-dimensional geomechanical characteristic space is constructed, and a three-dimensional tunnel characteristic tensor with a unified structure is output, providing data-driven support for subsequent digital modeling and mechanical response analysis.

[0098] Preferably, generating the initial digital model of the tunnel based on the three-dimensional tunnel characteristic tensor in step S2 includes:

[0099] Based on the three-dimensional lane feature tensor, the spatial pattern is learned through a preset convolution layer to obtain the lane spatial feature data;

[0100] The 3D tunnel characteristic tensor is used to reconstruct the surrounding rock mass geometry to obtain a 3D tunnel structure model;

[0101] The geometric strain state attributes of the concrete lining are reconstructed from the 3D tunnel characteristic tensor to obtain a 3D support structure sub-model.

[0102] The spatial characteristic data of the tunnel are used to simulate the contact surface force transmission mechanism of the three-dimensional tunnel structure model and the three-dimensional support structure model to obtain the initial digital model of the tunnel.

[0103] In an embodiment of the present invention, after the three-dimensional tunnel feature tensor is constructed, it needs to be structurally analyzed to support the construction of a digital model. First, a multi-layer three-dimensional convolutional neural network (3D-CNN) is set up to learn spatial features of the tensor data. The convolution operation takes the spatial local region of the feature tensor as input and extracts the spatial local stress, displacement, and strain coupling patterns on multiple channels through a kernel function. After ReLU activation and maximum pooling operations, high-dimensional tunnel spatial feature data is formed to characterize the spatial distribution of structural response. Subsequently, based on the data channels related to the surrounding rock mass in the three-dimensional feature tensor, a voxel-based geometric reconstruction method is used to reconstruct the surrounding rock mass in three dimensions. Specifically, isosurface extraction (such as the Marching Cubes algorithm) or implicit function-based reconstruction technology is used to extract the surrounding rock mass boundary from the dense tensor field to generate a three-dimensional tunnel structure volume model that fully expresses its geometric morphology and material property distribution. For the concrete lining structure, regions containing high-gradient strain data in the tensor are selected. Deformation gradient tensor decomposition and material constitutive inversion are used to reconstruct the geometric strain state properties of the lining structure under different loading conditions. A three-dimensional support structure submodel is generated by combining engineering primitives (such as ellipsoid fitting and tetrahedron meshing). After obtaining the structural and support models, the contact state information and mechanical property parameters contained in the spatial feature data are used to model the contact force transmission mechanism between the two. This modeling utilizes a surface coupling method based on finite element contact analysis, setting contact stiffness, friction coefficient, and slip boundary conditions. The normal and tangential forces at the contact nodes between the different structures are numerically solved and dynamically balanced to construct a complete contact force field distribution. Ultimately, the structural model, support structure model, and contact surface mechanical relationships are integrated to form an initial digital model of the roadway. This model, which includes complete geometric configuration, material response characteristics, and mechanical interaction information, is used in subsequent mechanical simulations and repair strategy simulations in the form of volume data or graph structures.

[0104] Preferably, step S3 includes the following steps:

[0105] Step S31: simulating the repair and support scheme using the initial digital model of the tunnel to obtain tunnel design repair and support simulation data;

[0106] Step S32: generating physical abrasive tool parameters based on the tunnel design repair and support simulation data, and performing a physical model test to obtain physical test data;

[0107] Step S33: Perform digital-physical error analysis using the physical test data and the tunnel design repair and support simulation data to obtain model compensation parameters; perform parameter correction on the initial tunnel digital model based on the model compensation parameters to obtain an optimized tunnel digital model.

[0108] In an embodiment of the present invention, after the initial digital model of the tunnel is constructed, it is first necessary to carry out a virtual simulation of the repair support scheme based on the model. Specifically, by setting multiple sets of support scheme parameter sets, covering multiple design variables such as anchor type and layout, arch stiffness, and shotcrete thickness, and using the three-dimensional stress-strain-displacement tensor field in the digital model as the input boundary condition, the multi-physics field coupled finite element simulation method is used to model and calculate the mechanical response process under each support scheme, and output tunnel design repair support simulation data including surrounding rock stress redistribution, support structure stress state change and plastic zone evolution. Subsequently, according to the key structural parameters and boundary loading conditions extracted from the simulation data, the mold parameters of the scaled model for physical testing are automatically generated using a CAD / CAE integrated modeling method, including the geometric surface of the support structure, the material hierarchical structure and the loading path, etc., to control the dimensional accuracy and stress equivalence principle, complete the corresponding scaled structure loading experiment on the physical test platform, and collect the obtained stress, displacement and crack extension data to form the physical test data. After acquiring these two types of data, digital-physical error analysis was performed by establishing spatial grid node correspondences and matching functions for similar physical quantities. Root mean square error (RMSE), structural similarity (SSIM), and maximum error point difference were used to quantitatively evaluate the differences in the responses of the two models at key monitoring points and along key loading paths. The error analysis results were extracted by constructing a residual field map to extract the spatial distribution characteristics of model deviations. The backpropagation optimization algorithm was then used to calculate the compensation parameter set required for the digital model, including material stiffness correction coefficients, boundary condition adjustment factors, and local mesh refinement compensation parameters. Ultimately, the compensation parameters were fed back to the initial digital model of the tunnel for global or local parameter updates, recalibrating its mechanical response characteristics and obtaining an optimized digital model of the tunnel that more closely matches the actual engineering response.

[0109] It is particularly important that step S31 includes the following steps:

[0110] Step S311: using the initial digital model of the tunnel to load boundary conditions and obtain an equivalent mechanical environment;

[0111] Step S312: Parameterize and define using an equivalent mechanical environment to generate a support simulation scheme space;

[0112] Step S313: Perform multi-physics coupling simulation based on the support simulation scheme space, and perform simulation safety criterion evaluation to obtain tunnel design repair support simulation data.

[0113] In an embodiment of the present invention, boundary condition loading processing is performed based on the constructed initial digital model of the tunnel. Specifically, it is necessary to determine the key boundary inputs according to the tunnel geological environment, surrounding rock type, historical stress evolution data and field monitoring data, including top vertical load, lateral ground pressure, earthquake motion simulation disturbance and structure-surrounding rock contact friction characteristics. By embedding these mechanical boundary conditions into the initial tunnel model in the form of tensor boundary constraints, an equivalent mechanical environment with engineering representativeness is constructed. After obtaining the equivalent environment, the parameter space definition method is further used to construct a support parameter vector space around dimensions such as anchor length, spacing, angle, anchor strength, lining thickness and material modulus. Latin hypercube sampling, Sobol sequence or uniform design method are used to generate a high-dimensional support simulation parameter set to form a support simulation scheme space. Subsequently, a multi-physics field simulation platform, such as FLAC3D, ABAQUS or UDEC, is used to perform coupled simulation calculations on each set of parameter combinations. The simulation process covers physical behaviors such as nonlinear deformation of surrounding rock, anchor-lining interface contact, material fracture and yield criterion. The simulation results generate multi-dimensional output data, including vertical displacement of the roof, horizontal convergence of the sidewalls, anchor stress distribution, and lining structure strain contours. Based on these outputs, each set of simulation outputs is quantitatively evaluated using pre-defined simulation safety criteria, such as the maximum displacement limit of the roof, the ultimate strain of the lining, or the yield rate of the anchors. Results that do not meet engineering safety requirements are eliminated, while those that meet the requirements are retained as "tunnel design repair and support simulation data." This dataset provides physically consistent, highly reliable input for subsequent physical experimental modeling and error compensation.

[0114] Preferably, step S32 includes the following steps:

[0115] Step S321: performing scaling model size processing based on the tunnel design repair support simulation data to obtain tunnel scaling model parameters, wherein the geometric similarity ratio of the scaling model size processing is 1:50 and the stress similarity ratio is 1:75;

[0116] Step S322: extract anchor position points from the initial digital model of the tunnel to obtain model anchor position data; reserve anchor points based on the model anchor position data to obtain hole density parameter data;

[0117] Step S323: Perform an equivalent boundary load physical model experiment using the tunnel scaling model parameters and the hole density parameter data to obtain physical test data.

[0118] In this embodiment of the present invention, based on the tunnel design and repair support simulation data, the digital model was first scaled to meet the scale requirements of the physical test. Specifically, a geometric similarity ratio of 1:50 and a stress similarity ratio of 1:75 were used. Based on similarity theory, the three-dimensional geometric dimensions and mechanical loads were proportionally adjusted to ensure the equivalence of the scaled models in terms of spatial structure and mechanical behavior. The geometric similarity ratio conversion linearly scaled the three-dimensional mesh node coordinates of the tunnel digital model to adjust the model dimensions to meet the physical constraints of the experimental platform. Simultaneously, the stress similarity ratio was processed based on the principle of similarity in material mechanics, adjusting the amplitude and distribution of the loading boundary conditions to maintain the similarity of the stress field and ensure effective representativeness of the test data. Subsequently, for anchor bolt configuration, the spatial layout information of the anchor bolts in the initial tunnel digital model was utilized. Point cloud extraction and spatial indexing algorithms were used to accurately identify anchor bolt locations, generating an anchor bolt location dataset. Based on this dataset, hole reservation calculations were performed. The spatial spacing and distribution patterns between anchor bolts were calculated using hole density analysis, generating hole density parameter data to ensure that the anchor bolt layout in the physical model accurately reflects the density and distribution characteristics of the actual support arrangement. Finally, the scaled geometric model parameters and hole density parameters are input into the physical experimental device. Based on the experimental scheme of equivalent boundary load, a multi-point loading system is used to impose representative mechanical boundary conditions on the model. The stress, deformation and failure data of the model during the loading process are collected in real time to form a complete physical test data set to support subsequent digital-physical error analysis and model optimization.

[0119] Preferably, step S33 includes the following steps:

[0120] Step S331: using the physical test data and the tunnel design repair and support simulation data to calculate the displacement relative error, and obtain the tunnel displacement relative error data; using the physical test data and the tunnel design repair and support simulation data to calculate the strain relative error, and obtain the tunnel strain relative error data;

[0121] Step S332: When the relative error data of the roadway displacement is greater than 10% and the relative error data of the roadway strain is less than 0.85, a genetic algorithm is used to optimize the parameters of the surrounding rock constitutive model to obtain model compensation parameters;

[0122] Step S333: modifying the parameters of the initial digital model of the tunnel based on the model compensation parameters to obtain an optimized digital model of the tunnel.

[0123] In this embodiment of the present invention, this step first extracts corresponding displacement and strain measurements based on physical test data and tunnel design repair and support simulation data. The relative displacement and strain errors are then calculated through comparative analysis. Specifically, the relative displacement error is calculated by the ratio of the absolute value of the point-by-point displacement difference to the physical test displacement value, forming an error distribution dataset encompassing all monitoring points. The relative strain error is calculated by the ratio of the simulated and experimental data at the corresponding strain sensing points, yielding a statistical index of the overall strain error, which is further expressed as an error matrix reflecting its spatial distribution characteristics. Subsequently, under the conditions that the relative displacement error exceeds 10% and the relative strain error index is less than 0.85, a parameter optimization method based on a genetic algorithm is used to modify the parameters of the surrounding rock constitutive model. The genetic algorithm uses the constitutive model parameters as genetic encoding, sets an initial population, and uses an error function as a fitness evaluation metric. Through selection, crossover, and mutation operations, iterative optimization is performed to reduce the error between the simulated and physical data, ultimately obtaining an optimal set of compensation parameters, including the elastic modulus, Poisson's ratio, and the plastic yield criterion correction factor. Finally, the compensation parameters are fed back to the initial digital model of the tunnel, and the parameters describing its mechanical behavior are corrected and updated, and the material constitutive relationship and boundary conditions are adjusted, thereby improving the fitting accuracy of the digital model to the actual engineering behavior. This forms an optimized digital model of the tunnel that has undergone error correction and parameter optimization, and has more accurate mechanical response prediction capabilities, supporting subsequent structural safety assessment and design optimization work.

[0124] As an example of the present invention, refer to Figure 2 As shown, in this example, step S4 includes:

[0125] Preferably, step S41 includes:

[0126] Based on the preset judgment criteria, the repair effect prediction of the tunnel optimization digital model is carried out to generate support optimization prediction data, where the repair effect prediction includes design variable prediction and constraint condition prediction;

[0127] When predicting design variables, the preset judgment criteria include anchor bolt horizontal spacing of 0.5-1.5m and lining thickness of 0.1-0.3m. When the support optimization prediction data is not, the remaining data will be marked as scrapped data;

[0128] When predicting the constraint conditions, the preset judgment criteria include top plate displacement less than 50mm, side displacement less than 30mm, and when the support optimization prediction data is the rest, the data will be marked as scrapped.

[0129] In an embodiment of the present invention, based on the digital model of tunnel optimization, the repair effect is systematically predicted by pre-set judgment criteria to generate support optimization prediction data. In the design variable prediction stage, by limiting the numerical range of key design parameters in the model, such as the horizontal spacing of anchor rods and the lining thickness, the horizontal spacing of anchor rods is specifically set between 0.5 meters and 1.5 meters, and the lining thickness is set between 0.1 meters and 0.3 meters, and the interval constraint method is used to filter the simulation output data. When the support optimization prediction data obtained by simulation exceeds the above-mentioned design variable range, the system automatically marks the data as scrapped data to eliminate abnormal results that do not meet the design specifications. In the constraint prediction stage, based on the tunnel structure displacement control requirements, the maximum allowable displacement of the top plate is set to 50 mm, and the maximum allowable displacement of the side is set to 30 mm. Threshold comparison and judgment are performed based on the structural displacement time series data. When the predicted displacement data exceeds the corresponding threshold, it is determined to violate the constraint and is also marked as scrapped data. During data processing, a rule-based judgment algorithm is used to verify the compliance of the predicted data. A logical screening mechanism is then used to categorize and manage abnormal and compliant data, forming a structured support optimization prediction dataset. This process ensures accurate filtering and quality control of the digital model output, supports subsequent full-process modeling and evaluation, and provides an effective data foundation that meets engineering standards for report generation.

[0130] In this specification, a modeling system for a lane model is provided, which is used to execute the above-mentioned modeling method for a lane model. The modeling system for a lane model includes:

[0131] The multi-source data acquisition module is used to deploy a multi-source sensor array on-site in the tunnel project to collect surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data, and perform data preprocessing to construct a multi-source original data set for the tunnel;

[0132] The spatiotemporal fusion and 3D modeling module is used to perform spatiotemporal fusion processing on the multi-source original data sets of the roadway to generate a synchronized roadway data set; construct a 3D geomechanical feature space using the synchronized roadway data set and generate a 3D roadway feature tensor; and generate an initial digital model of the roadway based on the 3D roadway feature tensor;

[0133] The physical feedback and model correction module is used to simulate the repair and support scheme using the initial digital model of the tunnel, conduct physical model tests, and obtain physical test data; use the physical test data to perform digital-physical error analysis to obtain model compensation parameters; and based on the model compensation parameters, perform parameter correction on the initial digital model of the tunnel to obtain the optimized digital model of the tunnel;

[0134] The modeling evaluation and report generation module is used to predict the repair effect based on the tunnel optimization digital model and generate support optimization prediction data; perform full-process modeling evaluation based on the support optimization prediction data, and generate a report to obtain a full-process model construction report for the tunnel model.

[0135] The beneficial effect of the present invention is that, by deploying a multi-source sensor array at the tunnel engineering site, surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data are collected in real time, forming a multi-dimensional, multi-spatiotemporal resolution data foundation, providing detailed and comprehensive raw data support for subsequent data fusion and model construction. The multi-source raw data sets are synchronized using a spatiotemporal fusion algorithm to eliminate spatiotemporal asynchrony and dimensional differences between the data, ensuring that the spatial positions and timestamps of different types of data are highly consistent, thereby constructing an accurate synchronized tunnel data set. On this basis, combined with the principles of geomechanics, the fused data is mapped to a three-dimensional feature space, and a high-dimensional characterization of the multi-physical quantity characteristics of the complex tunnel structure is achieved through tensor expression, effectively capturing the coupling relationship between the surrounding rock and the support structure, and then generating an initial digital model that accurately reflects the mechanical state of the tunnel. By combining digital models with physical model tests, the digital and physical test data are compared using error analysis methods, and model compensation parameters are extracted to correct the systematic deviations in the digital model, thereby achieving optimization and precision of the digital model. Based on the optimized digital model, the effectiveness of the repair support scheme is predicted. Relying on structured support optimization prediction data, a full-process modeling evaluation is conducted. The feasibility and safety of the repair scheme are systematically quantified, and a detailed evaluation report is generated, forming a closed-loop data-driven design and verification process. This process not only enables dynamic and detailed monitoring of the tunnel structure status, but also improves the accuracy and reliability of the model through data-driven methods, promoting scientific and intelligent support design.

[0136] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0137] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A modeling method applied to a roadway model, characterized in that: The following steps are involved: Step S1: deploying a multi-source sensing device array at the tunnel engineering site to collect surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data, and performing data preprocessing to construct a multi-source original data set for the tunnel; Step S2: performing spatiotemporal fusion processing on the multi-source original data sets of the roadway to generate a synchronized roadway data set; constructing a three-dimensional geomechanical feature space using the synchronized roadway data set, and generating a three-dimensional roadway feature tensor; Generate an initial digital model of the tunnel based on the 3D tunnel characteristic tensor. The synchronized tunnel dataset is used to construct a 3D geomechanical characteristic space and generate the tunnel characteristic tensor, including: Extract stress distribution based on synchronized roadway data sets, mark rock stress maximums, and generate roadway rock stress characteristic data; Based on the synchronized roadway data set, the roadway surface point displacement convergence analysis is performed, and the surface curvature change is measured to generate roadway point cloud feature data; Based on the synchronized roadway data set, the support structure axis deformation analysis is performed, and the internal shear deformation of the structure is calculated to generate the internal structural characteristic data of the roadway; The roadway rock stress characteristic data, roadway point cloud characteristic data and roadway internal structure characteristic data are mapped to a three-dimensional spatial grid according to stress × displacement × strain to construct a three-dimensional geomechanical characteristic space and generate a three-dimensional roadway characteristic tensor; Among them, generating the initial digital model of the tunnel based on the three-dimensional tunnel characteristic tensor includes: Based on the three-dimensional lane feature tensor, the spatial pattern is learned through a preset convolution layer to obtain the lane spatial feature data; The 3D tunnel characteristic tensor is used to reconstruct the surrounding rock mass geometry to obtain a 3D tunnel structure model; The geometric strain state attributes of the concrete lining are reconstructed from the 3D tunnel characteristic tensor to obtain a 3D support structure sub-model. Using the tunnel spatial characteristic data, the contact surface force transmission mechanism of the 3D tunnel structure sub-model and the 3D support structure sub-model is simulated to obtain the initial digital model of the tunnel; Step S3: Using the initial digital model of the tunnel to simulate the repair support scheme, and conducting physical model tests to obtain physical test data; using the physical test data to perform digital-physical error analysis to obtain model compensation parameters; based on the model compensation parameters, modifying the parameters of the initial digital model of the tunnel to obtain an optimized digital model of the tunnel; Step S4: Predict the repair effect based on the tunnel optimization digital model and generate support optimization prediction data; perform full-process modeling evaluation based on the support optimization prediction data, and generate a report to obtain a full-process model construction report for the tunnel model.

2. The modeling method for a laneway model according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: deploying fiber Bragg grating sensor arrays on the roadway roof and both sides with a sampling frequency of 10 Hz, a measurement range of 0-50 MPa, and an accuracy of ±0.1 MPa to collect surrounding rock deformation data; Step S12: Arrange a laser scanner on the tunnel surface with a scanning accuracy of 0.1 mm and collect deformation point cloud data of the tunnel surface at intervals of 30 minutes; Step S13: embedding strain sensors inside the anchor rods and lining structure with a sampling frequency of 5 Hz to collect support structure status data; Step S14: using wavelet transform to perform noise reduction processing on the surrounding rock deformation data, the roadway surface deformation point cloud data, and the support structure status data to obtain a multi-source roadway noise reduction data set; Step S15: performing data cleaning on the multi-source laneway denoising dataset to generate a laneway multi-source original dataset.

3. The modeling method for a laneway model according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: simulating the repair and support scheme using the initial digital model of the tunnel to obtain tunnel design repair and support simulation data; Step S32: generating physical mold parameters based on the tunnel design repair and support simulation data, and performing a physical model test to obtain physical test data; Step S33: Perform digital-physical error analysis using the physical test data and the tunnel design repair and support simulation data to obtain model compensation parameters; perform parameter correction on the initial tunnel digital model based on the model compensation parameters to obtain an optimized tunnel digital model.

4. The modeling method for a tunnel model according to claim 3, characterized in that: Step S32 includes the following steps: Step S321: performing scaling model size processing based on the tunnel design repair support simulation data to obtain tunnel scaling model parameters, wherein the geometric similarity ratio of the scaling model size processing is 1:50 and the stress similarity ratio is 1:75; Step S322: extract anchor position points from the initial digital model of the tunnel to obtain model anchor position data; reserve anchor points based on the model anchor position data to obtain hole density parameter data; Step S323: Perform an equivalent boundary load physical model experiment using the tunnel scaling model parameters and the hole density parameter data to obtain physical test data.

5. The modeling method for a laneway model according to claim 3, characterized in that: Step S33 includes the following steps: Step S331: Calculate the displacement relative error using the physical test data and the tunnel design repair and support simulation data to obtain tunnel displacement relative error data; calculate the strain relative error using the physical test data and the tunnel design repair and support simulation data to obtain tunnel strain relative error data; Step S332: When the relative error data of the roadway displacement is greater than 10% and the relative error data of the roadway strain is less than 0.85, a genetic algorithm is used to optimize the parameters of the surrounding rock constitutive model to obtain model compensation parameters; Step S333: modifying the parameters of the initial digital model of the tunnel based on the model compensation parameters to obtain an optimized digital model of the tunnel.

6. The modeling method for a laneway model according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: Predicting the repair effect based on the tunnel optimization digital model and generating support optimization prediction data; Step S42: Perform full-process modeling evaluation based on the support optimization prediction data, and update the mold design parameters to obtain full-process roadway modeling data; Step S43: constructing a full-process model construction report of the lane model based on the full-process lane model modeling data.

7. The modeling method for a laneway model according to claim 6, characterized in that: Step S41 includes: Based on the preset judgment criteria, the repair effect prediction of the tunnel optimization digital model is carried out to generate support optimization prediction data, where the repair effect prediction includes design variable prediction and constraint condition prediction; When predicting design variables, the preset judgment criteria include anchor bolt horizontal spacing of 0.5-1.5m and lining thickness of 0.1-0.3m. When the support optimization prediction data is not, the remaining data will be marked as scrapped data; When predicting the constraint conditions, the preset judgment criteria include top plate displacement less than 50mm, side displacement less than 30mm, and when the support optimization prediction data is the rest, the data will be marked as scrapped.

8. A modeling system applied to a roadway model, characterized in that: For executing the modeling method applied to a laneway model according to claim 1, the modeling system applied to the laneway model comprises: The multi-source data acquisition module is used to deploy a multi-source sensor array on-site in the tunnel project to collect surrounding rock deformation data, tunnel surface deformation point cloud data, and support structure status data, and perform data preprocessing to construct a multi-source original data set for the tunnel; The spatiotemporal fusion and 3D modeling module is used to perform spatiotemporal fusion processing on the multi-source original data sets of the roadway to generate a synchronized roadway data set; construct a 3D geomechanical feature space using the synchronized roadway data set and generate a 3D roadway feature tensor; and generate an initial digital model of the roadway based on the 3D roadway feature tensor; The physical feedback and model correction module is used to simulate the repair and support scheme using the initial digital model of the tunnel, conduct physical model tests, and obtain physical test data; use the physical test data to perform digital-physical error analysis to obtain model compensation parameters; and based on the model compensation parameters, perform parameter correction on the initial digital model of the tunnel to obtain the optimized digital model of the tunnel; The modeling evaluation and report generation module is used to predict the repair effect based on the tunnel optimization digital model and generate support optimization prediction data; perform full-process modeling evaluation based on the support optimization prediction data, and generate a report to obtain a full-process model construction report for the tunnel model.

Images