Road geological disaster risk early warning method and system based on BIM and numerical simulation

Abstract

The invention relates to a highway geological disaster risk early warning method and system based on BIM and numerical simulation, and the method mainly combines an unmanned plane approaching photogrammetry technology with a BIM technology, and builds a three-dimensional geological model of a highway and a geological environment along the highway. ABAQUS software and PFC discrete elements are used for simulating different working conditions to carry out high-speed analysis on the geological disaster damage range under external conditions. Unmanned aerial vehicle surveying and mapping, a BIM technology, numerical simulation (ABAQUS and PFC) and a GIS platform (ArcGIS) are integrated, and accurate and dynamic risk early warning of the in-service expressway is achieved. According to the method, multi-level geological disaster risk early warning from a static structure to a dynamic environment can be achieved, the damage range and the movement process of geological disasters are analyzed, the mutual relation between the influence range of the geological disasters and the expressway is displayed in the real scene model, and the method has remarkable innovation advantages especially in the aspects of side slope landslide, roadbed settlement and the like.

Description

Technical Field

[0001] The present invention relates to the fields of engineering geology and transportation engineering research, and particularly relates to a highway geological disaster risk early warning method and system based on BIM and numerical simulation. This method combines unmanned aerial vehicle (UAV) close-range photogrammetry and then uses BIM and numerical simulation technologies to conduct risk early warning on geological disasters of in-service expressways. Background Art

[0002] With the rapid development of transportation infrastructure, expressways have become an important part of the national economy. However, expressways that have been in a complex natural environment for a long time are easily affected by various factors such as geological disasters, climate change, and traffic flow, resulting in certain risks to their safety and stability. Therefore, real-time and accurate risk monitoring and early warning of in-service expressways have become an important means to ensure their safe operation.

[0003] In recent years, photogrammetry technology based on unmanned aerial vehicles has made remarkable progress in the field of geographic information collection. By carrying high-precision cameras and global positioning systems and other devices, UAVs can quickly obtain high-quality image data and generate three-dimensional models of the target area through three-dimensional modeling technology. On the other hand, Building Information Modeling (BIM) technology, as an advanced engineering design and management tool, has been widely used in highway engineering design and road geological exploration.

[0004] Regarding the surveying and mapping modeling of in-service expressways, there are still limitations in the current UAV oblique photography and BIM technologies. For expressways that are often in complex terrains such as mountains and hills, most studies only use the method of UAV oblique photography for modeling. However, due to large terrain undulations or dense vegetation, the accuracy of UAV oblique photography will be poor, and due to flight altitude limitations, it is impossible to comprehensively cover the entire survey area. For example, in "Research on the Highway Slope Data Acquisition Mode Based on Unmanned Aerial Vehicle System", when Yang Yanwei used UAVs to collect highway slope data, due to weak signals, the obtained data was blurred, and the accuracy of subsequent data restoration was not completely accurate. In addition, the existing technology has not been used for risk early warning of the geology of expressways and surrounding facilities, but only for the modeling of expressways themselves.

[0005] In the face of complex terrain areas, traditional methods have limitations such as long data collection cycles, low accuracy, and poor real-time performance, making it difficult to achieve large-scale and dynamic real-time monitoring, and being easily affected by external factors such as climate and terrain. Existing methods often rely on fixed monitoring points and cannot flexibly adapt to terrain changes and emergencies. Therefore, when an expressway is in a high-risk area of geological disasters, the early warning effect of existing means is difficult to guarantee. Summary of the Invention

[0006] Aiming at the deficiencies of the existing technologies, the technical problem to be solved by the present invention is to provide a highway geological disaster risk early warning method and system based on BIM and numerical simulation. The aim is to solve the problems of difficult and inaccurate acquisition of highway terrain data, mismatch between design and actual terrain environment, and untimely risk early warning in the current expressways. It is of great significance for enhancing the early warning accuracy and reliability of expressway risk early warning, and for calculating the stability and movement range of geological bodies around the expressway under rainfall conditions using numerical simulation technology subsequently.

[0007] The technical solution adopted by the present invention to solve the above technical problem is as follows:

[0008] In the first aspect, the present invention provides a highway geological disaster risk early warning method based on BIM and numerical simulation. The steps of the early warning method are as follows:

[0009] Step 1: Obtain high-precision original point cloud data and image data of the area along and around the target highway through close-range photogrammetry by an unmanned aerial vehicle (UAV).

[0010] Step 2: Import the original point cloud data collected by the UAV into Context Capture software. Denoise, optimize, and adjust the accuracy of the original point cloud data in ContextCapture, remove invalid points and duplicate points, splice and integrate the data of different flight batches to generate a complete three-dimensional point cloud model, and adjust the spatial coordinate system according to project requirements to ensure its consistency with the coordinates of the subsequent established BIM model, thus completing the preprocessing of the point cloud data.

[0011] Step 3: Based on the data preprocessed in Step 2, generate a high-precision real-scene three-dimensional terrain of the in-service road and its surrounding area with the help of Smart 3D software, and simultaneously convert the data format of the real-scene three-dimensional terrain data.

[0012] Step 4: Import the point cloud data preprocessed in Step 2 into Revit software for three-dimensional BIM modeling operations. Use the terrain modeling tool in Revit to generate a terrain surface, accurately reflecting the actual terrain features of the expressway and its surrounding area. Based on the existing design data of the expressway infrastructure, use Revit to create an infrastructure model of the expressway. At the same time, combine the ground object features (such as trees, valleys, etc.) in the point cloud data to refine the model details, avoid inaccurate modeling caused by occlusion or errors. Through the comparative analysis of the ground object features in the point cloud data and the infrastructure model, verify and optimize the size and position of the BIM model to ensure its complete match with the actual terrain and facility environment, and thus obtain a BIM model of the in-service road and surrounding houses based on the point cloud data.

[0013] Step 5: Compare and adjust the BIM model with the high-precision real-scene three-dimensional terrain generated by Smart3D software to ensure seamless connection between the facilities in the BIM model and the terrain morphology in the real-scene three-dimensional terrain. Then, integrate the BIM model with the high-precision real-scene three-dimensional terrain generated by Smart3D software and optimize the geometric details of the BIM model to ensure that the BIM model not only meets the design requirements but also truly reflects the actual terrain conditions. Export the qualified BIM model in Revit to a format suitable for processing by ArcGIS software;

[0014] Step 6: Export the BIM model processed in Step 5 from Revit to STL format or OBJ format for ABAQUS numerical simulation and analysis. Among them, the STL format data is used for three-dimensional numerical simulation modeling and analysis, and the OBJ format is used for two-dimensional cross-section numerical simulation modeling and analysis. Export the DXF format from Revit for PFC numerical simulation and analysis;

[0015] For two-dimensional modeling analysis, use the OBJ format data. First, establish a complete slope geometric model through ABAQUS software for finite element analysis, set the physical properties and boundary conditions of the slope, simulate the stress field and displacement field of the slope under rainfall conditions, and identify the potentially unstable area, that is, the instability range, when the convergence condition is reached;

[0016] Import the DXF format exported from Revit into PFC to generate a two-dimensional cross-section model of the landslide. Divide the sliding mass and sliding bed of the slope according to the instability range determined by ABAQUS under rainfall conditions. Use PFC software to conduct discrete element analysis on the potentially unstable area (i.e., the sliding mass) to dynamically simulate the movement of particles during the landslide process, apply dynamic conditions (such as rainfall, earthquake, etc.), and simulate the sliding path, accumulation range, displacement of particles, and kinematic characteristics under earthquake action;

[0017] For three-dimensional modeling analysis, use the STL format data. First, establish a complete slope geometric model through ABAQUS software for finite element analysis, set the physical properties and boundary conditions of the slope, simulate the stress field and displacement field of the slope under rainfall conditions, and identify the potentially unstable area when the convergence condition is reached;

[0018] Import the DXF format exported from Revit into PFC to generate a three-dimensional model of the landslide. Divide the sliding mass and sliding bed of the slope according to the instability range determined by ABAQUS under rainfall conditions. Use PFC software to conduct discrete element analysis on the potentially unstable area (i.e., the sliding mass) to dynamically simulate the movement of particles during the landslide process, apply dynamic conditions (such as rainfall, earthquake, etc.), and simulate the sliding path, accumulation range, displacement of particles, and kinematic characteristics under earthquake action;

[0019] Step 7: Export the numerical simulation results of PFC into a format suitable for processing by ArcGIS software, and import them into the ArcGIS platform. At the same time, import the BIM model in Step 5 into the ArcGIS platform, and set the BIM model to share the same coordinate system with the landslide data imported by PFC, so that the two can be correctly displayed and three-dimensionally superimposed within the same spatial range;

[0020] Step 8: Analyze the relative relationship between the sliding path of landslide particles, the accumulation area and the facilities in the BIM model through the analysis tools of ArcGIS, intuitively display the interaction between the landslide area, the accumulation location and the road facilities, and finally form early warning analysis and maintenance suggestions to generate a visual risk early warning map.

[0021] Furthermore, the method can perform dynamic early warning. The data collected by the drone communicates with the Context Capture software, the Context Capture software communicates with the Smart 3D software and the Revit software respectively, the Revit software communicates with the ABAQUS software and the PFC software respectively, and at the same time the ABAQUS software and the PFC software communicate. The Revit software and the PFC software both communicate with the ArcGIS platform. The drone dynamically collects data, enabling the BIM model to be continuously updated according to the periodic data collection of the drone, and combining the simulation results of PFC for early warning adjustment, realizing the dynamics and accuracy maintenance of the BIM model, and ensuring that the early warning information always conforms to the current environment.

[0022] The ABAQUS software also communicates with the ArcGIS platform to display the numerical simulation results of the ABAQUS software.

[0023] In the said Step 5, the facilities in the BIM model include roads, bridges, tunnels and surrounding mountains.

[0024] The process of processing point cloud data with the Smart 3D software includes: matching homologous points using a feature-based matching algorithm, performing multi-view image joint adjustment using a bundle adjustment algorithm for regional network, and completing the preliminary construction of high-density point clouds according to the clustering algorithm and the patch-based dense matching algorithm; finally, combining the TIN network model constructed based on the point clouds with texture mapping to realize the construction of a high-precision real-scene three-dimensional model.

[0025] In the second aspect, the present invention provides a highway geological disaster risk early warning system based on BIM and numerical simulation, and the system includes:

[0026] A drone data collection module, which uses a drone to collect high-precision original point cloud data and image data of the area along and around the target highway;

[0027] A data preprocessing module for preprocessing the data collected by the UAV data acquisition module;

[0028] A BIM modeling module for performing BIM modeling based on the preprocessed data and optimizing the BIM model using the real - scene three - dimensional terrain generated by Smart 3D software to obtain a BIM model that not only meets the design requirements but also truly reflects the actual terrain conditions;

[0029] A numerical simulation module for performing finite - element analysis and discrete - element analysis based on the BIM model output by the BIM modeling module. Using finite - element analysis to determine the potentially unstable areas when reaching the convergence condition, and using discrete - element analysis to determine the sliding paths, accumulation ranges, displacement conditions of the particles in the potentially unstable areas, and the kinematic characteristics under seismic action;

[0030] A GIS data integration module for integratively displaying the output results of the BIM modeling module and the numerical simulation module, analyzing the relative relationship between the sliding paths of landslide particles, the accumulation areas and the facilities in the BIM model, and visually displaying the interaction between the landslide area, the accumulation position and the road facilities;

[0031] A risk warning and feedback module for performing risk assessment on the results output by the GIS data integration module, generating a visual risk warning map according to the finally formed warning analysis and maintenance suggestions.

[0032] Compared with the prior art, the beneficial effects of the present invention are:

[0033] 1. Aiming at the problem of insufficient risk warning for geological disasters along in - service expressways at present, the present invention mainly combines the UAV close - range photogrammetry technology and BIM technology to establish a three - dimensional geological model of the expressway and its surrounding geological environment, and uses ABAQUS software and PFC discrete element to simulate different working conditions for high - speed analysis of the geological disaster damage range under external conditions.

[0034] 2. The present invention effectively integrates UAV close - range photogrammetry, BIM technology, numerical simulation tools (such as ABAQUS and PFC) and GIS together. By using UAV close - range photogrammetry to obtain high - precision terrain data along the expressway in real time and generate a high - precision BIM model to meet the geological monitoring requirements of complex terrains, it realizes the all - round dynamic monitoring of expressway slopes. Combining the simulation analysis of ABAQUS and PFC, the present invention can realize multi - level geological disaster risk warning from static structures to dynamic environments, analyze the damage range and movement process of geological disasters, and display the mutual relationship between the geological disaster influence range and the expressway in the real - scene model, especially having significant innovative advantages in aspects such as slope landslides and subgrade settlements.

[0035] 3. The close-range photogrammetry technology of drones can quickly generate a 3D model of the highway through efficient and accurate topographic data collection. After integrating the aerial survey-based model with BIM, the geometric accuracy and information richness of the model can be further improved, ensuring a perfect fit between the design and the actual topographic environment. This combination can reduce manual modeling errors and improve efficiency and data accuracy.

[0036] 4. Combining Smart 3D software with BIM can not only accurately display topographic details but also integrate information such as the design, construction, and operation and maintenance of road engineering onto a unified visualization platform. Such an integrated display facilitates relevant personnel to monitor and analyze the safety status of the highway from multiple dimensions. The high-precision original point cloud and image data obtained by the close-range photogrammetry of drones provide accurate 3D topographic support for the BIM model. Combining with the full-life-cycle data management ability of BIM technology, the monitoring information can be real-time fed back into the BIM system, and dynamic real-time risk warning is achieved through the numerical simulations of ABAQUS and PFC. Through the overlay analysis of the BIM model, stress field, and particle movement by ArcGIS, a multi-dimensional warning model is formed, providing data support for the full-life-cycle management of the highway.

[0037] 5. In the numerical simulation stage, the topographic part of the BIM model is exported in formats suitable for ABAQUS and PFC (such as STL, DXF). ABAQUS simulates the slope stress field and displacement field under different rainfall conditions through finite element analysis, while PFC generates a particle model based on the BIM model profile for landslide particle flow analysis, dynamically simulating the landslide movement process. This combination method integrates the static geometric model of BIM with the dynamic analysis of numerical simulation, realizing the dynamic monitoring of slopes under complex geological conditions. It improves the risk prediction ability of the highway during its in-service stage. This data linkage and real-time warning help to detect potential road safety hazards early and take timely measures to ensure the operation safety of the highway.

[0038] 6. Flexibility and adaptability: The present invention still has efficient data collection capabilities in complex terrains, is applicable to the monitoring of highways in different geological environments, and provides higher adaptability for disaster warning.

[0039] 7. Cost and efficiency improvement: By quickly collecting large-scale data with drones, the labor and time costs are reduced, the monitoring efficiency is improved, and early warning for large-scale areas is achieved. Description of the Drawings

[0040] Figure 1 : Numerical simulation analysis diagram of geological disasters along the in-service highway.

[0041] Figure 2: Schematic diagram of the method of the present invention for risk warning of expressways affected by geological disasters in AcrGIS.

[0042] Figure 3 : Schematic diagram of the traditional method for warning expressways and houses along the line.

[0043] Figure 4 : Schematic diagram of the process of the in-service expressway risk warning system of the present invention. Detailed implementation manners

[0044] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Of course, the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0045] In the present invention, UAV + BIM is mainly used for modeling and extracting terrain information, and ABAQUS and PFC are used to analyze the influence range of geological disasters along the line. Through the distance and speed between geological disasters and expressways in the BIM model, the geological disaster risks of different expressway sections are warned, and the warning levels can be accurately distinguished. The prior art can only roughly judge the influence range of geological disasters, roughly give the warning level and issue warning information, and the accuracy of this application is significantly improved compared with the prior art.

[0046] The present invention realizes precise and dynamic risk warning of in-service expressways by innovatively integrating UAV surveying and mapping, BIM technology, numerical simulation (ABAQUS and PFC) and GIS platform (ArcGIS). The following is the organic combination mode of each software in this application:

[0047] ·Combination of UAV surveying and mapping and BIM model creation

[0048] The UAV collects high-precision original point cloud data and images of the expressway and its surrounding terrain to generate a basic terrain model. After preprocessing such as filtering and denoising of the original point cloud data, the processed point cloud data is generated and directly imported into BIM software (such as Revit or Navisworks) for creating three-dimensional models such as terrain, roads, and bridges. The BIM model combines road design and UAV measured information to ensure that the generated model highly matches the actual terrain, laying an accurate data foundation for subsequent numerical simulation.

[0049] ·Combination of BIM model and ABAQUS, PFC

[0050] In the numerical simulation stage, the terrain part of the BIM model is exported in formats suitable for ABAQUS and PFC (such as STL, DXF). ABAQUS simulates the slope stress field and displacement field under different rainfall conditions through finite element analysis. Based on the analysis results of ABAQUS and the BIM model data, PFC generates a particle model for landslide particle flow analysis to dynamically simulate the landslide movement process. This combination method integrates the static geometric model of BIM with the dynamic analysis of numerical simulation, realizing the dynamic monitoring of slopes under complex geological conditions.

[0051] ·Combination of Numerical Simulation and GIS Platform

[0052] After completing the numerical simulation, the analysis results of ABAQUS and PFC (such as stress, displacement, particle movement path, etc.) can be imported into the ArcGIS platform and superimposed with the BIM model for three-dimensional visualization analysis. ArcGIS compares the spatial relationship between the simulation results and road facilities, intuitively shows the interaction between the landslide area, deposition location and road facilities, and finally forms early warning analysis and maintenance suggestions. The GIS platform provides a global view to help decision-makers dynamically understand the spatial distribution of landslide risks.

[0053] ·Data Format and Compatibility Issues

[0054] The data formats of different software vary greatly, such as the LAS format of raw drone point cloud data, the IFC format of BIM software, the STL format of ABAQUS, the DXF format of PFC, etc. This application solves the compatibility problem between different formats through data format conversion. For example, by converting the raw point cloud into the OBJ or STL format, the data can be seamlessly docked in BIM modeling and finite element analysis.

[0055] ·Data Accuracy and Real-time Challenges

[0056] Traditional risk early warning models are difficult to provide high-precision and real-time risk early warnings in complex terrains. The present invention utilizes high-precision data sources of drones, real-time collects road and surrounding terrain data through close-range photogrammetry to generate a fine BIM model, and conducts dynamic analysis in combination with numerical simulation. In particular, the flow simulation of landslide particles by PFC can reflect the impact of terrain changes on highways in real time, providing a reliable basis for real-time early warning in complex terrains.

[0057] ·Integration and Display of Multidimensional Data

[0058] There are differences among multi-dimensional data such as topographic data collected by drones, BIM models, ABAQUS stress fields, and PFC landslide granular flows. These data are WS superimposed on the ArcGIS platform and uniformly displayed through three-dimensional visualization analysis. This integration method intuitively shows the spatial relationship between the landslide area and road facilities, enabling risk early warning to have the ability to integrate multi-dimensional information.

[0059] · Model dynamic update and accuracy maintenance

[0060] In practical applications, the terrain and facilities along highways may change, and traditional static models are difficult to reflect these dynamic information. The present invention enables the BIM model to be continuously updated according to the periodic data collection of drones and adjusted in combination with the simulation results of ABAQUS and PFC, realizing the dynamic nature and accuracy maintenance of the model, and ensuring that the early warning information always conforms to the current environment.

[0061] · The present invention realizes precise and real-time monitoring of highway slopes in service through the multi-software combination of drones, BIM, numerical simulation, and GIS. Compared with traditional methods, this patent has a highly automated data collection and processing ability, significantly improving the monitoring accuracy and real-time performance.

[0062] · Multi-dimensional information integration is achieved in complex terrains, upgrading the risk early warning of highways from static monitoring to dynamic early warning, fully considering the real-time changes of geological disasters such as landslides and settlements, and providing more scientific decision-making support for road management.

[0063] The highway geological disaster risk early warning system based on BIM and numerical simulation of the present invention, the system includes:

[0064] UAV data collection module, which uses UAVs to collect high-precision original point cloud data and image data of the areas along and around the target highway.

[0065] Data preprocessing module, which is used to preprocess the data collected by the UAV data collection module.

[0066] BIM modeling module, which is used to perform BIM modeling according to the preprocessed data and optimize the BIM model using the real three-dimensional terrain generated by Smart 3D software to obtain a BIM model that not only meets the design requirements but also truly reflects the actual terrain conditions.

[0067] Numerical simulation module, which is used to perform finite element analysis and discrete element analysis according to the BIM model output by the BIM modeling module, determine the potentially unstable areas when the convergence condition is reached using finite element analysis, and determine the sliding paths, accumulation ranges, displacement conditions of the particles in the potentially unstable areas, and the kinematic characteristics under seismic action using discrete element analysis.

[0068] The GIS data integration module is used to integratively display the output results of the BIM modeling module and the numerical simulation module, analyze the relative relationships between the sliding paths and accumulation areas of landslide particles and the facilities in the BIM model, and visually display the interaction between the landslide area, the accumulation location and the road facilities;

[0069] The risk warning and feedback module is used to conduct risk assessment on the results output by the GIS data integration module, generate a visual risk warning map according to the finally formed warning analysis and maintenance suggestions.

[0070] The in-service highway risk warning system of this application is a complete dynamic monitoring and warning system, aiming to realize the whole-process management from data collection to risk assessment. Each module is connected and closely combined with each other, forming a multi-dimensional and multi-functional dynamic risk assessment system.

[0071] The UAV data collection module uses a UAV equipped with LiDAR and a high-resolution camera to efficiently survey the terrain along and around the highway. This module ensures data accuracy by setting up image control points and GPS devices, and collects the original point cloud data (LAS format), high-resolution image data (GeoTIFF format) and digital elevation model (DEM). The automated flight path planning and fast coverage capabilities of the UAV provide a convenient and efficient solution for data collection under complex terrains.

[0072] The BIM modeling module imports the original point cloud data collected by the UAV into BIM software (such as Revit) to generate a three-dimensional terrain model that matches the actual terrain height. By integrating design data and measured terrain data, this module constructs a fine three-dimensional model including subgrades, bridges, mountains and drainage systems, providing an accurate geometric basis for subsequent numerical simulation analysis. The three-dimensional visualization ability of the BIM model can visually reflect the structural relationship between road facilities and the surrounding terrain.

[0073] In the numerical simulation module, ABAQUS and PFC respectively undertake the tasks of static and dynamic risk analysis. ABAQUS simulates the stress field and displacement field of the slope through finite element analysis to predict the potential landslide instability area under working conditions such as rainfall; PFC dynamically simulates the flow path and accumulation range of landslide particles based on the cross-section data of the BIM model. By using the two tools jointly, the system can accurately quantify the impact of landslides on road infrastructure, providing strong technical support for risk warning under complex geological conditions.

[0074] The GIS data integration module imports the above data into the ArcGIS platform, combines the BIM model with the numerical simulation results, and conducts the integration and analysis of multi-dimensional data. The GIS platform uses tools such as 3DAnalyst to superimpose the BIM model, stress distribution, and particle movement trajectories in the three-dimensional space, visually displaying the specific impact area of the landslide on highway facilities. The system also generates a landslide area distribution map and risk level identification, helping managers quickly grasp the spatial relationship of the landslide area and improving the decision-making efficiency.

[0075] The risk early warning and feedback module combines the numerical simulation results and real-time monitoring devices to conduct intelligent early warning. Through the real-time data collection of displacement monitoring sensors and rain gauges, the ABAQUS software obtains this real-time data, which can update the output results of the numerical simulation module, and then be displayed in real time in the GIS platform and update the risk early warning map. The system can quickly detect high-risk areas and trigger the alarm mechanism. At the same time, this module generates maintenance suggestions according to the simulation results to guide the implementation of disaster prevention measures. The early warning information has strong timeliness and can effectively reduce the threat of highway disasters to operation.

[0076] The system of the present invention organically combines UAV surveying and mapping, BIM modeling, numerical simulation, and GIS visual analysis to form a full-process closed loop from data collection to risk assessment. The high-precision data collection, efficient dynamic simulation, and multi-dimensional data integration capabilities are applicable to different scenarios such as highways, bridges, and tunnels, and can provide accurate and efficient disaster risk early warning solutions for transportation infrastructure, thereby improving operation safety and management science.

[0077] Embodiment 1

[0078] The method for highway geological disaster risk early warning based on BIM and numerical simulation of the present invention specifically comprises the following steps:

[0079] S1. Collect topographic information by UAV close-range photogrammetry technology

[0080] First, use UAV close-range photogrammetry technology to collect topographic information of the highway and its surrounding areas. The UAV is equipped with devices such as a high-precision camera, lidar (LiDAR), and global positioning system (GPS), and continuously conducts aerial photography of the target area according to the preset flight path to obtain high-precision original point cloud data and image data. The collected data includes the topographic and geomorphic features around the road, and the accurate spatial positions of infrastructure such as bridges and tunnels, generating a three-dimensional original point cloud model containing high-precision textures and real geographical locations.

[0081] Specific operation steps: (1) The quality of aerial flight is mainly affected by flight lines, relative flight altitude, and overlap. To obtain high-quality aerial images, these parameters need to be set during the aerial flight plan. In this embodiment, a DJI drone Mavic 3E is used for measurement, and the required number of groups of images and corresponding Pos point data, as well as the width and height of the images, are measured.

[0082] (2) Acquisition of image control point data: Determine the number of control points according to the actual situation of the measured highway, and conduct field measurements on the coordinates and elevation values of the control points. Based on the flight line planning and image data, the regional network method is mainly used for the point layout scheme. According to the layout principle of control points, select appropriate control points within the measurement area as image control points to participate in the solution.

[0083] (3) Accuracy requirements for image control points: Relative to adjacent basic image control points, the mapping error of planar image control points and planimetric and height image control points does not exceed 0.2 mm. The elevation error of elevation image control points and planimetric and height image control points does not exceed 0.1 m. The position of the image control points is marked on the image with a "crosshair" of appropriate size and described in detail in words. After shooting, import the drone photos with Pos data into Pix4D software for stitching, establish a dense original point cloud, aerial triangulation, generate a mesh and texture, and finally obtain a digital orthophoto map (DOM), a digital surface model (DSM), and a three-dimensional real scene model.

[0084] S2. Preprocess the original point cloud data using Context Capture

[0085] In S1, the original point cloud data collected by the drone is imported into the Context Capture software, ensuring that the data format is LAS, LAZ, or PLY. These original point cloud data contain high-precision information about the highway and its surrounding terrain and are the basis for subsequent data processing. Denoise, optimize, and adjust the accuracy of the original point cloud data in Context Capture, remove invalid points and duplicate points, and ensure the clarity and accuracy of the data. If the drone collects data in multiple flights, the data of different flight batches can be stitched and integrated through the software to generate a complete three-dimensional point cloud model, and the spatial coordinate system can be adjusted according to the project requirements to ensure its consistency with the coordinates of the subsequent established BIM model.

[0086] S3. Generate a three-dimensional terrain model map based on the preprocessed point cloud data using Smart 3D software

[0087] To ensure the accuracy and precision of the 3D real - scene model, the pre - processed point cloud data obtained in S2 is used to generate a high - precision real - scene 3D terrain of the in - service road location with the help of Smart3D software. At the same time, the data format of the real - scene 3D terrain is converted. On this basis, the image control points before aerial survey are compared with the corresponding position points after modeling to complete the accuracy analysis.

[0088] Based on the computer vision 3D reconstruction algorithm, Smart3D software can calculate the accurate exterior orientation elements only by combining the Pos data obtained during the flight process. And it uses the feature - based matching algorithm to match homologous points, adopts the bundle adjustment method for multi - view image joint adjustment, and completes the preliminary construction of the point cloud after high - density processing according to the clustering algorithm and the patch - based dense matching algorithm. Finally, with the TIN network model constructed based on the processed point cloud and texture mapping, the construction of a high - precision real - scene 3D terrain is realized (as Figure 2 shown).

[0089] The data formats generated by Smart3D software include various types such as s3c, fbx, osgb, dae, obj, stl, etc. Among them: the osgb, obj, and fbx formats can be applied to a variety of modeling software for easy later modification; they are also applicable to a variety of display platforms and can be quickly published on the network for sharing. The present invention selects the obj format to combine with BIM technology.

[0090] S4. Combine the pre - processed point cloud data with BIM software to generate road and building models

[0091] (1) Import the pre - processed point cloud data in S2 into Revit software for 3D BIM modeling operations. First, create a terrain surface model through the elevation information in the point cloud to accurately reflect the actual terrain features of the highway and its surrounding areas. On this basis, create an infrastructure model of the highway according to the design scheme, including structures such as roads, bridges, and tunnels. At the same time, combine the ground feature characteristics (such as trees, valleys, etc.) in the point cloud data to refine the model details and avoid inaccurate modeling caused by occlusion or errors. Through the comparative analysis of the ground feature characteristics in the point cloud data and the infrastructure model, verify and optimize the size and position of the BIM model to ensure its complete match with the actual terrain and facility environment.

[0092] (2) Export the BIM model from Revit to the OBJ format for further analysis and visual display in other software (such as ArcGIS). Through this step, the efficient combination of the pre - processed point cloud data and BIM technology is successfully realized, and an accurate 3D real - scene and infrastructure design model is constructed, providing accurate and reliable data support and technical support for landslide risk warning, disaster assessment, and highway maintenance management.

[0093] It should be noted that step S3 and step S4 can be carried out simultaneously, both based on the preprocessed point cloud data obtained in S2. In S3, smart3D is mainly used to model the terrain around the highway, while in S4, Revit is mainly used to model the highway and the surrounding houses because of its better accuracy. smart3D has better accuracy for natural terrain modeling, so smart3D is used to model the terrain except for the highway and houses.

[0094] S5. Combine the BIM model and the real - scene 3D terrain generated by Smart3D

[0095] S3 and S4 have respectively obtained a high - precision real - scene 3D terrain and a BIM model of the highway and its surrounding houses. In order to achieve the purpose of risk warning, the two need to be combined under the same platform. The specific steps are as follows:

[0096] (1) Format compatibility and data import:

[0097] Export the real - scene 3D terrain generated by Smart3D to OBJ or OSGB format (supporting the storage of high - precision data details), and then import it into Revit as a reference basis. In Revit, the real - scene 3D terrain is displayed as a complete real - scene 3D background, which can intuitively show the terrain features of the highway and its surrounding areas.

[0098] (2) Alignment of terrain and BIM model:

[0099] In Revit, through the adjustment of the coordinate system, align the real - scene 3D terrain generated by Smart3D with the BIM model in space. The two need to share the same coordinate reference system to ensure that the facilities in the BIM model (such as roads, bridges, tunnels, etc.) are seamlessly connected with the terrain morphology in the real - scene 3D terrain.

[0100] (3) Model integration and analysis:

[0101] In Revit, integrate the BIM model and the real - scene 3D terrain to generate a comprehensive model containing infrastructure design and actual terrain. At this time, through the comparison and analysis function, verify the accuracy of the BIM design model to ensure that it correctly reflects the location and scale of the facilities in the 3D terrain. For example, by comparing the roadbed of the highway with the actual slope in the real - scene 3D terrain, detect whether there are height differences or offsets in the model and make adjustments as needed.

[0102] (4) Detail optimization and integration:

[0103] Using the Revit tool, the real - scene 3D terrain of Smart3D is used as a background reference to further optimize the geometric details of the BIM model. For example, in combination with the actual undulation of the terrain, adjust the road slope design, the foundation height of the bridge, or the excavation angle of the tunnel, etc. This optimization ensures that the BIM model not only meets the design requirements but also truly reflects the actual terrain conditions.

[0104] S6. The combination and simulation analysis of the established model with ABAQUS and PFC software

[0105] (1) Convert the model obtained in S5 into a 3D geometric format that can be used by ABAQUS and PFC. The OBJ format exported from Revit can be used for ABAQUS numerical simulation and analysis, and the DXF format exported from Revit can be used for PFC numerical simulation and analysis.

[0106] (2) In the ABAQUS software, import the optimized BIM model in step S5 to establish a complete slope geometric model. Through the mesh generation tool built in ABAQUS, finely divide the mesh to ensure the calculation accuracy of the slope stress field and displacement field. Then, according to the engineering requirements, set the property parameters of the geological materials (such as the Mohr - Coulomb model) to simulate the slope response under different external loads, such as rainfall and earthquake. Especially under the rainfall condition, ABAQUS can simulate the influence of pore water pressure on the slope and predict the potential areas where landslides may occur. Through the calculation of ABAQUS, the stress distribution, displacement distribution, and plastic zone distribution of the slope under different working conditions can be obtained, which helps to identify the areas that may become unstable. When the finite - element simulation reaches the convergence condition, the areas that may become unstable can be identified, and at this time, it is considered that the simulation field and stress field simulated by the ABAQUS software are accurate.

[0107] (3) Use the PFC software to conduct a discrete - element analysis on the areas that may become unstable to dynamically simulate the movement of particles during the landslide process. This discrete - element analysis is carried out on the premise that the simulation field and stress field of ABAQUS are accurate. After importing the DXF format exported from Revit into PFC, a two - dimensional profile model of the landslide is generated. Fill the particles by the random generation method, set the sliding area and the sliding bed part to generate a particle model with a certain porosity. Subsequently, apply rainfall conditions in the model, set the pore water pressure boundary, and track the movement path, speed, and accumulation position of the landslide particles through PFC. In addition, PFC can also simulate the influence of earthquake on the slope stability by applying acceleration conditions, so as to analyze the kinematic characteristics of the landslide particles under earthquake action.

[0108] S7. Import of the PFC two - dimensional model and the BIM model into the ArcGIS software

[0109] (1) PFC model data export: The PFC model is mainly a landslide or geological body model composed of particles. These particles and their movement trajectories are the core of the PFC model. Export the PFC model into a format suitable for GIS software processing. PFC can convert the PFC simulation results (such as particle positions, movement trajectories, etc.) into DXF, STL, or CSV formats through export tools. STL is suitable for the visualization of 3D models, while CSV can store the position information of particles for the visualization of point data. Use the wallexport and geometry export commands to export the landslide boundary and particle positions as DXF or STL files, and export the spatial position information of landslide particles as CSV format (including the X, Y, and Z coordinates of the particles), which is suitable for the visualization of 3D spatial point data in ArcGIS.

[0110] (2) BIM model data export: The BIM model mainly includes highways and the surrounding buildings and terrain. Export the BIM model from Revit software into common 3D model formats, including IFC, OBJ, and FBX formats. These formats can save complex geometric structures and are suitable for 3D model comparison and overlay in ArcGIS. Simplify the BIM model before export, retain the structural information such as roads, bridges, and tunnels, and delete irrelevant details to improve processing efficiency.

[0111] (3) Import the PFC and BIM models into ArcGIS:

[0112] 3.1) Import the PFC model. Use the Add Data function in ArcGIS to import the STL or DXF file exported from PFC into ArcGIS. The STL file can be used to represent the 3D positions of landslide particles, while the DXF file is used to represent the landslide boundary. Then import the CSV file to import the particle spatial position information of PFC as point data. Through the AddXY Data function in ArcGIS, convert the coordinate data in the CSV file into point features. Set the coordinate system (such as the UTM coordinate system) to ensure that the point data is correctly displayed in 3D space. These points represent the positions and movement trajectories of landslide particles, and the sliding process and accumulation area of the particles can be visualized in ArcGIS.

[0113] 3.2) Import the BIM model. ArcGIS supports directly importing the BIM model in OBJ or FBX format. Use the Add Data function to import the BIM model and ensure that the model has correct georeference information (such as elevation and position) for alignment with the PFC model.

[0114] 3.3) Ensure that the BIM model and the landslide data imported by PFC share the same coordinate system (e.g., UTM coordinate system) so that the model can be correctly displayed and overlaid within the same spatial range. Overlay the PFC model and the BIM model through the 3D Analyst tool of ArcGIS. Ensure that both are displayed in the same three-dimensional coordinate system and check the relative positional relationship between the landslide area and the highway and its surrounding facilities. If the particles in the PFC model slide into areas such as roads and bridges in the BIM model, the potential impact of the landslide on the infrastructure can be further analyzed.

[0115] S8. Compare and analyze the PFC and BIM models in ArcGIS to issue early warnings for geological disasters

[0116] (1) Spatial analysis and visualization. Through analysis tools such as Hillshade and Slope in ArcGIS, analyze the relative relationship between the sliding path of the landslide particles, the accumulation area and the facilities in the BIM model, whether the landslide effect endangers the BIM model, or how large a landslide can have an impact on the BIM. The Volume tool can be used to calculate the accumulation volume of the landslide and evaluate its impact on roads and bridges. In addition, set different colors or transparencies for the landslide model of PFC to help clearly distinguish the landslide body and structures such as roads and bridges in the BIM model. In the 3D Viewer, the model can be rotated and zoomed to view how the landslide body contacts or accumulates with the infrastructure.

[0117] (3) Drawing of schematic diagrams of analysis results. Use the mapping tools in ArcGIS to generate two-dimensional or three-dimensional schematic diagrams of the overlay effect of the PFC landslide model and the BIM road model. Mark the sliding path of the landslide body, the intersection points of the accumulation area with roads and bridges in the figure, and analyze the threat level of the landslide to highway facilities. Set different layer symbols and colors to distinguish the landslide body, roads and other facilities, and arrows can be used to mark the movement direction of the landslide body. Finally, output the drawings. The final schematic diagrams can be exported in PDF, JPEG or PNG formats for reporting or presenting the risk analysis of the landslide and the road. The generated three-dimensional schematic diagrams can show how the landslide particles enter the highway area and provide visual risk warning information.

[0118] (4) Key points of the schematic diagrams:

[0119] · Location and movement path of the PFC landslide body: Show the initial position, sliding path and final accumulation area of the landslide particles in the schematic diagram.

[0120] · BIM highway model: Show the design structure of the highway, especially the areas vulnerable to landslides, such as roadbeds, bridges, tunnel entrances, etc.

[0121] · Overlay comparison: Clearly show the area where landslide particles enter the highway, highlighting the infrastructure that may be affected by the accumulation of the landslide body.

[0122] · Color and symbols: Distinguish the PFC landslide model from the BIM facility model with different colors, transparencies, and symbols to ensure that the schematic diagram is clear and easy to understand.

[0123] Example 2

[0124] In this research example, slope stability analysis and risk warning were carried out on a section of about 3 kilometers long of a highway in Hebei Province. Through the combination of unmanned aerial vehicle (UAV) surveying and mapping, BIM modeling, and numerical simulation (ABAQUS and PFC), the risk warning of slope landslide was finally achieved.

[0125] Step S1: UAV surveying and mapping and data collection

[0126] First, a 3-kilometer section of a highway in Hebei Province with potential landslide risks was selected, mainly including the roadbed, bridge, and the nearby hillside area. An unmanned aerial vehicle equipped with a lidar (LiDAR) and a high-resolution camera was used for low-altitude flight surveying and mapping.

[0127] · Flight parameter setting: The flight altitude of the UAV was set to 120 meters to cover the terrain of this 3-kilometer section. The lateral overlap was set to 70%, and the forward overlap was set to 80% to ensure the surveying and mapping accuracy. The flight speed was set to 5 m / s to obtain high-precision images and original point cloud data.

[0128] · Data collection: The main data collected by the UAV during flight included original point cloud data (stored in LAS format), high-resolution image data (orthophoto map), and digital elevation model (DEM, stored in Geo TIFF format).

[0129] Step S2: Data processing and 3D modeling

[0130] Use Context Capture software to process the original point cloud data collected by the UAV: Import the LAS format point cloud data obtained by the UAV into Context Capture and perform data cleaning to remove noise points, duplicate points, and error data.

[0131] Step S3: Generate a detailed real-scene 3D terrain based on the processed point cloud data, including topographic features such as hillsides, bridges, and roadbeds along this section. Finally, export it in STL and OBJ formats for subsequent numerical simulation and modeling.

[0132] Step S4: Establishment of BIM model

[0133] Import the preprocessed point cloud data in Step 2 into Revit software for 3D BIM modeling operations. Use Autodesk Revit software to create a 3D BIM model of this section of the road, generate it using the terrain modeling tool in Revit, and establish infrastructure models of roads, bridges, and drainage systems based on highway design data. Focus on modeling key structures such as the roadbed, bridges, and culverts of the highway to ensure that the geometric accuracy of the model is consistent with the actual situation. At the same time, combine the ground object features (such as trees, valleys, etc.) in the point cloud data to refine the model details, avoid inaccurate modeling caused by occlusion or errors, verify and optimize the size and position of the BIM model through the comparative analysis of the ground object features in the point cloud data and the infrastructure model, and ensure that it perfectly matches the actual terrain and facility environment, so as to obtain a BIM model of the in-service road and surrounding houses based on the point cloud data;

[0134] Step S5: Compare and adjust the BIM model with the high-precision real-scene 3D terrain generated by Smart3D software to ensure seamless connection between the facilities in the BIM model and the terrain morphology in the real-scene 3D terrain. Then integrate the BIM model with the high-precision real-scene 3D terrain generated by Smart3D software and optimize the geometric details of the BIM model to ensure that the BIM not only meets the design requirements but also truly reflects the actual terrain conditions; Export the BIM model that meets the requirements in Revit to a format suitable for processing by ArcGIS software;

[0135] At this time, the BIM model integrates both the design data of the road and the on-site terrain features, providing a basis for subsequent analysis.

[0136] Step S6: Numerical simulation analysis

[0137] Export the BIM model processed in Step 5 from Revit in STL format for ABAQUS numerical simulation and analysis, and export it in DXF format from Revit for PFC numerical simulation and analysis;

[0138] (1) To analyze the stability of the surrounding slopes of this section of the road under different rainfall conditions, use ABAQUS for finite element analysis:

[0139] First, import the terrain part (STL format) in the BIM model into ABAQUS and generate a 2D profile model to simulate the key geometric features of the slope. Then, perform mesh division on the imported slope geometric model, using four-node plane strain elements (CPE4P elements). Mesh encryption is carried out in key areas (such as the contact part between the landslide body and the roadbed) to improve the simulation accuracy.

[0140] Material parameter setting: According to the geological exploration data, the physical parameters of the soil mass on the hillside of this section of the road are: soil mass density: 1600 kg / m 3; Angle of internal friction: 30°; Cohesion: 20 kPa; Modulus of deformation: 5×10⁷ Pa; Poisson's ratio: 0.3;

[0141] Rainfall condition simulation: Set the rainfall condition with a return period of 50 years, rainfall amount of 120 mm, and duration of 3 days. Set the pore water pressure boundary condition to simulate the influence of rainfall infiltration on the slope stress and displacement.

[0142] Analysis results: The simulation results show that when the rainfall lasts for 48 hours, stress concentration occurs at the outer edge of the slope, the maximum stress is 120 kPa, the maximum displacement is located below the roadbed and reaches 0.6 m, the plastic zone develops to the middle of the slope, and the landslide risk increases.

[0143] (2) To further simulate the movement process of landslide particles, discrete element analysis is carried out using PFC software:

[0144] After importing the DXF format exported from Revit into PFC2D, a two-dimensional profile model of the landslide is generated. Use PFC software to perform discrete element analysis on the potentially unstable area to dynamically simulate the movement of particles during the landslide process, apply dynamic conditions (such as rainfall, earthquake, etc.), and simulate the sliding path, accumulation range, displacement of particles, and kinematic characteristics under earthquake action.

[0145] Specifically, set the porosity of the sliding area to 0.25 for particle generation, use the random generation method to generate landslide particles, set the particle radius to 0.5 - 0.7 m, and use the same method to generate smaller particles for the sliding bed part with a particle radius of 0.3 - 0.5 m. A total of about 8000 particles are generated for simulating the landslide process. For the rainfall load condition, apply the same rainfall load as in ABAQUS, set the pore water pressure boundary, and simulate the influence of rainfall on particle movement. Finally, the PFC simulation results show that 48 hours after rainfall, the particles in the upper part of the sliding mass begin to be unstable and slide, the maximum displacement of the sliding mass is 1.2 m, and the landslide particles finally accumulate at the edge of the highway, and some particles invade the roadbed.

[0146] S7: Comprehensive analysis and risk warning

[0147] Use ArcGIS software for visual analysis and risk warning. Export the simulation results of ABAQUS and PFC as STL, DXF, and CSV formats respectively and import them into ArcGIS. Perform overlay analysis on the stress field and displacement field generated by ABAQUS and the landslide particle movement path of PFC. Then, through the 3D analysis tool of ArcGIS, overlay the BIM model, the stress field of the slope, and the landslide particle model to analyze the influence of the landslide body on the highway.

[0148] The analysis shows that in the most severe cases of the landslide, some of the accumulated materials will affect the highway subgrade. Combining the analysis results of ABAQUS and PFC, safety thresholds for stress and displacement are set. When the slope displacement reaches 0.5 meters or the pore water pressure exceeds 30 kPa, the system triggers an early warning to remind the maintenance department to carry out emergency treatment. In addition, based on the influence range of the debris flow obtained from the simulation, users can accurately determine which sections of the highway are at high risk levels and which are at low risk. Similarly, the risk levels to which the houses along the highway are exposed can be determined and specific risk levels can be distinguished.

[0149] However, in Figure 3 , if the patented technology of the present invention is not used, users can only generally determine that the mountains along the highway are unstable and disasters such as landslides and debris flows may occur, but they cannot accurately predict which sections of the road are at high risk of geological disasters and which sections are at low risk. At this time, the relevant management departments cannot issue accurate early warning information. For safety reasons, usually the entire section of the road is set to a high risk level, but it consumes a large amount of manpower and material resources.

[0150] The present invention obtains high-precision original point cloud data and image data of the area along and around the target highway through UAV close-range photogrammetry. After the data processed with high precision is imported into BIM software (such as Revit, Navisworks), a BIM model of the road, bridge and surrounding mountains is generated to ensure the terrain matching and structural integrity of the model. The BIM model is exported from Revit in OBJ format and imported into ABAQUS for finite element analysis. The physical properties and boundary conditions of the slope are set to simulate the stress distribution and displacement response under rainfall conditions. At the same time, PFC (particle flow simulation tool) is used to generate a particle model of the landslide. By applying dynamic conditions (such as rainfall, earthquake, etc.), the sliding path, accumulation range and displacement of the particles are simulated. After the analysis is completed, the numerical simulation results are imported into ArcGIS for three-dimensional overlay analysis with the BIM model to generate a visual risk early warning map, providing a scientific basis for maintenance and decision-making. There are differences among the multi-dimensional data such as the terrain data collected by the UAV, the BIM model, the ABAQUS stress field, and the PFC landslide particle flow. The present application overlays these data on the ArcGIS platform and realizes unified display through three-dimensional visual analysis. This integration method intuitively shows the spatial relationship between the landslide area and the road facilities, enabling the risk early warning to have the ability to integrate multi-dimensional information.

[0151] Matters not described in the present invention are applicable to the prior art.

Claims

1. A highway geological disaster risk early warning method based on BIM and numerical simulation, characterized in that: The steps of the early warning method are: Step 1: Use drone close-up photogrammetry to obtain high-precision original point cloud data and image data along the target highway and surrounding areas; Step 2: Import the original point cloud data collected by the drone into the Context Capture software, perform denoising, optimization, and precision adjustment on the original point cloud data in Context Capture, remove invalid points and duplicate points, splice and integrate data from different flight batches, generate a complete 3D point cloud model, and adjust the spatial coordinate system according to project requirements to ensure that it is consistent with the coordinates of the subsequently established BIM model, thus completing the preprocessing of the point cloud data; Step 3: Based on the data pre-processed in step 2, high-precision real-life 3D terrain of the in-service road and surrounding areas is generated by using Smart 3D software, and data format conversion is performed on the real-life 3D terrain data; Step 4: Import the point cloud data preprocessed in step 2 into Revit software for 3D BIM modeling operations. Use Revit's terrain modeling tool to generate a terrain surface to accurately reflect the actual terrain features of the highway and its surrounding areas. Based on the existing design data of the highway infrastructure, use Revit to create a highway infrastructure model. At the same time, combine the terrain features in the point cloud data to refine the model details to avoid inaccurate modeling caused by occlusion or errors. Through comparative analysis of the terrain features in the point cloud data and the infrastructure model, verify and optimize the size and position of the BIM model to ensure that it fully matches the actual terrain and facility environment, thereby obtaining a BIM model of the in-service road and surrounding houses based on the point cloud data. Step 5: Compare and adjust the BIM model with the high-precision real-life 3D terrain generated by Smart3D software to ensure that the facilities in the BIM model are seamlessly connected with the terrain in the real-life 3D terrain. Then, integrate the BIM model with the high-precision real-life 3D terrain generated by Smart3D software and optimize the geometric details of the BIM model to ensure that the BIM model meets the design requirements and truly reflects the actual terrain conditions. In Revit, export the BIM model that meets the requirements into a format suitable for processing by ArcGIS software. Step 6: Export the BIM model processed in step 5 from Revit to STL format or OBJ format for ABAQUS numerical simulation and analysis, where STL format data is used for three-dimensional numerical simulation modeling and analysis, OBJ format is used for two-dimensional section numerical simulation modeling and analysis, and DXF format is exported from Revit for PFC numerical simulation and analysis; A complete slope geometry model was established through ABAQUS software for finite element analysis. The physical properties and boundary conditions of the slope were set to simulate the slope stress field and displacement field under rainfall conditions. When the convergence conditions were reached, the possible unstable area, i.e. the unstable range, was identified. The DXF format exported from Revit was imported into PFC to generate a landslide model. The sliding body and sliding bed of the slope were divided according to the instability range under rainfall conditions determined by ABAQUS. PFC software was used to perform discrete element analysis on the possible unstable areas to dynamically simulate the movement of particles during the landslide process. Dynamic conditions were applied to simulate the sliding path, accumulation range and displacement of particles, as well as the kinematic characteristics under earthquake action. Step 7: Export the numerical simulation results of PFC into a format suitable for ArcGIS software processing and import it into the ArcGIS platform. Meanwhile, import the BIM model of step 5 into the ArcGIS platform, and set the BIM model and the landslide data imported by PFC to share the same coordinate system so that the two can be correctly displayed and three-dimensionally superimposed in the same spatial range. Step 8: Analyze the sliding path of landslide particles, the relative relationship between the accumulation area and the facilities in the BIM model through ArcGIS analysis tools, intuitively display the interaction between the landslide area, accumulation location and road facilities, and finally form early warning analysis and maintenance recommendations to generate a visual risk warning map.

2. The method according to claim 1, characterized in that The method can provide dynamic early warning. The data collected by the drone communicates with the Context Capture software. The Context Capture software communicates with the Smart 3D software and the Revit software respectively. The Revit software communicates with the ABAQUS software and the PFC software respectively. Meanwhile, the ABAQUS software communicates with the PFC software. The Revit software and the PFC software both communicate with the ArcGIS platform. The drone dynamically collects data, so that the BIM model can be continuously updated according to the periodic data collection of the drone, and the early warning adjustment is performed in combination with the simulation results of the PFC, so that the dynamicity and accuracy of the BIM model are maintained, and it is ensured that the early warning information always conforms to the current environment.

3. The method according to claim 2, characterized in that The ABAQUS software also communicates with the ArcGIS platform to display the numerical simulation results of the ABAQUS software.

4. The method according to claim 1, characterized in that: In step 5, the facilities in the BIM model include roads, bridges, tunnels and surrounding mountains.

5. The method according to claim 1, characterized in that: The process of processing point cloud data with the help of Smart 3D software includes: matching points with the same name using a feature-based matching algorithm, performing joint adjustment of multi-view images using a bundle method regional block adjustment algorithm, and completing the preliminary construction of a high-density point cloud based on a clustering algorithm and a dense matching algorithm based on patches; finally, the TIN network model constructed based on the point cloud is combined with texture mapping to achieve the construction of a high-precision real-scene three-dimensional model.

6. A highway geological disaster risk early warning system based on BIM and numerical simulation, characterized in that: The system comprises: The UAV data acquisition module uses UAVs to collect high-precision original point cloud data and image data along the target highway and surrounding areas; A data preprocessing module is used to preprocess the data collected by the UAV data collection module; The BIM modeling module is used to perform BIM modeling based on the preprocessed data and optimize the BIM model using the real-life three-dimensional terrain generated by Smart 3D software to obtain a BIM model that meets the design requirements and truly reflects the actual terrain conditions; The numerical simulation module is used to perform finite element analysis and discrete element analysis based on the BIM model output by the BIM modeling module. The finite element analysis is used to determine the area that may become unstable when the convergence condition is reached. The discrete element analysis is used to determine the sliding path, accumulation range and displacement of particles in the area that may become unstable, as well as the kinematic characteristics under earthquake action; GIS data integration module is used to integrate and display the output results of the BIM modeling module and the output results of the numerical simulation module, analyze the relative relationship between the sliding path of landslide particles, the accumulation area and the facilities in the BIM model, and intuitively display the interaction between the landslide area, accumulation location and road facilities; The risk warning and feedback module is used to conduct risk assessment based on the output of the GIS data integration module, and to generate a visual risk warning map based on the final warning analysis and maintenance recommendations.

Images