Engineering construction digital twinborn simulation deduction method

Abstract

The invention discloses an engineering construction digital twinborn simulation deduction method, and relates to the technical field of engineering construction, and the method comprises the deduction steps: S1, virtual scene construction: converting a construction site into a virtual three-dimensional scene through a digital twinborn simulation technology, configuring a digital construction equipment assembly and a digital component assembly in the virtual three-dimensional scene, the spatial relationship with the surrounding construction environment is reflected; s2, rapid acquisition and three-dimensional modeling: based on an oblique photography technology of the unmanned aerial vehicle, planning a route and an image acquisition action of the unmanned aerial vehicle in a preset modeling range; according to the method, low-cost rapid modeling is realized by adopting an unmanned aerial vehicle oblique photography technology, and high hardware investment is not needed; high-threshold compound talents are not needed, and popularization and application of small and medium-sized enterprises are facilitated; through combination of a centimeter-level precision three-dimensional model, mechanical dynamics analysis and a physical engine, nonlinear changes in an extreme environment and the stress state of a complex component are accurately simulated, and the problem of deviation of hoisting deduction of the extreme working condition and the complex component is effectively solved.

Description

Technical Field

[0001] This invention relates to the field of engineering construction technology, specifically to a digital twin simulation method for engineering construction. Background Technology

[0002] The current engineering equipment hoisting construction simulation technology is in a stage of rapid digital and intelligent development, and has achieved certain research results in terms of technology integration and application, and scenario implementation. However, due to constraints such as standards, costs, and technology compatibility, there are still many problems that need to be solved: First, there are barriers to data interoperability. There is a lack of unified modeling and data transmission standards in the industry. BIM models, equipment data, and progress data generated by different software are often scattered in different systems, forming "data silos." Cross-enterprise and cross-process collaboration requires a lot of time for data conversion, which not only reduces efficiency but may also affect the accuracy of the simulation due to data loss. At the same time, the simulation parameter setting standards are different for different projects, resulting in poor technology reusability. The application threshold is high for small and medium-sized enterprises. On the one hand, the initial investment in simulation technology is large, including high-precision modeling, sensor deployment, and software procurement. On the one hand, it requires high costs; on the other hand, the technology demands highly skilled professionals who possess both engineering hoisting knowledge and BIM and simulation software operation skills. However, small and medium-sized construction companies often lack such multi-skilled personnel, resulting in the technology being concentrated in key projects of large state-owned enterprises and hindering its widespread adoption across the industry. Furthermore, the technology lacks the ability to simulate extreme working conditions. While existing technologies can simulate most conventional scenarios, they exhibit significant limitations in extreme environments: for example, in offshore wind power hoisting, the nonlinear changes in waves and tides are difficult to model accurately; the impact of low air pressure and strong gusts on hoisting in high-altitude areas requires improvement in the accuracy of simulation model predictions; in addition, errors exist in the digital mapping of the complex stress state of irregularly shaped and ultra-large components during hoisting, potentially leading to discrepancies between the simulation results and actual working conditions. To address these issues, we propose a digital twin simulation method for engineering construction. Summary of the Invention

[0003] To address the aforementioned technical problems, this technical solution provides a digital twin simulation method for engineering construction, thus resolving the problems described above.

[0004] To achieve the above objectives, the technical solution adopted in this invention is: a digital twin simulation and deduction method for engineering construction, the deduction steps of which are as follows: S1. Virtual Scene Construction: The construction site is transformed into a virtual three-dimensional scene through digital twin simulation technology. Digital construction equipment components and building components are configured in the virtual three-dimensional scene to reflect the spatial relationship with the surrounding construction environment. S2. Rapid data acquisition and 3D modeling: Based on UAV oblique photogrammetry technology, the flight path and image acquisition actions of the UAV are planned for the predetermined modeling range. The UAV completes image acquisition according to the solution requirements of aerial triangulation. The coordinates of the corresponding points between the images are substituted into the collinearity equation to solve the stereo image pairs and obtain the 3D model of the predetermined modeling range as the data base of the environmental model. S3. Simulation and Pre-operation of Construction Operations: Commonly used construction components are parameterized to construct component model A, and commonly used construction equipment is parameterized and bound with interactive actions to construct construction equipment component model B. The resulting 3D model is used as environment model C. The mechanical actions of construction equipment component model B are manipulated to simulate the on-site construction and installation process, and overload, boundary crossing and collision situations during the construction process are detected in real time. The motion and stress state of the construction equipment are analyzed in combination with mechanical dynamics, and the safe operating angle and safe distance of the construction equipment are deduced. The safe operating range is displayed in a visual way to obtain the simulation and pre-operation results. S4. Adjust the selection of construction equipment, work layout plan and site rectification content according to the simulation and pre-run results, formulate safety technical guidance measures and form a construction technical plan.

[0005] Preferably, the step of converting to a virtual 3D scene in step S1 is as follows: using ground 3D laser scanning, UAV oblique photography and GIS to collect multi-source data on construction terrain, surrounding environment, engineering drawings and equipment components, and performing preprocessing; based on the preprocessed data, constructing an integrated basic scene of terrain and surrounding environment through a 3D modeling engine, and combining it with BIM model to perform refined modeling of the construction area.

[0006] Preferably, in step S1, the construction equipment components and component components are deployed in space according to the construction plan after parameterization adjustment. The virtual and real spatial relationship is established and conflicts are checked through spatial topology analysis. An IoT sensor network is deployed, a virtual and real linkage channel of edge computing + cloud transmission is built, and simulation rules for equipment movement and component assembly are embedded to verify the scene accuracy and function.

[0007] Preferably, the collinear equation in step S2 represents the spatial geometric relationship where the ground point, the image point, and the camera center lie on the same straight line during exposure, and its formula is: ; ; Where: (x, y) are the pixel coordinates of the image point; (x0, y0) are the coordinates of the principal point of the image; f is the camera focal length; (X, Y, Z) represent the coordinates of the ground point in the object space coordinate system; (X0, Y0, Z0) represent the position of the camera projection center in the object space coordinate system; m ij The elements of the rotation matrix M are calculated from the attitude angles (ω, Φ, κ) of the camera rotation around the three coordinate axes; The rotation matrix M is: .

[0008] Preferably, the mechanical dynamics in step S3 includes Newton's second law and Euler's rotation equation, wherein the formula for Newton's second law is: ;in The external force acting on an object; The mass of the object itself. Let x be the acceleration produced by an object under the action of an external force; Euler's rotation equation is: ;in The net external torque acting on the object. Let be the moment of inertia of the object. Let be the angular acceleration of the object.

[0009] Preferably, in step S2, the flight path of the UAV for the predetermined modeling range is planned, including flight altitude, flight speed and obstacle avoidance settings. The flight altitude is set based on the size of the modeling range, the flight speed is set to ≤8m / s, and the obstacle avoidance setting is a safe distance of ≥5 meters. Image acquisition uses RTK / PPK differential positioning technology to obtain the position and attitude data of the UAV in real time, and the positioning data and images are stored synchronously. After the acquisition is completed, the image data and the corresponding POS data are exported for data backup and filtering.

[0010] Preferably, in step S2, the 3D model within the predetermined modeling range is obtained by first preprocessing the acquired images and POS data, matching corresponding points based on the SIFT / ORB algorithm, completing aerial triangulation using the bundle method, and substituting the collinearity equation to solve the stereo image pairs; based on the stereo image pairs, multi-view dense matching is performed to generate a high-density point cloud, a 3D mesh is constructed through Poisson surface reconstruction, the model is optimized to unify the coordinate system, and exported and saved in a general format as the data base for the simulation environment.

[0011] Preferably, in step S3, the model components are constructed by parametrically building commonly used construction components into component model A. The outer contour of component model A is modeled using simple shapes such as cuboids, triangular prisms, and cylinders, and the length and thickness parameters are adjusted. The construction equipment is parameterized and bound to interactive actions to construct construction equipment component model B. The core actions of the construction equipment are decomposed, and parameters are set one by one. The position of the equipment is adjusted by dragging and rotating operations.

[0012] Preferably, in step S3, overload detection calculates the ratio of component weight to equipment rated load in real time. When the ratio exceeds the limit by ≥10%, a red warning is triggered, and the equipment lifting action is locked simultaneously. Boundary crossing detection delineates the construction boundary based on environmental model C. When the equipment or component exceeds the boundary by ≥0.5m, an audible and visual warning is issued, and the boundary crossing location and violation type are marked. Collision detection uses a bounding box + fine collision dual algorithm. When the distance between the equipment boom, component and environmental obstacles is ≤ the safety threshold, the collision risk is automatically calculated and classified. In low-risk situations, avoidance is prompted, and in high-risk situations, the simulation is forcibly paused.

[0013] Preferably, in step S3, a dynamic analysis model of equipment and components is constructed by combining mechanical dynamics. The tensile force and inertial force, bending stress and rotational torque of the equipment boom during the component hoisting process are calculated in real time, and a stress distribution cloud map is output, marking high stress areas. For complex environments such as high altitude and strong winds, the dynamic parameters are corrected to simulate the component swing amplitude and equipment force changes under extreme working conditions. Based on the force data, the force balance of the equipment outriggers and the stability of component hoisting are deduced, and a stability risk report is output in real time.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention achieves efficient integration and interoperability of construction data through a unified oblique photogrammetry modeling standard and parametric component model design, avoiding the problem of "data silos," reducing data conversion time and loss risk, and improving the accuracy and reusability of simulations. It utilizes UAV oblique photogrammetry technology to achieve low-cost and rapid modeling without requiring significant hardware investment. The virtual scene is easy to operate, requiring no highly skilled professionals, making it suitable for widespread application by small and medium-sized enterprises. Combining centimeter-level precision 3D models, mechanical dynamics analysis, and a physics engine, it accurately simulates nonlinear changes under extreme environments and the stress state of complex components, effectively solving the problem of deviations in simulations for extreme working conditions and complex component hoisting. Through multi-dimensional safety detection, visualization of safety indicators, and optimization of multiple schemes, it identifies potential construction hazards in advance, formulates scientific safety measures, reduces construction risks, and improves construction efficiency. The 3D model can serve as an archive for engineering data and a record of construction history, providing complete data support for project management. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the deductive steps of the present invention. Detailed Implementation

[0016] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0017] Reference Figure 1As shown, the digital twin simulation method for engineering construction includes the following steps: S1. Virtual Scene Construction: The construction site is transformed into a virtual three-dimensional scene through digital twin simulation technology. Digital construction equipment components and building components are configured in the virtual three-dimensional scene to reflect the spatial relationship with the surrounding construction environment. S2. Rapid data acquisition and 3D modeling: Based on UAV oblique photogrammetry technology, the flight path and image acquisition actions of the UAV are planned for the predetermined modeling range. The UAV completes image acquisition according to the solution requirements of aerial triangulation. The coordinates of the corresponding points between the images are substituted into the collinearity equation to solve the stereo image pairs and obtain the 3D model of the predetermined modeling range as the data base of the environmental model. S3. Simulation and Pre-operation of Construction Operations: Commonly used construction components are parameterized to construct component model A, and commonly used construction equipment is parameterized and bound with interactive actions to construct construction equipment component model B. The resulting 3D model is used as environment model C. The mechanical actions of construction equipment component model B are manipulated to simulate the on-site construction and installation process, and overload, boundary crossing and collision situations during the construction process are detected in real time. The motion and stress state of the construction equipment are analyzed in combination with mechanical dynamics, and the safe operating angle and safe distance of the construction equipment are deduced. The safe operating range is displayed in a visual way to obtain the simulation and pre-operation results. S4. Adjust the selection of construction equipment, work layout plan and site rectification content according to the simulation and pre-run results, formulate safety technical guidance measures and form a construction technical plan.

[0018] This application utilizes a 3D environment model constructed from aerial triangulation to solve stereo image pairs, achieving high-precision reconstruction of spatial relationships at the construction site. Parametric component and equipment models support repeated use and flexible adjustment. Combined with mechanical dynamics analysis, it overcomes the limitations of traditional construction simulations that only focus on process simulation, achieving an integrated technical loop of motion simulation, force analysis, and safety boundary deduction, thus improving the scientific rigor and accuracy of construction plans. Before construction, it can detect risks such as overload, boundary violations, and collisions in real time, deducing safe operating angles and distances for equipment, visually defining safe operating ranges, and developing targeted safety technical measures in advance. This prevents personnel and equipment safety accidents during construction from the source, reducing the difficulty of safety management. It optimizes equipment selection, work arrangement, and site rectification without relying on on-site trial and error, significantly shortening the development cycle of construction technical plans. Virtual simulation pre-drafts replace some physical testing stages, improving the efficiency of pre-construction preparation and accelerating the overall project progress.

[0019] The steps in step S1 to transform into a virtual 3D scene are as follows: 3D laser scanning of the ground, oblique photography by UAVs, and GIS retrieval are used to collect multi-source data on construction terrain, surrounding environment, engineering drawings, and equipment components for preprocessing; based on the preprocessed data, a basic scene integrating terrain and surrounding environment is constructed through a 3D modeling engine, and a detailed model of the construction area is carried out in combination with a BIM model.

[0020] This application integrates high-precision terrain data from 3D laser scanning, stereoscopic image data from UAV oblique photography, surrounding environmental data retrieved from GIS, and component and equipment parameters from engineering drawings. It covers the core elements of the construction site from multiple dimensions, effectively compensating for the information blind spots of a single data source and ensuring a high degree of consistency between the spatial topology of the virtual 3D scene and the real construction site. By eliminating redundant and erroneous information through data preprocessing, the computational load of the 3D modeling engine is reduced, and the scene construction efficiency is improved. Combined with the BIM model, the construction area is refined and modeled, which can accurately map details such as component assembly relationships and equipment deployment locations, providing high-precision and interactive basic scene support for the subsequent S3 step construction simulation and pre-visualization.

[0021] In step S1, construction equipment components and component components are deployed in space according to the construction plan after parametric adjustment. The relationship between virtual and real space is established and conflicts are checked through spatial topology analysis. An IoT sensor network is deployed to build a virtual-real linkage channel of edge computing + cloud transmission. Equipment movement and component assembly simulation rules are embedded to verify the accuracy and functionality of the scene.

[0022] This application significantly improves the accuracy, interoperability, and functionality of virtual 3D construction scenes through a combination of parametric deployment, topology analysis, virtual-real linkage, and rule embedding optimization. After parametric adjustment, equipment and components are deployed strictly according to the construction plan, flexibly adapting to different construction conditions and ensuring a high degree of alignment between the spatial position of virtual components and the construction plan, avoiding simulation distortion caused by layout deviations. By establishing virtual-real spatial relationships and verifying conflicts through spatial topology analysis, the application can identify spatial interference risks between equipment, components, and the surrounding environment in advance during the scene construction phase, reducing the cost of adjusting construction plans from the outset. Through the deployment of an IoT sensor network, a virtual-real linkage channel combining edge computing and cloud transmission is established, enabling real-time data interaction between the physical site and the virtual scene. This provides accurate data support for subsequent dynamic construction simulation. Embedding simulation rules for equipment movement and component assembly and verifying scene accuracy ensures that the virtual scene possesses complete construction simulation capabilities, laying a solid technical foundation for construction pre-simulation, safety risk detection, and scheme optimization.

[0023] The collinear equation in step S2 represents the spatial geometric relationship where the ground point, image point, and camera center lie on the same straight line during exposure. Its formula is: ; ; Where: (x, y) are the pixel coordinates of the image point; (x0, y0) are the coordinates of the principal point of the image; f is the camera focal length; (X, Y, Z) represent the coordinates of the ground point in the object space coordinate system; (X0, Y0, Z0) represent the position of the camera projection center in the object space coordinate system; m ij The elements of the rotation matrix M are calculated from the attitude angles (ω, Φ, κ) of the camera rotation around the three coordinate axes; The rotation matrix M is: .

[0024] The collinearity equations in this application serve as the core basis for solving stereo image pairs in aerial triangulation, providing crucial support for 3D modeling of construction sites. They offer significant advantages, accurately depicting the collinear geometric relationship between ground points, image points, and the camera center. By constructing a standardized theoretical model, they achieve direct mapping between image-side pixel coordinates and object-side spatial coordinates, ensuring the rigor of coordinate transformation logic and avoiding errors in empirical modeling. Incorporating multi-dimensional parameters such as camera parameters, rotation matrix, and attitude angles, they can quantify and correct the effects of flight attitude deviations and camera distortion, significantly improving the accuracy of the 3D model. This provides a solid foundation of high-precision environmental data for virtual scenes. Furthermore, the high-precision model ensures the accuracy of calculations such as equipment movement and collision detection in subsequent construction simulations, preventing simulation distortion.

[0025] The mechanical dynamics in step S3 include Newton's second law and Euler's rotation equation, where Newton's second law is formulated as follows: ;in The external force acting on an object; The mass of the object itself. Let x be the acceleration produced by an object under the action of an external force; Euler's rotation equation is: ;in The net external torque acting on the object. Let be the moment of inertia of the object. Let be the angular acceleration of the object.

[0026] This application uses Newton's second law to quantify the relationship between external force and acceleration in the translational state of construction equipment, and Euler's rotation equations to accurately characterize the relationship between net external torque, moment of inertia, and angular acceleration during rotation. The combination of these two methods enables comprehensive force and motion analysis of the equipment in both translational and rotational states. This approach overcomes the limitations of traditional experience-based judgments, quantifies the overload critical point and motion limit posture of the equipment, accurately predicts safe operating angles and distances, and improves the accuracy of risk detection such as collisions and boundary crossings. At the same time, the quantitative output of dynamic parameters ensures that the virtual simulation closely matches the actual construction conditions, providing reliable data support for equipment selection and work layout adjustments, avoiding deviations in the scheme caused by simulation distortion, and reducing subsequent on-site trial and error costs.

[0027] In step S2, the flight path of the UAV for the predetermined modeling range is planned, including flight altitude, flight speed and obstacle avoidance settings. The flight altitude is set based on the size of the modeling range, the flight speed is set to ≤8m / s, and the obstacle avoidance safety distance is set to ≥5 meters. Image acquisition uses RTK / PPK differential positioning technology to obtain the position and attitude data of the UAV in real time, and the positioning data and images are stored synchronously. After the acquisition is completed, the image data and the corresponding POS data are exported for data backup and filtering.

[0028] This application's flight altitude adapts to the modeling range size, ensuring the integrity of image coverage; a flight speed ≤8m / s avoids image blurring caused by high-speed flight, ensuring clear image point information; an obstacle avoidance safety distance ≥5m effectively avoids the risk of collision between the UAV and surrounding obstacles, ensuring equipment operation safety; combined with RTK / PPK differential positioning technology, it acquires and stores UAV position, attitude data, and image POS data in real time, providing accurate parameter support for solving collinearity equations for stereo image pairs, and improving the spatial positioning accuracy of the 3D model.

[0029] In step S2, the 3D model within the predetermined modeling range is obtained by first preprocessing the acquired images and POS data, matching corresponding points based on the SIFT / ORB algorithm, completing aerial triangulation using the bundle method, and substituting the collinearity equation to solve the stereo image pairs; based on the stereo image pairs, multi-view dense matching is performed to generate a high-density point cloud, and a 3D mesh is constructed through Poisson surface reconstruction, optimizing the model to unify the coordinate system, and exporting and saving it in a general format as the data base for the simulation environment.

[0030] This application first preprocesses the imagery and POS data, then uses the SIFT / ORB algorithm to achieve accurate matching of corresponding points. This algorithm has strong anti-interference capabilities and can effectively cope with matching errors caused by changes in illumination and angle, providing reliable data support for bundled aerial triangulation. Bundled aerial triangulation is then used to solve the collinearity equations for stereo image pairs, enabling joint adjustment of multiple images, reducing the accumulated errors of traditional step-by-step solutions, and improving the rigor and accuracy of coordinate solutions. Based on multi-view dense matching of stereo image pairs, a high-density point cloud is generated, and a complete 3D mesh is constructed by combining Poisson surface reconstruction. The coordinate system is optimized to ensure the spatial uniformity of the model. Exporting a universal format enhances the compatibility of the model with subsequent digital twin and BIM systems.

[0031] In step S3, the model components are constructed by parametrically building component model A, which uses simple shapes such as cuboids, triangles and cylinders to model the outer contour of component model A and adjusts the length and thickness parameters. The construction equipment is then parametrically built by binding interactive actions to construct construction equipment component model B. The core actions of the construction equipment are decomposed, and parameters are set one by one. The position of the equipment is adjusted by dragging and rotating operations.

[0032] The parametric construction and lightweight modeling design of component and equipment models in step S3 of this application can significantly improve the efficiency, flexibility and adaptability of virtual simulation. Component model A adopts simple geometric shapes such as cuboids and triangular prisms and supports parameter debugging, which can reduce the computational load of the model, speed up the modeling speed, and adapt to the simulation requirements of different specifications of components through flexible adjustment of length and thickness parameters. Construction equipment component model B achieves accurate mapping of equipment actions and convenient adjustment of position by parametrically binding core actions and supporting drag and rotate interactive operations, so as to restore the working posture of the equipment without complex modeling.

[0033] In step S3, overload detection calculates the ratio of component weight to equipment rated load in real time. When the ratio exceeds 10%, a red alert is triggered, and the equipment lifting action is locked simultaneously. Boundary crossing detection delineates construction boundaries based on environmental model C. When equipment or components exceed the boundary by ≥0.5m, an audible and visual warning is issued, and the location of the boundary crossing and the type of violation are marked. Collision detection uses a dual algorithm of bounding box and fine collision. When the distance between the equipment boom, component, and environmental obstacles is ≤ the safety threshold, the collision risk is automatically calculated and classified. In low-risk situations, avoidance is prompted, and in high-risk situations, the simulation is forcibly paused.

[0034] This application's overload detection uses the ratio of component weight to equipment rated load to make quantitative judgments. When the ratio is ≥10%, a red warning is triggered and the lifting action is locked, accurately avoiding the hidden dangers of equipment overload operation. Boundary crossing detection relies on the environmental model to delineate the construction boundary. When equipment or components exceed the boundary by ≥0.5m, an audible and visual warning is given and the location and type of violation are marked, promptly restricting the work area. Collision detection adopts a dual algorithm of bounding box + fine collision, classifying risk levels according to safety thresholds. Low-risk situations prompt avoidance, while high-risk situations force a pause in simulation, balancing detection efficiency and accuracy.

[0035] In step S3, a dynamic analysis model of equipment and components is constructed by combining mechanical dynamics. The tensile force and inertial force, bending stress and rotational torque of the equipment boom are calculated in real time during the component hoisting process. The stress distribution cloud map is output and high stress areas are marked. For complex environments such as high altitude and strong winds, the dynamic parameters are corrected to simulate the component swing amplitude and equipment stress changes under extreme working conditions. Based on the stress data, the stress balance of the equipment outriggers and the stability of component hoisting are deduced, and a stability risk report is output in real time.

[0036] The model in this application can calculate the lifting tension, inertial force, and bending stress and rotational torque of the equipment boom in real time, output stress distribution cloud map and mark high stress areas, accurately grasp the mechanical state of the equipment and components, avoid structural overload damage, correct dynamic parameters for complex environments such as high altitude and strong winds, simulate the swing amplitude of components and changes in equipment force under extreme working conditions, fill the gap in conventional simulation coverage of special working conditions, deduce the force balance of the outriggers and the lifting stability based on force data, and output stability risk reports in real time to identify potential instability hazards in advance.

[0037] Example: Digital Twin Simulation of Power Installation Engineering Construction Virtual scene construction: A virtual 3D scene of the power installation construction site is built using the Unity3D platform. Digital components of cranes, tensioners, and cross-frame construction equipment, as well as digital components of steel structure columns and steel beams are imported. All components restore the operating conditions and spatial state of real equipment and components. The crane can be controlled by the handle to complete actions such as platform rotation, boom extension and retraction, and hook lifting and lowering. Rapid data acquisition and 3D modeling: A 200m×200m power construction area was selected. A DJI drone was used to plan the flight path to ensure that the image overlap met the requirements of aerial triangulation. After image acquisition, DJI Terra software was used for data processing. The collinearity equation was substituted to solve the stereo image pairs. Combined with RTK technology, the accuracy of the 3D model reached 2cm. Irrelevant trees and vehicle elements in the scene were removed online to obtain the environment model C, which was archived as engineering data for future use. Simulation and pre-construction operation: Build component model A and adjust parameters to make the outer contour of the model consistent with the actual component size; build construction equipment component model B, bind the boom extension and hook lifting and lowering action parameters, and set the crane's rated load to 139.2 kg; By integrating models A and B with environmental model C, technicians can control the crane model to simulate the steel column hoisting process and use a physics engine to simulate the installation status of the steel column. The system monitors in real time whether there is overloading, boundary crossing, or collision during the hoisting process. It analyzes the force state of the crane based on Newton's second law, deduces the safe operating angle and safe distance of the crane, and displays the safe operating range through a virtual barrier wall. The measured spatial distance between the crane and the power line is 10.84 meters, and the horizontal distance is 10.84 meters, which meets the safety distance requirements. The process was observed and simulated from multiple perspectives. The placement of the crane and the hoisting route were adjusted, the work layout plan was optimized, safety technical guidance measures were formulated, and the final construction plan was formed.

[0038] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A digital twin simulation method for engineering construction, characterized in that, The deduction steps are as follows: S1. Virtual Scene Construction: The construction site is transformed into a virtual three-dimensional scene through digital twin simulation technology. Digital construction equipment components and building components are configured in the virtual three-dimensional scene to reflect the spatial relationship with the surrounding construction environment. S2. Rapid data acquisition and 3D modeling: Based on UAV oblique photogrammetry technology, the flight path and image acquisition actions of the UAV are planned for the predetermined modeling range. The UAV completes image acquisition according to the solution requirements of aerial triangulation. The coordinates of the corresponding points between the images are substituted into the collinearity equation to solve the stereo image pairs and obtain the 3D model of the predetermined modeling range as the data base of the environmental model. S3. Simulation and Pre-operation of Construction Operations: Commonly used construction components are parameterized to construct component model A, and commonly used construction equipment is parameterized and bound to interactive actions to construct construction equipment component model B. The resulting 3D model is used as environment model C. The mechanical actions of construction equipment component model B are manipulated to simulate the on-site construction and installation process, and overload, boundary crossing and collision situations during the construction process are detected in real time. The motion and stress state of the construction equipment are analyzed in combination with mechanical dynamics, and the safe operating angle and safe distance of the construction equipment are deduced. The safe operating range is displayed in a visual way to obtain the simulation and pre-operation results. S4. Adjust the selection of construction equipment, work layout plan and site rectification content according to the simulation and pre-run results, formulate safety technical guidance measures and form a construction technical plan.

2. The digital twin simulation and deduction method for engineering construction according to claim 1, characterized in that, The steps in step S1 to transform into a virtual 3D scene are as follows: 3D laser scanning of the ground, oblique photography by UAVs, and GIS retrieval are used to collect multi-source data on construction terrain, surrounding environment, engineering drawings, and equipment components for preprocessing; based on the preprocessed data, a basic scene integrating terrain and surrounding environment is constructed through a 3D modeling engine, and a detailed model of the construction area is carried out in combination with a BIM model.

3. The digital twin simulation method for engineering construction according to claim 1, characterized in that: In step S1, construction equipment components and component components are deployed in space according to the construction plan after parametric adjustment. The relationship between virtual and real space is established and conflicts are checked through spatial topology analysis. An IoT sensor network is deployed to build a virtual-real linkage channel of edge computing + cloud transmission. Equipment movement and component assembly simulation rules are embedded to verify the accuracy and functionality of the scene.

4. The digital twin simulation method for engineering construction according to claim 1, characterized in that: The collinear equation in step S2 represents the spatial geometric relationship where the ground point, image point, and camera center lie on the same straight line during exposure. Its formula is: ； ； Where: (x, y) are the pixel coordinates of the image point; (x0, y0) are the coordinates of the principal point of the image; f is the camera focal length; (X, Y, Z) represent the coordinates of the ground point in the object space coordinate system; (X0, Y0, Z0) represent the position of the camera projection center in the object space coordinate system; m ij The elements of the rotation matrix M are calculated from the attitude angles (ω, Φ, κ) of the camera rotation around the three coordinate axes; The rotation matrix M is: 。 5. The digital twin simulation method for engineering construction according to claim 1, characterized in that: The mechanical dynamics in step S3 include Newton's second law and Euler's rotation equation, where Newton's second law is formulated as follows: ;in The external force acting on an object; The mass of the object itself. The acceleration produced by an object under the action of an external force; Euler's rotation equation is as follows: ;in The net external torque acting on the object. Let be the moment of inertia of the object. Let be the angular acceleration of the object.

6. The digital twin simulation and deduction method for engineering construction according to claim 1, characterized in that: In step S2, the flight path of the UAV for the predetermined modeling range is planned, including flight altitude, flight speed and obstacle avoidance settings. The flight altitude is set based on the size of the modeling range, the flight speed is set to ≤8m / s, and the obstacle avoidance safety distance is set to ≥5 meters. Image acquisition uses RTK / PPK differential positioning technology to obtain the position and attitude data of the UAV in real time, and the positioning data and images are stored synchronously. After the acquisition is completed, the image data and the corresponding POS data are exported for data backup and filtering.

7. The digital twin simulation method for engineering construction according to claim 1, characterized in that: In step S2, the 3D model within the predetermined modeling range is obtained by first preprocessing the acquired images and POS data, matching corresponding points based on the SIFT / ORB algorithm, completing aerial triangulation using the bundle method, and substituting the collinearity equation to solve the stereo image pairs; based on the stereo image pairs, multi-view dense matching is performed to generate a high-density point cloud, and a 3D mesh is constructed through Poisson surface reconstruction, optimizing the model to unify the coordinate system, and exporting and saving it in a general format as the data base for the simulation environment.

8. The digital twin simulation method for engineering construction according to claim 1, characterized in that: In step S3, the model components are constructed by parametrically building component model A, which uses simple shapes such as cuboids, triangles and cylinders to model the outer contour of component model A and adjusts the length and thickness parameters. The construction equipment is then parametrically built by binding interactive actions to construct construction equipment component model B. The core actions of the construction equipment are decomposed, and parameters are set one by one. The position of the equipment is adjusted by dragging and rotating operations.

9. The digital twin simulation method for engineering construction according to claim 1, characterized in that: In step S3, overload detection calculates the ratio of component weight to equipment rated load in real time. When the ratio exceeds the limit by ≥10%, a red warning is triggered, and the equipment lifting action is locked simultaneously. Boundary crossing detection is based on environmental model C to delineate construction boundaries. When equipment or components exceed the boundary by ≥0.5m, an audible and visual warning is issued, and the location of the boundary crossing and the type of violation are marked. The collision detection uses a dual algorithm of bounding box and fine collision. When the distance between the equipment boom, components and environmental obstacles is less than or equal to the safety threshold, the collision risk is automatically calculated and classified. When the risk is low, the system prompts the user to avoid the collision, and when the risk is high, the simulation is forcibly paused.

10. The digital twin simulation and deduction method for engineering construction according to claim 1, characterized in that: In step S3, a dynamic analysis model of equipment and components is constructed by combining mechanical dynamics. The tensile force and inertial force, bending stress and rotational torque of the equipment boom are calculated in real time during the component hoisting process. The stress distribution cloud map is output and high stress areas are marked. For complex environments such as high altitude and strong winds, the dynamic parameters are corrected to simulate the component swing amplitude and equipment stress changes under extreme working conditions. Based on the stress data, the stress balance of the equipment outriggers and the stability of component hoisting are deduced, and a stability risk report is output in real time.

Images