Steel structure intelligent demolition method and system based on digital twinning and reverse construction analysis

Abstract

This invention discloses a method and system for intelligent steel structure demolition based on digital twins and reverse construction analysis, belonging to the field of civil engineering construction safety and control. The method first integrates point cloud data and drawings to construct a high-precision As-Built model, which is then calibrated to establish a digital twin. Next, the demolition stage is defined in reverse using finite element software, simulating component demolition and considering geometric nonlinearity to complete the simulation and extract risk indicators. Subsequently, an IoT sensor network is deployed, setting three levels of dynamic early warning thresholds and deploying a cloud-based platform. Finally, measured and simulated data are compared; when an early warning is triggered, the model is updated in reverse, the simulation is repeated, and construction instructions are dynamically optimized. Blockchain is used for data storage. The system has four functional modules to implement the above method. This invention constructs a BIM-FEA-IoT technology closed loop, solving problems such as model fragmentation and data silos in traditional demolition, improving construction safety and digitalization levels, and is applicable to various steel structure demolition and renovation projects.

Description

Technical Field

[0001] This invention belongs to the field of civil engineering construction safety and control technology, specifically involving the cross-software platform integrated application of building information modeling, structural finite element analysis, three-dimensional laser scanning and Internet of Things monitoring, and particularly involving a method and system for intelligent demolition of steel structures based on digital twin and reverse construction analysis. Background Technology

[0002] The demolition of existing steel structures is a typical high-risk, time-varying system operation. During construction, the structure's stress system and stiffness distribution continuously change as components are removed. Without accurate simulation analysis and real-time safety control, structural instability, collapse, and other safety accidents are highly likely. Traditional steel structure demolition methods have many significant shortcomings and can no longer meet the safety management requirements for the demolition of large and complex steel structures. (1) Model discontinuity problem: There is a lack of automatic geometric and attribute association between AutoCAD two-dimensional drawings, steel structure physical entities and finite element calculation models. Model construction relies on manual redrawing and parameter input, which is not only inefficient, but also prone to errors introduced by human operation, resulting in a large deviation between the calculation model and the actual structure. (2) Limitations of the analysis: Most demolition analyses only perform linear static verification on the initial or final state of the structure, which cannot systematically simulate the dynamic changes of component failure during the demolition process, and do not fully consider the influence of geometric nonlinearity on structural stability. The accuracy and reference value of the simulation results are insufficient. (3) Data silo problem: The demolition construction plan, simulation analysis results and on-site monitoring data are separated from each other, making it impossible to achieve real-time comparison between simulation prediction data and on-site measured data, and making it difficult to form a scientific decision-making closed loop; (4) Control lag problem: Construction safety judgment lacks quantitative early warning thresholds based on high-fidelity simulation, and it is impossible to realize immersive pre-play and real-time early warning of risk conditions in the visualization management platform. When abnormal situations occur on site, the formulation and implementation of safety control measures are significantly delayed, and the best time to deal with them is easily missed.

[0003] Therefore, there is an urgent need to develop a cross-software platform integration technology that can connect BIM modeling, FEA simulation and IoT monitoring, build a full-process digital management and control system, realize the calculability, transparency and control of steel structure demolition operations, and fundamentally improve the safety and digital management level of demolition construction. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned defects of the prior art and provide a method and system for intelligent demolition of steel structures based on digital twin and reverse construction analysis, construct a technical closed loop of deep integration of BIM-FEA-IoT, and realize high-precision simulation, real-time monitoring, intelligent early warning and dynamic control of the steel structure demolition process.

[0005] To achieve the objectives of this invention, the technical solution provided by this invention is as follows: A method for intelligent demolition of steel structures based on digital twin and reverse construction analysis includes the following steps: Step S1: Based on the 3D laser scanning point cloud data and existing design drawings, construct a high-precision parametric As-Built model in the building information modeling platform; import the model into finite element analysis software through a dedicated interface to generate a calculation model, and calibrate the calculation model using the dynamic characteristic data measured on site to establish a benchmark digital twin consistent with the physical structure. Step S2: In the finite element analysis software, based on the preset demolition plan, a construction stage sequence containing multiple stages is defined in reverse. The relevant actions of each demolition stage are simulated by setting the activation and passivation states of the components. Under the condition of considering geometric nonlinearity, the construction stage analysis is performed, and the displacement field, internal force redistribution and eigenvalue buckling stability coefficient of each stage are calculated simultaneously to extract key risk quantification indicators. Step S3: Based on the high-risk areas identified by the simulation results, deploy an IoT sensor network; set the displacement and strain time history data predicted by the simulation as dynamic early warning thresholds, and deploy a cloud data platform for real-time data aggregation and visualization; Step S4: During the demolition process, the real-time monitoring data of the IoT sensor network is dynamically compared with the simulation prediction threshold. When the data deviation exceeds the preset range, an early warning is triggered and the application programming interface of the finite element analysis software is automatically called to update the boundary conditions or material parameters of the digital twin model with the measured data. Based on the updated model, the subsequent demolition steps are quickly re-simulated, the subsequent construction instructions are dynamically optimized and issued, forming a closed-loop control of monitoring-simulation-decision-execution.

[0006] Among them, the high-precision parametric As-Built model constructed in step S1 is an as-built BIM model with a LOD of no less than 350. The construction process is as follows: use Bentley Context Capture or Autodesk ReCap software to process point cloud data, use Dynamo visual programming scripts in the Autodesk Revit platform to drive the instantiation of parametric structural families, automatically fit the geometry of components, and embed the attribute information of connection nodes.

[0007] In step S1, the specific method for importing the model into the finite element analysis software through a dedicated interface is as follows: using the MIDAS Gen for Revit dedicated plugin, the steel structure frame and load information in the Revit model are automatically mapped and converted into beam and column elements and load cases in the MIDAS Gen software; the specific method for model calibration is as follows: comparing the modal frequencies calculated by MIDAS Gen with the vibration test frequencies of the field environment, and adjusting the model parameters until the error is less than 5%.

[0008] In step S2, the specific method for considering the geometric nonlinearity of the construction stage analysis is as follows: enable the large displacement effect option in the analysis control of the MIDAS Gen software; the risk quantification index includes at least the maximum displacement δmax(i) of each stage, the maximum stress ratio of the component σratio(i), and the lowest order yield load factor λmin(i) of the structural system at that stage.

[0009] Step S2 also includes a visualization pre-simulation step: the deformation and stress results of each stage calculated by MIDAS Gen are exported in IFC format and imported into Autodesk Navisworks software to generate a 4D demolition process simulation animation that integrates the time dimension, which is used for safety briefings and intuitive risk display of steel structure demolition.

[0010] In step S3, the dynamic warning threshold is set in three levels: the yellow warning threshold is 70% of the simulation prediction value, the orange warning threshold is 90%, and the red warning threshold is 100% or reaches the standard allowable limit; the warning information is simultaneously highlighted in the cloud data platform cockpit interface and the Navisworks4D model.

[0011] In step S4, the specific method of updating the digital twin model with measured data is as follows: when the monitored displacement triggers an early warning, the measured displacement value δmeasured is applied as a forced displacement load to the corresponding construction stage CSI model through the API interface provided by the MIDAS Gen software. A reverse analysis is run once, and the software’s built-in sensitivity analysis tool is used to automatically identify and correct the parameters in the model that are most likely to deviate from reality by comparing the reaction force differences. The dynamic closed-loop control in step S4 also includes a blockchain evidence storage step: all construction instructions, real-time monitoring data, model correction records, early warning events and optimized solutions automatically issued by the system are encrypted and written into the blockchain network to generate an immutable digital archive of demolition project safety.

[0012] Before constructing the high-precision parameterized As-Built model in step S1, a multi-source data acquisition step is also included: using a three-dimensional laser scanning device to scan the steel structure to be demolished on-site to generate point cloud data with an accuracy better than ±3mm, and simultaneously using a total station to measure the coordinates of key control points as a reference for point cloud stitching and model calibration. In step S2, the simulation of the dismantling stage also includes the simulation of temporary supports: the element properties of the temporary supports are activated during the installation stage and the element properties are deactivated during the dismantling stage; and the stability and safety criterion in the construction stage analysis is set as the minimum buckling load factor λcri of the structural system ≥ 2.5.

[0013] Second aspect This invention provides an intelligent steel structure demolition system for implementing the aforementioned intelligent steel structure demolition method based on digital twin and reverse construction analysis, comprising: The reverse digital twin model construction and calibration module is used to construct a high-precision parametric As-Built model in the building information modeling platform based on 3D laser scanning point cloud data and existing design drawings; the model is imported into finite element analysis software through a dedicated interface to generate a calculation model, and the calculation model is calibrated using field-measured dynamic characteristic data to establish a benchmark digital twin consistent with the physical structure; The demolition process dynamic simulation and risk quantification module is used in the finite element analysis software to reverse define a construction stage sequence containing multiple stages based on a preset demolition plan. It simulates the relevant actions of each demolition stage by setting the activation and deactivation states of the components. Under the condition of considering geometric nonlinearity, it performs construction stage analysis, synchronously calculates the displacement field, internal force redistribution and eigenvalue buckling stability coefficient of each stage, and extracts key risk quantification indicators. The monitoring system construction and early warning threshold setting module is used to deploy an IoT sensor network in high-risk areas identified by simulation results; the displacement and strain time history data predicted by simulation are set as dynamic early warning thresholds, and a cloud data platform is deployed for real-time data aggregation and visualization. The real-time feedback and dynamic closed-loop control module is used to dynamically compare the real-time monitoring data of the IoT sensor network with the simulation prediction threshold during the demolition construction process. When the data deviation exceeds the preset range, an early warning is triggered and the application programming interface of the finite element analysis software is automatically called to update the boundary conditions or material parameters of the digital twin model with the measured data. Based on the updated model, the subsequent demolition steps are quickly re-simulated, the subsequent construction instructions are dynamically optimized and issued, forming a closed-loop control of monitoring-simulation-decision-execution.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By integrating three-dimensional laser scanning with BIM technology, an As-Built model of LOD 350 level was constructed, and the model was calibrated by combining on-site dynamic characteristic testing, so that the error between the digital twin and the physical structure was less than 5%, providing an accurate model basis for demolition simulation analysis and solving the model discontinuity problem of traditional methods. (2) Based on the construction stage analysis function of finite element analysis software, the demolition dynamic process of component failure is simulated, the geometric nonlinear effect is fully considered, and multi-dimensional risk quantification indicators are extracted. At the same time, 4D visualization technology is combined to realize intuitive display of risk, which solves the limitation problem of traditional analysis. (3) By integrating IoT monitoring technology with cloud data platform, the simulation prediction data and the actual measured data are aggregated, compared and visualized in real time, and a simulation-monitoring data linkage system is constructed, which solves the data silo problem of traditional methods; (4) Set three levels of dynamic early warning thresholds, realize the real-time update of digital twin model and dynamic optimization of demolition scheme based on actual measurement data, complete the closed loop of monitoring-simulation-decision-execution in milliseconds, and introduce blockchain evidence storage to realize the immutable management of data throughout the process, which solves the control lag problem of traditional methods; (5) This invention relies on general software platforms in the engineering industry such as Autodesk, Bentley, and MIDAS. The data interface and analysis module are all commercially mature technologies. No additional special software needs to be developed. It is easily accepted and applied by the engineering community and can be widely used in the demolition, renovation and reinforcement of steel structures such as large stadiums, industrial plants and bridges. It has broad prospects for industrial application. Attached Figure Description

[0015] Figure 1 The diagram shows a flowchart of the intelligent steel structure demolition method based on digital twin and reverse construction analysis of the present invention. Figure 2 The diagram shown is a block diagram of the real-time feedback control logic based on the cloud platform and Gen API of this invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0017] It should be noted that the acquisition of data and collection of information in this application are legal, compliant, or obtained with the consent of the subject of the data collection.

[0018] like Figures 1-2As shown, a smart steel structure demolition method based on digital twin and reverse construction analysis is implemented through the following software-coordinated steps: Phase 1: Construction of a High-Fidelity Reverse Digital Twin Model S101: Multi-source data acquisition and fusion modeling Autodesk ReCap was used to perform a 3D laser scan of the large-span steel structure to be demolished, generating a high-precision point cloud (accuracy better than ±3mm). Simultaneously, a total station was used to measure the coordinates of key control points, serving as the benchmark for point cloud stitching and model calibration.

[0019] In Autodesk Revit, using its Dynamo visual programming platform, scripts are written to drive the parametric family instantiation of structural columns, beams, and supports. Based on point clouds, the component axes and cross-sectional orientations are automatically fitted to generate a Level 350 As-BuiltBIM model. At the same time, the node connection information (bolts, welds) of the steel structure is embedded into the family instance through Revit's shared parameter function.

[0020] S102: Finite Element Model Transformation and Automatic Calibration for Analysis The MIDAS Gen for Revit plugin allows you to convert Revit models into MIDAS Gen analysis models with a single click. The plugin executes automatically. (1) Map the Revit structural frame to Gen beam elements; (2) Map the structural plate to plate units; (3) Transmitting material properties and loads.

[0021] In MIDAS Gen, run eigenvalue analysis to extract the first three modal frequencies f. model Simultaneously, wireless vibration sensors deployed on-site were used to collect the actual frequency f under environmental excitation. field Gen's design parameter adjustment function allows for fine-tuning of the material's elastic modulus or boundary spring stiffness until [f] is achieved. model -f field / f field <5%, complete model dynamics calibration.

[0022] Phase Two: Nonlinear Dynamic Simulation and Risk Quantification of the Demolition Process S201: Definition of Construction Phases Based on In-Software Objects In the MIDAS Gen construction phase analysis manager, create phase sequences CS0, CS1, CS2...CSn based on the demolition plan (such as block demolition or reverse demolition).

[0023] In each phase of CSi, the component groups to be dismantled and their associated surface loads are set to a passivated state via the group activation / passivation command. Simultaneously, temporary supports are simulated, activating their element properties during the installation phase and passivating them during the dismantling phase.

[0024] S202: Nonlinear Analysis and Stability Verification In the analysis control, select the geometric nonlinearity (large displacement) option to account for the P-Δ effect.

[0025] For each critical stage CSi, create a sub-load case, run buckling analysis, and calculate the minimum buckling load factor λ of the transient system. cri Establish stable and safe criteria: λ cri ≥2.5.

[0026] S203: Risk Indicator Extraction and Visualization After performing the analysis, use the results table function of MIDAS Gen to batch export all components for each stage: (1) Maximum combined stress σ max ; (2) Maximum displacement δ max ; (3) Stress ratio σ ratio =σ max / f y (f) y (yield strength).

[0027] Export the deformation results from MIDAS Gen in U3D format and link them to Autodesk Navisworks. Create a 4D construction simulation animation of the entire demolition process in Navisworks, visually demonstrating deformation development, stress cloud map changes, and highlighting of risky components.

[0028] Phase 3: Data-Driven Dynamic Monitoring and Feedback Control System S301: Monitoring System Deployment and Digital Threshold Setting Before on-site demolition work, vibrating wire strain gauges and robotic total station prisms were deployed in high-risk areas identified through simulation (such as areas with large deformation and high-stress members). Monitoring data was transmitted in real time to the cloud data platform via a 4G / 5G IoT gateway.

[0029] The theoretical displacement values ​​δ of each stage calculated by MIDAS Gen theory and strain value ε theory Write to the database as an early warning benchmark. Setting: Normal state (δ) measured≤ 0.7ε theory ); Huang Jing (δ) measured >0.7ε theory Orange Alert (δ)measured >0.9ε theory Red Alert (δ) measured ≥ε theory There are four states.

[0030] S302: Real-time Data Fusion and Model Reverse Update When the monitoring data triggers a yellow alert, the system automatically calls the MIDAS Gen API interface to update the current measured displacement δ. measured As a forced displacement load, it is applied to the model of the corresponding construction stage CSi for reverse analysis.

[0031] By comparing the measured reaction force with the model reaction force, the sensitivity analysis function of Gen is used to automatically identify and correct the model parameters most likely to deviate from reality, and update the digital twin model.

[0032] S303: Dynamic Optimization of Solutions and Issuance of Commands Based on the updated model, the system automatically and quickly resimulates the subsequent three demolition stages. If the predicted results are safe, the system sends a "continue construction" command to the site via external software; if the predicted risk increases, it sends a "pause" command along with a temporary reinforcement suggestion automatically generated by Gen.

[0033] All instructions, monitoring data, model versions, and early warning records are stored using blockchain technology to create an immutable digital archive of safe construction.

[0034] Test case Take the demolition project of a large-span irregular steel structure dome in a subway station as an example.

[0035] (1) Use Leica RTC360 to scan, process the point cloud in Context Capture, import it into Revit 2024, and use Dynamo script to automatically generate adaptive space frame components.

[0036] (2) The model was converted using the MIDAS Gen for Revit plugin. Five demolition stages were defined in Gen. During the simulation of demolishing CS3 (demolition of the large-span braced steel frame unit), the buckling analysis showed that the statically determinate system underwent stress changes, and λ cri =1.8, triggering a system warning.

[0037] (3) Scheme optimization: Before the demolition stage, a large-span load-distribution platform was added to the bottom of the dome; after a new full-process analysis, λ cri Increased to 3.2. At the same time, DT50 laser displacement gauges were installed at the support points.

[0038] (4) During actual construction up to CS5, the monitored displacement reached 85% of the theoretical value (triggering a red alert). The system automatically paused the operation and, through Gen API inversion analysis, found that the actual deformation of the rigid connection point between a steel frame unit and an adjacent unit was greater than the design due to corrosion. After updating the model, a re-simulation was conducted, suggesting the addition of a safety cable for lifting at the top (on-site implementation involved using a large crawler crane to assist in stabilizing the center of the structural unit throughout the entire process). After execution, subsequent construction monitoring data returned to the safe range.

[0039] This patent closely relies on common engineering software platforms such as Autodesk, Bentley, and MIDAS, and its data interfaces and analysis modules are all commercially mature technologies. The resulting system can be widely applied to the demolition, renovation, and reinforcement of steel structures in large stadiums, industrial plants, bridges, etc., significantly improving the handling of major risks. The level of digital management of operations has broad prospects for industrial application.

[0040] In addition, the present invention also provides an intelligent steel structure demolition system for realizing an intelligent steel structure demolition method based on digital twin and reverse construction analysis, comprising: The reverse digital twin model construction and calibration module is used to construct a high-precision parametric As-Built model in the building information modeling platform based on 3D laser scanning point cloud data and existing design drawings; the model is imported into finite element analysis software through a dedicated interface to generate a calculation model, and the calculation model is calibrated using field-measured dynamic characteristic data to establish a benchmark digital twin consistent with the physical structure; The demolition process dynamic simulation and risk quantification module is used in the finite element analysis software to reverse define a construction stage sequence containing multiple stages based on a preset demolition plan. It simulates the relevant actions of each demolition stage by setting the activation and deactivation states of the components. Under the condition of considering geometric nonlinearity, it performs construction stage analysis, synchronously calculates the displacement field, internal force redistribution and eigenvalue buckling stability coefficient of each stage, and extracts key risk quantification indicators. The monitoring system construction and early warning threshold setting module is used to deploy an IoT sensor network in high-risk areas identified by simulation results; the displacement and strain time history data predicted by simulation are set as dynamic early warning thresholds, and a cloud data platform is deployed for real-time data aggregation and visualization. The real-time feedback and dynamic closed-loop control module dynamically compares real-time monitoring data from the IoT sensor network with simulation prediction thresholds during demolition construction. When the data deviation exceeds the pre-set range, an early warning is triggered, and the application programming interface of the finite element analysis software is automatically invoked to update the boundary conditions or material parameters of the digital twin model with measured data. Based on the updated model, subsequent demolition steps are quickly re-simulated, dynamically optimized, and subsequent construction instructions are issued, forming a closed-loop control system of monitoring-simulation-decision-execution. Finally, it should be noted that the above embodiments are merely illustrative and explanatory of the present invention, and are not intended to limit the present invention to the scope of the described embodiments. Furthermore, those skilled in the art will understand that the present invention is not limited to the above embodiments, and many more variations and modifications can be made based on the teachings of the present invention, all of which fall within the scope of protection claimed by the present invention.

Claims

1. A method for intelligent demolition of steel structures based on digital twin and reverse construction analysis, characterized in that, Includes the following steps: Step S1: Based on the 3D laser scanning point cloud data and existing design drawings, construct a high-precision parametric As-Built model in the building information modeling platform; The model is imported into finite element analysis software through a dedicated interface to generate a calculation model. The calculation model is then calibrated using field-measured dynamic characteristic data to establish a benchmark digital twin consistent with the physical structure. Step S2: In the finite element analysis software, based on the preset demolition plan, a construction stage sequence containing multiple stages is defined in reverse. The relevant actions of each demolition stage are simulated by setting the activation and passivation states of the components. Under the condition of considering geometric nonlinearity, the construction stage analysis is performed, and the displacement field, internal force redistribution and eigenvalue buckling stability coefficient of each stage are calculated simultaneously to extract key risk quantification indicators. Step S3: Based on the high-risk areas identified by the simulation results, deploy an IoT sensor network; set the displacement and strain time history data predicted by the simulation as dynamic early warning thresholds, and deploy a cloud data platform for real-time data aggregation and visualization; Step S4: During the demolition process, the real-time monitoring data of the IoT sensor network is dynamically compared with the simulation prediction threshold. When the data deviation exceeds the preset range, an early warning is triggered and the application programming interface of the finite element analysis software is automatically called to update the boundary conditions or material parameters of the digital twin model with the measured data. Based on the updated model, the subsequent demolition steps are quickly re-simulated, the subsequent construction instructions are dynamically optimized and issued, forming a closed-loop control of monitoring-simulation-decision-execution.

2. The intelligent steel structure demolition method based on digital twin and reverse construction analysis according to claim 1, characterized in that, The high-precision parametric As-Built model constructed in step S1 is an as-built BIM model with a LOD of no less than 350. The construction process is as follows: use Bentley Context Capture or Autodesk ReCap software to process point cloud data, and in the Autodesk Revit platform, use Dynamo visual programming scripts to drive the instantiation of parametric structural families, automatically fit the geometry of components, and embed the attribute information of connection nodes.

3. The intelligent steel structure demolition method based on digital twin and reverse construction analysis according to claim 1 or 2, characterized in that, In step S1: The specific method for importing the model into the finite element analysis software through the dedicated interface is as follows: using the MIDAS Gen for Revit dedicated plugin, the steel structure frame and load information in the Revit model are automatically mapped and converted into beam and column elements and load cases in the MIDAS Gen software; the specific method for model calibration is as follows: comparing the modal frequencies calculated by MIDAS Gen with the vibration test frequencies of the field environment, and adjusting the model parameters until the error is less than 5%.

4. The intelligent steel structure demolition method based on digital twin and reverse construction analysis according to claim 1, characterized in that, In step S2: The specific method for considering the geometric nonlinearity of the construction stage analysis is as follows: enable the large displacement effect option in the analysis control of the MIDAS Gen software; the risk quantification index includes at least the maximum displacement δmax(i) of each stage, the maximum stress ratio of the component σratio(i) and the lowest order yield load factor λmin(i) of the structural system at that stage.

5. The intelligent steel structure demolition method based on digital twin and reverse construction analysis according to claim 1, characterized in that, Step S2 also includes a visualization pre-simulation step: the deformation and stress results of each stage calculated by MIDAS Gen are exported in IFC format and imported into Autodesk Navisworks software to generate a 4D demolition process simulation animation that integrates the time dimension, which is used for safety briefings and intuitive risk display of steel structure demolition.

6. The intelligent steel structure demolition method based on digital twin and reverse construction analysis according to claim 1, characterized in that, The dynamic warning thresholds mentioned in step S3 are set in three levels: the yellow warning threshold is 70% of the simulation prediction value, the orange warning threshold is 90%, and the red warning threshold is 100% or reaches the standard allowable limit; the warning information is simultaneously highlighted in the cloud data platform cockpit interface and the Navisworks4D model.

7. The intelligent steel structure demolition method based on digital twin and reverse construction analysis according to claim 1, characterized in that, The specific method for updating the digital twin model with measured data in step S4 is as follows: when the monitored displacement triggers an early warning, the measured displacement value δmeasured is applied as a forced displacement load to the corresponding construction stage CSI model through the API interface provided by the MIDAS Gen software. A reverse analysis is run once, and the software’s built-in sensitivity analysis tool is used to automatically identify and correct the parameters in the model that are most likely to deviate from reality by comparing the reaction force differences. The dynamic closed-loop control in step S4 also includes a blockchain evidence storage step: all construction instructions, real-time monitoring data, model correction records, early warning events and optimized solutions automatically issued by the system are encrypted and written into the blockchain network to generate an immutable digital archive of demolition project safety.

8. The intelligent steel structure demolition method based on digital twin and reverse construction analysis according to claim 1, characterized in that, Before constructing the high-precision parameterized As-Built model in step S1, a multi-source data acquisition step is also included: using a three-dimensional laser scanning device to scan the steel structure to be demolished on-site to generate point cloud data with an accuracy better than ±3mm, and simultaneously using a total station to measure the coordinates of key control points as a reference for point cloud stitching and model calibration.

9. The intelligent steel structure demolition method based on digital twin and reverse construction analysis according to claim 1, characterized in that, In step S2, the simulation of the dismantling stage also includes the simulation of temporary supports: the element properties of the temporary supports are activated during the installation stage and the element properties are deactivated during the dismantling stage; and the stability and safety criterion in the construction stage analysis is set as the minimum buckling load factor λcri of the structural system ≥ 2.

5.

10. A steel structure intelligent demolition system for implementing the steel structure intelligent demolition method based on digital twin and reverse construction analysis as described in any one of claims 1-9, characterized in that, include: The reverse digital twin model construction and calibration module is used to build a high-precision parametric As-Built model in the building information modeling platform based on 3D laser scanning point cloud data and existing design drawings. The model is imported into finite element analysis software through a dedicated interface to generate a calculation model. The calculation model is then calibrated using field-measured dynamic characteristic data to establish a benchmark digital twin consistent with the physical structure. The demolition process dynamic simulation and risk quantification module is used in the finite element analysis software to reverse define a construction stage sequence containing multiple stages based on a preset demolition plan. It simulates the relevant actions of each demolition stage by setting the activation and deactivation states of the components. Under the condition of considering geometric nonlinearity, it performs construction stage analysis, synchronously calculates the displacement field, internal force redistribution and eigenvalue buckling stability coefficient of each stage, and extracts key risk quantification indicators. The monitoring system construction and early warning threshold setting module is used to deploy an IoT sensor network in high-risk areas identified by simulation results; the displacement and strain time history data predicted by simulation are set as dynamic early warning thresholds, and a cloud data platform is deployed for real-time data aggregation and visualization. The real-time feedback and dynamic closed-loop control module is used to dynamically compare the real-time monitoring data of the IoT sensor network with the simulation prediction threshold during the demolition construction process. When the data deviation exceeds the preset range, an early warning is triggered and the application programming interface of the finite element analysis software is automatically called to update the boundary conditions or material parameters of the digital twin model with the measured data. Based on the updated model, the subsequent demolition steps are quickly re-simulated, the subsequent construction instructions are dynamically optimized and issued, forming a closed-loop control of monitoring-simulation-decision-execution.

Images