Chemical steel structure multi-module collaborative installation control method based on BIM and digital twinning

Abstract

This invention discloses a collaborative installation control method for multi-module chemical steel structures based on BIM and digital twins, belonging to the field of building construction control technology. This method addresses the technical problems existing in the installation of multi-module chemical steel structures, such as the reliance on discrete static measurements for installation accuracy control, difficulties in coordinating the progress of multiple modules, and the disconnect between the physical site and the digital model. The key points of the solution are: creating a digital twin model integrating multi-module design information based on Building Information Modeling (BIM); installing sensor groups on the modules to collect real-time position and attitude data; transmitting the data to a digital twin platform via a wireless network; fusing the data to generate a dynamic digital twin model; and dynamically determining the allowable threshold for position and attitude deviations based on the real-time installation stage of the modules, calculating the deviation between the real-time pose and the designed pose. This method is mainly used to achieve real-time, accurate, and collaborative control of the multi-module installation process of chemical steel structures.

Description

Technical Field

[0001] This invention relates to the field of building construction control technology. More specifically, this invention relates to a multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins. Background Technology

[0002] In the construction of facilities in the chemical and energy sectors, the construction mode of using prefabricated functional modules in the factory and then installing them as a whole on site has become increasingly common for large steel structures. While this multi-module construction method can increase the proportion of factory manufacturing, the on-site installation phase still faces a series of systemic technical challenges, which can be summarized into the following three interrelated problems: First, the precision control during installation relies on discrete, static measurements and experience-based decision-making, making it difficult to achieve systematic, adaptive, and precise control. Modular installation demands extremely high positional accuracy, especially in situations involving pipeline connections and the placement of large equipment. Traditional methods primarily rely on total stations and other surveying instruments for manual spot checks at key points, comparing the measurement data with two-dimensional drawings or static three-dimensional models. This method has significant limitations: First, the measurements are intermittent, failing to provide continuous trajectory tracking and real-time attitude awareness throughout the entire process of module hoisting, placement, and adjustment, resulting in control blind spots. Second, control standards are often static and uniform, such as using the same set of allowable deviation thresholds throughout the hoisting or adjustment process. However, the installation process itself is phased, from coarse aerial positioning to fine-tuning of the docking points, with each stage having drastically different tolerances for accuracy and actual requirements. Using fixed thresholds may be overly sensitive in the coarse-scale stage, leading to unnecessary frequent adjustments, while in the fine-tuning stage, the thresholds may be too wide to meet the final fixing requirements. This static control logic is incompatible with the dynamic construction process, resulting in low control efficiency. Furthermore, the final accuracy heavily depends on the technical skills and experience of the surveyors, making it difficult to guarantee quality stability.

[0003] Secondly, coordinating the progress of multiple modules during parallel installation is difficult due to the lack of dynamic scheduling and intervention mechanisms based on real-time status. Multiple modules are typically installed in parallel by different hoisting units on-site, often resulting in physical space interference, conflicts in hoisting resources (such as large cranes), and structural dependencies. Traditional progress management relies on pre-defined construction plans, but on-site conditions are complex and changeable. The installation of a single module may be delayed due to positioning difficulties, connector matching problems, or unforeseen circumstances. Because of the lack of precise and automatic perception of the real-time installation progress of each module, project managers often struggle to promptly identify the cascading effects of progress deviations. When a module lags behind, adjustments are usually only made passively afterward, potentially leading to waiting times for subsequent modules, resource idleness, or process conflicts, and even safety risks arising from modules already in place being in a temporary, fixed state for extended periods. The root of the problem lies in the fact that the "progress" of a module in traditional management is a macroscopic and vague concept, unable to be quantified into real-time, micro-parameters directly related to the current installation action. This lack of a data foundation for dynamic coordination results in planning adjustments that are both delayed and arbitrary.

[0004] Third, the lack of a real-time, accurate, and unified mapping between physical site information and digital models leads to a disconnect between the "virtual" and the real, resulting in unreliable decision-making basis. Although Building Information Modeling (BIM) technology has been widely applied in the design phase, in construction control, BIM models are mostly static design blueprints, failing to truly connect with dynamically changing physical entities. The main obstacles are: First, inconsistent coordinate systems. On-site surveying equipment (such as total stations and GPS) operates based on its own established measurement coordinate system, while BIM models exist in an independent design coordinate system. Comparing manually recorded coordinates of discrete measurement points with model coordinates requires tedious and error-prone manual conversions, and the on-site control points themselves may experience slight displacements due to environmental temperature and foundation settlement, further weakening the long-term reliability of the comparison benchmark. Second, dispersed and heterogeneous data sources. Key information such as location, attitude, and progress status may originate from different sensors and manual records, with varying formats, frequencies, and accuracies, lacking an effective fusion mechanism and failing to form a complete and consistent state description of modules on a unified platform. This disconnect between the virtual and the real world means that digital models cannot accurately reflect the actual situation on site, and the effectiveness of any simulations, predictions, or decisions based on the models is greatly reduced. Advanced digital technologies have failed to substantially empower real-time control of the installation process.

[0005] In summary, existing technologies face three interconnected core challenges in multi-module collaborative installation: mismatch between control logic and dynamic processes, lack of real-time data support for collaborative scheduling, and separation of physical and digital spatial information. The root of these challenges lies in the fact that existing methods fail to treat the installation process as a continuous system capable of real-time sensing, data-driven operation, and closed-loop control. The difficulty in solving these challenges lies in constructing a comprehensive technical framework that can automatically fuse multi-source heterogeneous data from the field, dynamically correlate it with a high-precision model, and intelligently make and execute decisions based on process status. This places extremely high demands on the real-time performance and accuracy of the data, the reliability of the system, and its engineering practicality. Summary of the Invention

[0006] One objective of this invention is to provide a collaborative installation control method for multi-module chemical steel structures based on BIM and digital twins, which is mainly used to achieve real-time, accurate and collaborative control of the multi-module installation process of chemical steel structures.

[0007] To achieve these objectives and other advantages of the present invention, according to one aspect of the present invention, a multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins is provided, comprising the following steps: S1. Based on the building information model of chemical steel structure, a digital twin model integrating multi-module design information is created in the digital twin platform. Sensor groups, including global positioning system sensors and attitude sensors, are installed on each prefabricated steel structure module in the factory. S2. The digital twin platform receives real-time position and attitude data collected by the sensor group, integrates pre-stored design information, and updates the model to form a dynamic digital twin model. S3. The platform determines the installation stage of each module based on its real-time position and signals from the crane and adjustment device. The installation stages include the hoisting and aerial movement stage, the preliminary positioning and adjustment stage, and the final fixing and welding stage. Based on the determined installation stage, the platform calls the preset position deviation allowable threshold and attitude deviation allowable threshold corresponding to that stage. At the same time, the dynamic digital twin model calculates the position deviation between the real-time position and the design position of each module, and the attitude deviation between the real-time attitude and the design attitude. S4. When the position deviation or attitude deviation exceeds the allowable threshold for position deviation or attitude deviation corresponding to the current installation stage, the platform generates an adjustment command and sends it to the corresponding installation equipment; the installation equipment coordinates its actions under the drive of the adjustment command to correct the module's position and attitude. S5. The platform uses the ratio of the number of installed connecting components to the total number of designed connecting components as the real-time installation progress of the module, and determines the progress coordination threshold according to the installation stage. When the difference between the real-time installation progress ratios of any two modules exceeds the progress coordination threshold set for the module with the earlier installation stage, the platform re-plans the subsequent installation sequence and generates coordination control instructions. The process involves repeatedly executing steps S2 to S4 until each module is installed; and continuously executing step S5 to collaboratively monitor and dynamically adjust the installation progress of multiple modules.

[0008] Preferably, the digital twin platform has a pre-installed automatic coordinate unification and data calibration module; On the main structural foundation at the installation site, at least four physical control points are set up, and a target is fixedly installed on each physical control point; in the digital twin model, model control points corresponding to the physical control points are created, and the three-dimensional coordinates of the target in the BIM design coordinate system are entered. Before the start of each construction cycle, a high-precision total station is used to measure the real-time three-dimensional coordinates of all targets in the field measurement coordinate system; the automatic coordinate unification and data calibration module calculates coordinate transformation parameters based on the coordinates of the same target in the BIM design coordinate system and the field measurement coordinate system. The automatic coordinate unification and data calibration module uses the coordinate transformation parameters to convert the real-time position data from the GPS sensors and automatic total station of the sensor group to the BIM design coordinate system. The automatic coordinate unification and data calibration module performs residual verification on the real-time position data after coordinate transformation. When the coordinate transformation residual of any target is greater than 3 mm to 5 mm, the calibration is deemed to have failed and an alarm is issued. The digital twin platform compares the real-time position data that has passed the residual verification and has been transformed to the BIM design coordinate system with the design position in the digital twin model.

[0009] Preferably, the digital twin platform is configured with a dynamic threshold management module; The dynamic threshold management module presets independent position deviation thresholds and attitude deviation thresholds for each installation stage. During the aerial movement phase of the hoisting, the position deviation threshold is 10 mm to 20 mm, and the attitude deviation threshold is 3° to 5°. For the initial positioning adjustment stage, the position deviation threshold is 5 mm to 10 mm, and the attitude deviation threshold is 1° to 2°; For the final fixed welding stage, the position deviation threshold is 2 mm to 5 mm, and the attitude deviation threshold is 0.5° to 1°. The digital twin platform determines the current installation stage of each module based on its real-time height, horizontal movement speed, and force status signals from the crane and adjustment device. After calculating the position and attitude deviations, the dynamic digital twin model calls the dynamic threshold management module to obtain a threshold that matches the current installation stage for judgment. When the module is in the hoisting and aerial movement phase, the progress coordination threshold is 30% to 40%. When the module is in the initial positioning and adjustment phase, the progress coordination threshold is 20% to 30%. When the module is in the final fixed welding stage, the progress coordination threshold is 10% to 15%.

[0010] Preferably, it also includes establishing a multi-source fusion positioning system; A positioning source management module is constructed within the digital twin platform, with its inputs connected to a GPS sensor, an automatic total station, and an ultra-wideband positioning base station. At least three ultra-wideband positioning base stations are deployed between the top and middle layers of the frame at the installation site, forming a network covering the hoisting path and the placement area. Ultra-wideband positioning tags are installed near the hoisting points and center of gravity of each steel structure module. The positioning source management module receives raw position data from the GPS sensor, the automatic total station, and the ultra-wideband positioning tags, and determines the signal obstruction level based on the real-time height of the module and the surrounding structural information in the digital twin model. When the module is in the open area at the top of the frame, the positioning source management module uses the fused Global Positioning System data as the main positioning source; When the module descends to the middle interlayer area of ​​the frame, the positioning source management module switches the main positioning source to ultra-wideband positioning data; When the module enters the designed positioning point and the ultra-wideband positioning tag is within the line of sight of the total station, the positioning source management module will switch the main positioning source to the automatic total station measurement data; The location source management module transmits the real-time location data of the selected primary location source to the digital twin platform.

[0011] Preferably, it also includes establishing an attitude data anti-interference processing procedure; The attitude sensor group installed on each steel structure module includes a microelectromechanical system inertial measurement unit and a dual-axis digital inclinometer; the angular velocity data output by the inertial measurement unit and the angle data output by the digital inclinometer are synchronously transmitted to the digital twin platform; The Kalman filter configured in the digital twin platform uses the angle data of the digital tilt meter as the observation vector and the angular velocity integral result of the inertial measurement unit as the state prediction vector to perform data fusion and output the real-time pitch angle and real-time roll angle. The digital twin platform performs a confidence check before calculating the attitude deviation. The check method is to compare the difference between the acceleration calculated by the inertial measurement unit and the gravitational acceleration component. When the difference exceeds 0.05g to 0.1g for 3s to 5s, the data is judged to be abnormal. When the data is abnormal, the digital twin platform switches to using only the data from the digital tiltmeter to calculate the attitude deviation and issues a sensor check command.

[0012] Preferably, it also includes the deployment of a hierarchical field industrial wireless network; The network comprises a core layer and multiple access layers; the core layer consists of industrial-grade wireless access points deployed at high points, forming a 5G dedicated network or high-speed wireless local area network covering the entire work area; the access layers consist of multiple IoT gateways, each IoT gateway being bound to a hoisting unit, the hoisting unit comprising a crane, the module to be installed, and adjustment devices. An IoT edge computing node is integrated within the sensor group of each steel structure module; the IoT edge computing node collects the raw sensor data of this module via wired connection and connects to the IoT gateway of this hoisting unit via a first wireless link; the IoT gateway accesses the core layer network via a second wireless link and uploads the data to the digital twin platform; the adjustment instructions and collaborative control instructions issued by the digital twin platform are sent to the device controller via the core layer network and the IoT gateway; The IoT edge computing node sets priority tags for data, marking location data and emergency alarm data as the highest priority, attitude data as the medium priority, and sensor status data as the low priority; the data queue manager in the IoT gateway schedules the uploaded data according to the priority tags.

[0013] Preferably, it also includes a human-machine collaborative instruction execution system; The system includes a first command terminal installed in the crane operator's cab and a second command terminal installed near the adjustment device. The first and second command terminals are connected to the digital twin platform via a wireless network. The adjustment commands generated by the digital twin platform are simultaneously sent to the crane controller, the first command terminal, the adjustment device controller, and the second command terminal. The adjustment commands include a positional deviation value and an adjustment vector. Both the first and second command terminals are equipped with a display screen and a confirmation button. The crane's movements are manually executed by the operator based on the adjustment vector displayed on the first command terminal. The adjustment device operates in two modes based on the type of instruction: for fine-tuning instructions with a displacement adjustment range of 2 mm to 10 mm and a current load pressure of 80% to 90% below its rated pressure, the adjustment device controller executes automatically; for instructions with a displacement adjustment exceeding 10 mm or a load pressure exceeding 80% to 90% above the rated pressure, the operator executes manually according to the adjustment vector displayed on the second instruction terminal; after the operator performs the manual operation, they must press the confirmation button on the corresponding terminal; upon receiving the confirmation signal from the terminal, the digital twin platform updates the execution status of the instruction.

[0014] Preferably, an installation sequence dynamic planning module is built within the digital twin platform; The installation sequence dynamic planning module is pre-set with a rule base containing rules for resource conflict, structural dependency, and hoisting path conflict. When the difference between the real-time installation progress ratios of any two modules exceeds the current progress coordination threshold, the installation sequence dynamic planning module is triggered. The installation sequence dynamic planning module filters all uninstalled modules according to the rule base and identifies associated modules that have resource conflicts, structural dependencies, or hoisting path conflicts with the currently lagging module. The installation sequence dynamic planning module calculates a first adjustment scheme and a second adjustment scheme. The first adjustment scheme is to postpone the installation sequence of related modules, and the second adjustment scheme is to swap the installation sequence of the delayed module and the next preparation module. The installation sequence dynamic planning module simulates the execution process of the first adjustment scheme and the second adjustment scheme based on a digital twin model, and predicts the overall project duration extension caused by each scheme. The installation sequence dynamic planning module selects the scheme with the smaller overall project duration extension as the recommended scheme, generates a collaborative control instruction containing the new installation sequence, and sends it to the instruction terminal of the corresponding hoisting unit.

[0015] Preferably, it also includes establishing a control point stability monitoring and automatic calibration mechanism; At each physical control point deployed on the main foundation of the structure, a target and a high-precision displacement sensor are installed. The measurement direction of the displacement sensor is consistent with the foundation settlement direction. The displacement sensor is connected to the automatic coordinate unification and data calibration module via a wireless network. The automatic coordinate unification and data calibration module continuously receives the real-time displacement data from each displacement sensor and compares it with the initial value. The automatic coordinate unification and data calibration module sets a displacement change threshold ranging from 0.5 mm to 2 mm. When the change in displacement sensor data at any physical control point exceeds the displacement change threshold, the automatic coordinate unification and data calibration module determines that the coordinates of that control point are invalid. The automatic coordinate unification and data calibration module sends an alarm to the digital twin platform and suspends the use of coordinate transformation parameters related to the control point. The digital twin platform schedules surveyors to use a high-precision total station to remeasure the control point whose coordinates have failed, and obtain new field measurement coordinates. The automatic coordinate unification and data calibration module uses the newly measured coordinates and the coordinates of other unfailed physical control points to recalculate the coordinate transformation parameters. The automatic coordinate unification and data calibration module performs residual verification on the recalculated coordinate transformation parameters, and resumes data transformation using the updated parameters after the verification passes. The total number of physical control points deployed on the main foundation of the structure shall not be less than six, including at least two backup control points; when the number of failed control points results in fewer than four remaining effective control points, the system shall automatically enable the coordinates of the backup control points to participate in the calculation of the coordinate transformation parameters.

[0016] Preferably, in step S3, the digital twin platform determines the installation stage of the module based on the following specific conditions: when the real-time height of the module is greater than 10 m and the horizontal moving speed is greater than 0.1 m / s, it is determined to be in the hoisting and aerial movement stage; when the real-time height of the module is greater than 1.5 m and less than or equal to 10 m, the horizontal moving speed is less than or equal to 0.1 m / s, and the pressure signal received from the adjustment device is greater than 5 MPa, it is determined to be in the preliminary positioning and adjustment stage; when the real-time height of the module is less than or equal to 1.5 m and the first fastener installation completion signal is received from the module connection node, it is determined to be in the final fixing and welding stage. In step S5, the number of installed connection components of the module is automatically obtained by sensors deployed at each connection node of the module; when the sensor detects that the connection component is installed in place, it sends a signal, and the digital twin platform receives and accumulates the signal to calculate the real-time installation progress.

[0017] This invention offers at least the following beneficial effects: The BIM- and digital twin-based multi-module collaborative installation control method for chemical steel structures described in this invention generates a series of interconnected and progressively beneficial effects by constructing a digital twin-based collaborative installation control system. First, by establishing a synchronized virtual and real dynamic digital twin model and setting phased dynamic thresholds, adaptive continuous control of the module installation process is achieved, moving from coarse to fine, effectively improving installation accuracy and first-pass yield while reducing absolute reliance on operator experience. Second, by introducing a real-time installation progress quantification mechanism and dynamic collaborative rules, this method enables timely perception and quantification of progress deviations during parallel construction of multiple modules, providing a clear basis for proactive intervention and scheduling optimization, helping to alleviate process conflicts and resource idleness, and ensuring overall construction efficiency. Third, by deploying a multi-source fusion positioning system and attitude anti-interference processing flow, the method overcomes signal obstruction and measurement noise problems in complex site environments, ensuring the continuity and reliability of pose perception data throughout the entire process from hoisting to placement, laying a solid data foundation for precise control. Fourth, by constructing a hierarchical industrial wireless network and setting data transmission priorities, the real-time, orderly, and reliable transmission of massive amounts of sensor data and control commands in the field environment is ensured, meeting the stringent requirements of closed-loop control for the timeliness and stability of information interaction. Fifth, by configuring a human-machine collaborative command execution system, necessary manual confirmation and manual operation interfaces are retained in key links, enabling advanced automatic control logic to be safely and flexibly adapted to existing construction equipment and process habits, enhancing the engineering practicality and transition adaptability of the technical solution. Sixth, by integrating an installation sequence dynamic planning module, feasible rescheduling schemes can be quickly generated based on preset rules and simulation when progress is unbalanced, transforming delayed passive responses into proactive optimizations, improving the overall coordination and predictability of multi-line construction. Finally, by establishing a control point stability monitoring and automatic calibration mechanism, the long-term accuracy of the transformation parameters between the field measurement coordinate system and the design model coordinate system can be continuously maintained, ensuring the stability and reliability of the entire system's spatial reference from the source, providing a fundamental guarantee for long-cycle, high-precision construction. These effects work together to form a complete closed loop from data perception and intelligent decision-making to precise execution, significantly improving the overall quality control level, construction collaboration efficiency, and process management capabilities of multi-module installation of chemical steel structures.

[0018] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0019] Figure 1 This is a flowchart of the multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in this invention. Detailed Implementation

[0020] The present invention will now be described in further detail with reference to specific embodiments, so that those skilled in the art can implement it based on the description.

[0021] It should be understood that terms such as “having,” “comprising,” and “including” as used herein do not exclude the presence or addition of one or more other elements or combinations thereof.

[0022] It should be noted that, unless otherwise specified, the experimental methods described in the following implementation plan are all conventional methods, and the reagents and materials described are all commercially available unless otherwise specified.

[0023] In the description of this invention, the terms "lateral", "longitudinal", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", and "outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0024] like Figure 1 As shown, this invention provides a multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins, including the following steps: S1. Based on the building information model of chemical steel structure, a digital twin model integrating multi-module design information is created in the digital twin platform. Sensor groups, including global positioning system sensors and attitude sensors, are installed on each prefabricated steel structure module in the factory. S2. The digital twin platform receives real-time position and attitude data collected by the sensor group, integrates pre-stored design information, and updates the model to form a dynamic digital twin model. S3. The platform determines the installation stage of each module based on its real-time position and signals from the crane and adjustment device. The installation stages include the hoisting and aerial movement stage, the preliminary positioning and adjustment stage, and the final fixing and welding stage. Based on the determined installation stage, the platform calls the preset position deviation allowable threshold and attitude deviation allowable threshold corresponding to that stage. At the same time, the dynamic digital twin model calculates the position deviation between the real-time position and the design position of each module, and the attitude deviation between the real-time attitude and the design attitude. S4. When the position deviation or attitude deviation exceeds the allowable threshold for position deviation or attitude deviation corresponding to the current installation stage, the platform generates an adjustment command and sends it to the corresponding installation equipment; the installation equipment coordinates its actions under the drive of the adjustment command to correct the module's position and attitude. S5. The platform uses the ratio of the number of installed connecting components to the total number of designed connecting components as the real-time installation progress of the module, and determines the progress coordination threshold according to the installation stage. When the difference between the real-time installation progress ratios of any two modules exceeds the progress coordination threshold set for the module with the earlier installation stage, the platform re-plans the subsequent installation sequence and generates coordination control instructions. The process involves repeatedly executing steps S2 to S4 until each module is installed; and continuously executing step S5 to collaboratively monitor and dynamically adjust the installation progress of multiple modules.

[0025] In the above technical solution, a digital twin model integrating multi-module design information is created based on the building information model of the chemical steel structure. The multi-module is a prefabricated steel structure functional module in the factory. A sensor group is installed on each module, which includes a global positioning system sensor and an attitude sensor.

[0026] In practical implementation, commercially available Building Information Modeling (BIM) software can be used for the 3D design of the chemical steel structure. The digital twin model can be created based on this design model and imported into a dedicated digital twin platform via a data interface. The GPS sensor in the sensor array can be a measurement receiver supporting real-time dynamic differential functionality, and the attitude sensor can be a dual-axis digital inclinometer. These sensors can be fixed to the main beam of the prefabricated steel structure module or a rigid component near the center of gravity using mounting brackets and bolted connections, ensuring no relative movement between the sensors and the main structure of the module. Regarding materials, the mounting brackets can be made of ordinary carbon steel, and the sensor housings are typically made of engineering plastics or metal alloys.

[0027] The real-time position and attitude data collected by the sensor group are transmitted to the digital twin platform via a wireless network. The digital twin platform integrates the real-time data with pre-stored design information and updates the model state to generate a dynamic digital twin model. Based on the dynamic digital twin model, the digital twin platform dynamically determines the corresponding position deviation allowable threshold and attitude deviation allowable threshold for each module during the real-time installation stage. The digital twin platform calculates the position deviation between the real-time position and the designed position of each module, as well as the attitude deviation between the real-time attitude and the designed attitude.

[0028] In practical implementation, industrial-grade wireless access points can be deployed to form wireless local area network coverage on-site. The digital twin platform can be deployed on a site server or a cloud server. After receiving sensor data, the platform first decodes and transforms the data, converting the geodetic coordinates of the GPS sensor into three-dimensional coordinates in the design coordinate system consistent with the building information model. Based on the module's real-time height, speed, and working status signals from the crane control system, the platform automatically determines whether the module is in the hoisting aerial movement, preliminary positioning adjustment, or final fixing and welding stage. To ensure the stability of control commands and prevent frequent threshold changes caused by short-term signal fluctuations switching back and forth at the installation stage boundaries, the digital twin platform monitors the stability of the stage determination results: when the determination result changes, the platform maintains the threshold corresponding to the original stage for a preset stable time window (e.g., 30 seconds). Only when the new stage determination result remains stable within this time window will the platform officially switch and apply the deviation allowance threshold and progress coordination threshold corresponding to the new stage.

[0029] When the position deviation exceeds the allowable threshold for position deviation corresponding to the current installation stage, or when the attitude deviation exceeds the allowable threshold for attitude deviation corresponding to the current installation stage, the platform generates an adjustment command. The adjustment command is sent to the installation equipment, where the crane and adjustment device work together to correct the module's attitude. The digital twin platform uses the ratio of the number of installed connected components to the total number of designed connected components as the real-time installation progress of the module, and dynamically determines the progress coordination threshold based on the installation stage. When the difference between the real-time installation progress ratios of any two modules exceeds the progress coordination threshold corresponding to the current installation stage, the platform re-plans the subsequent installation sequence and generates a coordination control command.

[0030] In practice, the adjustment instructions generated by the platform can include the direction and distance vectors to be adjusted. These instructions are sent via network to the display terminal and the controller of the adjustment device installed in the crane cab. For example, when the positional deviation exceeds a threshold, the instruction may instruct the crane to fine-tune the trolley or carriage, while simultaneously instructing the hydraulic jacks at the bottom of the module to lift or move it. Real-time installation progress can be obtained by installing simple contact sensors at the module connection nodes or by having installation workers scan the component's QR code using a handheld terminal to confirm installation completion and send a count signal to the platform. The progress coordination threshold can be set to 35% during the aerial movement phase and 25% during the initial positioning and adjustment phase. When the platform detects that the progress difference between two modules exceeds this threshold, it will activate a built-in scheduling algorithm, such as one based on priority rules or heuristic algorithms, to simulate and calculate the impact of adjusting the lifting sequence of the subsequent two to three modules on the overall project duration, select the option with the smaller delay, and generate a new task sequence to be issued to the corresponding lifting team.

[0031] This implementation method achieves adaptive and continuous control of the installation accuracy of steel structure modules by constructing a dynamic digital twin model synchronized with the physical installation process and dynamically setting differentiated control thresholds according to different installation stages, thereby improving the first-pass yield and efficiency of installation. By quantifying and monitoring the micro-installation progress of each module in real time and setting dynamic progress coordination thresholds, the method can automatically trigger optimization adjustments to the installation sequence in the early stages of progress imbalance, reducing waiting times and resource conflicts in subsequent processes caused by delays in individual modules, and enhancing the overall coordination and executability of the plan for parallel construction of multiple modules. The entire method forms a closed loop from real-time perception and intelligent analysis to proactive control, reducing over-reliance on the personal experience of operators and improving the controllability and standardization of the construction process.

[0032] In other technical solutions, the digital twin platform has a pre-built automatic coordinate unification and data calibration module; On the main structural foundation at the installation site, at least four physical control points are set up, and a target is fixedly installed on each physical control point; in the digital twin model, model control points corresponding to the physical control points are created, and the three-dimensional coordinates of the target in the BIM design coordinate system are entered. Before the start of each construction cycle, a high-precision total station is used to measure the real-time three-dimensional coordinates of all targets in the field measurement coordinate system; the automatic coordinate unification and data calibration module calculates coordinate transformation parameters based on the coordinates of the same target in the BIM design coordinate system and the field measurement coordinate system. The automatic coordinate unification and data calibration module uses the coordinate transformation parameters to convert the real-time position data from the GPS sensors and automatic total station of the sensor group to the BIM design coordinate system. The automatic coordinate unification and data calibration module performs residual verification on the real-time position data after coordinate transformation. When the coordinate transformation residual of any target is greater than 3 mm to 5 mm, the calibration is deemed to have failed and an alarm is issued. The digital twin platform compares the real-time position data that has passed the residual verification and has been transformed to the BIM design coordinate system with the design position in the digital twin model.

[0033] In the above technical solution, the implementation of the automatic coordinate unification and data calibration mechanism begins with on-site measurement and control point layout. On the main structural foundation at the installation site, four stable locations with good visibility are selected, such as the center of a concrete foundation or permanent column base. Four physical control points are precisely laid out using surveying instruments. At each control point, a high-precision spherical prism is fixedly installed as a target using a forced-centering base. In the corresponding BIM model on the digital twin platform, operators create four virtual model control points at the same coordinate positions according to the design drawings, and accurately input the theoretical three-dimensional coordinate values ​​of this set of targets in the project's unified BIM design coordinate system, in meters (m).

[0034] The coordinate data acquisition and transformation process is as follows. At each construction cycle, such as before the start of each day or during major process transitions, surveyors use a surveying robot or a high-precision automatic total station to sequentially and accurately measure the three-dimensional coordinates of the targets at four physical control points, using arbitrary on-site station setups, obtaining their measured values ​​in the on-site measurement coordinate system. This measured data is uploaded to the digital twin platform. The platform's automatic coordinate unification and data calibration module then activates. Using a least-squares algorithm, based on four sets of coordinate pairs for the same set of targets in two coordinate systems (BIM design coordinate system and on-site measurement coordinate system), it calculates a seven-parameter coordinate transformation model containing three translation parameters, three rotation parameters, and one scale parameter.

[0035] This module runs in the background in real time, uniformly converting all subsequently connected location data sources. Whether it's the WGS-84 coordinates from the GPS sensors on the module or the polar coordinate measurements of the targets on the module from an automatic total station, these seven parameters are immediately invoked and converted to a unified BIM design coordinate system, with the coordinate unit uniformly in mm. To ensure the long-term reliability of the converted model, the module performs residual verification after each application of the conversion parameters. During verification, the module uses the current parameters to back-calculate the measured coordinates of the four physical control points back to the BIM design coordinate system, then compares them with the theoretical coordinates entered in the model to calculate the positional deviation of each point.

[0036] When the system determines that the residual of any control point exceeds the preset 4 mm threshold, the automatic coordinate unification and data calibration module immediately determines that the current coordinate transformation relationship is invalid. The module then sends a prominent visual and audible alarm to the main interface of the digital twin platform and automatically suspends all real-time deviation calculations and instruction generation processes that rely on this transformation relationship. The platform will prompt the surveyors to review the control points involved in the alarm or remeasure all control points. Only after obtaining new measured coordinates, recalculating, and passing residual verification will the system resume operation using the new transformation parameters. All subsequent comparisons and calculations on the digital twin platform strictly use coordinate data verified through this process and located in the unified BIM design coordinate system, thereby ensuring the consistency of spatial benchmarks and the comparability of data from the source.

[0037] In some other technical solutions, the digital twin platform is configured with a dynamic threshold management module; The dynamic threshold management module presets independent position deviation thresholds and attitude deviation thresholds for each installation stage. During the aerial movement phase of the hoisting, the position deviation threshold is 10 mm to 20 mm, and the attitude deviation threshold is 3° to 5°. For the initial positioning adjustment stage, the position deviation threshold is 5 mm to 10 mm, and the attitude deviation threshold is 1° to 2°; For the final fixed welding stage, the position deviation threshold is 2 mm to 5 mm, and the attitude deviation threshold is 0.5° to 1°. The digital twin platform determines the current installation stage of each module based on its real-time height, horizontal movement speed, and force status signals from the crane and adjustment device. After calculating the position and attitude deviations, the dynamic digital twin model calls the dynamic threshold management module to obtain a threshold that matches the current installation stage for judgment. When the module is in the hoisting and aerial movement phase, the progress coordination threshold is 30% to 40%. When the module is in the initial positioning and adjustment phase, the progress coordination threshold is 20% to 30%. When the module is in the final fixed welding stage, the progress coordination threshold is 10% to 15%.

[0038] In the above technical solution, the dynamic threshold management module can serve as a parameter management and logic judgment unit within the digital twin platform software. This module is defined and configured during system initialization. It explicitly defines three sequential installation stages: the hoisting and aerial movement stage, the preliminary positioning and adjustment stage, and the final fixing and welding stage. For each stage, the module presets a set of independent position deviation threshold and attitude deviation threshold parameters.

[0039] The platform determines the current installation stage of the module through a logical process based on multi-source signal fusion. The digital twin platform continuously analyzes altitude and speed data from GPS sensors. When the module is at a high altitude and its horizontal movement speed is greater than 0.1 m / s, it is tended to be classified as being in the hoisting and aerial movement stage. When the module's height approaches the design elevation and its horizontal movement speed decreases significantly, and a pressure sensor signal from the adjustment device is received, indicating that it has begun to be subjected to force for fine-tuning, the platform determines that it has entered the preliminary positioning and adjustment stage. When the platform receives signals from the contact sensors or image recognition system at the module's connection nodes indicating "first bolt tightening" or "spot welding begins," the platform determines that the module has entered the final fixing and welding stage. These sensor signals from the crane, adjustment device, and module body collectively constitute the basis for stage determination.

[0040] After calculating the position and attitude deviations of the modules in real time, the digital twin platform immediately invokes the dynamic threshold management module. The platform queries this module for the position and attitude deviation thresholds corresponding to the currently determined stage. Subsequently, the platform compares the calculated real-time deviation value with the queried threshold. Regarding progress coordination, the platform calculates the installation progress ratio of each module in real time. When comparing any two modules, the platform makes a judgment based on the progress coordination threshold corresponding to the earlier stage of the two modules. If the difference in progress ratio exceeds this threshold, a replanning process is triggered. In this way, the control standards are dynamically matched with the actual progress of the installation operation, allowing control to maintain a certain degree of fault tolerance in the roughing-out stage to improve efficiency, while imposing strict standards in the precision stage to ensure final quality. The dynamic threshold management module has a stage judgment lag interval. When the real-time parameters of a module are in the critical interval between two installation stages, the threshold of the previous installation stage is maintained until all judgment parameters continuously and stably enter the next stage for more than a preset time (e.g., 10 seconds), at which point the threshold of the next stage is switched. The determination of the installation phase adopts a multi-parameter weighted voting algorithm and sets a minimum stable time window (such as 15 seconds) to avoid frequent phase switching caused by sensor noise.

[0041] Other technical solutions also include establishing a multi-source fusion positioning system; A positioning source management module is constructed within the digital twin platform, with its inputs connected to a GPS sensor, an automatic total station, and an ultra-wideband positioning base station. At least three ultra-wideband positioning base stations are deployed between the top and middle layers of the frame at the installation site, forming a network covering the hoisting path and the placement area. Ultra-wideband positioning tags are installed near the hoisting points and center of gravity of each steel structure module. The positioning source management module receives raw position data from the GPS sensor, the automatic total station, and the ultra-wideband positioning tags, and determines the signal obstruction level based on the real-time height of the module and the surrounding structural information in the digital twin model. When the module is in the open area at the top of the frame, the positioning source management module uses the fused Global Positioning System data as the main positioning source; When the module descends to the middle interlayer area of ​​the frame, the positioning source management module switches the main positioning source to ultra-wideband positioning data; When the module enters the designed positioning point and the ultra-wideband positioning tag is within the line of sight of the total station, the positioning source management module will switch the main positioning source to the automatic total station measurement data; The location source management module transmits the real-time location data of the selected primary location source to the digital twin platform.

[0042] In the above technical solution, the establishment of the multi-source fusion positioning system first involves the deployment of hardware devices and the configuration of software modules. Within the digital twin platform, a positioning source management software module can be built using software development tools. The software interface of this module connects to the receiving terminals of three types of physical positioning devices via wired or wireless network protocols: the GPS sensor receiver installed on the module, the automatic total station control unit deployed on-site, and the base station network controller covering the entire ultra-wideband positioning system on-site.

[0043] For hardware deployment, three to four ultra-wideband (UWB) positioning base stations are deployed on the top of the chemical steel structure frame at the installation site in unobstructed areas, such as the parapet wall of the highest floor or on specially erected poles. Simultaneously, two to three additional base stations are deployed in open locations on the equipment floor or walkway platform in the middle of the frame to ensure that the base station network signal can cover all potential paths from the high-altitude hoisting to the placement points on each floor. These base stations are typically bolted to pre-welded supports, which can be made of angle steel. For each steel structure module to be installed, an UWB positioning tag is installed near the upper lifting lug and at the lower part near the theoretical center of gravity. The tags can be temporarily fixed using magnetic bases or bolts. In terms of materials, the base station and tag housings are typically made of reinforced plastic or aluminum alloy to balance strength and wireless signal penetration.

[0044] The system operates as follows: The positioning source management module continuously receives latitude, longitude, and elevation data from GPS sensors, 3D coordinate data of module tags from the UWB (Ultra-Wideband) positioning system, and tracking measurement data of targets on specific modules from an automatic total station. The module has a pre-defined set of logical judgment rules: it reads the module's height data in real time and retrieves structural models (such as the distribution density of beams, columns, and slabs) around that height plane from the digital twin model to comprehensively assess the signal obstruction level. Based on these judgments, the module performs positioning source switching: in "open areas," differentially corrected GPS data is prioritized and can be weighted and fused with UWB data as the final position output; in "signal-obstructed areas," degraded GPS data is discarded, and the primary positioning source is switched to pure UWB positioning data; in "precision measurement usable areas," the primary positioning source is switched to more accurate automatic total station measurement data. The positioning source management module pushes the finally selected and verified primary positioning source coordinate data to the core processing unit of the digital twin platform in real time for subsequent deviation calculation. The entire switching process is designed to be smooth to ensure the continuity of position tracking.

[0045] Other technical solutions also include establishing an anti-interference processing flow for attitude data; The attitude sensor group installed on each steel structure module includes a microelectromechanical system inertial measurement unit and a dual-axis digital inclinometer; the angular velocity data output by the inertial measurement unit and the angle data output by the digital inclinometer are synchronously transmitted to the digital twin platform; The Kalman filter configured in the digital twin platform uses the angle data of the digital tilt meter as the observation vector and the angular velocity integral result of the inertial measurement unit as the state prediction vector to perform data fusion and output the real-time pitch angle and real-time roll angle. The digital twin platform performs a confidence check before calculating the attitude deviation. The check method is to compare the difference between the acceleration calculated by the inertial measurement unit and the gravitational acceleration component. When the difference exceeds 0.05g to 0.1g for 3s to 5s, the data is judged to be abnormal. When the data is abnormal, the digital twin platform switches to using only the data from the digital tiltmeter to calculate the attitude deviation and issues a sensor check command.

[0046] In the above technical solution, the implementation of the attitude data anti-interference processing flow first requires the configuration of hardware and software. The attitude sensor group installed on each steel structure module can include a commercially available microelectromechanical system (MEMS) inertial measurement unit (IMU) module and an independent dual-axis digital inclinometer module. In implementation, these two sensors can be integrated onto a single circuit board or installed side-by-side as two independent devices via rigid connectors. During installation, they need to be bolted to a selected rigid reference surface on the main structure of the module, such as the flange of the module's main beam or the surface of a reinforced node plate, using a mounting base plate to ensure that the sensor axis is aligned with the module's structural axis and to minimize measurement errors caused by structural deformation. The mounting base plate can be made of stainless steel or aluminum, and the sensor housing is typically made of plastic or metal.

[0047] The process works as follows: The microelectromechanical system (MEMS) inertial measurement unit (IMSU) mounted on the module outputs raw data of triaxial angular velocity and triaxial acceleration at a relatively high frequency, such as 100 Hz. The dual-axis digital inclinometer outputs tilt angle data in both pitch and roll directions at a relatively low frequency, such as 10 Hz. An integrated data acquisition and transmission unit reads the data from both sensors via a wired connection, timestamps and synchronizes them, and then transmits these synchronized data packets to the digital twin platform via a wireless network.

[0048] Within the digital twin platform, a Kalman filter is configured in the software. This filter uses the real-time angular velocity data output by the microelectromechanical system's inertial measurement unit (MEMS) to predict the module's attitude at the next moment through integration. Simultaneously, it uses the less noisy but potentially dynamically lagging angle data output by the dual-axis digital inclinometer as observations of the predicted state. The filter continuously performs optimal weighted fusion of the predicted and observed attitude angles through its algorithm. Specifically, during the dynamic lifting process, the filter relies more on the rapid response of the inertial measurement unit; during the relatively static fine-tuning phase, it relies more on the absolute accuracy of the digital inclinometer. After fusion calculation, the filter outputs smoothed and denoised real-time pitch and roll angles for subsequent deviation calculations.

[0049] Before performing formal attitude deviation calculations, the platform first verifies the confidence level of the fused attitude data. This verification is done using cross-validation based on the redundancy information of the inertial measurement unit (IMU). The platform calculates the currently perceived gravitational acceleration component based on the real-time triaxial acceleration vectors measured by the IMU, combined with the fused attitude angles. Theoretically, this component should be approximately 1g (9.8 m / s²). 2During implementation, the platform continuously calculates the absolute value of the difference between the calculated gravitational acceleration and 1g. When this difference consistently exceeds 0.08g and lasts for 4 seconds, the platform determines that the data from the microelectromechanical system's inertial measurement unit may have severe drift or malfunction, leading to a decrease in the confidence level of the fusion result. Once an anomaly is detected, the platform automatically switches the data processing logic, temporarily stopping the use of the Kalman filter fusion result and instead using only the raw angle data output by the slightly slower but more stable digital inclinometer for attitude deviation calculation. Simultaneously, the platform generates a sensor check command and sends it to the handheld terminal of the on-site maintenance personnel, prompting them to check the sensor group of the specified module.

[0050] Other technical solutions also include deploying hierarchical field industrial wireless networks; The network comprises a core layer and multiple access layers; the core layer consists of industrial-grade wireless access points deployed at high points, forming a 5G dedicated network or high-speed wireless local area network covering the entire work area; the access layers consist of multiple IoT gateways, each IoT gateway being bound to a hoisting unit, the hoisting unit comprising a crane, the module to be installed, and adjustment devices. An IoT edge computing node is integrated within the sensor group of each steel structure module; the IoT edge computing node collects the raw sensor data of this module via wired connection and connects to the IoT gateway of this hoisting unit via a first wireless link; the IoT gateway accesses the core layer network via a second wireless link and uploads the data to the digital twin platform; the adjustment instructions and collaborative control instructions issued by the digital twin platform are sent to the device controller via the core layer network and the IoT gateway; The IoT edge computing node sets priority tags for data, marking location data and emergency alarm data as the highest priority, attitude data as the medium priority, and sensor status data as the low priority; the data queue manager in the IoT gateway schedules the uploaded data according to the priority tags.

[0051] In the above technical solution, the deployment of the hierarchical on-site industrial wireless network is a systematic project. Its implementation begins with network planning and hardware preparation. The core layer can be constructed by deploying a dedicated network based on 5G technology or a high-bandwidth, low-latency industrial wireless LAN. During implementation, industrial-grade wireless access points or 5G small base stations conforming to relevant standards can be selected as core equipment. These devices need to be installed at high points on the construction site, such as the top of existing tower cranes, the outer perimeter of the highest level of completed structures, or on specially erected communication poles, to ensure that the signal can overlook and cover the entire hoisting operation area, reducing blind spots. These access points are connected to the central computer room or cloud gateway of the deployed digital twin platform via fiber optic or high-bandwidth wireless backhaul links.

[0052] The implementation of the access layer is tightly coupled with the hoisting operation unit. For each independent hoisting operation unit on site (typically comprising a crane, a module currently being hoisted, and associated adjustment devices), an industrial IoT gateway is configured. This gateway can be installed in the crane's cab or in a waterproof electrical control box on the crane body. Simultaneously, an IoT edge computing node needs to be integrated into the sensor group of each steel structure module. This node is an embedded hardware integrating a microprocessor, local storage, and wireless communication module, which can be directly connected to the module's GPS sensors, attitude sensors, etc., via cables. This edge node communicates with the hoisting unit's IoT gateway via a first wireless link, which can be a high-speed, short-range wireless connection based on Wi-Fi 6 or a proprietary protocol, forming a small-scale device subnet.

[0053] The network workflow is as follows: Raw data generated by various sensors installed on the module is first collected in real time by the integrated IoT edge computing nodes via wired interfaces. The edge nodes perform preliminary timestamp alignment and format encapsulation on the raw data, and assign priority tags based on data type. For example, its real-time 3D coordinate data and emergency out-of-tolerance alarms are marked as highest priority; the module's pitch and roll angle data are marked as medium priority; while the sensor's own temperature, voltage, and other status monitoring data are marked as low priority. Subsequently, the edge nodes send the packaged data to the IoT gateway of the hoisting unit they are bound to via the first wireless link.

[0054] The IoT gateway runs data queue management software. This manager continuously receives data from its subordinate edge nodes, as well as from the controllers of cranes and adjustment devices. It sorts and schedules these data packets based on their inherent priority tags. When sending to the core network, the manager ensures that all "highest priority" packets are placed in the transmission queue first, possibly with shorter transmission intervals, while "low priority" data may be sent when the network is idle or allowed some buffering. The gateway uploads the scheduled data stream to a remote digital twin platform server via its second wireless link (i.e., the link connecting to the core 5G or high-speed wireless LAN).

[0055] The command flow in the opposite direction is also transmitted through this network architecture. Adjustment or collaborative control commands generated by the digital twin platform are first sent to the IoT gateway of the target hoisting unit via the core layer network. Upon receiving the command, the gateway forwards it to the corresponding crane controller, adjustment device controller, or module edge node (for display or confirmation) based on the device identifier contained in the command. This layered, bound network structure enables the orderly aggregation of massive amounts of sensor data and the precise distribution of control commands, ensuring the real-time requirements of the control system and effectively managing wireless channel congestion. The deployment of the entire network enhances the communication reliability of the system in complex industrial environments.

[0056] Other technical solutions also include configuring a human-machine collaborative instruction execution system; The system includes a first command terminal installed in the crane operator's cab and a second command terminal installed near the adjustment device. The first and second command terminals are connected to the digital twin platform via a wireless network. The adjustment commands generated by the digital twin platform are simultaneously sent to the crane controller, the first command terminal, the adjustment device controller, and the second command terminal. The adjustment commands include a positional deviation value and an adjustment vector. Both the first and second command terminals are equipped with a display screen and a confirmation button. The crane's movements are manually executed by the operator based on the adjustment vector displayed on the first command terminal. The adjustment device operates in two modes based on the type of instruction: for fine-tuning instructions with a displacement adjustment range of 2 mm to 10 mm and a current load pressure of 80% to 90% below its rated pressure, the adjustment device controller executes automatically; for instructions with a displacement adjustment exceeding 10 mm or a load pressure exceeding 80% to 90% above the rated pressure, the operator executes manually according to the adjustment vector displayed on the second instruction terminal; after the operator performs the manual operation, they must press the confirmation button on the corresponding terminal; upon receiving the confirmation signal from the terminal, the digital twin platform updates the execution status of the instruction.

[0057] In the above technical solution, the system hardware can include a first command terminal and a second command terminal. The first command terminal can be an industrial tablet computer, installed in the crane cab for easy operator observation and operation. The second command terminal can also be an industrial tablet computer or a handheld terminal, installed near the adjustment device, such as next to the hydraulic pump station control box. The terminal casing can be made of aluminum alloy, and the screen can be a tempered glass touchscreen. The digital twin platform can be deployed on a field server. The first and second command terminals can connect to the field-deployed 5G dedicated network or industrial Wi-Fi network through built-in wireless communication modules, thereby establishing a data connection with the digital twin platform. The adjustment device controller and the crane controller are usually built-in programmable logic controllers. The working process is as follows: After the digital twin platform generates adjustment commands based on real-time pose deviations, these commands are simultaneously and in parallel sent to the crane controller, the first command terminal, the adjustment device controller, and the second command terminal. The transmission method can be TCP / IP protocol. The data structure of the adjustment command includes specific pose deviation values, such as +20 mm in the X direction, and an adjustment vector for guiding operation.

[0058] All crane movements are performed manually by the operator. The operator primarily relies on the adjustment vectors displayed on the primary command terminal screen. These vectors are visually represented by a combination of arrow graphics and text descriptions, such as a left-pointing arrow accompanied by the text "Move 20 mm in the -X direction." Following this guidance, the operator manually operates the handles or buttons of the crane's travel, slewing, luffing, or hoisting mechanisms to move the crane's modules in the specified direction. This process is entirely based on human judgment and operation; the platform does not directly control the crane's power system.

[0059] The adjustment device operates in two modes. When the adjustment command is a fine-tuning command, the adjustment device controller executes automatically. A fine-tuning command requires two conditions to be met simultaneously: first, the displacement adjustment amount must be within the range of 2 mm to 10 mm; second, the current pressure signal value of the adjustment device must be lower than 80% to 90% of its rated working pressure. For example, for a hydraulic jack with a rated pressure of 50 MPa, when the platform command requires a 5 mm lift and the real-time pressure is below 40 MPa, the automatic mode is triggered. After receiving the command, the adjustment device controller drives a servo motor or proportional valve to complete the precise displacement.

[0060] For non-fine-tuning commands, i.e., when the displacement adjustment exceeds 10 mm, or the load pressure exceeds 80% to 90% of the rated pressure, the system switches to manual mode. At this time, the field operator needs to observe the adjustment vector displayed on the second command terminal. Following this guidance, the operator manually operates the control panel of the adjustment device, such as jogging the hydraulic valve button or rotating the handwheel, to complete adjustments under large displacements or high loads. The pressure threshold is used to determine whether the adjustment resistance is normal, preventing overload during automatic execution.

[0061] After manual execution is complete, both the crane operator and the adjustment device operator must press the confirmation button on their respective terminals. The confirmation button can be a physical button on the terminal or a virtual button on the touchscreen. Upon pressing, the first or second command terminal will immediately generate a confirmation signal and transmit it back to the digital twin platform via the wireless network. Upon receiving the confirmation signal, the digital twin platform will update the status of the corresponding adjustment command in its task management log from "Pending Execution" or "Executing" to "Manually Executed and Confirmed." The color of the command entry on the platform interface may change accordingly, for example, from yellow to green, thus providing dispatchers with clear status feedback.

[0062] This human-machine collaborative command execution system achieves a balance between accuracy and safety during installation by combining automated decision-making with manual operation. The system generates and distributes commands uniformly through a digital twin platform, ensuring centralized and coordinated control logic and avoiding multiple command structures. Different execution modes are designed for different equipment and working conditions: key crane actions are entirely performed manually by experienced operators, fully utilizing human judgment; the adjustment device intelligently switches between automatic fine-tuning and manual intervention based on the adjustment amount and load, improving fine-tuning accuracy and efficiency while retaining a manual safety valve for complex or high-risk operations. A clear confirmation mechanism ensures the traceability of each manual step, enabling the platform to accurately grasp the on-site execution status and providing a reliable basis for subsequent scheduling and decision-making. The overall design allows the advanced digital twin control logic to adapt to existing construction equipment and operating habits, enhancing the system's practicality and adaptability in real-world engineering environments.

[0063] In other technical solutions, an installation sequence dynamic planning module is constructed within the digital twin platform; The installation sequence dynamic planning module is pre-set with a rule base containing rules for resource conflict, structural dependency, and hoisting path conflict. When the difference between the real-time installation progress ratios of any two modules exceeds the current progress coordination threshold, the installation sequence dynamic planning module is triggered. The installation sequence dynamic planning module filters all uninstalled modules according to the rule base and identifies associated modules that have resource conflicts, structural dependencies, or hoisting path conflicts with the currently lagging module. The installation sequence dynamic planning module calculates a first adjustment scheme and a second adjustment scheme. The first adjustment scheme is to postpone the installation sequence of related modules, and the second adjustment scheme is to swap the installation sequence of the delayed module and the next preparation module. The installation sequence dynamic planning module simulates the execution process of the first adjustment scheme and the second adjustment scheme based on a digital twin model, and predicts the overall project duration extension caused by each scheme. The installation sequence dynamic planning module selects the scheme with the smaller overall project duration extension as the recommended scheme, generates a collaborative control instruction containing the new installation sequence, and sends it to the instruction terminal of the corresponding hoisting unit.

[0064] In the above technical solution, the construction of the dynamic planning module for installation sequence is part of the software functionality of the digital twin platform. In implementation, a dedicated software module can be developed within the platform, containing an editable rule base database. Resource conflict rules can be defined as "only one module can be hoisted by the same large crane within the same time period"; structural dependency rules can be defined as "module B can only begin installation after module A, which provides support for it, is installed and fixed"; hoisting path conflict rules can be defined as "the overlap rate of the virtual hoisting path spatial projection of two modules exceeds 70%". These rules are entered into the system at the beginning of the project based on the construction organization design. The real-time installation progress of the modules is continuously calculated using the ratio of the number of installed connecting components to the total number designed.

[0065] The operation of this module is triggered by a progress coordination threshold. When the digital twin platform detects that the difference in the real-time installation progress ratio between any two modules reaches, for example, 28%, and exceeds the 25% threshold set for the current stage, the dynamic planning module for the installation sequence will be automatically triggered. After triggering, this module first retrieves all rules from the rule base, and then scans the list of all modules that have not yet started installation or are being installed. It makes logical judgments based on the rules, identifying those modules that are directly related to the currently lagging module. For example, it identifies subsequent modules that plan to use the same crane (resource conflict), or upper modules that are structurally dependent on the lagging module (structural dependency). Subsequently, the module starts the simulation calculation engine. Based on the current state of the digital twin model, it simulates the execution of the first adjustment plan: postponing the planned installation time of all identified related modules by a fixed time interval. Then, it simulates the second adjustment plan: swapping the installation order of the lagging module with the next module scheduled for hoisting. During the simulation, the system considers the reasonable time required for crane relocation, re-hooking, and other procedures, and virtually extrapolates the impact of the adjusted new sequence on the critical path of the entire project. After the simulation is complete, the module will output the predicted overall project duration extension values ​​for the two scenarios.

[0066] The dynamic planning module for installation sequence automatically selects the option with the smaller overall project duration extension as the final recommended option. Once selected, the module generates a structured collaborative control instruction. This instruction includes the new module installation sequence, the adjusted planned start time for each module, and the allocation information for the involved hoisting equipment. This instruction is directly sent to the first and second instruction terminals corresponding to the affected hoisting units, as well as the main control interface of the project scheduling center, through the communication interface of the digital twin platform, thereby enabling dynamic and rapid adjustments to the installation plan.

[0067] Other technical solutions also include establishing a control point stability monitoring and automatic calibration mechanism; At each physical control point deployed on the main foundation of the structure, a target and a high-precision displacement sensor are installed. The measurement direction of the displacement sensor is consistent with the foundation settlement direction. The displacement sensor is connected to the automatic coordinate unification and data calibration module via a wireless network. The automatic coordinate unification and data calibration module continuously receives the real-time displacement data from each displacement sensor and compares it with the initial value. The automatic coordinate unification and data calibration module sets a displacement change threshold ranging from 0.5 mm to 2 mm. When the change in displacement sensor data at any physical control point exceeds the displacement change threshold, the automatic coordinate unification and data calibration module determines that the coordinates of that control point are invalid. The automatic coordinate unification and data calibration module sends an alarm to the digital twin platform and suspends the use of coordinate transformation parameters related to the control point. The digital twin platform schedules surveyors to use a high-precision total station to remeasure the control point whose coordinates have failed, and obtain new field measurement coordinates. The automatic coordinate unification and data calibration module uses the newly measured coordinates and the coordinates of other unfailed physical control points to recalculate the coordinate transformation parameters. The automatic coordinate unification and data calibration module performs residual verification on the recalculated coordinate transformation parameters, and resumes data transformation using the updated parameters after the verification passes. The total number of physical control points deployed on the main foundation of the structure shall not be less than six, including at least two backup control points; when the number of failed control points results in fewer than four remaining effective control points, the system shall automatically enable the coordinates of the backup control points to participate in the calculation of the coordinate transformation parameters.

[0068] In the above technical solution, the establishment of the control point stability monitoring and automatic calibration mechanism enhances and maintains the coordinate unification system. Its implementation begins with the modification of the physical control points. On the forced alignment base of each physical control point already deployed on the main foundation of the structure, in addition to the original measurement target, a high-precision displacement sensor is installed in parallel. This displacement sensor can be an inductive or laser micrometer, and its sensitive axis is precisely adjusted to be consistent with the estimated main settlement direction of the foundation, typically vertical. The sensor is connected via a signal line to an embedded data acquisition box with wireless transmission capabilities, which is fixed in a protective enclosure near the control point. Each acquisition box transmits the real-time measurement data from its connected displacement sensor to the backend server of the digital twin platform via the on-site industrial wireless network.

[0069] The platform's automatic coordinate unification and data calibration module has added a data monitoring and processing thread. This thread continuously receives data streams from the displacement sensors at each control point. The module records the first stable reading of the displacement sensor for each control point after the system's initial calibration as the initial value. Subsequently, the module compares the received real-time displacement data with the corresponding initial value at a certain frequency, such as once per minute, and calculates the cumulative change. A displacement change threshold is set within the module as the criterion for determining whether a control point has experienced significant displacement.

[0070] When the monitoring logic detects that the change in displacement sensor data at any physical control point exceeds a set threshold, the automatic coordinate unification and data calibration module immediately executes a series of actions. First, it internally marks the control point as "coordinate failure." Then, the module sends a clear alarm message to the main control interface of the digital twin platform, specifying the number of the failed control point and its displacement change. Simultaneously, the module automatically removes or isolates the coordinate data pair corresponding to the failed control point from the currently used coordinate transformation parameter calculation model, and suspends any real-time data transformations based on transformation parameters calculated from the old complete control point set to prevent the introduction of system errors.

[0071] Once the alarm is issued, the task management function of the digital twin platform generates a field measurement task and pushes it to the handheld terminals of the relevant surveyors. Following the instructions, the surveyors use a high-precision total station to remeasure the target at the control point where the coordinates have failed, obtaining its new coordinate values ​​in the current field measurement coordinate system. After the new coordinates are uploaded, the automatic coordinate unification and data calibration module immediately initiates the recalculation process. Using these newly measured coordinates, along with the original coordinates of all other valid physical control points, it re-runs the least squares algorithm to calculate a new set of coordinate transformation parameters.

[0072] After the new parameters are calculated, the module uses the remaining, unaffected control point coordinates to perform residual verification on the new parameters. If the verification passes, the module replaces the old parameters with the new set of transformation parameters, unsuspends the real-time transformation function, and the system resumes full-function operation. Through this mechanism, automated monitoring and closed-loop maintenance of the stability of the spatial measurement benchmark are achieved, significantly reducing the risk of a decrease in the accuracy of the entire system due to imperceptible displacement of control points.

[0073] In other technical solutions, in step S3, the digital twin platform determines the installation stage of the module based on the following specific conditions: when the real-time height of the module is greater than 10 m and the horizontal moving speed is greater than 0.1 m / s, it is determined to be in the hoisting and aerial movement stage; when the real-time height of the module is greater than 1.5 m and less than or equal to 10 m, the horizontal moving speed is less than or equal to 0.1 m / s, and the pressure signal received from the adjustment device is greater than 5 MPa, it is determined to be in the preliminary positioning and adjustment stage; when the real-time height of the module is less than or equal to 1.5 m and the first fastener installation completion signal is received from the module connection node, it is determined to be in the final fixing and welding stage. In step S5, the number of installed connection components of the module is automatically obtained by sensors deployed at each connection node of the module; when the sensor detects that the connection component is installed in place, it sends a signal, and the digital twin platform receives and accumulates the signal to calculate the real-time installation progress.

[0074] In the above technical solution, to achieve automatic determination during the installation phase, corresponding sensors and data acquisition systems need to be configured. Real-time altitude data of the module can be acquired through a GNSS receiver installed on the module. This receiver can be a measurement model supporting RTK functionality, and its antenna can be installed in an open area on top of the module. Horizontal movement speed can be calculated using position differential data output from the same GNSS receiver, or it can be fused with data from an inertial measurement unit. The pressure signal from the adjustment device can be directly read from the pressure transmitter built into the hydraulic jack or pusher. This transmitter is typically installed on the interface of the oil circuit near the actuator, transmitting a 4-20mA current signal to the acquisition terminal via cable. The pressure threshold used for the determination phase can be set to 5 MPa. The digital twin platform can be deployed on an industrial computer, and its internal software continuously receives the above data streams. Its operation process is as follows: The platform software scans the sensor data of each module at a fixed frequency. For a given module, the software first reads its altitude value. If the height is greater than 10 m and the calculated instantaneous horizontal velocity is greater than 0.1 m / s, the software marks the module as being in the hoisting and aerial movement stage within that cycle. When the module descends to a height between 1.5 m and 10 m and the horizontal velocity drops below 0.1 m / s, the software further checks the pressure signal of the corresponding adjustment device. If the pressure value consistently exceeds 5 MPa, the module is determined to have entered the preliminary positioning and adjustment stage. When the module height drops to 1.5 m or below, and the platform receives a trigger signal indicating that the first fastener installation is complete from the module, the stage is immediately updated to the final fixing and welding stage, regardless of the pressure status.

[0075] To automate the quantification of installation progress, sensing elements need to be deployed at the designed connection nodes of each steel structure module. These sensing elements can be miniature limit switches or magnetic proximity switches. For high-strength bolt connections, a miniature limit switch can be fixed to one side of the connection plate, with its probe aligned with the bolt hole; when the bolt is installed in place and the nut is tightened to contact the switch, the switch contacts close. Alternatively, a magnet can be embedded in the bolt tail or a dedicated washer, and a magnetic proximity switch can be installed at the corresponding position for non-contact detection. The switch housing material can be stainless steel or engineering plastic to withstand the field environment. These switches are networked via a fieldbus or wireless sensor network. The operation is as follows: the switch at each connection node is normally open when not triggered. When the installer completes the installation of a connection component at that node and meets the preset tightening requirements, the switch is physically triggered, changing its state from open to closed. This state change is immediately captured by nearby IoT data acquisition nodes and encoded into a short data packet representing the completion of a node. Data packets from all nodes are aggregated to the digital twin platform via a wireless network. A counter is maintained for each module within the platform, with an initial value of zero. Each time a completion signal is received from a node within this module, the counter increments by one. The platform compares this count with the total number of designed connecting components for this module (a preset constant) in real time; the ratio of the two values ​​represents the real-time installation progress of the module. This progress value is dynamically updated on the platform interface in the form of numbers or a progress bar.

[0076] This solution achieves precise and automated judgment of the module installation phase by setting clear and quantifiable multi-parameter conditions, reducing the subjectivity and lag of manual judgment and enabling subsequent deviation control and schedule coordination to be based on more accurate phase information. Simultaneously, by deploying simple and reliable sensors at each connection node on-site, the originally ambiguous construction progress is transformed into real-time, automatically collected micro-data, providing the digital twin platform with progress feedback accurate to each component. The combination of these two elements constructs a highly granular and timely on-site status perception system, providing a solid and reliable data input foundation for the platform's dynamic threshold management, deviation closed-loop control, and schedule coordination scheduling functions, thereby improving the automation level and decision reliability of the entire installation control process.

[0077] Example: Module installation and application in the diesel hydrotreating reactor area of ​​a large-scale refining and chemical project This project is a diesel hydrorefining unit. Its reactor area employs modular construction, comprising 32 prefabricated steel structure functional modules. The largest module (R-201 reactor frame) weighs 480 tons and is installed at a height of 42.5 meters. Traditional installation methods face challenges such as complex hoisting paths, high positioning accuracy requirements (within ±5mm), and difficulties in coordinating multiple cranes (one 800-ton crawler crane and two 400-ton truck cranes). Delays in any single module could easily lead to overall project delays.

[0078] To implement the method of this invention, the following hardware and software systems were built: Software platform: A detailed BIM model created based on Tekla Structures 21.0 was imported into Siemens Teamcenter via the IFC standard interface. ® XR Digital Twin Platform (Version 2022). This platform activates core modules such as "Automatic Coordinate Unification," "Dynamic Threshold Management," "Location Source Management," and "Dynamic Planning of Installation Sequence."

[0079] Sensing and positioning hardware: Trimble is installed on each module ® The basic sensor group consists of an R12i GNSS receiver (supporting RTK) and a Spectra Sensors® N500 dual-axis digital inclinometer. A total of eight Pozyx sensors are deployed at the highest point of the structure and on the central platform. TM UWB positioning base stations form a layered coverage network. Each module has two UWB positioning tags installed near the lifting point and center of gravity. Customized micro-switch sensors are integrated into the gaskets of the key connection nodes of the modules (a total of 2,560 high-strength bolts are planned) for automatic counting.

[0080] On-site network: Deployed a hierarchical wireless network based on Huawei's 5G dedicated network (core layer) and ADLINK industrial IoT gateways (access layer, one per crane). Advantech edge computing nodes are embedded in the sensor groups of each module.

[0081] Control point layout: Six physical control points (including two backup points) are set up on the concrete foundation platform in the reactor area. Each point is equipped with a Leica sensor. TM 360° prism and KEYENCE™ LK-G500 laser displacement sensor.

[0082] Complete the following key configurations in the digital twin platform: Six control points were measured using a Leica TS16 total station. The platform automatically calculated a seven-parameter transformation model, and the residual verification passed (maximum residual 2.1 mm).

[0083] Dynamic threshold setting: Installation phase Judgment conditions Position deviation threshold Attitude deviation threshold Progress coordination threshold hoisting and aerial movement Height > 10m and horizontal speed > 0.1m / s 15 mm 4° 35% Initial positioning adjustment Height ≤ 10m and > 1.5m, speed ≤ 0.1m / s, and bottom lifting pressure > 5MPa 8 mm 1.5° 25% Final fixation welding Height ≤ 1.5m and first bolt tightening signal received 3 mm 0.8° 12% The input rules are as follows: "An 800-ton crawler crane can only serve one module at a time", "Module R-205 can only be hoisted after R-204 is installed and fixed", and "The spatial spacing of the hoisting path must be more than 3 meters".

[0084] Complete execution and linkage of the control process (taking the R-205 module as an example) The following demonstrates how the system works in tandem during actual installation: (1) Lifting and initial correction (S2-S4 are executed in a cycle) After the R-205 module was lifted, the platform determined through GNSS data that it had entered the "lifting and aerial movement phase".

[0085] Event: When the module was hoisted to a height of 25 meters, the platform detected its positional deviation as follows: X=+22mm, Y=-5mm (design threshold 15mm).

[0086] Platform decision: Determines that the X-axis deviation exceeds the limit.

[0087] Command issuance: Generate adjustment command ("Slightly move the crane 20mm in the -X direction") and send it to the first command terminal in the 800-ton crawler crane cab via 5G network.

[0088] Manual execution and confirmation: The crane operator performs the operation according to the terminal instructions and presses the confirmation button upon completion.

[0089] Closed-loop verification: The platform continuously receives updated GNSS data, and the calculated deviation is reduced to X=+3mm and Y=-4mm, which are below the threshold. The closed-loop adjustment is then complete.

[0090] (2) Phase switching and fine-tuning When the module descends to a height of 9 meters, the speed drops to 0.05 m / s, and the pressure sensor of the bottom hydraulic jack returns an 8 MPa signal, the platform determines that it has entered the "preliminary positioning adjustment stage" and automatically switches the control threshold to the more stringent 8 mm and 1.5°.

[0091] Event: The platform discovered that the module's Y-axis deviation was -10mm (the ultra-new threshold is 8mm) by integrating UWB data.

[0092] The platform generates a fine-tuning instruction ("Synchronously lift hydraulic jacks No. 2 and No. 3 in the Y+ direction by 12mm").

[0093] Because the adjustment amount (12mm) is greater than 10mm, the command is sent simultaneously to the adjustment device controller and the second command terminal of the field operator.

[0094] The operator starts the lifting machine according to the terminal instructions and confirms the completion.

[0095] The platform confirmed through UWB and inclinometer data that the final pose was stable at X=+1.5mm, Y=+2mm, and levelness 0.5°, meeting the positioning requirements. The platform then updated the stage to "final fixed welding".

[0096] (3) Progress coordination and dynamic scheduling trigger (execute S5) While R-205 was undergoing precision adjustments, the platform monitored the progress of all modules in parallel. The real-time progress of R-205 (counted via microswitches) was 15%, while the adjacent R-206 module, which was using the same crane, had reached 50% progress.

[0097] Event: The progress difference has reached 35%, exceeding the 25% coordination threshold for the current "initial in place" phase of R-205.

[0098] Trigger Planning Module: The platform automatically triggers the "Installation Sequence Dynamic Planning Module".

[0099] Simulation and Decision-Making: Based on the rule base, the module identifies a crane resource conflict between R-206 and R-207. The simulation shows that, according to the original plan, R-206 can only start after R-205 is fixed (estimated to take another 4 hours), which will delay the total project time by 3.5 hours; if the order of R-207 and R-205 is swapped, the total project time will only be delayed by 1.2 hours.

[0100] Command generation and issuance: The platform automatically selects the second option, generates a new collaborative control command ("The subsequent hoisting sequence is adjusted to: R-207 -> R-205 -> R-206"), and issues it to the command terminals of the relevant hoisting teams and the central dispatch room.

[0101] (4) Emergency response to control point failure During the installation process, the platform detected a sudden change in the displacement sensor data at control point 3 by +1.8mm (exceeding the threshold of 1.5mm).

[0102] The platform immediately issued an alert and automatically removed the data for point 3 from the coordinate transformation model.

[0103] Since there are still more than 4 valid points (5), the system uses the coordinates of backup control point No. 4 to participate in the calculation, seamlessly generating new transformation parameters, and the residual verification is passed (maximum 2.5mm).

[0104] The system continued to operate without interruption and simultaneously sent a work order to the surveyor to verify point 3.

[0105] 5. Implementation Results After applying the method of this invention, the final placement accuracy of all 32 modules in this project met the design requirements (±5mm) 100%, with an average accuracy of ±2.3mm. The project was completed 5 days ahead of schedule, and dynamic scheduling reduced potential delays by an average of 2.1 hours per instance. The reliance on the on-site experience of surveyors and schedulers was significantly reduced.

[0106] The above embodiments illustrate in detail the application of the present invention in a typical project. For those skilled in the art, based on the teachings of these embodiments, adjusting specific hardware and software models and threshold parameters to adapt to chemical steel structure projects of different scales (e.g., number of modules, weight) and accuracy requirements all fall within the scope of protection of this invention. For example, in small and medium-sized projects, more economical sensor combinations (e.g., using only UWB and inclinometers) can be used, or the rule base can be simplified, while the core method logic of "phased dynamic threshold determination - real-time closed-loop control - progress collaborative triggering planning" remains unchanged.

[0107] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and embodiments shown and described herein.

Claims

1. A multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins, characterized in that, Includes the following steps: S1. Based on the building information model of chemical steel structure, a digital twin model integrating multi-module design information is created in the digital twin platform. Sensor groups, including global positioning system sensors and attitude sensors, are installed on each prefabricated steel structure module in the factory. S2. The digital twin platform receives real-time position and attitude data collected by the sensor group, integrates pre-stored design information, and updates the model to form a dynamic digital twin model. S3. The platform determines the installation stage of each module based on its real-time location and signals from the crane and adjustment device. The installation phase includes the hoisting and aerial movement phase, the preliminary positioning and adjustment phase, and the final fixing and welding phase. Based on the determined installation phase, the preset position deviation allowable threshold and attitude deviation allowable threshold corresponding to that phase are invoked. At the same time, the dynamic digital twin model calculates the position deviation between the real-time position and the design position of each module, and the attitude deviation between the real-time attitude and the design attitude. S4. When the position deviation or attitude deviation exceeds the allowable threshold for position deviation or attitude deviation corresponding to the current installation stage, the platform generates an adjustment command and sends it to the corresponding installation equipment; the installation equipment coordinates its actions under the drive of the adjustment command to correct the module's position and attitude. S5. The platform uses the ratio of the number of installed connecting components to the total number of designed connecting components as the real-time installation progress of the module, and determines the progress coordination threshold according to the installation stage. When the difference between the real-time installation progress ratios of any two modules exceeds the progress coordination threshold set for the module with the earlier installation stage, the platform re-plans the subsequent installation sequence and generates coordination control instructions. The process involves repeatedly executing steps S2 to S4 until each module is installed; and continuously executing step S5 to collaboratively monitor and dynamically adjust the installation progress of multiple modules.

2. The multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in claim 1, characterized in that, The digital twin platform has a pre-installed automatic coordinate unification and data calibration module; On the main structural foundation at the installation site, at least four physical control points are set up, and a target is fixedly installed on each physical control point; in the digital twin model, model control points corresponding to the physical control points are created, and the three-dimensional coordinates of the target in the BIM design coordinate system are entered. Before the start of each construction cycle, a high-precision total station is used to measure the real-time three-dimensional coordinates of all targets in the field measurement coordinate system; the automatic coordinate unification and data calibration module calculates coordinate transformation parameters based on the coordinates of the same target in the BIM design coordinate system and the field measurement coordinate system. The automatic coordinate unification and data calibration module uses the coordinate transformation parameters to convert the real-time position data from the GPS sensors and automatic total station of the sensor group to the BIM design coordinate system. The automatic coordinate unification and data calibration module performs residual verification on the real-time position data after coordinate transformation. When the coordinate transformation residual of any target is greater than 3 mm to 5 mm, the calibration is deemed to have failed and an alarm is issued. The digital twin platform compares the real-time position data that has passed the residual verification and has been transformed to the BIM design coordinate system with the design position in the digital twin model.

3. The multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in claim 1, characterized in that, The digital twin platform is equipped with a dynamic threshold management module; The dynamic threshold management module presets independent position deviation thresholds and attitude deviation thresholds for each installation stage. During the aerial movement phase of the hoisting, the position deviation threshold is 10 mm to 20 mm, and the attitude deviation threshold is 3° to 5°. For the initial positioning adjustment stage, the position deviation threshold is 5 mm to 10 mm, and the attitude deviation threshold is 1° to 2°; For the final fixed welding stage, the position deviation threshold is 2 mm to 5 mm, and the attitude deviation threshold is 0.5° to 1°. The digital twin platform determines the current installation stage of each module based on its real-time height, horizontal movement speed, and force status signals from the crane and adjustment device. After calculating the position and attitude deviations, the dynamic digital twin model calls the dynamic threshold management module to obtain a threshold that matches the current installation stage for judgment. When the module is in the hoisting and aerial movement phase, the progress coordination threshold is 30% to 40%. When the module is in the initial positioning and adjustment phase, the progress coordination threshold is 20% to 30%. When the module is in the final fixed welding stage, the progress coordination threshold is 10% to 15%.

4. The multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in claim 1, characterized in that, This also includes establishing a multi-source fusion positioning system; A positioning source management module is constructed within the digital twin platform, with its inputs connected to a GPS sensor, an automatic total station, and an ultra-wideband positioning base station. At least three ultra-wideband positioning base stations are deployed between the top and middle layers of the frame at the installation site, forming a network covering the hoisting path and the placement area. Ultra-wideband positioning tags are installed near the hoisting points and center of gravity of each steel structure module. The positioning source management module receives raw position data from the GPS sensor, the automatic total station, and the ultra-wideband positioning tags, and determines the signal obstruction level based on the real-time height of the module and the surrounding structural information in the digital twin model. When the module is in the open area at the top of the frame, the positioning source management module uses the fused Global Positioning System data as the main positioning source; When the module descends to the middle interlayer area of ​​the frame, the positioning source management module switches the main positioning source to ultra-wideband positioning data; When the module enters the designed positioning point and the ultra-wideband positioning tag is within the line of sight of the total station, the positioning source management module will switch the main positioning source to the automatic total station measurement data; The location source management module transmits the real-time location data of the selected primary location source to the digital twin platform.

5. The multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in claim 4, characterized in that, This also includes establishing an anti-interference processing flow for attitude data; The attitude sensor group installed on each steel structure module includes a microelectromechanical system inertial measurement unit and a dual-axis digital inclinometer; the angular velocity data output by the inertial measurement unit and the angle data output by the digital inclinometer are synchronously transmitted to the digital twin platform; The Kalman filter configured in the digital twin platform uses the angle data of the digital tilt meter as the observation vector and the angular velocity integral result of the inertial measurement unit as the state prediction vector to perform data fusion and output the real-time pitch angle and real-time roll angle. The digital twin platform performs a confidence check before calculating the attitude deviation. The check method is to compare the difference between the acceleration calculated by the inertial measurement unit and the gravitational acceleration component. When the difference exceeds 0.05g to 0.1g for 3s to 5s, the data is judged to be abnormal. When the data is abnormal, the digital twin platform switches to using only the data from the digital tiltmeter to calculate the attitude deviation and issues a sensor check command.

6. The multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in claim 1, characterized in that, This also includes deploying a tiered field industrial wireless network; The network comprises a core layer and multiple access layers; the core layer consists of industrial-grade wireless access points deployed at high points, forming a 5G dedicated network or high-speed wireless local area network covering the entire work area; the access layers consist of multiple IoT gateways, each IoT gateway being bound to a hoisting unit, the hoisting unit comprising a crane, the module to be installed, and adjustment devices. An IoT edge computing node is integrated within the sensor group of each steel structure module; the IoT edge computing node collects the raw sensor data of this module via wired connection and connects to the IoT gateway of this hoisting unit via a first wireless link; the IoT gateway accesses the core layer network via a second wireless link and uploads the data to the digital twin platform. The adjustment instructions and collaborative control instructions issued by the digital twin platform are sent to the device controller through the core layer network and the IoT gateway; The IoT edge computing node sets priority tags for data, marking location data and emergency alarm data as the highest priority, attitude data as the medium priority, and sensor status data as the low priority; the data queue manager in the IoT gateway schedules the uploaded data according to the priority tags.

7. The multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in claim 1, characterized in that, It also includes configuring a human-machine collaborative instruction execution system; The system includes a first command terminal installed in the crane operator's cab and a second command terminal installed near the adjustment device. The first and second command terminals are connected to the digital twin platform via a wireless network. The adjustment commands generated by the digital twin platform are simultaneously sent to the crane controller, the first command terminal, the adjustment device controller, and the second command terminal. The adjustment commands include a positional deviation value and an adjustment vector. Both the first and second command terminals are equipped with a display screen and a confirmation button. The crane's movements are manually executed by the operator based on the adjustment vector displayed on the first command terminal. The adjustment device operates in two modes based on the type of instruction: for fine-tuning instructions with a displacement adjustment range of 2 mm to 10 mm and a current load pressure of 80% to 90% below its rated pressure, the adjustment device controller executes automatically; for instructions with a displacement adjustment exceeding 10 mm or a load pressure exceeding 80% to 90% above the rated pressure, the operator executes manually according to the adjustment vector displayed on the second instruction terminal; after the operator performs the manual operation, they must press the confirmation button on the corresponding terminal; upon receiving the confirmation signal from the terminal, the digital twin platform updates the execution status of the instruction.

8. The multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in claim 1, characterized in that, An installation sequence dynamic planning module is constructed within the digital twin platform; The installation sequence dynamic planning module is pre-set with a rule base containing rules for resource conflict, structural dependency, and hoisting path conflict. When the difference between the real-time installation progress ratios of any two modules exceeds the current progress coordination threshold, the installation sequence dynamic planning module is triggered. The installation sequence dynamic planning module filters all uninstalled modules according to the rule base and identifies associated modules that have resource conflicts, structural dependencies, or hoisting path conflicts with the currently lagging module. The installation sequence dynamic planning module calculates a first adjustment scheme and a second adjustment scheme. The first adjustment scheme is to postpone the installation sequence of related modules, and the second adjustment scheme is to swap the installation sequence of the delayed module and the next preparation module. The installation sequence dynamic planning module simulates the execution process of the first adjustment scheme and the second adjustment scheme based on a digital twin model, and predicts the overall project duration extension caused by each scheme. The installation sequence dynamic planning module selects the scheme with the smaller overall project duration extension as the recommended scheme, generates a collaborative control instruction containing the new installation sequence, and sends it to the instruction terminal of the corresponding hoisting unit.

9. The multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in claim 2, characterized in that, This also includes establishing a control point stability monitoring and automatic calibration mechanism; At each physical control point deployed on the main foundation of the structure, a target and a high-precision displacement sensor are installed. The measurement direction of the displacement sensor is consistent with the foundation settlement direction. The displacement sensor is connected to the automatic coordinate unification and data calibration module via a wireless network. The automatic coordinate unification and data calibration module continuously receives the real-time displacement data from each displacement sensor and compares it with the initial value. The automatic coordinate unification and data calibration module is set with a displacement change threshold ranging from 0.5 mm to 2 mm. When the change in displacement sensor data at any physical control point exceeds the displacement change threshold, the automatic coordinate unification and data calibration module determines that the coordinates of that control point are invalid. The automatic coordinate unification and data calibration module sends an alarm to the digital twin platform and suspends the use of coordinate transformation parameters related to the control point. The digital twin platform schedules surveyors to use a high-precision total station to remeasure the control point whose coordinates have failed, and obtain new field measurement coordinates. The automatic coordinate unification and data calibration module uses the newly measured coordinates and the coordinates of other unfailed physical control points to recalculate the coordinate transformation parameters. The automatic coordinate unification and data calibration module performs residual verification on the recalculated coordinate transformation parameters, and resumes data transformation using the updated parameters after the verification passes. The total number of physical control points deployed on the main foundation of the structure shall not be less than six, including at least two backup control points; when the number of failed control points results in fewer than four remaining effective control points, the system shall automatically enable the coordinates of the backup control points to participate in the calculation of the coordinate transformation parameters.

10. The multi-module collaborative installation control method for chemical steel structures based on BIM and digital twins as described in claim 1, characterized in that, In step S3, the digital twin platform determines the installation stage of the module based on the following specific conditions: when the real-time height of the module is greater than 10 m and the horizontal moving speed is greater than 0.1 m / s, it is determined to be in the hoisting and aerial movement stage; when the real-time height of the module is greater than 1.5 m and less than or equal to 10 m, the horizontal moving speed is less than or equal to 0.1 m / s, and a pressure signal from the adjustment device is greater than 5 MPa, it is determined to be in the preliminary positioning and adjustment stage; when the real-time height of the module is less than or equal to 1.5 m and a signal indicating that the first fastener installation is complete is received from the module connection node, it is determined to be in the final fixing and welding stage. In step S5, the number of installed connection components of the module is automatically obtained by sensors deployed at each connection node of the module; when the sensor detects that the connection component is installed in place, it sends a signal, and the digital twin platform receives and accumulates the signal to calculate the real-time installation progress.

Images