Disclosed is a disc-type scaffold modularization erection scheme intelligent generation system.

Abstract

This invention relates to the field of intelligent building construction technology, specifically disclosing an intelligent generation system for modular scaffolding erection schemes. The system includes an engineering geometric parameter analysis device, a wind field dynamic simulation device, a physical information neural network prediction device, a wind-resistant reinforcement zone generation device, and a real-time meteorological data access device. By integrating computational fluid dynamics with a neural network embedded with physical constraints, based on a three-dimensional building model and real-time meteorological data, it dynamically predicts wind-induced vibration risks and automatically generates wind-resistant reinforcement schemes conforming to the modular rules of scaffolding. By employing the above technical solution, this invention enables high-precision wind vibration early warning and intelligent optimization of scaffolding systems in extreme environments such as super high-rise buildings or cross-sea structures, improving their inherent safety level and construction feasibility.

Description

Technical Field

[0001] This invention belongs to the field of intelligent building construction technology, specifically relating to an intelligent generation system for modular erection schemes of disc-lock scaffolding. Background Technology

[0002] With the rapid development of industrialized construction, modular scaffolding, as an efficient and safe support system, has been widely used in the construction of super high-rise buildings, cross-sea bridges, and complex steel structures. Modular erection schemes, as a core element in ensuring construction safety and project progress, achieve high load-bearing capacity and rapid assembly and disassembly of the scaffolding system through standardized component combinations and predefined structural modules. This industrialized construction mode plays a crucial technical supporting role in improving the support stability of large and complex projects and reducing the uncertainties of manual on-site operations.

[0003] The intelligent generation technology for modular scaffolding erection schemes is a current research hotspot in construction informatization. It aims to automatically analyze engineering geometric parameters through a computer system and output a member distribution and node scheme that meets mechanical equilibrium according to industry design specifications. To ensure structural safety in variable environments, the system needs to accurately calculate the comprehensive stress conditions, including self-weight, construction live load, and external wind load, and generate scientific reinforcement suggestions for specific construction scenarios.

[0004] Existing technologies for generating structural reinforcement schemes largely rely on static wind load calculation models in specifications, neglecting the vortex-induced resonance and nonlinear dynamic damage effects generated by local wind fields around complex structures. Traditional design systems lack in-depth analytical capabilities of fluid dynamic characteristics, resulting in a lag in predicting structural safety hazards under extreme weather conditions and difficulty in capturing in real-time fatigue damage caused by instantaneous wind pressure to the structural members. Existing scheme generation logic lacks deep integration of physical and mechanical equations and deep learning models, failing to adjust the coordinates of the reinforcement zone in real-time based on dynamically changing meteorological data, leading to a lack of specificity and timeliness in wind-resistant reinforcement measures.

[0005] There is an urgent need for an intelligent generation system for modular scaffolding erection schemes. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent generation system for modular erection schemes of disc-lock scaffolding, which can solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A modular scaffolding erection scheme intelligent generation system includes an engineering geometric parameter analysis device, a wind field dynamic simulation device, a physical information neural network prediction device, a wind-resistant reinforcement zone generation device, and a real-time meteorological data access device, as follows: The engineering geometric parameter analysis device is configured to receive three-dimensional building model data of the target construction area, and extract the boundary conditions, height distribution, facade outline and node coordinate information required for scaffolding erection from it. Based on the preset modular rules of the disc fastener components, a foundation erection scheme that meets the structural bearing requirements is initially generated. The wind field dynamic simulation device is configured to construct a computational fluid dynamics model of the local wind field at high altitude based on the scaffold geometry defined by the basic erection scheme, simulate the separation, reattachment and vortex shedding process of the airflow around the scaffold under different wind directions and speeds, and output the instantaneous wind pressure distribution and wind-induced vibration response characteristics of each member surface. The physical information neural network prediction device is configured to embed the Navier-Stokes equations and continuity equations as physical constraints into a deep neural network architecture, and to jointly train it using historical wind field simulation data and measured vibration response to establish a mapping relationship between wind load input and structural dynamic response, and to predict the wind vibration risk level within a predetermined time in the future based on the current wind field state. The wind-resistant reinforcement zone generation device is configured to receive the wind vibration risk distribution map output by the physical information neural network prediction device, identify high-risk areas that exceed the preset threshold, and automatically calculate the specific spatial coordinates of the diagonal braces or reinforcement nodes that need to be added based on the connection module characteristics of the disc-lock scaffold, and generate wind-resistant optimization instructions that include the type, quantity and installation coordinates of the reinforcement components. The real-time meteorological data access device is configured to acquire real-time wind speed, wind direction, atmospheric stability, and turbulence intensity data of the construction site from an external meteorological monitoring network, and dynamically input the data into the wind field dynamic simulation device and the physical information neural network prediction device to drive the entire system to perform closed-loop correction of the construction plan.

[0008] Preferably, the wind field dynamic simulation device uses an unsteady Reynolds-averaged Navier-Stokes equation solver, combined with the large eddy simulation method, to perform refined modeling of the local wind pressure gradient on the windward and leeward sides of the scaffold, in order to capture the vortex-induced resonance phenomenon caused by abrupt changes in structural geometry.

[0009] Furthermore, the neural network architecture in the physical information neural network prediction device includes multiple spatiotemporal convolutional layers and long short-term memory units, which are used to process the spatial distribution characteristics and temporal evolution of the wind field, and force the network output to satisfy the physical consistency of mass conservation and momentum conservation through a loss function.

[0010] Furthermore, when determining the reinforcement location, the wind-resistant reinforcement zone generation device comprehensively considers the standard spacing module of the disc-locking nodes, the stress redundancy of existing members, and construction operability constraints to ensure that the generated reinforcement scheme not only meets the mechanical performance requirements but can also be quickly implemented on site.

[0011] Preferably, the engineering geometry parameter analysis device is also configured to identify irregular areas such as cantilevered, recessed, or curved surfaces in the building facade, and automatically increase the initial erection density in these areas as an enhanced input basis for subsequent dynamic wind load analysis.

[0012] Furthermore, the real-time meteorological data access device is equipped with a data filtering and outlier removal unit, which is used to smooth the raw meteorological data, eliminate instantaneous pulse interference, and ensure that the data input to the wind field model is representative and timely.

[0013] Furthermore, the system also includes a scheme verification feedback device, which is configured to import the final generated erection scheme into a virtual construction simulation environment, apply dynamic wind load excitation and monitor the displacement response of the overall structure and the stress change of the members. If the monitoring results exceed the safety tolerance, a re-optimization process is triggered until the scheme meets the dynamic stability requirements.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. The intelligent generation system for modular erection scheme of disc-lock scaffolding provided by the present invention breaks through the limitations of traditional static wind load design. For the first time, it deeply integrates computational fluid dynamics and physical information neural network, and realizes high-precision prediction of the dynamic response of scaffolding under complex wind fields at high altitudes.

[0015] 2. Based on real-time meteorological data, the system can proactively identify potential vortex-induced resonance and wind-induced fatigue risk areas, and automatically generate wind-resistant reinforcement schemes that conform to modular construction specifications, thereby improving the disaster resistance and intrinsic safety level of scaffolding systems in extreme environments such as super high-rise buildings and cross-sea bridges.

[0016] 3. By embedding physical equations into the artificial intelligence model, the mechanical rationality and engineering credibility of the prediction results are ensured, avoiding the physical distortion problem that may occur in a purely data-driven model. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall technical solution architecture of the present invention; Figure 2 This is a schematic diagram of the core principle framework of the physical constraint embedded neural network in the physical information neural network prediction device of this invention; Figure 3 This is a logical flowchart of the foundation erection scheme generated based on the analysis of engineering geometric parameters and the modular rules of the disc fastener components in this invention. Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow of wind field dynamic simulation and risk prediction driven by real-time meteorological data access in this invention; Figure 5This is a logical flowchart of the wind-resistant strengthening zone generation and scheme verification feedback based on wind vibration risk distribution identification in this invention. Detailed Implementation

[0018] Example 1: Please refer to the appendix Figure 1 To be continued Figure 5 To make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments.

[0019] A modular scaffolding erection scheme intelligent generation system includes an engineering geometric parameter analysis device, a wind field dynamic simulation device, a physical information neural network prediction device, a wind-resistant reinforcement zone generation device, a real-time meteorological data access device, and a scheme verification feedback device. The engineering geometric parameter analysis device is used to receive three-dimensional building model data of the target construction area, extract the boundary conditions, height distribution, facade outline and node coordinate information required for scaffolding erection, and generate a preliminary foundation erection plan based on the preset modular rules of the disc fastener components. The wind field dynamic simulation device is used to construct a computational fluid dynamics model of a local wind field at high altitude, simulate the characteristics of airflow around the scaffold under different wind directions and speeds, and output the instantaneous wind pressure distribution and wind-induced vibration response characteristics of the surface of each member. The physical information neural network prediction device is used to establish a mapping relationship between wind load input and structural dynamic response, and to predict the wind vibration risk level based on the current wind field status. The wind-resistant reinforcement zone generation device is used to identify high-risk areas, calculate the specific spatial coordinates of the diagonal braces or reinforcement nodes that need to be added, and generate wind-resistant optimization instructions. The real-time meteorological data access device is used to acquire real-time meteorological data at the construction site and dynamically input it into the wind field dynamic simulation device and the physical information neural network prediction device to drive the entire system to perform closed-loop correction of the erection scheme. The scheme verification feedback device is used to import the final generated erection scheme into a virtual construction simulation environment for verification, ensuring that the scheme meets the dynamic stability requirements.

[0020] The engineering geometric parameter analysis device includes a BIM data access unit, a geometric feature extraction unit, a modular mapping unit, and a preliminary scheme generation unit. The BIM data access unit is configured to establish a high-speed data link with an external building information modeling platform, using a general industrial basic format or a specific 3D design software interface to retrieve the geometric shape data of the construction target in real time. The geometric feature extraction unit is configured to perform topological analysis on the 3D mesh data to identify the flatness, concavity / convexity dimensions, and opening locations of the building facade, particularly identifying irregular areas such as cantilevered, recessed, or curved surfaces on the building facade. The modular mapping unit stores a database of standard disc-lock scaffolding specifications, including the length sequence of uprights, the modular step distance of horizontal bars, and the geometric matching rules for diagonal bars. This unit discretizes the physical space into 3D mesh nodes that meet the modular requirements based on the extracted building outline. The preliminary scheme generation unit is used to automatically increase the initial erection density in irregular areas, determine the step distance and longitudinal distance of the scaffolding in combination with the building height distribution, and preliminarily arrange the uprights, horizontal bars and scaffold boards according to the preset static safety threshold, as the input benchmark for subsequent dynamic analysis.

[0021] The wind field dynamic simulation device includes a fluid domain construction unit, a solver configuration unit, a vortex feature identification unit, and a wind pressure mapping unit. The fluid domain construction unit is configured to extend outwards from the scaffold geometry in the foundation erection scheme to establish a computational fluid domain including the atmospheric boundary layer, and to perform local mesh refinement processing for the dense scaffold members. The solver configuration unit is configured to use an unsteady Reynolds-averaged Navier-Stokes equation solver, combined with the large eddy simulation method. Its internal algorithm logic is set to perform spatiotemporal discretization of the mass and momentum conservation of fluid flow, paying particular attention to the local wind pressure gradient formed by the pressure difference between the windward and leeward sides. The vortex feature identification unit is used to capture vortex-induced resonance phenomena caused by abrupt changes in structural geometry, extracting the frequency and intensity of detached vortices in the flow field by analyzing the separation and reattachment points of the airflow. The wind pressure mapping unit converts the calculated spatial flow pressure field into time-history wind pressure loads applied to the surface of specific scaffold members, providing a refined excitation source for structural response analysis.

[0022] The physical information neural network prediction device includes a neural network architecture unit, a physical constraint embedding unit, a spatiotemporal feature fusion unit, and a risk classification assessment unit. The neural network architecture unit contains multiple spatiotemporal convolutional layers and long short-term memory units. The spatiotemporal convolutional layers capture the topological characteristics of wind pressure distribution in space, while the long short-term memory units record the time dependence of wind load evolution. The physical constraint embedding unit is configured to transform the Navier-Stokes equations, continuity equations, and structural dynamics equations into loss function components of the neural network. By constraining the parameter update process of the neural network, its output not only conforms to the trend of historical training data but also must comply with the laws of energy and mass conservation in physics. The loss function is defined as the root mean square error between the network's predicted value and the simulated true value, plus a weighted sum of the residual terms of the physical equations, ensuring that the prediction results maintain mechanical rationality even in data-sparse regions. The spatiotemporal feature fusion unit uses historical wind field simulation data and measured vibration responses for joint training to establish a deep mapping between multidimensional input vectors and structural displacement and stress response. The risk grading assessment unit automatically determines the wind-induced vibration risk level at different heights and locations by comparing the predicted response value with the preset fatigue strength threshold and instability critical load.

[0023] The wind-resistant reinforcement zone generation device includes a risk distribution visualization unit, a modular reinforcement calculation unit, a node coordinate mapping unit, and an optimization instruction generation unit. The risk distribution visualization unit receives data from the physical information neural network prediction device and generates a dynamic risk heat map covering the entire structure. The modular reinforcement calculation unit is configured to calculate the required type of additional support components for areas exceeding the risk threshold, taking into account the standard spacing module of the snap-fit ​​nodes and the stress redundancy of existing members; for example, adding vertical diagonal braces or horizontal reinforcing rings of specific specifications. The node coordinate mapping unit uses a three-dimensional spatial projection algorithm to accurately map the reinforcement requirements to the modular coordinate points of the snap-fit ​​frame, ensuring that the added components can be installed using existing snap-fit ​​nodes, meeting construction operability constraints. The optimization instruction generation unit summarizes all reinforcement points and generates a digital instruction set including a component material list, spatial installation coordinate sequence, and construction sequence instructions.

[0024] The real-time meteorological data access device includes a sensor array communication unit, a data filtering unit, an anomaly removal unit, and a dynamic correction drive unit. The sensor array communication unit is configured to establish synchronous connections with ultrasonic anemometers, barometers, and humidity sensors deployed at various heights on the construction site to acquire real-time wind speed, wind direction, atmospheric stability, and turbulence intensity data. The data filtering unit employs an adaptive Kalman filter algorithm to smooth the raw data and filter out high-frequency random noise. The anomaly removal unit is configured to identify instantaneous pulse interference or sensor malfunction data through deviation analysis and supplement it using historical trends. The dynamic correction drive unit transmits the processed meteorological parameters in real-time to the wind field dynamic simulation device and the physical information neural network prediction device, triggering the system's rolling prediction and online closed-loop correction of the scheme to ensure that the construction scheme always adapts to the constantly evolving meteorological environment.

[0025] The scheme verification feedback device includes a virtual environment modeling unit, a dynamic excitation loading unit, a multi-criteria judgment unit, and an iterative optimization unit. The virtual environment modeling unit converts the optimized scheme into a structural finite element model. The dynamic excitation loading unit applies pulsating wind load excitation generated by a wind field simulation device. The multi-criteria judgment unit monitors the overall structure's displacement response, rotational deviation, and member stress changes to determine if they are within the safe design range. If the iterative optimization unit finds that the monitoring results exceed the safety tolerance, it feeds back the deviation characteristics to the wind-resistant reinforcement zone generation device, triggering a re-optimization process until the generated scheme performs stably under extreme working conditions.

[0026] Through the coordinated operation of the aforementioned devices, this system achieves a process from engineering geometric analysis to refined wind load simulation, then to deep learning prediction based on physical laws, and finally automatically generates wind-resistant reinforcement schemes that meet construction specifications. The entire process realizes a deep integration of physical mechanisms and artificial intelligence, overcoming the shortcomings of insufficient accuracy and lag in static wind load calculation in traditional design, and improving the safety and intelligence level of scaffolding systems in complex environments.

[0027] Example 2: Based on the intelligent generation system for modular erection scheme of disc-lock scaffolding described in Example 1, this example provides a variant of the system architecture based on edge computing and distributed processing, which aims to further improve the real-time response and data processing throughput of the system in large-scale building construction.

[0028] A smart generation system for modular scaffolding erection schemes is provided, whose hardware architecture adopts a distributed system consisting of a cloud-based central processing cluster and multiple on-site edge computing nodes.

[0029] The engineering geometric parameter analysis device performs preliminary lightweight BIM data processing at the edge computing node. This edge computing node is deployed in a local computer room at the construction site and integrates a high-performance graphics processing unit specifically designed for lightweight conversion of large-scale 3D point cloud data or BIM models. The geometric feature extraction unit is configured to use a voxel-mesh-based feature recognition algorithm to quickly identify the main outlines of the building facade and transform complex surface geometry into a set of discrete feature points that the edge node can process. This edge-side preprocessing method reduces the data bandwidth requirements for transmission to the cloud and improves the efficiency of initial scheme generation.

[0030] The wind field dynamic simulation device is divided into distributed simulation sub-modules. For the complex local flow fields in super high-rise buildings, the system divides the overall computational fluid domain into multiple overlapping sub-regions, each assigned to different computing cores in the cloud cluster for parallel solving. A data exchange mechanism at the interface between sub-regions ensures the continuity of fluid flow and pressure balance between different computing cores. The solver configuration unit is optimized for GPU-accelerated architectures, shortening the iteration time of the unsteady Reynolds-averaged Navier-Stokes equations through massively parallel processing.

[0031] The physical information neural network prediction device employs a federated learning and incremental training model. Edge computing nodes deployed at each construction site are responsible not only for collecting real-time meteorological data but also for executing the inference operations of the physical information neural network. The neural network architecture unit uses a lightweight model optimized by pruning, capable of running in embedded neural network computing units at the edge. After the cloud-based central processing cluster updates the model weights based on broader global wind field data, it distributes the new physical constraint logic to each edge node through a parameter distribution mechanism. The physical constraint embedding unit calculates the deviation between the structural response and physical laws in real time at the edge and fine-tunes the model parameters according to the unique local topographic features, making the prediction results more closely match the micro-meteorological characteristics of the specific construction area.

[0032] The real-time meteorological data access device integrates a local data hub at the edge. The sensor array communication unit adopts a low-power wide-area network communication protocol, enabling access to hundreds of miniature meteorological sensing nodes deployed throughout the construction site. The data filtering unit is completed at the edge, utilizing local computing resources to achieve microsecond-level data denoising. The dynamic correction drive unit can trigger emergency warning commands at the edge based on instantaneous extreme wind conditions, bypassing cloud processing, and driving the wind-resistant reinforcement zone generation device to output temporary reinforcement suggestions, thus shortening the system response time in emergency situations.

[0033] The wind-resistant reinforcement zone generation device interacts with construction workers' terminals using augmented reality technology. The generated spatial coordinates of the reinforcement components are encoded as visual guidance information and pushed to the construction workers' wearable augmented reality glasses or handheld smart terminals via a wireless network. The node coordinate mapping unit automatically aligns the geographic coordinates of the reinforcement points with the reference coordinate system of the construction site, overlaying the specific locations and installation angles of the diagonal braces to be added onto the terminal screen, guiding the construction workers to accurately erect the reinforcement according to the system-generated plan.

[0034] The scheme verification feedback device incorporates a real-time comparison function based on digital twins during the verification process. The system acquires real-time stress data of the scaffolding during actual construction by deploying wireless strain gauges and displacement sensors at key nodes of the scaffolding. The multi-criteria judgment unit compares the predicted response of the verified scheme with the measured response in real time. If the residual between the two exceeds a preset tolerance limit, the system automatically initiates a self-calibration procedure to correct the wind field simulation parameters or the weighting coefficients of the neural network. This closed-loop optimization mechanism based on real-time feedback ensures a high degree of consistency between the virtual simulation environment and the physical entity.

[0035] The system in this embodiment also includes a material supply chain collaborative management module. This module is connected to the wind-resistant reinforcement zone generation device. When the system generates a reinforcement command, the module automatically calculates the quantity of the required additional disc-lock rods, fasteners, and supporting components, and retrieves the inventory status at the construction site in real time. If the inventory is insufficient, the module is configured to automatically send a transfer or procurement request to the project material management system and optimize the logistics route according to the construction schedule to ensure that the reinforcement materials can arrive at the designated installation layer on time.

[0036] Through the optimization of the distributed architecture described above, this embodiment can better adapt to large-scale complex projects with multiple towers, achieving a reasonable allocation of computing power and a qualitative leap in emergency response speed, providing higher redundancy and robust security support for industrialized construction.

[0037] Example 3: In this example, the logic details of the intelligent generation system for modular scaffolding erection scheme are described in detail when dealing with extreme dynamic conditions (such as strong typhoons and complex gust fields), as well as the textual logic implementation of physical laws embedded in the physical information neural network prediction device.

[0038] The core of the physical information neural network prediction device lies in its physical soft constraint mechanism implemented through a loss function. During training, the neural network is configured not only to fit the given input-output data pair, but also for its output physical field variables (such as wind speed vector, pressure gradient, and rod deflection) to satisfy a set of partial differential equations in the spatiotemporal domain. The physical constraint embedding unit transforms the nonlinear convection terms, viscous dissipation terms, and pressure gradient terms in the Navier-Stokes equations into algebraic logic described in pure Chinese text: that is, at any calculation point, the sum of the rate of change of fluid density and the divergence of fluid momentum should approach zero; the rate of change of momentum per unit volume over time should be equal to the algebraic sum of the surface forces, volume forces, and inertial forces generated by fluid flow applied to that volume.

[0039] During the forward propagation of the neural network, the spatiotemporal feature fusion unit is configured with a multi-layer residual network structure, and gradient vanishing is prevented by introducing a skip connection mechanism. The spatiotemporal convolutional layer uses convolutional kernels with different receptive fields to extract multi-scale features from large-scale cyclones to small-scale pulsating vortices in the wind field. The long short-term memory unit automatically filters historical wind load sequence information that has the greatest impact on the current structural safety through its internal forget gate, input gate, and output gate logic.

[0040] The wind-resistant reinforcement zone generation device employs a sensitivity analysis method based on energy contribution when performing reinforcement calculations. The modular reinforcement calculation unit is configured to: assume a virtual stiffness increment is applied at each selectable modular point of the system, and calculate the contribution rate of this increment to the reduction of the structure's total potential energy. The node coordinate mapping unit prioritizes arranging reinforcing diagonal braces at the locations with the highest sensitivity (i.e., the locations that most significantly improve the overall wind resistance stability of the structure) according to the contribution rate from highest to lowest. When determining coordinates, the system automatically avoids existing construction passages, material hoist interfaces, and reserved personnel entrances and exits, ensuring that the reinforcement scheme logically meets the spatial avoidance rules of the construction organization design.

[0041] The wind field dynamic simulation device employs an adaptive time-domain integration strategy when handling vortex-induced resonance. When the vortex feature identification unit detects that the frequency of the shedding vortex is close to a certain natural frequency of the scaffolding system, the solver configuration unit automatically reduces the time step of the time-domain integration to capture more refined nonlinear fluid-structure interaction features. This localized refinement of a specific frequency range can accurately predict resonance damage that may occur at lower wind speeds, compensating for the blind spot of traditional static analysis, which only focuses on maximum wind speeds.

[0042] The engineering geometric parameter analysis device also integrates a knowledge base-based erection rule verification unit. This unit stores safety standards for modular scaffolding from multiple countries and industries. For each generated plan, the system automatically performs hundreds of compliance checks, including checking the slenderness ratio of uprights, calculating the strength of modular scaffold connections, and checking the continuity of scissor bracing. If the generated reinforcement plan is found to be mechanically feasible but has flaws in terms of regulations, the system will trigger an automatic correction mechanism to adjust the component types or spacing to enforce compliance with regulatory requirements.

[0043] This system also includes an online construction quality auxiliary evaluation module. This module is connected to the wind-resistant reinforcement zone generation device and uses a depth vision recognition algorithm to scan the erected scaffolding by calling on the visual sensing devices deployed on-site. The system performs a three-dimensional matching between the physical condition and the generated intelligent plan, calculates installation errors (such as verticality deviation of uprights, missing diagonal braces, and loose fasteners), and re-inputs the resulting load-bearing capacity reduction into the physical information neural network prediction device for real-time risk reassessment. This full life-cycle management logic, from design, simulation, optimization to physical feedback, constitutes the closed-loop safety barrier of this invention.

[0044] At the data storage and exchange protocol level, the system internally employs a highly compressed binary message format for inter-module communication. The real-time meteorological data access device encapsulates the processed meteorological sequence into a standard data packet containing timestamps, geographic coordinates, and multi-level wind speed vectors. This data packet undergoes symmetric encryption during transmission, ensuring data security in the open wireless environment of a construction site.

[0045] In summary, this embodiment demonstrates how neural networks acquire physical "intuition" through a textualized description of physical equations, and how the system generates scientific, compliant, and efficient solutions under complex engineering constraints. This deep interdisciplinary integration addresses the long-standing drawbacks of empirical design in scaffolding engineering, providing a solid technical path to achieving inherent safety.

[0046] Example 4: This example further illustrates the special processing logic of this system when facing the complexity of the high-altitude microenvironment in a mega-city cluster.

[0047] In the design of scaffolding for super high-rise buildings, the snagging effect and the shading effect of surrounding buildings can alter the local wind field. The fluid domain construction unit in the aforementioned wind field dynamic simulation device is configured to have large-scale scene modeling capabilities, including not only the geometric model of the project itself but also existing buildings and terrain undulations within a 500-meter radius by receiving Geographic Information System (GIS) data. The fluid domain construction unit utilizes automated Boolean operation logic to simplify surrounding buildings into roughness equivalents, while retaining the scaffolding model of the project as a high-fidelity entity.

[0048] In this complex environment, the physical constraint embedding unit of the physical information neural network prediction device is further configured to include conservation constraints of the Reynolds stress tensor. When dealing with turbulent characteristics, the system no longer relies on a single viscosity coefficient, but instead uses the spatially heterogeneous turbulent energy distribution output by the neural network to ensure that the flow field prediction results accurately reflect the shear stress generated by the altitude gradient at high altitudes. The risk grading assessment unit sets a special "aerodynamic instability early warning window" for the top area of ​​ultra-high-rise scaffolding (usually the area with the highest wind speed and turbulence intensity). When the predicted instantaneous displacement amplitude exceeds a predetermined proportion of the pole spacing, the system will forcibly activate the highest level of wind-resistant reinforcement logic.

[0049] The wind-resistant reinforcement zone generation device introduces a multi-objective optimization evolution algorithm to address such complex working conditions. When generating reinforcement instructions, the modular reinforcement calculation unit no longer merely pursues maximizing structural stiffness, but is configured to optimize two mutually exclusive objectives: minimizing the wind-induced vibration response of the structure and minimizing the total weight (i.e., cost) of the reinforcement components. The system automatically recommends a safe and economical reinforcement scheme to the user by calculating the Pareto optimal solution set. The node coordinate mapping unit automatically adjusts the setting coordinates of the horizontal wall ties according to the changes in the scaffolding step distance, utilizing the main building structure as additional lateral stiffness support to further improve overall stability.

[0050] The real-time meteorological data access device enhances the predictive foresight in this scenario. The system obtains weather trends for the next 24 to 72 hours by accessing the numerical weather prediction (NWP) interface of meteorological satellites. The dynamic correction drive unit weightedly fuses long-term forecast data with short-term real-time monitoring data, utilizing the long short-term memory unit of the physical information neural network to pre-calculate the cumulative fatigue damage to the structure caused by potential storm surges. This predictive analysis allows the installation of all wind-resistant reinforcement components to be completed on the construction site hours, or even a day, before strong winds arrive, transforming traditional passive defense into proactive prevention.

[0051] The scheme verification feedback device also serves as a dynamic demonstration of construction feasibility. By outputting the finalized modular scheme to the 3D visualization engine, the system can simulate the hoisting path of the reinforcement components, the on-site assembly sequence, and the safe working space for construction personnel. The virtual environment modeling unit embeds construction dynamics operators, which can simulate the swaying and positioning difficulties of workers installing diagonal braces in windy conditions, optimize the installation process, and ensure that the reinforcement scheme can be executed safely and accurately even under extreme weather conditions.

[0052] This embodiment demonstrates the system's comprehensive support capabilities in dealing with extreme difficulties such as the construction of super high-rise buildings and large-scale municipal engineering projects by introducing larger-scale environmental information and more refined physical models, ensuring that the layout of each member has undergone in-depth calculation and physical verification.

[0053] Example 5: This example elaborates on the logic of generating a disc-lock scaffolding erection scheme in special environments such as cross-sea bridges and riverbanks where there are strong winds, high humidity, and easy corrosion.

[0054] In the dynamic simulation of wind fields under strong wind conditions, the solver configuration unit of the wind field dynamic simulation device is configured to introduce bidirectional interactive logic of fluid-structure interaction. For scaffolding deployed in a large-span space, the small displacements of the members under strong winds will, in turn, change the surrounding flow field distribution. The solver performs two iterations in each time step: calculates the wind pressure distribution based on the current structural position, then calculates the new displacement of the structure based on the wind pressure, and updates the fluid mesh. This deeply coupled textual logic description ensures extremely high accuracy of the simulation results even under conditions where wind force is sufficient to cause large structural deformations.

[0055] In this type of working condition, the physical information neural network prediction device incorporates a damping constraint term based on dynamic characteristics into its physical constraint embedding unit. Considering the frictional force at the scaffold nodes (disc-lock connection points) and the structural damping caused by the initial gaps, the neural network is configured to learn and output the nonlinear stiffness matrix of each node. The neural network architecture unit establishes a correlation model between humidity, wind speed, and the probability of disengagement of disc-lock nodes through in-depth analysis of historical measured data from similar projects. When the real-time meteorological data input device reports that the ambient humidity exceeds a predetermined threshold (e.g., 85%) and the wind speed continues to fluctuate, the system automatically raises the risk level and instructs the wind-resistant reinforcement zone generation device to add additional pin locking check instructions or anti-loosening reinforcement components at key stress nodes.

[0056] The wind-resistant reinforcement zone generation device optimizes the arrangement logic of diagonal braces for this special geographical environment. The modular reinforcement calculation unit is configured to prioritize the generation of "spatial truss" reinforcement structures, that is, at the windward side and corners of the scaffolding, a closed force-bearing loop with extremely high torsional stiffness is formed through the geometric combination of diagonal braces. The node coordinate mapping unit pays special attention to the uplift resistance of the outriggers when calculating coordinates. For scaffolding erected on bridge piers or water platforms, the system automatically calculates the specific locations of the additional counterweights or ground anchors and adapts these non-standard reinforcement components to the standard modular design.

[0057] The real-time meteorological data access device employs multi-source heterogeneous data fusion technology in aquatic environments. In addition to anemometers, the system also incorporates parameters for correcting air density caused by wave fluctuations, water flow velocity, and temperature differences. When processing this multi-dimensional data, the data filtering unit uses correlation analysis to eliminate false disturbances caused by sensor oscillations. The anomaly removal unit is configured with self-diagnostic capabilities; when it detects a statistically significant difference between the meteorological data at a given node and surrounding nodes, it automatically isolates that node and switches it to a backup forecasting link.

[0058] The scheme verification feedback device incorporates a fatigue life assessment function during the verification phase. For cyclic loads caused by periodic gusts, the multi-criteria judgment unit processes the stress time history of key uprights based on the rainflow counting method to calculate the cumulative damage to the structure within the predetermined construction period. If the predicted cumulative damage may exceed the fatigue limit of the material before the project ends, the iterative optimization unit will force the addition of redundant supports to extend the overall service life of the structure by distributing the load paths.

[0059] This system also integrates an environmental suitability analysis unit, which is configured to automatically select the most suitable set of physical parameters (such as air viscosity and density variation curves with altitude and temperature) from the database based on the climate characteristics of different climate zones. This enables the system to be applied not only in typhoon-prone areas along the southeast coast of China, but also to provide the same high-precision support for generating construction plans in inland arid and windy areas and high-altitude low-pressure areas.

[0060] Through the organic combination of the aforementioned devices and units, the system described in this invention is not only a geometric layout tool, but also a comprehensive construction safety decision-making center integrating advanced fluid mechanics theory, nonlinear dynamics analysis, deep learning algorithms, and multi-source sensing technology. It breaks through the limitations of traditional scaffolding design, which relies on experience, standards, and static calculations, and truly achieves tailored safety protection for specific buildings, climates, and construction stages.

[0061] The above specific embodiments are merely detailed descriptions of the technical solution of the present invention, intended to enable those skilled in the art to better understand the technical essence of the present invention and implement it accordingly. The present invention is not limited to the specific embodiments described above. Any equivalent substitutions, improvements, or modifications made based on the technical concept of the present invention, utilizing common knowledge or conventional technical means in the art, or without departing from the spirit of the present invention, whether it be partial adjustments to the system architecture, replacement of hardware components, or textual equivalent transformations of algorithm logic, should be included within the scope of protection of the present invention. The scope of protection claimed by the present invention should be determined by the content of the claims. The specific details described in the specification and drawings are only for assisting in the interpretation of the claims and should not be considered as limitations on the scope of protection of the claims. In practical engineering applications, the functions of each module of this system can be merged, split, or recombined according to the specific hardware environment. These changes do not affect the completeness and advancement of the technical solution of the present invention.

Claims

1. A smart generation system for modular erection schemes of disc-lock scaffolding, characterized in that, include: The engineering geometric parameter analysis device is configured to receive three-dimensional building model data of the target construction area, extract the boundary conditions, height distribution, facade outline and node coordinate information required for scaffolding erection, and then generate a preliminary foundation erection scheme that meets the structural bearing requirements based on the preset modular rules of the disc fastener components. The wind field dynamic simulation device is configured to construct a computational fluid dynamics model of the local wind field at high altitude based on the scaffold geometry defined by the foundation erection scheme, simulate the separation, reattachment and vortex shedding process of the airflow around the scaffold under different wind directions and speeds, and output the instantaneous wind pressure distribution and wind-induced vibration response characteristics of each member surface. The physical information neural network prediction device is configured to embed physical constraint equations into a deep neural network architecture, use historical wind field simulation data and measured vibration response for joint training, establish a mapping relationship between wind load input and structural dynamic response, and predict the wind vibration risk level within a predetermined time based on the current wind field state. The wind-resistant reinforcement zone generation device is configured to receive the wind vibration risk distribution information output by the physical information neural network prediction device, identify high-risk areas that exceed a preset threshold, and automatically calculate the specific spatial coordinates of the reinforcement components to be added based on the connection module characteristics of the disc-lock scaffold, and generate wind-resistant optimization instructions. The real-time meteorological data access device is configured to acquire real-time meteorological data from the construction site from an external monitoring network and dynamically input the data into the wind field dynamic simulation device and the physical information neural network prediction device to drive the entire system to perform closed-loop correction of the construction plan.

2. The intelligent generation system for modular erection schemes of disc-lock scaffolding according to claim 1, characterized in that, The engineering geometric parameter analysis device includes: a building information model access unit, configured to retrieve the geometric shape data of the construction target in real time through a 3D design software interface; The geometric feature extraction unit is configured to perform topological analysis on the geometric shape data, identify the flatness, concavity and convexity of the building facade and the location of openings, and identify cantilevered, recessed or curved irregular areas in the building facade. The modular mapping unit internally stores a specification database of standard disc-lock scaffolding. The specification database includes the length sequence of uprights, the modular step distance of horizontal bars, and the geometric matching rules of diagonal bars. The modular mapping unit is configured to discretize the physical space corresponding to the irregular area into three-dimensional mesh nodes that meet the modular requirements. The preliminary scheme generation unit is configured to automatically increase the initial erection density in the irregular area, determine the step distance and longitudinal distance of the scaffolding in combination with the building height distribution, and initially arrange the uprights, horizontal bars and scaffold boards according to the preset static safety threshold, as the input benchmark for subsequent dynamic analysis.

3. The intelligent generation system for modular erection schemes of disc-lock scaffolding according to claim 2, characterized in that, The wind field dynamic simulation device includes: The fluid domain construction unit is configured to extend outward from the scaffold geometric skeleton in the basic erection scheme to establish a computational fluid domain containing an atmospheric boundary layer, and to perform local mesh refinement processing for the dense rod characteristics of the scaffold. The solver configuration unit is configured to use an unsteady Reynolds-averaged Navier-Stokes equation solver, and combined with the large eddy simulation method, to perform spatiotemporal discretization of mass conservation and momentum conservation of fluid flow, and calculate the local wind pressure gradient on the windward and leeward sides. The vortex feature recognition unit is configured to capture vortex-induced resonance phenomena caused by abrupt changes in structural geometry. By analyzing the separation and reattachment points of the airflow, the frequency and intensity of the detached vortices in the flow field are extracted. The wind pressure mapping unit is configured to convert the calculated spatial flow field pressure field into time-history wind pressure loads applied to the surface of specific scaffold members.

4. The intelligent generation system for modular erection schemes of disc-lock scaffolding according to claim 3, characterized in that, The physical information neural network prediction device includes: The neural network architecture unit contains multiple spatiotemporal convolutional layers and long short-term memory units. The spatiotemporal convolutional layers are used to capture the topological features of wind pressure in spatial distribution, and the long short-term memory units are used to record the time dependence of wind load evolution. A physical constraint embedding unit is configured to transform the Navier-Stokes equations, continuity equations, and structural dynamics equations into loss function components of a neural network. In the loss function components, physical equation residual terms are defined, and the logical expression of the physical equation residual terms is: at any calculation point, the sum of the rate of change of fluid density and the divergence of fluid momentum should approach zero, and the rate of change of momentum per unit volume with time should be equal to the algebraic sum of the surface forces, body forces, and inertial forces generated by fluid flow applied to that volume. The spatiotemporal feature fusion unit is configured to use historical wind field simulation data and measured vibration response for joint training to establish a deep mapping between multidimensional input vectors and structural displacement and stress response. The risk grading assessment unit is configured to compare the predicted response value with preset fatigue strength thresholds and instability critical loads to determine the wind-induced vibration risk level at different heights and locations.

5. The intelligent generation system for modular erection schemes of disc-lock scaffolding according to claim 4, characterized in that, The wind-resistant reinforcement zone generation device includes: A risk distribution visualization unit is configured to receive data output from the physical information neural network prediction device and generate a dynamic risk heat map covering the entire structure. The modular reinforcement calculation unit is configured to use the energy contribution sensitivity analysis method to calculate the contribution rate of applying virtual stiffness increments to the reduction of the total potential energy of the structure after applying them to each selectable modular point of the system, and to determine the type of support member to be added in order of the contribution rate from high to low for areas exceeding the risk threshold. The node coordinate mapping unit is configured to use a three-dimensional spatial projection algorithm to accurately map the reinforcement requirements to the modular coordinate points of the disc-lock frame, and automatically avoid the original construction passage, material hoist interface and reserved personnel entrance and exit. The instruction generation unit is optimized and configured to summarize all reinforcement points and generate a digital instruction set that includes a component material list, spatial installation coordinate sequence, and construction sequence description.

6. The intelligent generation system for modular erection schemes of disc-lock scaffolding according to claim 5, characterized in that, The real-time meteorological data access device includes: The sensor array communication unit is configured to establish a synchronous connection with meteorological sensors deployed at various heights on the construction site to acquire real-time data on wind speed, wind direction, atmospheric stability, and turbulence intensity. The data filtering unit is configured to use an adaptive Kalman filter algorithm to smooth the raw data and filter out high-frequency random noise. An anomaly removal unit is configured to identify transient pulse interference through deviation analysis and to complete the data using historical trends; The dynamic correction drive unit is configured to transmit the processed meteorological parameters to the wind field dynamic simulation device and the physical information neural network prediction device in real time, triggering the system's rolling prediction and online closed-loop correction of the scheme.

7. The intelligent generation system for modular erection schemes of disc-lock scaffolding according to claim 6, characterized in that, The modular scaffolding erection scheme intelligent generation system also includes a scheme verification and feedback device, which includes: The virtual environment modeling unit is configured to convert the optimized construction scheme into a structural finite element model; the dynamic excitation loading unit is configured to apply the pulsating wind load excitation generated by the wind field dynamic simulation device to the structural finite element model. The multi-criteria judgment unit is configured to monitor the displacement response, rotational deviation and stress change of the overall structure and determine whether they are within the preset safety design value range. The iterative optimization unit is configured to feed back the deviation characteristics to the wind-resistant reinforcement zone generation device when the monitoring results exceed the safety tolerance, triggering a re-optimization process until the generated construction scheme meets the dynamic stability requirements.

8. The intelligent generation system for modular erection schemes of disc-lock scaffolding according to claim 7, characterized in that, The intelligent generation system for the modular erection scheme of the disc-lock scaffolding adopts a distributed architecture and includes: Edge computing nodes deployed at the construction site are configured to perform lightweight transformation of 3D point cloud data and generate preliminary solutions based on the lightweight data. A central processing cluster deployed in the cloud is configured to solve the tasks of the wind field dynamic simulation device in parallel. The central processing cluster divides the overall computational fluid domain into multiple overlapping sub-regions. Each sub-region is assigned to a different computing core. The fluid flow continuity and pressure balance between the sub-regions are maintained through an interface data exchange mechanism. The edge computing node is also configured to receive physical constraint logic issued by the central processing cluster and fine-tune the parameters of the neural network model according to the unique terrain features of the site.

9. The intelligent generation system for modular erection schemes of disc-lock scaffolding according to claim 8, characterized in that, The intelligent generation system for modular scaffolding erection schemes also includes an online construction quality auxiliary evaluation module, which is configured to: The on-site visual sensing device is invoked, and a depth vision recognition algorithm is used to scan the erected scaffolding, and the physical state is matched with the generated erection plan in three dimensions. Calculate the installation error, which includes the verticality deviation of the upright, the omission of diagonal braces, and the abnormal tightness of the disc buckle connection. The load-bearing capacity attenuation parameters caused by the installation error are fed back to the physical information neural network prediction device for real-time risk reassessment.

10. The intelligent generation system for modular erection schemes of disc-lock scaffolding according to claim 9, characterized in that, When the wind field dynamic simulation device handles fluid-structure interaction conditions under strong wind environments, the solver configuration unit is configured as follows: Two iterative calculations are performed within each preset time step. First, the wind pressure distribution is calculated based on the current structural spatial location. Then, the displacement of the structure is calculated based on the calculated wind pressure distribution, and the topology of the fluid mesh is updated. When the vortex feature identification unit detects that the frequency of the shedding vortex in the flow field is close to the natural frequency of the scaffolding system, the solver configuration unit automatically reduces the time step of the time domain integration in order to capture the nonlinear fluid-structure interaction features.

Images