Large-span roof construction method based on intelligent model and monitoring system

Abstract

The invention relates to the technical field of building construction, and discloses a large-span roof construction method and monitoring system based on an intelligent model, and the method comprises the steps: carrying out the multi-objective optimization and structure verification of roof slope finding parameters in a BIM platform, so as to generate a final construction blueprint model, formulating a laser-guided construction scheme according to the model, in the paving operation, self-adaptive paving of the slope making layer is completed through real-time sampling inspection and feedback, finally, a completion point cloud model is generated through three-dimensional laser scanning and compared with a blueprint model, and closed-loop verification of the construction quality is completed; the system comprises a model generation module, a construction strategy module, a self-adaptive construction control module and a quality verification module. According to the method, multi-objective optimization of the design front end, dynamic closed-loop feedback of the construction process and digital verification of the completion stage are deeply fused, and digital closed-loop management and control of the whole process from design, construction to acceptance are achieved.

Description

Technical Field

[0001] This invention relates to the field of building construction technology, and in particular to a construction method and monitoring system for large-span roofs based on intelligent models. Background Technology

[0002] Large-span roofs are key structural units in modern large-scale public buildings and are widely used in stadiums, convention centers and transportation hubs. One of the core technical challenges is how to accurately construct the slope layer to achieve efficient drainage over a large area and strictly control the impact of new loads on the main structure. Therefore, the construction quality of the slope layer directly determines the long-term safety and performance of the roof system.

[0003] In existing large-span roofing projects, a conventional technical approach is generally followed. In the design phase, technicians determine key control parameters such as slope and elevation on two-dimensional drawings based on relevant specifications and engineering experience. In the construction phase, on-site personnel use these parameters to set up discrete control points using tools such as levels or laser leveling instruments, and the paving operation is carried out based on these points. In the final acceptance phase, quality assessment often adopts a random sampling method, using straightedges or leveling to verify the flatness and elevation of local areas.

[0004] This inherited operating system has inherent technical limitations when dealing with the complex requirements of large-span roofs. First, the design lacks optimal scientific basis at the source. The experience-based design method struggles to find the optimal solution when facing multiple objectives such as drainage efficiency and the sensitive impact of additional loads from the slope layer on the large-span structure below, often leading to unnecessary structural redundancy or potential performance risks. Second, the construction process is a static open-loop control, completely ignoring the settlement and deformation caused by the physical properties of materials such as concrete. Static construction guidance cannot capture and compensate for this dynamic process, ultimately causing a systematic deviation between the completed entity and the design model. Finally, the evaluation methods for final acceptance have limitations. For a vast work area of ​​tens of thousands of square meters, discrete sampling inspections cannot statistically represent the whole, and the conclusions are subject to chance, leaving undetected local quality defects as long-term hidden dangers. Summary of the Invention

[0005] The purpose of this invention is to provide a construction method and monitoring system for large-span roofs based on intelligent models, which solves the problems in traditional large-span roof construction, such as lack of scientific optimization in the design stage, inability to dynamically correct deviations during construction, and one-sided and inaccurate means of final verification.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The construction method for large-span roofs based on intelligent models includes the following steps:

[0008] Step S1: In the BIM platform, the optimization algorithm is used to automatically optimize the composite slope parameters of the roof through multiple objectives, and the structural performance correlation of the optimized scheme is verified to generate a final construction blueprint model that takes into account both drainage efficiency and structural performance.

[0009] Step S2: Using the final construction blueprint model, calculate the spacing of laser control points and formulate a multi-equipment collaborative operation strategy to form a complete laser-guided construction plan;

[0010] Step S3: Set up a laser leveling instrument according to the laser-guided construction plan, and dynamically correct the laser-guided surface through real-time sampling and feedback during the paving operation to complete the adaptive paving operation of the slope-finding layer;

[0011] Step S4: Use a 3D laser scanner to scan the slope-finding layer formed after the adaptive paving operation to generate an as-built point cloud model, and compare the deviation between the as-built point cloud model and the final construction blueprint model to complete the closed-loop verification of construction quality.

[0012] Preferably, step S1 specifically includes the following steps:

[0013] A genetic algorithm is used as the optimization algorithm to iteratively optimize the roof composite slope-finding parameters to generate a set of optimized roof composite slope-finding parameters.

[0014] Based on the optimized roof composite slope-finding parameters, a structural finite element model was established, and it was compared and analyzed with a structural finite element model using a traditional single slope-finding scheme under the same design load to complete the structural performance correlation verification.

[0015] After verifying the effectiveness of the optimized roof composite slope parameters through comparative analysis, a three-dimensional model was generated in the BIM platform using the optimized roof composite slope parameters to obtain the final construction blueprint model.

[0016] Preferably, the comparative analysis verifies the effectiveness of the optimized roof composite slope-finding parameters, specifically referring to:

[0017] When the internal force values ​​and long-term deflection values ​​of the key structural components in the structural finite element model established based on the optimized roof composite slope-finding parameters are both less than the corresponding values ​​in the structural finite element model using the traditional single slope-finding scheme, the validity verification is deemed to have passed.

[0018] Preferably, step S2 specifically includes the following steps:

[0019] The allowable deviation value of surface flatness is extracted from the final construction blueprint model, and the maximum spacing of laser control points is calculated by combining the automatic leveling accuracy of the laser leveling instrument used.

[0020] Plan the placement of at least two laser leveling transmitters so that the signal ranges of the laser leveling transmitters overlap and cover adjacent construction areas.

[0021] By integrating the maximum deployment spacing and the deployment location, a laser-guided construction scheme containing control point coordinates and equipment coordination instructions is generated.

[0022] Preferably, the method for calculating the maximum spacing of the laser control points is as follows:

[0023] Divide the allowable deviation of the surface flatness by the automatic leveling accuracy of the equipment, and determine the quotient as the maximum spacing of the laser control points.

[0024] Preferably, step S3 specifically includes the following steps:

[0025] A laser leveling instrument is set up according to the laser-guided construction plan described above;

[0026] During the paving operation, a handheld 3D laser scanner is used to perform real-time spot checks on the paved area, and the deviation data generated by the spot checks is transmitted in real-time to the central controller of the laser leveling instrument via a wireless data link.

[0027] Based on the deviation data, a correction instruction is generated to dynamically correct the laser-guided surface in order to complete the adaptive paving operation of the slope-finding layer.

[0028] Preferably, the dynamic correction of the laser guiding surface specifically refers to:

[0029] The paved area is scanned in real time using a portable scanning device to generate a real-time point cloud, which is then compared with the final construction blueprint model to calculate the elevation deviation caused by concrete settlement.

[0030] An elevation compensation value is generated based on the elevation deviation, and this elevation compensation value is fed back to the laser leveling instrument control system to adjust the laser guiding surface of the subsequent unpaved area.

[0031] Preferably, step S4 specifically includes the following steps:

[0032] Under a unified coordinate system, the as-built point cloud model and the final construction blueprint model are spatially registered and superimposed.

[0033] Calculate the normal spatial deviation value between each point on the as-built point cloud model and the surface of the final construction blueprint model to generate a set of deviation datasets;

[0034] A three-dimensional deviation cloud map analysis report is generated based on the deviation dataset, and the closed-loop verification of the construction quality is completed based on the report.

[0035] Preferably, the closed-loop verification of the construction quality based on the report specifically refers to:

[0036] The maximum deviation value is extracted from the deviation dataset and compared with a preset quality acceptance threshold. When the maximum deviation value is not greater than the quality acceptance threshold, the construction quality is deemed qualified.

[0037] A large-span roof construction monitoring system based on intelligent models includes:

[0038] The model generation module is configured to perform multi-objective automatic optimization of roof composite slope parameters in the BIM platform, and integrates a structural verification unit to verify the optimization scheme and generate the final construction blueprint model.

[0039] The construction strategy module is used to receive the final construction blueprint model generated by the model generation module, and calculate the layout spacing of laser control points and the multi-equipment collaborative operation strategy accordingly to generate the laser-guided construction scheme.

[0040] An adaptive construction control module is configured to deploy a laser leveling device on site according to the laser-guided construction scheme, and communicate with a portable scanning device to receive real-time sampling data, thereby generating dynamic correction instructions to control the adaptive paving operation.

[0041] The quality verification module is used to receive the as-built point cloud model generated by the 3D laser scanner and call the final construction blueprint model to perform deviation comparison analysis, so as to output the closed-loop verification result of construction quality.

[0042] In summary, the present invention has at least one of the following beneficial technical effects:

[0043] 1. This invention adopts a scheme that combines multi-objective automatic optimization with structural performance verification, realizing the scientific quantification and pre-optimization of roof slope design. This changes the existing design mode that relies on engineers' experience for repeated trial calculations and adjustments, and helps to solve the inherent defects of traditional design, which is highly subjective, difficult to take into account multiple performance indicators, and prone to structural redundancy or risk.

[0044] 2. This invention introduces a closed-loop control method based on real-time sampling and feedforward compensation, which gives the paving operation adaptive correction capability, helps to improve the accuracy and intelligence level of the construction process. Compared with the existing technology that follows static scheme construction and error lag detection, it helps to overcome construction deviations caused by physical factors such as concrete settlement, and avoids expensive rework and quality disputes in the later stage.

[0045] 3. This invention constructs a full-process digital closed-loop management and control system from design optimization to final verification, achieving a comprehensive and objective quantitative assessment of the construction quality of large-span roofs. Compared with the traditional quality inspection methods that use manual sampling for testing, this invention helps to solve the problems of low efficiency, insufficient coverage, and susceptibility to human factors in traditional testing methods, providing high-fidelity digital support for project delivery. Attached Figure Description

[0046] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0047] Figure 2 This is a flowchart illustrating the multi-objective optimization and structural verification process of the present invention.

[0048] Figure 3 This is a schematic diagram illustrating the planning of laser signal coverage and control point layout according to the present invention;

[0049] Figure 4 This is a schematic diagram illustrating the closed-loop control principle of the adaptive paving operation of the present invention.

[0050] Figure 5 This is a schematic diagram illustrating the deviation comparison and analysis between the as-built point cloud and the blueprint model of the present invention.

[0051] Figure 6 This is a system framework diagram of the present invention. Detailed Implementation

[0052] The following is in conjunction with the appendix Figure 1 - Appendix Figure 6 The present invention will be further described in detail below.

[0053] This invention provides a construction method for large-span roofs based on intelligent models, comprising the following steps:

[0054] Step S1: In the BIM platform, the optimization algorithm is used to automatically optimize the composite slope parameters of the roof through multiple objectives, and the structural performance correlation of the optimized scheme is verified to generate a final construction blueprint model that takes into account both drainage efficiency and structural performance.

[0055] Specifically, step S1 is initiated in the BIM platform. The BIM platform serves as the central data integration and processing environment for roof structure and building information, providing a unified data foundation for all subsequent calculations and analyses.

[0056] In the BIM platform, for the design of large-span roofs, it is necessary to perform multi-objective automatic optimization of the composite slope-finding parameters of the roof. These parameters include, but are not limited to, the slope, aspect, layout of drainage gutters, and thickness distribution of slope-finding layer materials in each area.

[0057] The optimization process aims to achieve two optimization objectives:

[0058] That is, to maximize the overall drainage efficiency of the roof and minimize the additional structural load caused by the slope layer.

[0059] To achieve these dual objectives, this implementation method preferably uses a genetic algorithm as the optimization algorithm. The genetic algorithm first generates an initial population, in which each individual represents a set of randomly generated composite roof slope-finding parameters encoded as chromosomes.

[0060] Subsequently, the genetic algorithm evaluates the fitness of each individual. The fitness is quantified based on the two optimization objectives mentioned above, and a multi-objective fitness function can be constructed. The function is represented as follows:

[0061] ;

[0062] In the formula, This represents the overall fitness function value of an individual. A set of composite roof slope-finding parameters that includes information such as slope and aspect. For parameter-based The evaluation function for overall roof drainage efficiency is derived from fluid dynamics simulation calculations. For parameter-based Evaluation function for the new structural load of the slope-finding layer, derived from volume and density calculations; and These are the weighting coefficients for drainage efficiency and structural load, respectively.

[0063] Next, a new generation of population is generated through selection, crossover, and mutation operations. The selection operation employs a roulette wheel selection method to achieve higher [quality / efficacy]. Individuals with a higher value are more likely to be selected as parents.

[0064] Crossover operation uses single-point crossover, which means randomly selecting the chromosomes of two parent individuals and randomly selecting a crossover point, exchanging some gene segments after the crossover point, thereby generating two new offspring individuals.

[0065] Mutation operations involve randomly altering the value of a gene on the chromosome of offspring individuals with a predetermined low probability. For example, the slope parameter of a certain area can be fine-tuned within an allowable range to increase population diversity.

[0066] This iterative process is repeated until the function converges or the preset maximum number of iterations is reached, and finally a set of optimal roof composite slope-finding parameters is output, that is, the optimized roof composite slope-finding parameters.

[0067] After obtaining this set of optimization parameters, their structural performance correlation must be verified to ensure the safety and feasibility of the optimization results in engineering.

[0068] The verification process begins with establishing a structural finite element model within the BIM platform or associated structural analysis software, based on the optimized roof composite slope parameters.

[0069] Meanwhile, a structural finite element model using the traditional single slope-finding scheme is established as a comparison benchmark. The traditional scheme usually refers to the conventional design that only meets the minimum drainage slope requirement without multi-objective optimization.

[0070] The two structural finite element models were compared and analyzed under the same design load conditions. The design load should comprehensively cover the combination of dead load, live load, wind load and snow load that the roof may bear.

[0071] The comparative analysis focused on extracting and comparing the internal force values ​​and long-term deflection values ​​of key structural components in the two models. Key components included primary and secondary steel beams, roof trusses, and critical connection nodes.

[0072] The pass criteria for validity validation are strictly defined as follows:

[0073] When the internal force values ​​and long-term deflection values ​​of the key structural components in the finite element model established based on the optimized roof composite slope-finding parameters are both less than the corresponding values ​​in the finite element model using the traditional single slope-finding scheme, the verification is deemed successful.

[0074] Finally, after verifying the effectiveness of the optimized roof composite slope-finding parameters, the parameters were applied to the BIM platform to automatically generate a refined 3D model.

[0075] The specific process of generating a 3D model is as follows:

[0076] By calling the secondary development interface (API) of the BIM platform, a parametric modeling script is written. The parametric modeling script reads the optimized roof composite slope parameters and converts these numerical parameters into topological constraints on the roof geometry.

[0077] For example, the script calculates the three-dimensional coordinates of key control points on the top surface of the slope-finding layer based on the optimized thickness and slope parameters, and automatically generates a three-dimensional solid geometric model of the slope-finding layer based on these coordinate points, ultimately obtaining a final construction blueprint model that balances excellent drainage efficiency and reliable structural performance.

[0078] This implementation method constructs a dual-objective optimization process with genetic algorithm as the core and finite element verification, and uses parametric scripts to realize the automatic conversion of optimization results to BIM models. This achieves automated and optimized generation of slope-finding design for large-span roofs while ensuring structural safety, which helps to significantly improve design quality and efficiency, and reduce later construction risks and construction costs.

[0079] Step S2: Using the final construction blueprint model, calculate the spacing of laser control points and formulate a multi-equipment collaborative operation strategy to form a complete laser-guided construction plan;

[0080] Specifically, step S2 aims to utilize the previously generated final construction blueprint model to construct a complete and directly executable laser-guided construction scheme through precise calculations and planning.

[0081] First, from the final construction blueprint model, the key quality control indicators, namely the allowable deviation value of surface flatness, are accurately extracted through data interfaces or parametric scripts. This value is a preset engineering attribute in the BIM model.

[0082] At the same time, the automatic leveling accuracy of the laser leveling instrument used is obtained. This accuracy is usually expressed in angular units (such as arcseconds) or vertical deviation at a specific distance (such as mm / m). It is a key parameter characterizing the performance of the equipment and is provided by the equipment manufacturer.

[0083] Based on the two key parameters mentioned above, the maximum spacing between laser control points is calculated. This calculation is not just a simple numerical division, but a precise model based on geometric optics and error accumulation. The solution method can be expressed as follows:

[0084] ;

[0085] In the formula, This represents the maximum spacing between laser control points. These are the permissible deviations in surface flatness extracted from the final construction blueprint model; The equivalent angle value for the automatic leveling accuracy of the laser leveling instrument used; A safety factor less than 1, ranging from 0.8 to 0.95, is used to account for uncertainties such as environmental vibrations at the site.

[0086] Next, given the complex spatial characteristics of large-span roofs due to the numerous structural components (such as trusses and beams), a comprehensive multi-device collaborative operation strategy must be developed, including planning the placement of at least two laser leveling transmitters.

[0087] The planning process described above is as follows:

[0088] First, the three-dimensional geometric information of the structural components in the final construction blueprint model is imported into an optical simulation environment.

[0089] Secondly, a virtual model of the laser emitter is defined, and it is given physical attributes such as actual signal power, divergence angle, and effective range.

[0090] Then, line-of-sight analysis or ray tracing simulation is performed in this environment. By placing virtual transmitters at different candidate locations, the signal intensity distribution of the laser signal on the working surface of the roof slope-finding layer is calculated, and signal blind spots caused by structural obstruction are identified.

[0091] By iteratively optimizing the placement of the transmitters, the simulation results showed that the signal strength of the entire working area was higher than the preset minimum reception threshold, and a signal overlap coverage zone with a width of not less than 1 meter was formed between the construction areas of two adjacent transmitters.

[0092] Finally, the calculated maximum deployment spacing and the planned deployment location are integrated to generate a laser-guided construction scheme that includes control point coordinates and equipment coordination instructions. This scheme is stored in a structured data format such as XML or JSON.

[0093] The control point coordinates are generated by applying an automatic point placement algorithm in the BIM model. This algorithm uses... To establish a spacing, a uniform grid is generated on the three-dimensional surface of the roof slope layer, and the precise three-dimensional coordinates of all grid points are extracted to form a detailed set of control point coordinates.

[0094] Device coordination instructions are a set of instructions that define the logical relationships between devices, specifically including:

[0095] Each device's device ID, scheduled start and stop times, laser power levels for different construction areas, and handshake mechanisms and protocols for seamless master-slave signal switching when the paving equipment moves from one signal zone to an overlapping zone.

[0096] This implementation method, by applying a spacing calculation method based on a geometric optics model and a device layout planning based on ray tracing simulation, forms a laser-guided construction scheme containing a precise three-dimensional coordinate set and complex collaborative logic instructions. This helps to transfer design accuracy to the construction site without loss, laying a solid foundation for achieving high-quality automated paving operations.

[0097] Step S3: Set up a laser leveling instrument according to the laser-guided construction plan, and dynamically correct the laser-guided surface through real-time sampling and feedback during the paving operation to complete the adaptive paving operation of the slope-finding layer;

[0098] Specifically, step S3 is a key execution link connecting the design model and physical construction. The core of step S3 is to deploy on-site according to the laser-guided construction plan and establish a dynamic, adaptive closed-loop control system to complete the slope-finding layer paving operation.

[0099] First, based on the planned deployment locations in the laser-guided construction plan, at least two laser leveling instruments are set up at the construction site. On-site technicians use measuring equipment such as total stations to accurately lay out the three-dimensional coordinates of the instruments given in the plan on the site, and perform precision leveling and fixing on the base of each laser leveling instrument to ensure the stability of its physical position.

[0100] After the physical setup is completed, the laser level is calibrated. This process aligns and binds the internal working coordinate system of the laser level with the global coordinate system of the final construction blueprint model. This ensures that the laser guidance reference plane emitted by the equipment is completely consistent with the design elevation in the BIM model in space.

[0101] Subsequently, the equipment coordination instructions in the laser-guided construction plan are loaded into the central controller of the laser leveling instrument via a wireless data link. The central controller then manages the startup, signal switching, and power of each device, forming a collaborative laser-guided network.

[0102] Once the paving operation officially begins, the adaptive control loop is activated. While the paver is paving, the on-site quality inspector uses a portable 3D laser scanner to perform real-time spot checks on areas where paving has just been completed but the concrete has not yet fully set.

[0103] Real-time spot checks are conducted within 3 to 5 minutes after paving is completed to capture the initial settlement caused by the concrete's own weight and water bleeding. The scanner quickly acquires a high-density three-dimensional point cloud of the paved area surface, i.e., a real-time point cloud.

[0104] The aforementioned real-time point cloud is immediately transmitted to the central controller of the laser leveling instrument via a wireless data link (such as Wi-Fi or 5G network).

[0105] After receiving the real-time point cloud, the central controller first performs coordinate transformation on it and registers it to the global coordinate system of the final construction blueprint model.

[0106] Next, the controller performs a deviation comparison analysis, which calculates the normal spatial deviation between each point in the real-time point cloud and the design surface of the final construction blueprint model, thereby obtaining the precise elevation deviation caused by concrete settlement. .

[0107] Based on this elevation deviation, the central controller calculates and generates a feedforward elevation compensation value. This compensation value is calculated to preemptively offset the settlement that will occur in the subsequent paving area. The calculation method can be expressed as follows:

[0108] ;

[0109] In the formula, This is the final generated elevation compensation value; This refers to the elevation deviation caused by concrete settlement, calculated by comparing real-time point cloud data with the blueprint model. It is a preset settlement prediction coefficient, with a value ranging from 0.05 to 0.2, used to predict and compensate for the subsequent settlement of concrete from the initial setting to the final setting stage.

[0110] Finally, the central controller will adjust the elevation compensation value. As a correction instruction, the data is fed back to the control system of the laser leveling instrument. Based on this, the control system dynamically corrects the laser-guided surface of the unpaved areas. This correction is specifically achieved by fine-tuning the laser head's emission angle or applying a parameter to the laser signal receiver's processing program. This is achieved using an equivalent digital bias.

[0111] The laser receiver on the paving equipment will immediately lock onto this dynamically raised area. The laser guides the surface and controls the paver's scraper to automatically adjust to the new height. This ensures that the top surface of the newly paved concrete is higher than the pre-designed elevation. When it settles naturally After subsequent settlement, its final elevation will precisely match the design elevation, thus completing the adaptive paving operation of the slope-finding layer.

[0112] This implementation method constructs a closed-loop control system that includes real-time scanning, deviation calculation, and feedforward compensation. This system enables real-time and automatic correction of construction errors caused by dynamic physical factors such as concrete settlement. This helps to significantly improve the final construction accuracy and flatness of the slope-finding layer for large-span roofs and significantly reduces the reliance on manual retesting and subsequent rework.

[0113] Step S4: Use a 3D laser scanner to scan the slope-finding layer formed after the adaptive paving operation to generate an as-built point cloud model. Then, compare and analyze the deviation between the as-built point cloud model and the final construction blueprint model to complete the closed-loop verification of construction quality.

[0114] Specifically, step S4 constitutes the final quality verification and data archiving closed loop of the entire intelligent construction method. Its purpose is to conduct a thorough geometric quality inspection of the final solidified slope layer using millimeter-level precision three-dimensional digital means, and to strictly compare it with the original blueprint model that carries the design intent, thereby completing the final quantitative evaluation and acceptance of the physical construction results.

[0115] This step should be initiated when the concrete of the slope-finding layer has reached more than 90% of its design strength and the curing period has ended, to ensure that the scanning surface has reached the final geometrically stable state.

[0116] First, a high-precision terrestrial 3D laser scanner, such as a phase-based or time-of-flight scanner, is used to perform a comprehensive 3D scan of the entire roof slope layer formed after the adaptive paving operation. To ensure comprehensive data coverage of the complex structure of the large-span roof without blind spots, multiple station scans are required on-site based on a pre-planned network of stations. At the same time, the equipment is fully preheated and atmospheric parameters (such as temperature and humidity) are calibrated to eliminate the influence of environmental factors on the ranging accuracy.

[0117] Subsequently, the raw point cloud data with overlapping areas obtained from multiple stations are imported into the post-processing software for stitching and denoising preprocessing. The stitching process uses a common target sphere with high reflectivity pre-deployed in the scanning area as a reference control point. The Iterative Closest Point (ICP) algorithm is used to automatically calculate the accurate spatial transformation matrix between each station and seamlessly merge the point cloud data of each sub-station into a single, complete as-built point cloud model. The denoising process uses algorithms such as Statistical Outlier Removal to remove isolated noise points caused by personnel movement, equipment vibration, or airborne suspended particles.

[0118] Next, in order to conduct effective deviation comparison analysis, the preprocessed as-built point cloud model and the final construction blueprint model must be placed under an absolutely unified spatial reference. This process is as follows:

[0119] Using the advanced engineering control points shared in BIM modeling and on-site construction layout as absolute coordinate references, the as-built point cloud model and the final construction blueprint model are accurately spatially registered and rigidly transformed and superimposed to ensure that the position and orientation of the two digital models in the virtual space are completely consistent with the correspondence in the real world.

[0120] After the two models are accurately registered and superimposed, the system will start the core deviation calculation engine. This calculation takes each data point on the as-built point cloud model as the object and performs a fast neighborhood search using efficient data structures such as k-dimensional trees to find the nearest normal projection point on the 3D mesh surface of the final construction blueprint model. The Euclidean distance between these two points is then calculated, which is the normal spatial deviation value of this point. After traversing millions of points in the as-built point cloud model, a deviation dataset containing a massive amount of high-precision deviation values ​​is generated.

[0121] Based on this deviation dataset, the system will automatically generate a detailed 3D deviation cloud map analysis report. This report not only presents the construction quality of the entire roof in an intuitive color rendering format, making the areas exceeding the deviation limit immediately clear, but also includes rich statistical analysis results, such as the maximum positive and negative deviation values, average deviation, root mean square error (RMSE), and histograms of deviation value distribution. It can even output contour maps of flatness for specific areas.

[0122] Finally, based on the 3D deviation cloud map analysis report, a closed-loop verification of construction quality is completed. The verification process is a multi-dimensional and quantifiable comprehensive judgment process. Specifically, it refers to automatically retrieving and extracting the maximum normal space deviation value from the deviation dataset using an algorithm, and comparing it with a preset quality acceptance threshold. The judgment conditions can be expressed as follows:

[0123] ;

[0124] In the formula, This represents the absolute value of the maximum normal space deviation extracted from the deviation dataset; These are the quality acceptance thresholds preset according to national construction acceptance standards or engineering contracts.

[0125] The system can determine that the construction quality is qualified only when the absolute value of the maximum deviation is not greater than the quality acceptance threshold, and the percentage of the area of ​​the deviation exceeding the limit to the total area is lower than the preset secondary acceptance index.

[0126] The above verification results, the three-dimensional deviation cloud map, and all relevant statistical analysis data will together constitute a complete digital construction quality verification file, which will be permanently associated and stored in the corresponding component attributes of the BIM platform.

[0127] This implementation method achieves an objective, comprehensive, and refined quantitative assessment of the construction quality of large-span roofs through a complete process, from high-precision data acquisition and planning, point cloud fusion based on the iterative nearest point algorithm, to normal deviation calculation based on kd-tree search and multi-dimensional statistical analysis. This helps to establish a truly meaningful digital quality control closed loop from design to completion, providing an unprecedented high-fidelity data foundation for the digital delivery and subsequent operation and maintenance of the project.

[0128] A large-span roof construction monitoring system based on intelligent models includes:

[0129] The model generation module is configured to perform multi-objective automatic optimization of roof composite slope parameters in the BIM platform, and integrates a structural verification unit to verify the optimization scheme and generate the final construction blueprint model.

[0130] The construction strategy module receives the final construction blueprint model generated by the model generation module and calculates the layout spacing of laser control points and the multi-equipment collaborative operation strategy accordingly to generate a laser-guided construction plan.

[0131] The adaptive construction control module is configured to deploy the laser leveling instrument on site according to the laser-guided construction plan, and communicate with the portable scanning device to receive real-time sampling data, thereby generating dynamic correction instructions to control the adaptive paving operation.

[0132] The quality verification module receives the as-built point cloud model generated by the 3D laser scanner and calls the final construction blueprint model to perform deviation comparison analysis, so as to output the closed-loop verification results of construction quality.

[0133] Specifically, the system is deployed on one or more interconnected servers or high-performance workstations and communicates with the BIM platform, on-site construction equipment and surveying instruments through data interfaces. The system includes a model generation module, a construction strategy module, an adaptive construction control module and a quality verification module.

[0134] The model generation module is the data source for the entire intelligent process, and it further includes a parameterization unit, an optimization engine, a structure verification unit, and a model synthesis unit.

[0135] The parameterized unit configuration is used to connect to the BIM platform and read the geometric and topological information of the original roof design, transforming it into a set of composite slope-finding parameter variables that can be called by the optimization algorithm. The optimization engine has built-in the core logic of the genetic algorithm, and the configuration is used to execute genetic operators such as selection, crossover and mutation, and call the preset multi-objective fitness function to iteratively optimize the individuals in the population in order to output a set of theoretically optimal slope-finding parameters.

[0136] Next, the structural verification unit, as an independent simulation verification interface, is configured to receive the optimal parameters output by the optimization engine and automatically call the API interface of the third-party structural finite element analysis software to create refined finite element analysis models based on these parameters and traditional schemes.

[0137] Subsequently, equivalent design loads are applied and structural analysis is performed. Finally, the internal force and deflection values ​​of key components are compared through internal comparator logic to determine the structural effectiveness of the optimized scheme. Only after verification is passed will the model synthesis unit be activated. This unit is configured to execute parametric modeling scripts, which transform the verified optimal parameters into geometric constraints on the BIM model, thereby automatically generating the final construction blueprint model containing the optimized design intent.

[0138] The construction strategy module is a key bridge connecting digital design and physical construction schemes. It receives the final construction blueprint model generated by the model generation module. This module includes a data extraction unit, a spacing calculation unit, a layout simulation unit, and a scheme generation unit.

[0139] The data extraction unit is configured to accurately analyze the preset engineering attributes in the final construction blueprint model to obtain the allowable deviation value of surface flatness, while the spacing calculation unit is a calculation core that has solidified the optical and error model. It is configured to calculate the maximum theoretical layout spacing of laser control points based on the extracted deviation value and the preset laser leveling instrument accuracy.

[0140] The layout simulation unit constructs a virtual optical environment, which is configured to import all three-dimensional entities in the final construction blueprint model as potential obstructions. In this environment, the propagation path and intensity attenuation of the laser signal are simulated through ray tracing algorithms, and the optimal placement of the laser emitter is determined through iterative optimization algorithms to achieve full signal coverage and seamless overlap.

[0141] Finally, the scheme generation unit will integrate the maximum control point spacing output by the spacing calculation unit and the optimal equipment position output by the layout simulation unit, generate a three-dimensional control point coordinate set covering the entire roof through an automatic point placement algorithm, and combine it with a preset collaborative communication protocol to encapsulate a structured laser-guided construction scheme containing coordinates and instructions.

[0142] The adaptive construction control module is the execution center for realizing dynamic closed-loop correction of the construction process. Its configuration is used to deploy and issue commands to the laser leveling instrument on site according to the laser-guided construction plan. The module includes a deployment interface, a real-time data receiver, a deviation analysis engine, a compensation calculation unit, and a dynamic correction transmitter.

[0143] The deployment interface is responsible for sending the laser-guided construction plan to the central controller of the laser leveling instrument, while the real-time data receiver establishes a stable connection with the portable scanning equipment on site through a high-bandwidth wireless communication link to continuously receive local three-dimensional point cloud data generated by real-time sampling inspection.

[0144] Once the real-time point cloud is received, the deviation analysis engine is immediately activated. Its configuration is used to quickly register the coordinates of the point cloud and compare it with the final construction blueprint model, calculating the real-time elevation deviation caused by concrete settlement at a speed of milliseconds.

[0145] Subsequently, the compensation calculation unit will calculate a feedforward elevation compensation value based on the real-time deviation and an adjustable settlement prediction coefficient. Finally, the dynamic correction transmitter will send the compensation value as a correction command to the control system of the laser leveling instrument in real time, instructing it to accurately and dynamically raise the laser-guided surface of the subsequent unpaved area, thereby completing the adaptive cancellation of physical settlement.

[0146] The quality verification module is the terminal for completing the final quantitative assessment of construction quality and data closure. It is used to receive the as-built point cloud model covering the entire roof generated by the ground 3D laser scanner. The module includes an as-built data acquisition interface, a spatial registration unit, a comprehensive deviation analysis engine, and a reporting and visualization unit.

[0147] The as-built data acquisition interface is used to receive the high-precision as-built point cloud model after stitching and denoising, while the spatial registration unit uses advanced engineering control points as absolute references and achieves precise alignment between the as-built point cloud model and the final construction blueprint model through an iterative nearest-point algorithm.

[0148] Once the alignment is complete, the comprehensive deviation analysis engine will be activated. Its configuration is used to calculate the normal spatial deviation between each point in the as-built point cloud model and the surface of the final construction blueprint model, thereby generating a global, high-density deviation dataset.

[0149] Finally, the reporting and visualization unit will automatically generate a 3D deviation cloud map and statistical analysis charts based on the dataset, and determine the maximum deviation value according to the preset quality acceptance threshold. Finally, it will output a closed-loop verification result of construction quality that includes visualization results, statistical data and a conclusion on whether the work is qualified or not.

[0150] This system implementation method integrates design optimization, construction planning, adaptive control, and quality verification into a seamless digital workflow by constructing four deeply coupled and functionally specialized software modules. This achieves closed-loop management of the entire process from virtual design to physical construction and then to digital verification, which helps to ensure the accuracy, quality, and efficiency of large-span roof construction.

Claims

1. A method for construction of large span roof based on intelligent model, characterized in that, The method comprises the following steps: Step S1: In the BIM platform, the multi-objective automatic optimization of the roof composite slope parameters is performed by using an optimization algorithm, and the structure performance correlation verification of the optimized scheme is performed to generate a final construction blueprint model considering the drainage efficiency and the structure performance; Step S2: The final construction blueprint model is used to calculate the layout interval of the laser control points and to formulate a multi-device collaborative operation strategy to form a complete laser guided construction scheme; Step S3: The laser screed instrument is erected according to the laser guided construction scheme, and the laser guided surface is dynamically corrected through real-time sampling and feedback in the paving operation to complete the adaptive paving operation of the slope layer; Step S4: The three-dimensional laser scanner is used to scan the slope layer formed after the adaptive paving operation to generate a completion point cloud model, and the deviation comparison analysis between the completion point cloud model and the final construction blueprint model is performed to complete the closed-loop verification of the construction quality.

2. The smart model based long span roofing construction method according to claim 1, wherein, The step S1 specifically comprises the following steps: The genetic algorithm is used as the optimization algorithm to iteratively optimize the roof composite slope parameters to generate a set of optimized roof composite slope parameters; The structure finite element model is established based on the optimized roof composite slope parameters, and the comparative analysis is performed between the structure finite element model and the structure finite element model using the traditional single slope scheme under the same design load to complete the structure performance correlation verification; After the comparative analysis verifies the effectiveness of the optimized roof composite slope parameters, the three-dimensional model is generated in the BIM platform by using the optimized roof composite slope parameters to obtain the final construction blueprint model.

3. The smart model based long span roofing construction method according to claim 2, wherein, The comparative analysis verifies the effectiveness of the optimized roof composite slope parameters, which specifically means that: When the internal force value and the long-term deflection value of the key structure components in the structure finite element model based on the optimized roof composite slope parameters are less than the corresponding values in the structure finite element model using the traditional single slope scheme, it is determined that the effectiveness verification is passed.

4. The smart model based long span roofing construction method according to claim 1, wherein, The step S2 specifically comprises the following steps: The surface flatness allowable deviation value is extracted from the final construction blueprint model, and the maximum layout interval of the laser control points is calculated by combining the device automatic leveling accuracy of the laser screed instrument; The layout positions of at least two laser screed instrument emitters are planned to make the signal range of the laser screed instrument emitters form overlapping coverage in adjacent construction areas; The maximum layout interval and the layout positions are integrated to generate the laser guided construction scheme containing the control point coordinates and the device collaborative instructions. 5.The smart model based construction method of large span roof according to claim 4, wherein, The method for calculating the maximum layout interval of the laser control points is specifically: The surface flatness allowable deviation value is divided by the device automatic leveling accuracy, and the quotient value obtained is determined as the maximum layout interval of the laser control points. 6.The smart model based construction method of large span roof according to claim 1, wherein, The step S3 specifically comprises the following steps: The laser screed instrument is erected according to the laser guided construction scheme; In the paving operation, the handheld three-dimensional laser scanner is used to perform real-time sampling on the paved area, and the deviation data generated by the sampling is transmitted to the central controller of the laser screed instrument in real time through the wireless data link; Generate a correction instruction based on the deviation data to dynamically correct the laser guide surface to complete the adaptive paving of the slope layer.

7. The smart model based long span roofing construction method according to claim 6, wherein, The dynamic correction of the laser guide surface specifically refers to: Real-time scanning of the already-paved area by a portable scanning device to generate real-time point clouds and compare them with the final construction blueprint model to calculate the elevation deviation caused by concrete settlement; Generate an elevation compensation value based on the elevation deviation and feed it back to the laser screed control system to adjust the laser guide surface of the subsequent unpaved area. 8.The smart model based construction method of large span roof according to claim 1, wherein, The S4 step specifically includes the following steps: Under a unified coordinate system, spatially register and superimpose the as-built point cloud model and the final construction blueprint model; Calculate the normal spatial deviation value between each point on the as-built point cloud model and the surface of the final construction blueprint model to generate a set of deviation data; Generate a three-dimensional deviation cloud analysis report based on the deviation data set and complete the closed-loop verification of the construction quality according to the report. 9.The smart model based construction method of large span roof according to claim 8, wherein, The closed-loop verification of the construction quality according to the report specifically refers to: Extract the maximum deviation value from the deviation data set and compare it with the preset quality acceptance threshold. If the maximum deviation value is not greater than the quality acceptance threshold, the construction quality is determined to be acceptable.

10. A smart model based long span roof construction monitoring system for performing the smart model based long span roof construction method as claimed in any one of claims 1 to 9, characterized in that, It includes: A model generation module configured to perform multi-objective automatic optimization of roofing composite slope parameters in a BIM platform and integrate a structure verification unit to verify the optimization scheme to generate a final construction blueprint model; A construction strategy module for receiving the final construction blueprint model generated by the model generation module and calculating the layout distance of laser control points and multi-device collaborative operation strategy based on it to generate the laser guide construction scheme; An adaptive construction control module configured to deploy the laser screed on site according to the laser guide construction scheme and communicate with the portable scanning device to receive real-time sampling data, thereby generating dynamic correction instructions to control the adaptive paving operation; A quality verification module for receiving the as-built point cloud model generated by the three-dimensional laser scanner and calling the final construction blueprint model for deviation comparison analysis to output the closed-loop verification result of the construction quality.

Images