Assembly Optimization Method and System of Prefabricated Steel Structure Integrated with BIM Technology

Abstract

The present invention discloses an assembly optimization method and system for prefabricated steel structures integrating BIM technology, which relates to the technical field of data processing. The method includes: obtaining the design information of the prefabricated steel structure of the target assembly project and constructing a digital twin model of the prefabricated steel structure; collecting multi-angle images of the nodes to be assembled, extracting surface defect and connection quality features, and determining the node quality factor; performing an assemblability certification based on the quality factor. If the certification is passed, then a simulation is carried out through the digital twin model of the prefabricated steel structure, and the assembly plan is optimized in combination with real-time environmental data. The present invention solves the technical problems in the prior art that the node quality assessment in traditional steel structure assembly is inaccurate, the assembly plan lacks intelligent optimization and dynamic adjustment, which affects the assembly accuracy and construction efficiency, and achieves the technical effect of accurately assessing the node quality and optimizing the assembly plan in real time by integrating BIM technology, improving the assembly accuracy and construction efficiency of the steel structure, and reducing the construction risk.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to an assembly optimization method and system for prefabricated steel structures integrating BIM technology. Background Art

[0002] With the rapid development of the construction industry, prefabricated steel structures have become an important application in the modern construction field due to their high efficiency, sustainability, and flexibility. However, in the traditional steel structure assembly process, problems such as low assembly accuracy, low operation efficiency, and worker operation errors often occur, affecting the overall construction quality and progress.

[0003] The traditional assembly process lacks intelligent quality monitoring and real-time adjustment mechanisms, and often cannot flexibly respond to dynamic changes in on-site construction, such as environmental factors and node quality fluctuations, affecting the overall efficiency and quality of prefabricated steel structure projects. Summary of the Invention

[0004] This application provides an assembly optimization method and system for prefabricated steel structures integrating BIM technology, which is used to solve the technical problems in the prior art that the node quality assessment in traditional steel structure assembly is inaccurate, the assembly plan lacks intelligent optimization and dynamic adjustment, and affects the assembly accuracy and construction efficiency.

[0005] In the first aspect of this application, an assembly optimization method for prefabricated steel structures integrating BIM technology is provided. The method includes: obtaining the design information of the prefabricated steel structure of the target assembly project, inputting the design information of the prefabricated steel structure into the BIM model for analysis and simulation, and constructing a digital twin model of the prefabricated steel structure; calling a high-definition industrial camera to collect multi-angle images of Q steel structure nodes to be assembled, obtaining Q node image sets, where Q is a positive integer greater than or equal to 1; extracting and analyzing features of the Q node image sets from two dimensions of surface defects and connection quality, in combination with the digital twin model of the prefabricated steel structure, to determine Q node quality factors; based on the magnitudes of the Q node quality factors, performing an assemblability certification on the Q steel structure nodes to be assembled. If the assemblability certification passes, performing an assembly simulation based on the digital twin model of the prefabricated steel structure, and optimizing the assembly plan in combination with the actual assembly environment information to obtain a target assembly plan; performing the assembly of the prefabricated steel structure of the target assembly project based on the target assembly plan.

[0006] In the second aspect of the present application, an assembly optimization system for prefabricated steel structures integrating BIM technology is provided. The system includes: a digital twin model construction module, which is used to obtain the design information of the prefabricated steel structure of the target assembly project, input the design information of the prefabricated steel structure into the BIM model for analysis and simulation, and construct a digital twin model of the prefabricated steel structure; a multi-angle image acquisition module, which is used to call high-definition industrial cameras to collect multi-angle images of Q steel structure nodes to be assembled, and obtain Q sets of node images, where Q is a positive integer greater than or equal to 1; a feature extraction and analysis module, which is used to extract and analyze the features of the Q sets of node images from two dimensions of surface defects and connection quality, in combination with the digital twin model of the prefabricated steel structure, to determine Q node quality factors; an assembly plan optimization module, which is used to perform assemblability certification on the Q steel structure nodes to be assembled based on the magnitudes of the Q node quality factors. If the assemblability certification passes, perform assembly simulation based on the digital twin model of the prefabricated steel structure, and optimize the assembly plan in combination with the actual assembly environment information to obtain a target assembly plan; an assembly plan execution module, which is used to execute the assembly of the prefabricated steel structure of the target assembly project based on the target assembly plan.

[0007] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0008] The assembly optimization method and system for prefabricated steel structures integrating BIM technology provided in the present application relate to the technical field of data processing. By constructing a digital twin model of the prefabricated steel structure through BIM technology, using high-definition industrial cameras to collect node images, extracting quality features and performing quality assessment, and performing assemblability certification based on the assessment results. If it passes, perform assembly simulation and optimize the assembly plan in combination with real-time environmental data. It solves the technical problems in the prior art that the node quality assessment in traditional steel structure assembly is inaccurate, the assembly plan lacks intelligent optimization and dynamic adjustment, which affects the assembly accuracy and construction efficiency, and realizes the technical effect of accurately assessing the node quality and optimizing the assembly plan in real time by integrating BIM technology, improving the assembly accuracy and construction efficiency of steel structures, and reducing construction risks. Description of the Drawings

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0010] Figure 1Schematic flow chart of the assembly optimization method for prefabricated steel structures integrating BIM technology provided by the embodiments of the present application;

[0011] Figure 2 Schematic structural diagram of the assembly optimization system for prefabricated steel structures integrating BIM technology provided by the embodiments of the present application.

[0012] Explanation of reference numerals: Digital twin model construction module 11, multi-angle image acquisition module 12, feature extraction and analysis module 13, assembly plan optimization module 14, assembly plan execution module 15. Detailed implementation manners

[0013] The present application provides an assembly optimization method and system for prefabricated steel structures integrating BIM technology, which are used to solve the technical problems in the prior art that the node quality assessment in traditional steel structure assembly is inaccurate, the assembly plan lacks intelligent optimization and dynamic adjustment, and the assembly accuracy and construction efficiency are affected.

[0014] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0015] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned accompanying drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described here can be implemented in an order different from those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0016] Embodiment 1, as Figure 1 shown, the present application provides an assembly optimization method for prefabricated steel structures integrating BIM technology, and the method includes:

[0017] P10: Obtain the design information of the prefabricated steel structure of the target assembly project, input the design information of the prefabricated steel structure into the BIM model for analysis and simulation, and construct a digital twin model of the prefabricated steel structure.

[0018] Optionally, it is first necessary to obtain the prefabricated steel structure design information of the target assembly project. The prefabricated steel structure design information includes detailed parameters such as various components, dimensions, connection methods, and material specifications of the steel structure. Among them, the design information is usually provided by structural engineers and generated based on architectural design drawings or design software. By integrating Building Information Modeling (BIM) technology, these design information can be digitized and input into the BIM system for processing. BIM technology is a 3D modeling-based technology that can integrate all relevant data throughout the building life cycle, enabling information visualization and collaborative work.

[0019] In the BIM model, the design information is not only static geometric data but also includes multi-dimensional information such as physical properties, construction techniques, time schedules, and cost budgets. A dynamic and three-dimensional digital twin model of the prefabricated steel structure is constructed through these data. Digital twin refers to a technology that precisely maps a physical object or system in the real world through a virtual model. In this step, the digital twin model of the prefabricated steel structure can reflect various design elements of the prefabricated steel structure and possible dynamic changes during the construction process in real time, such as the mechanical response of the steel structure, structural force changes, and material consumption.

[0020] Input the prefabricated steel structure design information into the BIM model for analysis and simulation. The system will verify and optimize the structure through technologies such as Computer-Aided Design (CAD) and Finite Element Analysis (FEA), and construct a digital twin model of the prefabricated steel structure. For example, finite element analysis can be used to analyze the stress distribution, deformation conditions, and possible structural failure risks of each node, so as to identify potential structural problems before actual assembly. Through these simulations, the stability, strength, and durability of the steel structure can be predicted to ensure the feasibility of the design scheme.

[0021] In addition, the digital twin model of the prefabricated steel structure will also serve as a data platform for the entire assembly project, running through all stages of the assembly process. Through the digital twin model of the prefabricated steel structure, project managers can obtain information such as assembly progress, resource consumption, and personnel scheduling in real time, so as to carry out efficient management and scheduling.

[0022] P20: Call a high-definition industrial camera to collect multi-angle images of Q steel structure nodes to be assembled, and obtain a set of Q node images, where Q is a positive integer greater than or equal to 1.

[0023] Specifically, a high-definition industrial camera is called to collect multi-angle images of the steel structure nodes to be assembled, obtaining high-resolution image data of each node, providing the original image basis for subsequent analysis and quality assessment. To ensure the accuracy and comprehensiveness of the collected data, each steel structure node to be assembled should be photographed from multiple different angles. This not only avoids image distortion caused by angle problems but also comprehensively presents the details of the node surface and connection parts, ensuring the accuracy of subsequent analysis.

[0024] In this process, the high-definition industrial camera used usually has the characteristics of high resolution and high frame rate, and can clearly capture the minute defects on the node surface, such as surface quality problems like cracks, deformations, corrosion, etc. The selection of the high-definition camera is crucial, especially in an industrial environment, which requires it to maintain stable image acquisition quality under complex on-site conditions. In addition, a high-definition industrial camera is called to collect multi-angle images of Q steel structure nodes to be assembled, forming a database containing Q node image sets, where Q represents the number of nodes to be assembled, and Q is a positive integer greater than or equal to 1. Each image in the image set corresponds to the capture of the node from different angles and different fields of view. To avoid any possible occlusion or perspective dead angle, it should be ensured that there are sufficient image records of the key parts of each node during the acquisition process. These images will become the basic data source for subsequent feature extraction and quality analysis.

[0025] After the image acquisition is completed, these image sets will be transmitted to the backend system for further processing and analysis. These images not only provide an intuitive feedback on the current state of the nodes but also provide accurate data support for quality assessment and assembly certification in the subsequent steps.

[0026] P30: From the two dimensions of surface defects and connection quality, combined with the digital twin model of the prefabricated steel structure, perform feature extraction and analysis on the Q node image sets to determine Q node quality factors.

[0027] Furthermore, step P30 of the embodiment of the present application further includes:

[0028] P31: Use the surface defect analysis network layer to identify each of the Q sets of node images, obtaining Q surface quality coefficients; P32: Based on the digital twin model of the prefabricated steel structure, identify the key connection points of the Q steel structure nodes to be assembled, obtaining Q sets of key connection points of the nodes, and use the feature extraction algorithm to traverse the Q sets of key connection points of the nodes to extract the processing features of the connection points, obtaining Q clusters of connection point processing features; P33: Traverse the Q clusters of connection point processing features to identify the connection quality, obtaining Q node connection coefficients; P34: Based on the Q surface quality coefficients and the Q node connection coefficients, conduct comprehensive quality identification on the Q steel structure nodes to be assembled, determining Q node quality factors; P35: Determine whether the Q node quality factors are less than the preset node quality factor. If so, the assembly certification fails, generate a warning message, and send the warning message to the staff for renovation of the steel structure nodes; if not, the assembly certification passes.

[0029] It should be understood that starting from the two core dimensions of surface defects and connection quality, combining the digital twin model and image recognition technology, using modern technical means to extract features and calculate the quality factors of the nodes, comprehensively evaluate the quality of each steel structure node to be assembled.

[0030] First, use a surface defect analysis network (such as a deep convolutional neural network) to process the images of each steel structure node to be assembled, identifying possible defects on the node surface. Surface defects include cracks, rust, scratches, deformation, etc., and these problems may affect the overall performance of the steel structure. Through the automated learning and image analysis of the network layer, the system can efficiently extract defect information from a large amount of image data, and then calculate the surface quality coefficient of each node. The surface quality coefficient is a quantitative indicator that reflects the severity of the surface defects of the node. The higher the value, the better the surface quality of the node and the fewer the defects.

[0031] Next, the connection quality of the steel structure nodes is directly related to the stability and safety of the entire structure. Therefore, use the digital twin model of the prefabricated steel structure to identify the key connection points of each steel structure node to be assembled. Locate the key connection points of each node based on the design data to determine the connection parts that need to be focused on for inspection, obtaining Q sets of key connection points of the nodes. Further, use the feature extraction algorithm to traverse the Q sets of key connection points of the nodes to extract the processing features of the connection points, including important parameters such as the shape, size, and processing accuracy of the connection. These features form Q clusters of connection point processing features, which are used to describe the processing quality and process status of the connection points.

[0032] Furthermore, the Q extracted connection point machining features are used to process feature clusters for further connection quality identification. The core of this process is to evaluate the machining quality of each connection point to identify whether it meets the design requirements. The identification process may include detecting aspects such as dimensions, angles, and welding quality of the connection part, and generating a connection coefficient for each node, which is a quantitative index reflecting the connection quality of the node. The higher the connection coefficient, the better the connection quality and the more reliable the connection performance of the node.

[0033] Through the above two analyses, the surface quality coefficient and the node connection coefficient of each node are obtained, that is, Q surface quality coefficients and Q node connection coefficients. These two types of coefficients will be used as the basis for evaluating the overall quality of the node, and a comprehensive quality identification of each node will be carried out. For example, the surface quality coefficient and the connection coefficient are fused through a weighting algorithm to obtain Q comprehensive node quality factors. The node quality factor is a key index for evaluating whether a node is suitable for assembly. The larger the value, the higher the quality of the node and the more suitable it is for assembly.

[0034] Finally, the system determines whether the node quality factor of each node meets the requirements according to the set preset standard. If the quality factor of a certain node is less than the preset threshold, it indicates that there are relatively serious quality problems with this node and it cannot meet the assembly requirements. At this time, the system will fail the assembly certification and generate a warning message, and send this information to the staff in a timely manner to remind them to repair or replace this node. On the contrary, if the quality factor of the node is greater than or equal to the preset standard, the system determines that the node passes the assembly certification and is ready to enter the next stage of actual assembly.

[0035] Through the above analysis and judgment, a comprehensive evaluation of each node from surface quality to connection quality is realized, ensuring that the steel structure nodes in the assembly process can meet the design requirements, reducing the risk of quality problems in the assembly process, and improving the overall assembly accuracy and efficiency.

[0036] Furthermore, step P32 of the embodiment of the present application further includes:

[0037] P32-1: Traverse the Q steel structure nodes to be assembled for neighborhood matching in the digital twin model of the prefabricated steel structure to obtain the Q neighborhoods of the steel structure nodes to be assembled and the Q sets of assembly parameters of the steel structure nodes to be assembled; P32-2: Extract the Q design information of the Q steel structure nodes to be assembled, combine the Q sets of assembly parameters of the steel structure nodes to be assembled for frequent item mining of faults, and determine the risk coefficients of the Q steel structure nodes to be assembled according to the mining analysis results; P32-3: Analyze the number of key connection points based on the magnitudes of the risk coefficients of the Q steel structure nodes to be assembled to determine the number of key connection points of the Q; P32-4: Using the number of key connection points of the Q as the first constraint, combine the Q neighborhoods of the steel structure nodes to be assembled and the Q steel structure nodes to be assembled for identification of key connection points of the nodes to obtain the Q sets of key connection points of the nodes.

[0038] Specifically, the connection quality analysis of the steel structure nodes to be assembled can be further deepened to improve the accuracy of node quality assessment and ensure the smooth progress of the assembly process.

[0039] First, perform neighborhood matching on the Q steel structure nodes to be assembled in the digital twin model of the prefabricated steel structure. The goal of neighborhood matching is to identify the relationships between adjacent nodes, especially the assembly parameters such as their connection methods and assembly directions. Through spatial positioning and data association in the digital twin model, the system can accurately obtain the assembly information of adjacent nodes for each node during the assembly process. These assembly parameters include connection methods (such as bolt connection, welding, riveting, etc.), connection directions (such as horizontal, vertical, inclined, etc.), and possible connecting components (such as connection plates, number of bolts, etc.). These information is crucial for subsequent assembly analysis because the assembly methods of adjacent nodes may affect the overall structural stability and the interaction between nodes.

[0040] Next, extract the design information of each of the Q steel structure nodes to be assembled, and combine it with the above-obtained set of assembly parameters for frequent item mining of faults. Analyze the fault frequencies that occur during the design, processing, and assembly of different nodes through data mining techniques, such as connection failures and assembly errors. By analyzing historical data and fault cases of similar projects, mine the high-frequency items where each node may have faults. According to the mining analysis results, calculate a risk coefficient for each node to be assembled, and determine the risk coefficients of the Q steel structure nodes to be assembled. This coefficient reflects the probability of problems occurring during the assembly process. Nodes with high risk coefficients usually require more stringent quality control and monitoring.

[0041] Furthermore, based on the risk coefficient of each node, the analysis of the number of key connection points continues. The number of connection points directly affects the complexity during the assembly process and the stability of the nodes. Especially in nodes with complex connection methods, excessive connection points may lead to the accumulation of assembly errors. The system determines the number of key connection points for each node by analyzing the design and connection characteristics of the nodes, and based on the magnitude of the risk coefficient of the nodes, identifies the connection points that require special attention. Nodes with a high risk coefficient may require more attention to avoid potential problems during assembly.

[0042] Finally, combining the aforementioned number of key connection points and the node neighborhood information, key connection point identification is performed on the nodes. This process is constrained by the aforementioned number of key connection points and, in combination with the neighborhood matching results, accurately determines the set of key connection points for each node. During this identification process, the system examines the assembly relationship between the node and its adjacent nodes to ensure that the identified key connection points are the most critical parts of the entire assembly process. Through this method, quality problems caused by omissions or errors in connection points during node assembly can be effectively avoided.

[0043] Through the above steps, not only can the key connection points of each node be accurately identified, but also potential assembly problems can be predicted and addressed in advance during the design phase, significantly improving the efficiency and quality control level of the assembly process.

[0044] Furthermore, step P32-4 of the embodiment of the present application further includes:

[0045] P32-41: Using the neighborhoods of the Q steel structure nodes to be assembled and the indexes of the Q steel structure nodes to be assembled, in combination with the digital twin model of the prefabricated steel structure, perform neighborhood connection point identification to obtain Q sets of neighborhood connection points; P32-42: Traverse the Q steel structure nodes to be assembled to extract connection points, obtaining Q sets of node connection points; P32-43: Divide the number of connection points in each of the Q sets of neighborhood connection points by the number of connection points in each of the Q sets of node connection points respectively to obtain Q neighborhood connection point ratios, and use the Q neighborhood connection point ratios as the second constraint for key connection point identification; P32-44: Based on the first constraint and the second constraint, perform key connection point identification on the Q sets of node connection points to obtain the Q sets of node key connection points.

[0046] Optionally, further deepen the analysis of node connection points to ensure the accuracy of node connection and the quality of assembly. First, based on the digital twin model, by combining the neighborhood information of Q steel structure nodes to be assembled and the indexes of Q steel structure nodes to be assembled, identify the neighborhood connection points. The core of this step is to identify the connection points between each node to be assembled and its adjacent nodes. These connection points are the key connection parts between nodes and may include important connection positions such as bolt holes and welding points. Through the spatial data of the digital twin model, the system can accurately locate and identify these connection points and generate a set of Q neighborhood connection points. The identification of neighborhood connection points not only considers the spatial proximity relationship but also needs to combine the assembly drawing and design data to ensure that all possible connection points can be accurately identified during the actual assembly process.

[0047] Next, traverse the Q steel structure nodes to be assembled and extract the connection points for each node. This process automatically extracts all relevant connection points through in-depth analysis of the node design information, forming a set of Q node connection points. The connection points of each node may involve different types of connection methods (such as bolt connection, welding, riveting, etc.). The quality and position of these connection points are crucial for the stability of the overall assembly of the node. Therefore, the extraction of connection points not only focuses on the number of points but also needs to ensure the precise positioning of each connection point.

[0048] After completing the extraction of node connection points, compare the number of connection points in the set of Q neighborhood connection points and the set of Q node connection points, and calculate the neighborhood connection point ratio. Specifically, divide the number of connection points within the neighborhood of each node by the number of connection points in the set of connection points of that node to obtain the neighborhood connection point ratio. This ratio reflects the connection density of each node in the neighborhood environment. A higher ratio may indicate that the connection between the node and its adjacent nodes is relatively tight, which may affect the mutual relationship between nodes during the assembly process.

[0049] Finally, based on the aforementioned two constraints, namely the first constraint (the number of key connection points) and the second constraint (the neighborhood connection point ratio), identify the key connection points for the set of Q node connection points. Among them, the first constraint comes from the analysis of the number of key connection points in the previous step, which can ensure that the currently concerned node connection points have sufficient criticality, while the second constraint ensures that the relationship between the node and the neighborhood is reasonably considered through the neighborhood connection point ratio. Through the comprehensive application of these two constraints, the system can accurately identify the set of key connection points for each node, and these key connection points determine the stability of node connection and the success rate of assembly.

[0050] Furthermore, step P32-44 of the embodiment of the present application further includes:

[0051] P32 - 441: Randomly extract the Q sets of node connection points based on the first constraint and the second constraint to obtain Q initial sets of key node connection points; P32 - 442: Perform three - dimensional simulation according to the positions of the connection points in the Q initial sets of key node connection points to obtain Q initial key node connection point models; P32 - 443: Based on the digital twin model of the prefabricated steel structure, retrieve the three - dimensional models of the Q steel structures to be assembled to obtain Q steel structure models to be assembled; P32 - 444: Perform similarity matching on the Q initial key node connection point models and the Q steel structure models to be assembled to obtain Q matching fitness values; P32 - 445: When the Q matching fitness values are greater than or equal to the preset matching fitness threshold, use the Q initial sets of key node connection points as the Q sets of key node connection points.

[0052] In a possible embodiment of the present application, by combining model simulation, three - dimensional matching, and fitness evaluation, it is ensured that the key connection points of each node are correctly identified to support efficient assembly and quality control.

[0053] First, based on the previously determined constraint conditions, namely the first constraint (the number of key connection points) and the second constraint (the ratio of neighboring connection points), randomly extract the Q sets of node connection points, and obtain Q initial sets of key node connection points from them. The purpose of random extraction is to provide preliminary key connection points for further analysis. These sets of connection points represent the most critical positions in the assembly process. In this way, the preliminarily screened connection points will be used as the basis for subsequent simulation and verification.

[0054] Next, according to the positions of the connection points in the Q initial sets of key node connection points, perform three - dimensional simulation. Three - dimensional modeling technology can be used to map the positions of these connection points into the digital twin model to generate Q initial key node connection point models. Through three - dimensional simulation, it is possible to more intuitively display the positions and distributions of each connection point in three - dimensional space, which is conducive to subsequent connection point matching and fitness evaluation, and helps to check whether the connections between nodes meet the design requirements and whether there may be spatial conflicts or assembly problems.

[0055] After obtaining the initial key node connection point models, the system retrieves the three - dimensional models of the Q steel structures to be assembled based on the digital twin model of the prefabricated steel structure. These three - dimensional models are generated based on design and actual processing data and contain all the geometric information and physical properties of the nodes. Through these three - dimensional models, simulate the overall assembly process of the steel structure and identify the assembly requirements and environmental factors of each node.

[0056] After having Q initial node key connection point models and Q steel structure models to be assembled, the two sets of models are matched through a similarity matching algorithm. The goal of the matching is to evaluate the fitness between the initial key connection point model and the steel structure model to be assembled. The fitness evaluation can consider the geometry, dimensions, and accuracy of the connection parts of the models. Each model matching pair will be assigned a matching fitness, and this fitness value reflects the matching degree between the initial connection point model and the actual model to be assembled. The higher the matching fitness, the more consistent the position and state of the connection point are with the actual assembly requirements.

[0057] Finally, according to the preset matching fitness threshold, it is judged whether the Q matching fitnesses meet the requirements. When all the matching fitnesses are greater than or equal to the set threshold, it means that these initial node key connection points meet the assembly requirements and can be used as the final key connection points. If the matching fitness does not reach the threshold, it may be necessary to readjust the connection points or further optimize the models until an ideal matching effect is achieved.

[0058] When it is confirmed that the matching fitness meets the standard, the set of Q initial node key connection points is used as the final set of key connection points, which are the parts that need to be focused on during the assembly process, so as to achieve the comprehensive identification and optimization of the steel structure node connection points and ensure the assembly quality.

[0059] Furthermore, step P34 of the embodiment of the present application further includes:

[0060] Pre-build a quality recognition network layer, and use the quality recognition network layer to respectively map and recognize the Q surface quality coefficients and the Q node connection coefficients to obtain the Q node quality factors.

[0061] Optionally, a pre-built quality recognition network layer can be introduced, and a deep learning model is used to map and recognize the surface quality and connection quality of the nodes to be assembled, so as to calculate the quality factor of each node.

[0062] Before implementing this step, it is first necessary to pre-build a quality recognition network layer. The quality recognition network layer is a deep neural network structure dedicated to processing and analyzing quality data from different sensors or image acquisition systems. This network layer will be supervised and trained according to the surface quality coefficients and node connection coefficients of the nodes to learn how to extract the key features affecting the node quality from the original data. Among them, the surface quality coefficients are derived from historical image analysis data (such as the detection results of defects such as cracks and corrosion), and the node connection coefficients reflect the processing quality and assembly conditions of the node connection parts.

[0063] After the training of the quality recognition network layer is completed, the next step is to map and identify the Q surface quality coefficients and Q node connection coefficients. Specifically, the surface quality coefficient and the node connection coefficient are used as input, and feature extraction and pattern recognition are performed through different processing layers of the quality recognition network layer. This process is similar to the convolution operation in image recognition. The quality recognition network layer will automatically identify the key factors affecting the node quality based on the input quality data. For example, for surface quality, it is recognized that there are many cracks or deformation patterns on the surface of a node, and for node connection, it can identify problems that may be caused by the accuracy of the connection part or the connection method.

[0064] After mapping and identification, the node quality factor of each node to be assembled is generated based on the output results. For example, the corresponding node quality factor can be generated by quantitative calculation based on the severity and importance of each quality defect. The node quality factor is a comprehensive evaluation indicator that reflects the overall performance of the node in terms of surface defects and connection quality. The higher the value of the node quality factor, the better the quality of the node and the lower the risk of problems during the assembly process; conversely, nodes with lower quality factors may have larger defects or assembly problems and require further inspection or repair. This process can use regression models or classification models in deep learning, in which the surface quality coefficient and node connection coefficient of each input are regarded as feature vectors, which are trained and optimized through the network to automatically process large amounts of data and accurately evaluate the quality of the nodes, thereby improving the efficiency and accuracy of node quality analysis.

[0065] P40: Based on the sizes of the Q node quality factors, the Q steel structure nodes to be assembled are certified for assembly. If the assembly certification is passed, assembly simulation is performed based on the digital twin model of the prefabricated steel structure, and the assembly plan is optimized based on the actual assembly environment information to obtain the target assembly plan.

[0066] Furthermore, step P40 in this embodiment of the present application further includes:

[0067] P41: Environmental data is collected based on IoT devices and sensors deployed at the assembly site to obtain real-time assembly environment information; P42: Assembly simulation is performed based on the preset assembly plan, real-time assembly environment information and the digital twin model of the prefabricated steel structure, and the assembly status of Q steel structure nodes to be assembled is analyzed using finite element analysis to obtain analysis results; P43: The preset assembly plan is optimized based on the analysis results to obtain the target assembly plan.

[0068] It should be understood that based on the evaluation results of the quality factors of Q nodes in the early stage, it is decided whether to pass the assembly certification, and the digital twin model of the prefabricated steel structure is used for assembly simulation to optimize the assembly plan and ensure the accuracy and efficiency of node assembly.

[0069] First, by analyzing the magnitudes of the quality factors of Q nodes, assembly certification is carried out for each node to be assembled. The quality factor usually synthesizes multiple dimensions such as surface quality and connection quality. Therefore, the higher the quality factor of a node, the better its assembly conditions, and vice versa, there may be quality problems. If the quality factor of a node meets the preset standard, the certification passes, and the system will enter the subsequent assembly simulation and optimization process. If the node fails the certification, the system will issue a warning, prompting the staff to repair the node to ensure qualified quality.

[0070] When the node passes the assembly certification, next the system will collect real-time environmental data through Internet of Things devices and sensors deployed at the assembly site. These devices include sensors such as temperature, humidity, vibration, and air pressure, which can provide accurate on-site environmental information to evaluate the external conditions that may affect node assembly during the assembly process. For example, temperature changes may affect the thermal expansion and contraction of the steel structure, and humidity may affect the bonding quality or corrosion condition of the connection part. These environmental data provide an actual environmental basis for the subsequent simulation.

[0071] Next, using the digital twin model of the prefabricated steel structure and the real-time collected environmental data, assembly simulation is carried out. In this step, the finite element analysis technology plays a key role. Finite element analysis is to decompose the assembly structure into multiple small units and calculate the mechanical responses of each unit, and then predict the performance of the entire structure. In this step, finite element analysis is mainly used to analyze the assembly states of Q steel structure nodes to be assembled, and evaluate factors such as stress, deformation, and force transmission of the nodes during the actual assembly process. These analysis results can reveal possible assembly problems, such as whether the node will deform due to excessive assembly force, or whether the connection part can withstand the expected load.

[0072] According to the results obtained from the finite element analysis, the preset assembly plan is optimized. The optimization process includes adjusting the assembly sequence, improving the connection method, optimizing the selection of assembly tools and equipment, and adjusting process parameters. The optimized plan is the target assembly plan, which can maximize the assembly efficiency, ensure the accuracy of node assembly, and minimize the risks during the assembly process, and will provide detailed guidance for the construction site, making the assembly process more efficient, safe, and economical.

[0073] Through the above steps, the entire assembly process can not only ensure that the quality of each node meets the standards, but also make dynamic adjustments and optimizations according to the actual environment, thus ensuring the accuracy and efficiency of the assembly process.

[0074] P50: Perform the assembly of prefabricated steel structures for the target assembly project based on the target assembly plan.

[0075] Specifically, based on the target assembly plan optimized previously, start to execute the assembly task, and actually carry out the installation and assembly of prefabricated steel structures, including the determination of the assembly sequence, the scheduling of assembly equipment, the application of process parameters, etc., to ensure the precise assembly and efficient completion of the steel structure.

[0076] Through the target assembly plan, first determine the assembly sequence of nodes and components to reduce working hours and improve assembly efficiency in the optimal way. The assembly sequence is the result of comprehensive optimization of multi-dimensional factors such as the quality factor of each node, environmental data, and assembly difficulty. At the same time, select appropriate assembly equipment, such as cranes, hoists, welding equipment, bolt wrenches, etc. The selection of these equipment will be optimized according to the actual situation such as the type of nodes to be assembled, weight, and assembly location. And extract process parameters, such as connection methods, connection strengths, bolt torques, welding temperatures, etc. These parameters must meet the requirements of steel structure design and be strictly controlled during on-site construction.

[0077] During the actual execution of the assembly process, with the help of the digital twin model and real-time monitoring system of prefabricated steel structures, all assembly tasks will be accurately tracked in an environment that combines virtual and real. By using Internet of Things devices and on-site sensors, collect data from the assembly site in real time, and these data will be continuously fed back to the system to ensure that all work during the assembly process proceeds smoothly as planned.

[0078] During the actual assembly process, precise measuring tools and control systems can be used to ensure that each node and component meet the design accuracy requirements during installation. Through the automated control system, adjust equipment parameters and assembly processes in real time to prevent structural problems caused by assembly deviations or errors. For example, use laser measuring equipment or sensors to detect the accuracy of nodes and ensure that their positions and directions are accurate. During the connection process of steel structure nodes, through real-time data collection and feedback, ensure that the quality of each connection point meets the predetermined standards.

[0079] The entire assembly process requires strict safety management and risk control measures. Track the safety status of the site in real time and give early warnings for abnormal situations. Through dynamic monitoring equipment, monitor environmental changes (such as temperature, humidity, vibration, etc.) at the assembly site, and adjust operation strategies in time to avoid safety hazards caused by environmental changes.

[0080] In summary, the embodiments of the present application have at least the following technical effects:

[0081] The present application obtains the design information of the prefabricated steel structure of the target assembly project, constructs a digital twin model of the prefabricated steel structure, collects multi-angle images of the nodes to be assembled, extracts surface defect and connection quality features, determines the node quality factors, conducts an assemblability certification based on the quality factors. If the certification is passed, a simulation is carried out through the digital twin model of the prefabricated steel structure, and the assembly plan is optimized in combination with real-time environmental data.

[0082] It achieves the technical effects of accurately evaluating the node quality and optimizing the assembly plan in real time by integrating BIM technology, improving the accuracy and construction efficiency of steel structure assembly, and reducing construction risks.

[0083] Embodiment 2, based on the same inventive concept as the prefabricated steel structure assembly optimization method integrating BIM technology in the foregoing embodiment, as Figure 2 shown, the present application provides a prefabricated steel structure assembly optimization system integrating BIM technology. The system in the embodiments of the present application and the method embodiments are based on the same inventive concept. Among them, the system includes:

[0084] A digital twin model construction module 11, which is used to obtain the design information of the prefabricated steel structure of the target assembly project, input the design information of the prefabricated steel structure into the BIM model for analysis and simulation, and construct a digital twin model of the prefabricated steel structure.

[0085] A multi-angle image acquisition module 12, which is used to call a high-definition industrial camera to collect multi-angle images of Q steel structure nodes to be assembled, and obtain Q sets of node images, where Q is a positive integer greater than or equal to 1.

[0086] A feature extraction and analysis module 13, which is used to perform feature extraction and analysis on the Q sets of node images from two dimensions of surface defects and connection quality, in combination with the digital twin model of the prefabricated steel structure, to determine Q node quality factors.

[0087] An assembly plan optimization module 14, which is used to conduct an assemblability certification on the Q steel structure nodes to be assembled based on the magnitudes of the Q node quality factors. If the assemblability certification is passed, an assembly simulation is carried out based on the digital twin model of the prefabricated steel structure, and the assembly plan is optimized in combination with the actual assembly environment information to obtain a target assembly plan.

[0088] An assembly plan execution module 15, which is used to perform the assembly of the prefabricated steel structure of the target assembly project based on the target assembly plan.

[0089] Further, the feature extraction and analysis module 13 is further configured to perform the following steps:

[0090] Use the surface defect analysis network layer to identify the Q sets of node images respectively, and obtain Q surface quality coefficients; based on the digital twin model of the prefabricated steel structure, identify the key connection points of the Q steel structure nodes to be assembled, and obtain Q sets of key connection points of the nodes. Then use the feature extraction algorithm to traverse the Q sets of key connection points of the nodes to extract the processing features of the connection points, and obtain Q clusters of connection point processing features; traverse the Q clusters of connection point processing features to identify the connection quality, and obtain Q node connection coefficients; based on the Q surface quality coefficients and the Q node connection coefficients, perform comprehensive quality identification on the Q steel structure nodes to be assembled, and determine Q node quality factors; judge whether the Q node quality factors are less than the preset node quality factor. If so, the assembly certification fails, generate a warning message, and send the warning message to the staff for renovation of the steel structure nodes; if not, the assembly certification passes.

[0091] Further, the feature extraction and analysis module 13 is further configured to perform the following steps:

[0092] Traverse the Q steel structure nodes to be assembled in the digital twin model of the prefabricated steel structure for neighborhood matching, and obtain Q neighborhoods of the steel structure nodes to be assembled and Q sets of assembly parameters of the steel structure nodes to be assembled; extract the Q design information of the Q steel structure nodes to be assembled, combine the Q sets of assembly parameters of the steel structure nodes to be assembled to mine frequent fault items, and determine the risk coefficients of the Q steel structure nodes to be assembled according to the mining analysis results; perform analysis on the number of key connection points based on the magnitudes of the risk coefficients of the Q steel structure nodes to be assembled, and determine the number of key connection points of Q; use the number of key connection points of Q as the first constraint, combine the Q neighborhoods of the steel structure nodes to be assembled and the Q steel structure nodes to identify the key connection points of the nodes, and obtain the Q sets of key connection points of the nodes.

[0093] Further, the feature extraction and analysis module 13 is further configured to perform the following steps:

[0094] Using the neighborhoods of the Q steel structure nodes to be assembled and the indexes of the Q steel structure nodes to be assembled, and combining with the digital twin model of the prefabricated steel structure to identify neighborhood connection points, obtaining Q sets of neighborhood connection points; traversing the Q steel structure nodes to be assembled to extract connection points, obtaining Q sets of node connection points; respectively dividing the number of connection points in the Q sets of neighborhood connection points by the number of connection points in the Q sets of node connection points, obtaining Q ratios of neighborhood connection points, and taking the Q ratios of neighborhood connection points as the second constraint for critical connection point identification; based on the first constraint and the second constraint, identifying critical connection points for the Q sets of node connection points, obtaining the Q sets of node critical connection points.

[0095] Further, the feature extraction and analysis module 13 is further configured to perform the following steps:

[0096] Randomly extracting from the Q sets of node connection points based on the first constraint and the second constraint, obtaining Q initial sets of node critical connection points; performing three-dimensional simulation according to the positions of the connection points in the Q initial sets of node critical connection points, obtaining Q initial node critical connection point models; based on the digital twin model of the prefabricated steel structure, retrieving the three-dimensional models of the Q steel structures to be assembled, obtaining Q steel structure models to be assembled; performing similarity matching on the Q initial node critical connection point models and the Q steel structure models to be assembled, obtaining Q matching fitness degrees; when the Q matching fitness degrees are greater than or equal to a preset matching fitness threshold, taking the Q initial sets of node critical connection points as the Q sets of node critical connection points.

[0097] Further, the feature extraction and analysis module 13 is further configured to perform the following steps:

[0098] Pre-constructing a quality recognition network layer, and using the quality recognition network layer to respectively perform mapping recognition on the Q surface quality coefficients and the Q node connection coefficients, obtaining the Q node quality factors.

[0099] Further, the assembly scheme optimization module 14 is further configured to perform the following steps:

[0100] Collecting environmental data based on Internet of Things devices and sensors deployed at the assembly site, obtaining real-time assembly environment information; performing assembly simulation according to a preset assembly scheme, real-time assembly environment information and the digital twin model of the prefabricated steel structure, and using finite element analysis to analyze the assembly states of the Q steel structure nodes to be assembled, obtaining an analysis result; optimizing the preset assembly scheme according to the analysis result, obtaining the target assembly scheme.

[0101] It should be noted that the above order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the specific embodiments of this specification have been described. Moreover, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0102] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0103] This specification and the drawings are only exemplary descriptions of the present application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. An assembly optimization method for prefabricated steel structures integrating BIM technology, characterized in that, The method includes: Obtain the design information of the prefabricated steel structure of the target assembly project, input the design information of the prefabricated steel structure into the BIM model for analysis and simulation, and construct a digital twin model of the prefabricated steel structure; Call a high-definition industrial camera to collect multi-angle images of Q steel structure nodes to be assembled, and obtain Q sets of node images, where Q is a positive integer greater than or equal to 1; From the two dimensions of surface defects and connection quality, combined with the digital twin model of the prefabricated steel structure, perform feature extraction and analysis on the Q sets of node images to determine Q node quality factors; Based on the magnitudes of the Q node quality factors, perform an assemblability certification on the Q steel structure nodes to be assembled. If the assemblability certification passes, perform an assembly simulation based on the digital twin model of the prefabricated steel structure, and optimize the assembly plan in combination with the actual assembly environment information to obtain a target assembly plan; Perform the assembly of the prefabricated steel structure of the target assembly project based on the target assembly plan; The step of, from the two dimensions of surface defects and connection quality, combined with the digital twin model of the prefabricated steel structure, perform feature extraction and analysis on the Q sets of node images to determine Q node quality factors includes: Use the surface defect analysis network layer to identify each of the Q sets of node images to obtain Q surface quality coefficients; Based on the digital twin model of the prefabricated steel structure, identify the key connection points of the Q steel structure nodes to be assembled to obtain Q sets of key connection points of the nodes, and use a feature extraction algorithm to traverse the Q sets of key connection points of the nodes to extract the processing features of the connection points to obtain Q clusters of connection point processing features; Traverse the Q clusters of connection point processing features to identify the connection quality and obtain Q node connection coefficients; Based on the Q surface quality coefficients and the Q node connection coefficients, perform comprehensive quality identification on the Q steel structure nodes to be assembled to determine Q node quality factors; Judge whether the Q node quality factors are less than a preset node quality factor. If so, the assemblability certification fails, generate a warning message, and send the warning message to the staff for renovation of the steel structure nodes; If not, the assemblability certification passes.

2. The optimized assembly method of prefabricated steel structure integrating BIM technology according to claim 1, characterized in that, Based on the digital twin model of the prefabricated steel structure, identify the key connection points of the Q steel structure nodes to be assembled to obtain Q sets of key connection points of the nodes, including: Traverse the Q steel structure nodes to be assembled for neighborhood matching in the digital twin model of the prefabricated steel structure to obtain Q neighborhoods of the steel structure nodes to be assembled and Q sets of assembly parameters of the steel structure nodes to be assembled; Extract the Q design information of the Q steel structure nodes to be assembled, combine the Q sets of assembly parameters of the steel structure nodes to be assembled to perform frequent item mining of faults, and determine the risk coefficients of the Q steel structure nodes to be assembled according to the mining and analysis results; Based on the magnitudes of the risk coefficients of the Q steel structure nodes to be assembled, perform an analysis of the number of key connection points to determine the number of Q key connection points; Taking the number of the Q key connection points as the first constraint, combining the neighborhoods of the Q steel structure nodes to be assembled and the Q steel structure nodes to be assembled, and identifying the key connection points of the nodes to obtain the set of the Q key connection points of the nodes.

3. The optimized assembly method of prefabricated steel structure integrating BIM technology according to claim 2, characterized in that, Including: Taking the neighborhoods of the Q steel structure nodes to be assembled and the indexes of the Q steel structure nodes to be assembled, combining with the digital twin model of the prefabricated steel structure to identify the neighborhood connection points, and obtaining a set of Q neighborhood connection points; Traversing the Q steel structure nodes to be assembled to extract connection points, and obtaining a set of Q node connection points; Dividing the number of connection points in the set of Q neighborhood connection points by the number of connection points in the set of Q node connection points respectively to obtain Q neighborhood connection point ratios, and taking the Q neighborhood connection point ratios as the second constraint for key connection point identification; Based on the first constraint and the second constraint, identifying the key connection points of the set of Q node connection points to obtain the set of the Q key connection points of the nodes.

4. The optimized assembly method of prefabricated steel structure integrating BIM technology according to claim 3, characterized in that, Including: Randomly extracting from the set of Q node connection points based on the first constraint and the second constraint to obtain a set of Q initial key connection points of the nodes; Performing three-dimensional simulation according to the positions of the connection points in the set of Q initial key connection points of the nodes to obtain a set of Q initial key connection point models; Based on the digital twin model of the prefabricated steel structure, retrieving the three-dimensional models of the Q steel structures to be assembled to obtain a set of Q steel structures to be assembled models; Performing similarity matching on the set of Q initial key connection point models and the set of Q steel structures to be assembled models to obtain Q matching fitnesses; When the Q matching fitnesses are greater than or equal to the preset matching fitness threshold, taking the set of Q initial key connection points of the nodes as the set of the Q key connection points of the nodes.

5. The optimized assembly method of prefabricated steel structure integrating BIM technology according to claim 1, characterized in that Including: Collecting environmental data based on Internet of Things devices and sensors deployed at the assembly site to obtain real-time assembly environment information; Performing assembly simulation according to the preset assembly plan, the real-time assembly environment information and the digital twin model of the prefabricated steel structure, and analyzing the assembly states of the Q steel structure nodes to be assembled by using finite element analysis to obtain analysis results; Optimizing the preset assembly plan according to the analysis results to obtain the target assembly plan.

6. The optimized assembly method of prefabricated steel structure integrating BIM technology according to claim 1, characterized in that, Pre-constructing a quality identification network layer, and using the quality identification network layer to respectively perform mapping identification on the Q surface quality coefficients and the Q node connection coefficients to obtain the Q node quality factors.

7. An assembly optimization system for prefabricated steel structures integrating BIM technology, characterized in that, The system is used to execute the prefabricated steel structure assembly optimization method integrating BIM technology according to any one of claims 1 to 6, and the system includes: A digital twin model construction module, which is used to obtain the design information of the prefabricated steel structure of the target assembly project, input the design information of the prefabricated steel structure into the BIM model for analysis and simulation, and construct a digital twin model of the prefabricated steel structure; A multi-angle image acquisition module, which is used to call a high-definition industrial camera to perform multi-angle image acquisition on the Q steel structure nodes to be assembled to obtain a set of Q node images, where Q is a positive integer greater than or equal to 1; A feature extraction and analysis module, which is used to extract and analyze features of the Q node image sets from two dimensions of surface defects and connection quality, in combination with the digital twin model of the prefabricated steel structure, to determine Q node quality factors; An assembly plan optimization module, which is used to perform assembly certification on the Q steel structure nodes to be assembled based on the magnitudes of the Q node quality factors. If the assembly certification passes, perform assembly simulation based on the digital twin model of the prefabricated steel structure, and optimize the assembly plan in combination with the actual assembly environment information to obtain a target assembly plan; An assembly plan execution module, which is used to perform the prefabricated steel structure assembly of the target assembly project based on the target assembly plan.

Images