AI-based intelligent forming method and system for precast box girder steel reinforcement skeleton

Abstract

This application relates to the field of precast box girder reinforcement cage forming technology, and in particular to an AI-based intelligent forming method and system for precast box girder reinforcement cages. The method includes generating a 3D model and extracting geometric feature parameters using a BIM model building module, generating an optimal mesh division scheme using a deep learning training module, selecting modular meshes from a standardized mesh library, and completing the splicing using a robotic arm and automated welding equipment. This application can significantly improve the accuracy of reinforcement layout and construction efficiency, solve problems such as uneven reinforcement layout and missed welding points in complex beam end variable cross-section areas, while reducing labor costs and material waste. It also possesses flexibility and scalability, making it suitable for industrial production scenarios.

Description

Technical Field

[0001] This invention relates to the field of intelligent building engineering technology, and in particular to an AI-based intelligent forming method and system for precast box girder steel reinforcement skeleton. Background Technology

[0002] Existing precast box girder reinforcement cage forming processes mainly rely on manual binding and welding, which has limitations in terms of construction quality and efficiency. For example, in the variable cross-section area at the beam end, due to the high density and complex spatial arrangement of the reinforcement, manual operation is prone to problems such as uneven spacing, missed welding points, or positioning deviations. These problems may further lead to insufficient concrete cover thickness and inaccurate positioning of prestressed ducts, resulting in quality risks. With the expansion of bridge engineering construction scale and increasingly stringent construction period requirements, the industry has an urgent need for high-precision, automated forming technology that can adapt to variable cross-section structures.

[0003] Currently, precast beam yards generally use manual methods to form the steel reinforcement cage. This traditional method requires a significant investment of manpower and material resources and is time-consuming. Furthermore, the inherent uncertainty of manual operation makes it difficult to maintain consistent cage binding quality, resulting in low overall efficiency and failing to meet the demands of modern large-scale engineering construction. For example, when installing the lower layer of longitudinal reinforcement in the base slab, each reinforcement bar must be placed manually, which is not only time-consuming but also makes it difficult to ensure uniform spacing. Similarly, when binding the stirrups in the base slab, manual fixing of the stirrups to the longitudinal reinforcement is also necessary, easily leading to insecure binding or positional deviations. Although some institutions have attempted to introduce intelligent construction technology into the standard mid-span section of box girders to improve the forming quality and efficiency of the steel reinforcement cage, automated production still faces significant challenges in the variable cross-section areas at the beam ends due to their complex structure and dense reinforcement arrangement, and a mature and effective solution has not yet been developed. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide an AI-based intelligent forming method and system for precast box girder steel reinforcement skeleton, which addresses the above-mentioned defects of the prior art. The aim is to solve the problems of low efficiency and poor accuracy of manual binding and welding in the prior art, especially the complex arrangement of steel reinforcement in the variable cross-section area at the beam end and the high difficulty of automated production.

[0005] The technical solution adopted by this invention to solve the technical problem is as follows: In a first aspect, the present invention provides an AI-based intelligent forming method for the reinforcing steel skeleton of precast box girders, comprising: Build a BIM model and split it into mesh panels; Deep learning algorithms are introduced to train on historical beam end variable cross-section data to generate the optimal mesh partitioning scheme; Based on the slope and length parameters of the variable cross-section at the beam end, modular meshes from the standardized mesh library are dynamically combined to form the final steel reinforcement skeleton.

[0006] In some embodiments of this application, the construction of the BIM model and the mesh splitting include: A BIM model of the precast box girder is generated using 3D modeling software. Extract the geometric feature parameters of the variable cross-section region at the beam end, including slope, length, and width; The geometric feature parameters are input into a preset deep learning model to generate an initial mesh partitioning scheme; wherein, The deep learning model is based on a convolutional neural network architecture, and the training data comes from the reinforcement layout data of historical beam ends with variable cross sections.

[0007] In some embodiments of this application, the introduction of a deep learning algorithm to train historical beam end variable cross-section data and generate an optimal mesh partitioning scheme includes: Data on the arrangement of variable cross-section steel bars at the beam ends of multiple precast box girder projects were collected to form a training dataset. The training dataset is labeled, including the spacing of the reinforcing bars, the location of the intersections, and the boundaries of the mesh. The labeled dataset is input into the deep learning model for training, and the model weights are adjusted until the loss function converges. Output the optimal mesh partitioning scheme, which includes the size, shape, and connection point location of each mesh.

[0008] In some embodiments of this application, the step of dynamically combining modular meshes from a standardized mesh library to form the final reinforcing steel skeleton based on the slope and length parameters of the variable cross-section at the beam end includes: Establish a standardized mesh library, which contains various types of modular meshes, numbered A1 to A13; Obtain the slope parameter θ and length parameter L of the variable cross section at the beam end; Based on the slope parameter θ and length parameter L, a suitable mesh type is selected from the standardized mesh library; The selected mesh types are combined according to preset splicing rules to form a complete steel reinforcement skeleton; among which, The splicing rules include the overlap length between mesh panels and the location of fixing points.

[0009] In some embodiments of this application, establishing a standardized mesh library includes: Design various types of modular mesh panels, each with a fixed size and shape; Each type of mesh is numbered, and its geometric parameters and mechanical properties are recorded; The geometric parameters and mechanical properties of all wire meshes are stored in a database to form a standardized wire mesh library; among them, The geometric parameters include the aspect ratio of the mesh and the spacing of the reinforcing bars, and the mechanical properties include tensile strength and yield strength.

[0010] In some embodiments of this application, the step of combining the selected mesh types according to preset splicing rules to form a complete steel reinforcement skeleton includes: Calculate the angular deviation between adjacent mesh panels based on the slope parameter θ of the variable cross section at the beam end; Adjust the placement direction of the mesh according to the aforementioned angular deviation; The overlap length between the mesh panels is determined based on the length parameter L of the variable cross section at the beam end. The mesh panels are fixed in the designated position by a robotic arm, and the connection between the mesh panels is completed using an automatic welding device.

[0011] Secondly, embodiments of the present invention also provide an AI-based intelligent forming system for the reinforcing steel skeleton of precast box girders, wherein the system includes: The BIM model building module is used to generate BIM models of precast box girders and extract geometric feature parameters of variable cross sections at the beam ends. The deep learning training module is used to train on historical beam end variable cross-section data to generate the optimal mesh partitioning scheme; The mesh assembly module is used to dynamically combine modular meshes from a standardized mesh library based on the slope and length parameters of the variable cross-section at the beam end, forming the final steel reinforcement skeleton.

[0012] Thirdly, embodiments of the present invention also provide a smart terminal, wherein the smart terminal includes a memory, a processor, and a precast box girder steel reinforcement skeleton intelligent forming program stored in the memory and executable on the processor. When the processor executes the precast box girder steel reinforcement skeleton intelligent forming program, it implements the steps of the AI-based precast box girder steel reinforcement skeleton intelligent forming method as described in any of the above embodiments.

[0013] Fourthly, embodiments of the present invention also provide a computer-readable storage medium, wherein the computer-readable storage medium stores a precast box girder reinforcement skeleton intelligent forming program, and when the precast box girder reinforcement skeleton intelligent forming program is executed by a processor, it implements the steps of the AI-based precast box girder reinforcement skeleton intelligent forming method as described in any of the above claims.

[0014] The beneficial technical effects of this invention are as follows: This invention accurately extracts the geometric feature parameters of the variable cross-section area at the beam end through the BIM model construction module, and generates the optimal mesh division scheme by combining it with the deep learning training module, which significantly improves the accuracy and efficiency of steel reinforcement layout. The modular meshes in the standardized mesh library have been rigorously designed and verified to ensure the reliability of geometric parameters and mechanical properties, which greatly improves the efficiency of the work and the accuracy of manufacturing. Attached Figure Description

[0015] Figure 1 This is a flowchart of the AI-based intelligent forming method for the steel reinforcement skeleton of precast box girders in this embodiment of the invention; Figure 2 This is a functional block diagram of the AI-based intelligent forming system for precast box girder steel reinforcement skeleton according to an embodiment of the present invention. Detailed Implementation

[0016] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0017] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0018] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0019] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the communication between the inner sides of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0020] This invention provides an AI-based intelligent forming method and system for precast box girder steel reinforcement skeletons. Its core lies in the collaborative work of a BIM model construction module, a deep learning training module, and a mesh combination module, combined with modular meshes from a standardized mesh library, to achieve intelligent forming of the steel reinforcement skeleton in the variable cross-section area of ​​the beam end.

[0021] See Figure 1 As shown, this invention provides an AI-based intelligent forming method for the steel reinforcement skeleton of precast box girders, comprising: Step S101: Construct a BIM model and split the mesh; Step S102: Introduce a deep learning algorithm to train on historical beam end variable cross-section data to generate the optimal mesh partitioning scheme; Step S103: Based on the slope and length parameters of the variable cross section at the beam end, dynamically combine the modular meshes in the standardized mesh library to form the final steel reinforcement skeleton.

[0022] In one specific embodiment of this application, constructing a BIM model and performing mesh segmentation includes: A BIM model of the precast box girder is generated using 3D modeling software. Extract the geometric feature parameters of the variable cross-section region at the beam end, including slope, length, and width; The geometric feature parameters are input into a pre-defined deep learning model to generate an initial mesh partitioning scheme; among which, The deep learning model is based on a convolutional neural network architecture, and the training data comes from the reinforcement layout data of historical beam ends with variable cross sections.

[0023] In one specific embodiment of this application, a deep learning algorithm is introduced to train historical beam end variable cross-section data to generate an optimal mesh partitioning scheme, including: Data on the arrangement of variable cross-section steel bars at the beam ends of multiple precast box girder projects were collected to form a training dataset. The training dataset is labeled, including the spacing of the reinforcing bars, the location of intersections, and the boundaries of the mesh. The labeled dataset is input into the deep learning model for training, and the model weights are adjusted until the loss function converges. Output the optimal mesh partitioning scheme, which includes the size, shape, and connection point location of each mesh.

[0024] In one specific embodiment of this application, modular meshes from a standardized mesh library are dynamically combined based on the slope and length parameters of the variable cross-section at the beam end to form the final steel reinforcement skeleton, including: Establish a standardized mesh library, which contains various types of modular meshes, numbered A1 to A13; Obtain the slope parameter θ and length parameter L of the variable cross section at the beam end; Based on the slope parameter θ and the length parameter L, select the appropriate mesh type from the standardized mesh library; The selected mesh types are combined according to preset splicing rules to form a complete steel reinforcement skeleton; among which, The splicing rules include the overlap length between mesh panels and the location of the fixing points.

[0025] In one specific embodiment of this application, establishing a standardized mesh library includes: Design various types of modular mesh panels, each with a fixed size and shape; Each type of mesh is numbered, and its geometric parameters and mechanical properties are recorded; The geometric parameters and mechanical properties of all wire meshes are stored in a database to form a standardized wire mesh library; among them, Geometric parameters include the aspect ratio of the mesh and the spacing of the reinforcing bars, while mechanical properties include tensile strength and yield strength.

[0026] In one specific embodiment of this application, the selected mesh types are combined according to preset splicing rules to form a complete steel reinforcement skeleton, including: Calculate the angular deviation between adjacent mesh panels based on the slope parameter θ of the variable cross section at the beam end; Adjust the placement direction of the mesh according to the angular deviation; The overlap length between the mesh panels is determined based on the length parameter L of the variable cross section at the beam end. The mesh panels are fixed in the designated position by a robotic arm, and the connection between the mesh panels is completed using an automatic welding device.

[0027] See Figure 2 As shown, in practice, the BIM model building module first generates a BIM model of the precast box girder and extracts the geometric feature parameters of the variable cross-section area at the beam end. This process is completed using 3D modeling software, where geometric feature parameters such as the slope, length, and width of the variable cross-section area at the beam end are accurately extracted and input into the deep learning training module. The deep learning training module is based on a convolutional neural network architecture and uses historical beam end variable cross-section reinforcement layout data for training to generate the optimal mesh partitioning scheme. The training dataset comes from the beam end variable cross-section reinforcement layout data of multiple precast box girder projects and has been annotated, with annotations including reinforcement spacing, intersection locations, and mesh boundaries. The annotated dataset is input into the deep learning model for training, adjusting the model weights until the loss function converges, and finally outputting the optimal mesh partitioning scheme containing the size, shape, and connection point locations of each mesh.

[0028] After obtaining the optimal mesh partitioning scheme, the mesh assembly module dynamically selects suitable modular meshes from the standardized mesh library based on the slope parameter θ and length parameter L of the variable cross-section region 6 at the beam end. The standardized mesh library contains various types of modular meshes, numbered A1 to A13. Each mesh has a fixed size and shape, and its geometric parameters and mechanical properties are recorded.

[0029] Geometric parameters include the aspect ratio of the mesh panels and the spacing of the reinforcing bars, while mechanical properties include tensile strength and yield strength. These parameters are stored in a database for easy retrieval and matching. The mesh assembly module calculates the angular deviation between adjacent modular mesh panels based on the slope parameter θ and the length parameter L, and adjusts the placement of the mesh panels to ensure splicing accuracy. Simultaneously, it determines the overlap length between modular mesh panels based on the length parameter L to ensure the overall stability of the reinforcing steel skeleton.

[0030] The modular mesh assembly process involves a robotic arm fixing selected modular mesh panels in designated positions and then connecting the panels using automated welding equipment. The position and movement of the robotic arm are controlled by the mesh assembly modules, and its trajectory is pre-planned based on the geometric parameters and assembly rules of the modular mesh panels. The assembly rules include the overlap length between mesh panels and the location of fixing points; these rules ensure that the modular mesh panels can be seamlessly assembled to form a complete steel reinforcement skeleton.

[0031] The automatic welding equipment welds the connection points of the modular mesh according to preset welding parameters, including current intensity, welding time, and weld point distribution. These parameters are optimized based on the mechanical properties of the modular mesh to ensure welding quality.

[0032] Throughout the process, the BIM model building module, deep learning training module, and mesh assembly module interact with each other via data interfaces. The BIM model building module transmits extracted geometric feature parameters to the deep learning training module, which then sends the optimal mesh partitioning scheme generated by the deep learning training module to the mesh assembly module. The mesh assembly module selects modular meshes from a standardized mesh library based on the partitioning scheme and plans the splicing process. This modular design not only improves the system's flexibility but also facilitates subsequent functional expansion and technological upgrades.

[0033] Furthermore, this invention also provides an intelligent terminal, which includes a memory, a processor, and an intelligent forming program for the precast box girder reinforcement skeleton stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the aforementioned AI-based intelligent forming method for the precast box girder reinforcement skeleton. The intelligent terminal, through communication with the BIM model building module, the deep learning training module, and the mesh assembly module, monitors the reinforcement skeleton forming process in real time and adjusts parameter settings according to actual conditions to ensure forming accuracy and efficiency.

[0034] In practical applications, this invention can be applied to the precast box girder production stage in bridge construction. For example, in large-scale bridge projects, construction units need to produce a batch of precast box girders with complex variable cross-section areas at the beam ends. By using the intelligent forming system provided by this invention, the construction unit first uses the BIM model building module to generate a BIM model of the precast box girder and extracts the geometric feature parameters of the variable cross-section areas at the beam ends. Subsequently, the deep learning training module generates the optimal mesh partitioning scheme based on historical data, and the mesh combination module selects suitable modular meshes from a standardized mesh library according to the scheme. The steel reinforcement skeleton is then formed using a robotic arm and automatic welding equipment. The entire process requires no manual intervention, significantly improving production efficiency and forming accuracy while reducing labor costs and material waste.

[0035] The present invention stores a precast box girder reinforcement cage intelligent forming program on a computer-readable storage medium. When executed by a processor, this program implements the steps of the aforementioned AI-based precast box girder reinforcement cage intelligent forming method. The storage medium can be a common storage device such as a hard disk, solid-state drive, or flash memory, ensuring that the program can run on different hardware platforms. In this way, the present invention not only achieves automation and intelligence in reinforcement cage forming, but also provides a solid foundation for subsequent technical improvements and functional expansion.

[0036] To enable those skilled in the art to fully understand and implement this invention, the specific implementation principle of this invention is further explained below in conjunction with a specific application scenario.

[0037] In large-scale bridge construction projects, construction units need to produce a batch of precast box girders with complex variable cross-section areas at the beam ends. To achieve intelligent forming of the reinforcing steel skeleton, a BIM model of the precast box girder is first generated using a BIM model building module, and the geometric feature parameters of the variable cross-section areas at the beam ends are extracted. These parameters include key information such as slope, length, and width. After accurate extraction using 3D modeling software, they are transmitted to a deep learning training module for processing. The deep learning training module, based on a convolutional neural network architecture, uses historical data on the arrangement of variable cross-section reinforcing steel at the beam ends for training, generating the optimal mesh partitioning scheme. During training, annotations include steel bar spacing, intersection points, and mesh boundaries to ensure that the output partitioning scheme meets actual construction requirements. Subsequently, the mesh combination module dynamically selects suitable modular meshes from a standardized mesh library based on the slope parameter θ and length parameter of the variable cross-section area at the beam ends. The standardized mesh library contains various modular meshes numbered A1 to A13, each recording its geometric parameters and mechanical properties. Geometric parameters include the aspect ratio and steel bar spacing of the mesh, while mechanical properties cover tensile strength and yield strength. These parameters are stored in a database for quick retrieval and matching. The mesh assembly module calculates the angular deviation between adjacent modular meshes based on the slope parameter θ and adjusts the mesh placement to ensure splicing accuracy. Simultaneously, it determines the overlap length between modular meshes based on the length parameter L, thereby ensuring the overall stability of the reinforcing steel frame.

[0038] The modular mesh splicing process involves a robotic arm fixing selected modular mesh panels in designated positions and then connecting the panels using automated welding equipment. The position and movements of the robotic arm are controlled by the mesh assembly modules, and its trajectory is pre-planned based on the geometric parameters and splicing rules of the modular mesh panels. These splicing rules include the overlap length between mesh panels and the location of fixing points, ensuring that the modular mesh panels can be seamlessly spliced ​​to form a complete steel reinforcement skeleton. The automated welding equipment then welds the connection points of the modular mesh panels according to preset welding parameters, including current intensity, welding time, and weld point distribution. These parameters are optimized based on the mechanical properties of the modular mesh panels to guarantee welding quality.

[0039] Throughout the process, the BIM model building module, deep learning training module, and mesh assembly module interact with each other via a data interface. The BIM model building module transmits extracted geometric feature parameters to the deep learning training module, and the optimal mesh partitioning scheme generated by the deep learning training module is passed to the mesh assembly module. The mesh assembly module selects modular meshes from a standardized mesh library according to the partitioning scheme and plans the splicing process. This modular design not only improves the system's flexibility but also facilitates subsequent functional expansion and technology upgrades.

[0040] The intelligent terminal communicates with the BIM model building module, deep learning training module, and mesh assembly module to monitor the steel reinforcement cage forming process in real time and adjust parameter settings according to actual conditions to ensure forming accuracy and efficiency. For example, during the splicing process, if the angular deviation of a modular mesh exceeds the allowable range, the intelligent terminal will automatically adjust the robotic arm's trajectory and reposition the mesh. Furthermore, if uneven weld point distribution occurs during welding, the intelligent terminal will adjust the current intensity or welding time based on feedback from welding parameters to ensure welding quality meets requirements.

[0041] Through the above steps, the entire rebar cage forming process requires no manual intervention, significantly improving production efficiency and forming accuracy while reducing labor costs and material waste. For example, in the rebar arrangement in the variable cross-section area at the beam end, traditional manual tying methods easily lead to problems such as uneven rebar spacing or missed welding points. However, this invention effectively avoids these problems through the precise splicing of modular mesh and the optimized operation of automated welding equipment. Furthermore, because the geometric parameters and mechanical properties of the modular mesh have been rigorously designed and verified, the overall stability and stress performance of the rebar cage are also significantly improved.

[0042] The present invention stores a precast box girder reinforcement cage intelligent forming program on a computer-readable storage medium. When executed by a processor, this program implements the steps of the aforementioned AI-based precast box girder reinforcement cage intelligent forming method. The storage medium can be a common storage device such as a hard disk, solid-state drive, or flash memory, ensuring that the program can run on different hardware platforms. In this way, the present invention not only achieves automation and intelligence in reinforcement cage forming, but also provides a solid foundation for subsequent technical improvements and functional expansion.

[0043] This invention provides an AI-based intelligent forming method and system for the reinforcing steel skeleton of precast box girders. It accurately extracts geometric feature parameters of the variable cross-section region at the beam ends through a BIM model building module, and generates an optimal mesh division scheme using a deep learning training module, significantly improving the accuracy and efficiency of steel reinforcement layout. The modular meshes in the standardized mesh library have undergone rigorous design and verification, ensuring the reliability of geometric parameters and mechanical properties. The collaborative work of a robotic arm and automated welding equipment enables seamless splicing of the modular meshes, avoiding problems such as uneven steel reinforcement spacing or missed welding points common in traditional manual binding and welding operations. Furthermore, this invention improves system flexibility through modular design, facilitating subsequent functional expansion and technological upgrades, while reducing labor costs and material waste.

[0044] All content not described in detail in this specification is prior art known to those skilled in the art, and the model parameters of each electrical appliance are not specifically limited; conventional equipment can be used. Electrical control components not mentioned in this technical solution are not shown in the figures because they are prior art, and will not be described further here.

[0045] The above description is merely one embodiment of the present invention, but it cannot be used to limit the scope of the present invention. Any structural changes made based on the present invention, as long as they do not lose the essence of the present invention, should be considered to fall within the protection scope of the present invention and be subject to its restrictions.

[0046] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process and related descriptions of the system described above can be found in the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0047] It should be noted that the system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the modules or steps in the embodiments of the present invention can be further decomposed or combined. For example, the modules in the above embodiments can be merged into one module, or further divided into multiple sub-modules to complete all or part of the functions described above. The names of the modules and steps involved in the embodiments of the present invention are only for distinguishing the various modules or steps and are not considered as an improper limitation of the present invention.

[0048] Those skilled in the art will recognize that the modules and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. The programs corresponding to the software modules and method steps can be placed in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in the art. To clearly illustrate the interchangeability of electronic hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in electronic hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the invention.

[0049] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.

[0050] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.

[0051] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention.

Claims

1. An AI-based intelligent forming method for the reinforcing steel skeleton of precast box girders, characterized in that, include: Build a BIM model and split it into mesh panels; Deep learning algorithms are introduced to train on historical beam end variable cross-section data to generate the optimal mesh partitioning scheme; Based on the slope and length parameters of the variable cross-section at the beam end, modular meshes from the standardized mesh library are dynamically combined to form the final steel reinforcement skeleton.

2. The AI-based intelligent forming method for precast box girder reinforcement skeleton according to claim 1, characterized in that, The process of constructing the BIM model and splitting it into mesh panels includes: A BIM model of the precast box girder is generated using 3D modeling software. Extract the geometric feature parameters of the variable cross-section region at the beam end, including slope, length, and width; The geometric feature parameters are input into a preset deep learning model to generate an initial mesh partitioning scheme; wherein, The deep learning model is based on a convolutional neural network architecture, and the training data comes from the reinforcement layout data of historical beam ends with variable cross sections.

3. The AI-based intelligent forming method for precast box girder reinforcement cage according to claim 2, characterized in that, The introduction of a deep learning algorithm to train on historical beam end variable cross-section data to generate an optimal mesh partitioning scheme includes: Data on the arrangement of variable cross-section steel bars at the beam ends of multiple precast box girder projects were collected to form a training dataset. The training dataset is labeled, including the spacing of the reinforcing bars, the location of the intersections, and the boundaries of the mesh. The labeled dataset is input into the deep learning model for training, and the model weights are adjusted until the loss function converges. Output the optimal mesh partitioning scheme, which includes the size, shape, and connection point location of each mesh.

4. The AI-based intelligent forming method for precast box girder reinforcement cage according to claim 3, characterized in that, The process of dynamically combining modular meshes from a standardized mesh library based on the slope and length parameters of the variable cross-section at the beam end to form the final steel reinforcement skeleton includes: Establish a standardized mesh library, which contains various types of modular meshes, numbered A1 to A13; Obtain the slope parameter θ and length parameter L of the variable cross section at the beam end; Based on the slope parameter θ and length parameter L, a suitable mesh type is selected from the standardized mesh library; The selected mesh types are combined according to preset splicing rules to form a complete steel reinforcement skeleton; among which, The splicing rules include the overlap length between mesh panels and the location of fixing points.

5. The AI-based intelligent forming method for the reinforcing steel skeleton of precast box girders according to claim 4, characterized in that, The establishment of the standardized mesh library includes: Design various types of modular mesh panels, each with a fixed size and shape; Each type of mesh is numbered, and its geometric parameters and mechanical properties are recorded; The geometric parameters and mechanical properties of all wire meshes are stored in a database to form a standardized wire mesh library; among them, The geometric parameters include the aspect ratio of the mesh and the spacing of the reinforcing bars, and the mechanical properties include tensile strength and yield strength.

6. The AI-based intelligent forming method for precast box girder reinforcement cage according to claim 4, characterized in that, The step of combining the selected mesh types according to preset splicing rules to form a complete steel reinforcement skeleton includes: Calculate the angular deviation between adjacent mesh panels based on the slope parameter θ of the variable cross section at the beam end; Adjust the placement direction of the mesh according to the aforementioned angular deviation; The overlap length between the mesh panels is determined based on the length parameter L of the variable cross section at the beam end. The mesh panels are fixed in the designated position by a robotic arm, and the connection between the mesh panels is completed using an automatic welding device.

7. An AI-based intelligent forming system for the reinforcing steel skeleton of precast box girders, characterized in that, The system includes: The BIM model building module is used to generate BIM models of precast box girders and extract geometric feature parameters of variable cross sections at the beam ends. The deep learning training module is used to train on historical beam end variable cross-section data to generate the optimal mesh partitioning scheme; The mesh assembly module is used to dynamically combine modular meshes from a standardized mesh library based on the slope and length parameters of the variable cross-section at the beam end, forming the final steel reinforcement skeleton.

8. The AI-based intelligent forming system for precast box girder reinforcement cages according to claim 7, characterized in that, The system also includes a standardized mesh library containing various types of modular meshes, each with a fixed size and shape, and its geometric parameters and mechanical properties are recorded. The geometric parameters include the aspect ratio of the mesh and the spacing between the reinforcing bars, and the mechanical properties include tensile strength and yield strength.

9. A smart terminal, characterized in that, The intelligent terminal includes a memory, a processor, and an intelligent forming program for the precast box girder steel reinforcement skeleton stored in the memory and executable on the processor. When the processor executes the intelligent forming program for the precast box girder steel reinforcement skeleton, it implements the steps of the AI-based intelligent forming method for the precast box girder steel reinforcement skeleton as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a precast box girder reinforcement skeleton intelligent forming program. When the precast box girder reinforcement skeleton intelligent forming program is executed by a processor, it implements the steps of the AI-based precast box girder reinforcement skeleton intelligent forming method as described in any one of claims 1 to 6.

Images