Intelligent BIM (Building Information Modeling) creation method, system and device based on AI (Artificial Intelligence) algorithm and medium
By combining exchange cloud and deep learning models, the security problem of building data storage is solved, the accuracy of BIM models and construction precision are ensured, and dynamic, secure storage and complete protection of data are achieved.
Patent Information
- Application Number
- CN202511066002.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-31
- Publication Date
- 2025-10-28
AI Technical Summary
The storage security of building data faces the challenge of data leakage or malicious tampering, which affects the accuracy of BIM models and construction precision.
Securely store building data through Exchange Cloud, combine deep learning models to process data to generate building element information, and create models in BIM software, using Exchange Cloud's dynamic storage and authorized network to protect data integrity.
It ensures the security and integrity of building data, guarantees the accuracy of BIM models, prevents data leakage and tampering, and improves construction precision and design rationality.
Smart Images
Figure CN120850430A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, specifically to a method, system, device, and medium for intelligent creation of BIM models based on AI algorithms. Background Art
[0002] In recent years, AI technology has made significant progress in machine learning, deep learning, and computer vision, enabling efficient processing of massive amounts of building data and automatic extraction of features and patterns from building elements, thus promoting the intelligent creation of BIM. However, building data, as a core asset spanning the entire design and construction process, faces challenges in storage security: data leaks or malicious tampering can lead to information distortion, resulting in BIM model deviations. If the accuracy of the model is compromised, it will directly interfere with the rationality of the design, the precision of construction, and the efficiency of operation and maintenance, affecting the progress of various aspects of building work. Summary of the Invention
[0003] The purpose of this invention is to provide a method, system, device, and medium for intelligent creation of BIM models based on AI algorithms, so as to solve the problems in the background art.
[0004] To achieve the above objectives, the present invention provides the following technical solution: an intelligent BIM model creation method based on AI algorithms, comprising the following steps:
[0005] Building data is collected for the target building and stored in the creation platform. The building data is then securely stored through the exchange cloud within the creation platform.
[0006] The platform is designed to process building data using deep learning models to obtain building element information.
[0007] Input building element information into BIM software to create a BIM model.
[0008] In a preferred embodiment, the steps of collecting building data for the corresponding target building, storing the building data in the creation platform, and securely storing the building data through the exchange cloud in the creation platform include:
[0009] Identify the target building, set up a data acquisition port for the target building, and acquire the building data of the target building based on the data acquisition port.
[0010] Building data is transmitted to the creation platform via the acquisition port;
[0011] The building data is stored securely in the corresponding exchange cloud during the platform creation process.
[0012] In a preferred embodiment, the step of securely storing the building data in the corresponding exchange cloud within the creation platform includes:
[0013] Multiple switching clouds are set up in the platform creation, where each switching cloud includes a storage cell, two switching shards, and multiple switching cells;
[0014] Bind multiple exchange clouds to their corresponding collection ports;
[0015] Building data is stored in exchange cells within a bound exchange cloud. Multiple exchange cells circulate and exchange data between two exchange slices. A tracking graph is established between the exchange cells storing building data and the storage cells. Based on the tracking graph, a connection relationship is established between the authorized network and the exchange cells.
[0016] In a preferred embodiment, the step of setting up multiple exchange clouds in the creation platform includes:
[0017] In the creation platform, multiple storage cells are set up, and two switching slices are set up in the multiple storage cells. The switching slice is composed of multiple data points.
[0018] Establish connection channels between data points in two switching slices, and specify the transmission direction of multiple connection channels respectively;
[0019] A switching element is set up between two switching segments, and the switching element transmits and moves data through the data points in the two switching segments and the connection channel.
[0020] In a preferred embodiment, the step of establishing a tracking graph between the exchange element storing building data and the storage element, and establishing the connection relationship between the authorized network and the exchange element based on the tracking graph, includes:
[0021] The building data is divided into multiple building data segments. Multiple exchange elements corresponding to the number of building data segments are randomly selected. The building data segments are stored one-to-one in the exchange elements. The exchange element containing the building data segments is used as the target exchange element.
[0022] Set up a tracking graph in the storage element, and set up multiple tracking points in the tracking graph. The number of tracking points is the same as the number of target swap elements.
[0023] Bind the tracking points to the target exchange elements in a one-to-one correspondence;
[0024] The target switching element moves between two switching segments through the connection channel. The position of the corresponding target switching element is tracked by the tracking point, and the position of the target switching element is marked on the tracking map. The corresponding authorized network is marked on the tracking map.
[0025] When the authorized network accesses the switching cloud through the output port on the storage cell, the location of the target switching cell is provided to the authorized network through the tracking graph, and the authorized network is connected to the target switching cell to obtain building data.
[0026] In a preferred embodiment, the step of processing building data based on a deep learning model in the creation platform to obtain building element information includes:
[0027] The convolutional neural network is trained using training data, which includes training building data and corresponding building element information, to obtain a trained deep learning model.
[0028] The deep learning model is configured with an access port. The deep learning model requests access to the exchange cloud through the access port. The tracking graph provides the location of the target exchange element to the access port. The model obtains building data by connecting to the target exchange element through the access port. Based on the access port, the building data is input into the deep learning model to obtain building element information.
[0029] In a preferred embodiment, the step of inputting building element information into BIM software to create a BIM model includes:
[0030] Building element information is transferred from the deep learning model to the BIM software via an API interface;
[0031] BIM software automatically generates models based on received data and instructions.
[0032] This invention also provides an intelligent BIM model creation system based on AI algorithms, comprising:
[0033] The storage module is used to collect building data for the target building and store the building data in the creation platform. The building data is securely stored through the exchange cloud in the creation platform.
[0034] The data processing module, connected to the storage module, is used to process building data based on a deep learning model in the creation platform to obtain building element information;
[0035] The creation module, connected to the data processing module, is used to input building element information into the BIM software to create a BIM model.
[0036] The technical effects and advantages provided by the present invention in the above technical solution are as follows:
[0037] This invention enables dynamic storage of building data through cloud exchange, making it difficult for outsiders to obtain, access, or tamper with the building data. This ensures the security and integrity of the building data storage, providing excellent data protection. Accurate building data also guarantees the accuracy of BIM model creation. Attached Figure Description
[0038] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.
[0039] Figure 1 This is a flowchart of the method of the present invention.
[0040] Figure 2 This is a system block diagram of the present invention. Detailed Implementation
[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0042] Example 1, please refer to Figure 1 As shown in this embodiment, the intelligent BIM model creation method based on AI algorithms includes the following steps:
[0043] S1. Collect building data for the target building, store the building data in the creation platform, and securely store the building data through the exchange cloud in the creation platform;
[0044] S2. In the creation platform, the building data is processed based on a deep learning model to obtain building element information;
[0045] S3. Input building element information into BIM software to create a BIM model;
[0046] As described in steps S1-S3 above, the target building is determined. This target building can be a building under construction or a building that has not yet been constructed (representing a building that has been designed but not yet constructed). In order to better control the progress and cost, and to facilitate various management aspects during the construction process, such as the construction period, construction costs, and material usage, building data is collected and a BIM model is constructed. To ensure the accuracy of the BIM model creation, the accuracy of the building data must be guaranteed. Building data comes from many sources and is complex, requiring long-term unified storage of various types of data. During the storage process, it is difficult to protect the building data securely, and erroneous data can lead to the creation of incorrect BIM models. This application, through the exchange cloud, can dynamically store building data, making it difficult for external parties to obtain, access, or tamper with the building data. This ensures the security and integrity of the building data storage and has a good data protection effect.
[0047] In one embodiment, the step S1 of collecting building data for the corresponding target building, storing the building data in the creation platform, and securely storing the building data through the exchange cloud in the creation platform includes:
[0048] S11. Determine the target building, set up a data acquisition port for the target building, and acquire the building data of the target building based on the data acquisition port.
[0049] S12. Transmit building data to the creation platform based on the acquisition port;
[0050] S13. In the creation platform, store the building data to the corresponding exchange cloud for secure storage.
[0051] In one embodiment, step S13, which involves securely storing building data in a corresponding exchange cloud within the creation platform, includes:
[0052] S131. In the creation platform, multiple switching clouds are set up, wherein the switching cloud includes a storage cell, two switching slices and multiple switching cells;
[0053] S132. Bind the corresponding collection ports of multiple exchange clouds;
[0054] S133. Store the building data in the swap cell in the bound swap cloud. Multiple swap cells circulate and exchange between two swap slices. Establish a tracking graph between the swap cell storing the building data and the storage cell. Establish the connection relationship between the authorized network and the swap cell based on the tracking graph.
[0055] In one embodiment, step S131 of setting up multiple exchange clouds in the creation platform includes:
[0056] S1311. In the creation platform, set up multiple storage units (data storage space, or storage units set up in the creation platform), and set up two switching pieces in the multiple storage units, wherein the switching piece is a combination of multiple data points;
[0057] S1312. Establish a connection channel between data points (data network points in the data space, existing as data ports, used to establish transmission channels) in two switching slices, and specify the transmission direction of multiple connection channels respectively;
[0058] S1313. Set up a switching element (virtual machine, used to carry subsequent building data, capable of securely storing building data and difficult to access) between the two switching segments. The switching element is transmitted and moved through the data points and connection channels in the two switching segments.
[0059] In one embodiment, step S133, which involves establishing a tracking graph between the exchange element storing building data and the storage element, and establishing the connection relationship between the authorized network and the exchange element based on the tracking graph, includes:
[0060] S1331. Divide the building data into multiple building data segments, randomly select multiple exchange elements corresponding to the number of building data segments, store the building data segments one-to-one into the exchange elements, and use the exchange element storing the building data segments as the target exchange element.
[0061] S1332. Set up a tracking graph in the storage cell (in the data space of the storage cell, set up a virtual route with the same connection structure as the switching cell, the virtual route is the data network built), set up multiple tracking points in the tracking graph (the tracking points are virtual machines, which can track the corresponding target switch to determine the position of the target switch, and the tracking points can simulate the target switch moving on the virtual route), the number of tracking points is the same as the number of target switching cells;
[0062] S1333, Bind the tracking points to the target exchange elements in a one-to-one correspondence;
[0063] S1334. The target switching element and the switching element are transmitted and moved between two switching segments through the connection channel. The position of the corresponding target switching element is tracked by the tracking point, and the position of the target switching element is marked on the tracking map. The corresponding authorized network is marked on the tracking map.
[0064] S1335. When the authorized network accesses the switching cloud through the output port on the storage cell, the location of the target switching cell is provided to the authorized network through the tracking graph, and the authorized network is connected to the target switching cell to obtain building data.
[0065] As described in steps S11-S13 above, after determining the target building, a corresponding acquisition port needs to be set up for it. This port is used for inputting building data. Through the acquisition port, the building data related to the target building can be uniformly collected and transmitted to the creation platform. There are multiple exchange clouds in the creation platform, and one exchange cloud corresponds to one target building. That is, one acquisition port stores the collected building data in the corresponding exchange cloud. There is a one-to-one correspondence between the exchange cloud and the acquisition port. An exchange cloud is a combination of storage units, two exchange slices, and multiple exchange units. The storage unit is the data storage space, which is the storage device set in the creation platform. The exchange slices here are part of the storage unit. In the data space, there are two switching segments within a storage cell. Multiple storage locations are determined within the data space of each switching segment, and data points are anchored at these locations. This results in multiple data points within a single switching segment. These data points are called data network points and are used to connect and transmit data to subsequent switching cells. For example, in two switching cells, one is switching cell A and the other is switching cell B. Switching cell A contains three data points: A1, A2, and A3; switching cell B contains three data points: B1, B2, and B3. The connection channels between the data points in the two switching cells can be established, for example, by arranging A1, A2, and A3 sequentially, and B1, B2, and B3 sequentially.From the perspective of switch A: A1 is connected to B1 and B2 respectively, A2 is connected to B1, B2 and B3 respectively, and A3 is connected to B2 and B3 respectively. This results in multiple connection channels. The transmission direction of each connection channel is specified. In the example above, the transmission direction of the connection channel is: B1-A1-B2-A2-B3-A3-B2-A2-B1, forming a closed loop. This allows the switch to continuously transmit and move data between the two switching segments, preventing access from external networks and enabling secure storage of building data in the storage unit. The building data is divided into multiple building data segments. These segments are then divided sequentially, and multiple exchange elements corresponding to the number of building data segments are randomly selected. Each exchange element stores a one-to-one sequence of building data segments. The exchange element records the order in which the building data segments were split within the building data. For example, if three building data segments are divided and ordered as j1, j2, and j3, the corresponding target exchange element records the sequence number of the stored building data segments. This sequence number is then provided to the access port for subsequent combination and reconstruction of the building data segments. The exchange element containing the building data segments... As the target switching unit, a tracking graph is set up in the storage unit to facilitate the acquisition of building data by the authorized network. The tracking graph contains multiple tracking points, which are virtual machines capable of tracking the corresponding target switch to determine its location. The tracking points simulate the movement of the target switch along a virtual route. The number of tracking points is the same as the number of target switching units, and a one-to-one correspondence is established between tracking points and target switching units. The tracking graph also serves as part of the data storage space within the storage unit. It sets up corresponding network routes for the relationships between two switching segments, switching units, and connection channels. The tracking points in the tracking graph correspond to the target switching units (one-to-one), while other switching units do not. The tracking points are used to track the location of the target switching units. To facilitate the acquisition of building data by the authorized network, the tracking graph records the authorized network. When the authorized network accesses the system, the tracking graph is provided to it, allowing it to determine which switching unit (target switching unit) the building data segment is located in. The authorized network then connects to the target switching unit and retrieves the building data segment. The building data segment is then combined according to the order in which the building data is split to obtain the building data. Here, the access port corresponding to the deep learning model is set as an authorized network. This way, the tracing graph provides the location of the target exchange element to the access port. By connecting to the target exchange element through the access port to obtain building data, unauthorized networks can be prevented from accessing and tampering with the building data, providing good data protection and ensuring the accuracy of the building data. Only with accurate building data can an accurate BIM model be created.
[0066] In one embodiment, step S2, which involves processing building data based on a deep learning model in a creation platform to obtain building element information, includes:
[0067] S21. Train the convolutional neural network using training data, which includes training building data and corresponding building element information, to obtain a trained deep learning model.
[0068] S22. Configure the access port (authorized network) for the corresponding deep learning model. The deep learning model requests access to the exchange cloud through the access port. The tracking graph provides the location of the target exchange element to the access port. The building data is obtained by connecting with the target exchange element through the access port. Based on the access port, the building data is input into the deep learning model to obtain building element information.
[0069] As described in steps S21 and S22 above, the training data consists of a large amount of building data and corresponding building element information obtained through big data analytics. The building data includes design drawings, construction drawings, material specifications, and bills of quantities. For example, design drawings (such as floor plans, elevations, and sections) mainly describe the building's geometric shape and spatial layout. The corresponding building element information includes: geometric attributes: shape, dimensions (such as wall length, thickness, and height); spatial location (such as coordinates and elevation); topological relationships (such as the connection between walls and columns, and the nesting of doors and windows with walls); semantic attributes: element type (such as load-bearing walls, non-load-bearing walls, and fire doors); material information (such as concrete strength grade and brick type); functional attributes: usage scenarios (such as office areas and restrooms); and design specifications (such as minimum clear height and evacuation width). Image or point cloud data containing building elements (such as building floor plans, elevations, and 3D scan data) are collected and labeled with the corresponding building element information (such as walls, doors, windows, beams, and columns). For example, in a floor plan, the architectural element information of a wall might include: Geometric: Length 5 meters, height 3 meters, thickness 240 millimeters. Semantic: Type: "240mm thick aerated concrete block wall", material: "A5.0 aerated concrete". Function: Partition wall belonging to the "office area", fire rating: "Level II". Then, a convolutional neural network can be trained using training data to obtain a deep learning model; subsequently, the collected architectural data is input into the deep learning model for analysis to obtain the architectural element information.
[0070] In one embodiment, step S3, which involves inputting building element information into BIM software to create a BIM model, includes:
[0071] S31. Transfer building element information from the deep learning model to the BIM software via API interface;
[0072] S32, BIM software automatically generates models based on received data and instructions.
[0073] As described in steps S31 and S32 above, an API interface is designed, including input parameters (such as the output data of the deep learning model) and output parameters (such as data formats that the BIM software can recognize). Plugins or scripts are developed using the API provided by the BIM software to receive data transmitted through the API interface and automatically generate models. Data mapping and conversion logic is implemented in the BIM plugins or scripts to map the output data of the deep learning model to the corresponding elements and attributes in the BIM software. Based on the mapped data, a building model is automatically generated in the BIM software, thus completing the creation of the BIM model. This allows for the creation of a model of the target building, facilitating better analysis. The 3D visualized model enables better planning of building construction, effective planning and control of progress and construction costs, and effective comprehensive building management.
[0074] Example 2, please refer to Figure 2 As shown in this embodiment, the AI-based intelligent BIM model creation system includes:
[0075] The storage module is used to collect building data for the target building and store the building data in the creation platform. The building data is securely stored through the exchange cloud in the creation platform.
[0076] The data processing module, connected to the storage module, is used to process building data based on a deep learning model in the creation platform to obtain building element information;
[0077] The creation module, connected to the data processing module, is used to input building element information into the BIM software to create a BIM model.
[0078] It should be noted that the exchange cloud enables dynamic storage of building data, making it difficult for outsiders to obtain, access, or tamper with the data. This ensures the security and integrity of the stored building data and provides good data protection.
[0079] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
Claims
1. A method for intelligent creation of BIM models based on AI algorithms, characterized in that, The following steps are involved: Building data is collected for the target building and stored in the creation platform. The building data is then securely stored through the exchange cloud within the creation platform. The platform is designed to process building data using deep learning models to obtain building element information. Input building element information into BIM software to create a BIM model.
2. The intelligent BIM model creation method based on AI algorithm according to claim 1, characterized in that, The steps of collecting building data for the corresponding target building, storing the building data in the creation platform, and securely storing the building data through the exchange cloud in the creation platform include: Identify the target building, set up a data acquisition port for the target building, and acquire the building data of the target building based on the data acquisition port. Building data is transmitted to the creation platform via the acquisition port; The building data is stored securely in the corresponding exchange cloud during the platform creation process.
3. The intelligent BIM model creation method based on AI algorithm according to claim 2, characterized in that, The step of securely storing building data in the corresponding exchange cloud within the creation platform includes: Multiple switching clouds are set up in the platform creation, where each switching cloud includes a storage cell, two switching shards, and multiple switching cells; Bind multiple exchange clouds to their corresponding collection ports; Building data is stored in exchange cells within a bound exchange cloud. Multiple exchange cells circulate and exchange data between two exchange slices. A tracking graph is established between the exchange cells storing building data and the storage cells. Based on the tracking graph, a connection relationship is established between the authorized network and the exchange cells.
4. The intelligent BIM model creation method based on AI algorithm according to claim 3, characterized in that, The steps for setting up multiple exchange clouds in the creation platform include: In the creation platform, multiple storage cells are set up, and two switching slices are set up in the multiple storage cells. The switching slice is composed of multiple data points. Establish connection channels between data points in two switching slices, and specify the transmission direction of multiple connection channels respectively; A switching element is set up between two switching segments, and the switching element transmits and moves data through the data points in the two switching segments and the connection channel.
5. The intelligent BIM model creation method based on AI algorithm according to claim 4, characterized in that, The step of establishing a tracking graph between the exchange element storing building data and the storage element, and establishing the connection relationship between the authorized network and the exchange element based on the tracking graph, includes: The building data is divided into multiple building data segments. Multiple exchange elements corresponding to the number of building data segments are randomly selected. The building data segments are stored one-to-one in the exchange elements. The exchange element containing the building data segments is used as the target exchange element. Set up a tracking graph in the storage element, and set up multiple tracking points in the tracking graph. The number of tracking points is the same as the number of target swap elements. Bind the tracking points to the target exchange elements in a one-to-one correspondence; The target switching element moves between two switching segments through the connection channel. The position of the corresponding target switching element is tracked by the tracking point, and the position of the target switching element is marked on the tracking map. The corresponding authorized network is marked on the tracking map. When the authorized network accesses the switching cloud through the output port on the storage cell, the location of the target switching cell is provided to the authorized network through the tracking graph, and the authorized network is connected to the target switching cell to obtain building data.
6. The intelligent BIM model creation method based on AI algorithm according to claim 5, characterized in that, The steps for processing building data based on a deep learning model to obtain building element information in the creation platform include: The convolutional neural network is trained using training data, which includes training building data and corresponding building element information, to obtain a trained deep learning model. The deep learning model is configured with an access port. The deep learning model requests access to the exchange cloud through the access port. The tracking graph provides the location of the target exchange element to the access port. The model obtains building data by connecting to the target exchange element through the access port. Based on the access port, the building data is input into the deep learning model to obtain building element information.
7. The intelligent BIM model creation method based on AI algorithm according to claim 1, characterized in that, The steps of inputting building element information into BIM software to create a BIM model include: Building element information is transferred from the deep learning model to the BIM software via an API interface; BIM software automatically generates models based on received data and instructions.
8. An AI-based intelligent BIM model creation system, used to implement the AI-based intelligent BIM model creation method according to any one of claims 1-7, characterized in that, include: The storage module is used to collect building data for the target building and store the building data in the creation platform. The building data is securely stored through the exchange cloud in the creation platform. The data processing module, connected to the storage module, is used to process building data based on a deep learning model in the creation platform to obtain building element information; The creation module, connected to the data processing module, is used to input building element information into the BIM software to create a BIM model.
9. An intelligent BIM model creation device based on AI algorithms, characterized in that, include: one or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the steps of the AI-based intelligent BIM model creation method as described in any one of claims 1-7.
10. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the intelligent creation method for BIM models based on AI algorithms as described in any one of claims 1-7.
Citation Information
Cited By
Cloud platform data remote backup method
CN121173595A