A method for building a digital twin visualization screen

Abstract

This invention discloses a method for building a digital twin visualization screen. By extracting and classifying family components from the BIM model, it achieves effective data organization and management. It establishes a link between the backend marker configuration library and the family library, enhancing data traceability and consistency. Combining the BIM model with IoT data, it uses GUID codes and device ID codes to achieve precise association, ensuring data integrity and accuracy. The UE engine is used to transform the BIM model into a highly realistic digital twin model. An interactive interface enhances user customization capabilities and ease of operation. Data interaction between the IoT platform and the WEB visualization frontend enables real-time updates and dynamic display, improving system usability and user experience. The BIM model undergoes lightweight transformation in the IoT platform, improving data processing efficiency and response speed. Lightweight mobile terminals for IoT and BIM enable automated data collection and storage. The backend IoT database integrates complete data, providing support for intelligent analysis and decision-making.

Description

Technical Field

[0001] This invention relates to the field of digital twin visualization technology, and in particular to a method for building a digital twin visualization large screen. Background Technology

[0002] With the rapid development of information technology and the deepening of digital transformation, digital twin technology, as an important technical means, has been widely used in various fields. By establishing virtual models of physical objects, digital twin technology enables real-time monitoring, simulation, and optimization of the operational status of physical objects, greatly improving management efficiency and decision-making accuracy. In the construction industry, BIM (Building Information Modeling) serves as an important foundation for digital twin technology, providing rich building information data and strong support for the construction of digital twin models.

[0003] However, traditional BIM models have some problems in application. First, BIM models often contain a large amount of data and information, which can become complex and cumbersome to process and display, making it difficult to quickly and accurately obtain the required information. Second, the integration of BIM models with IoT technology is not close enough, and there are obstacles to the data collection of smart devices and the association with BIM models, resulting in serious data silos and failing to fully realize the value of the data. Finally, the visualization methods of BIM models are relatively simple, lacking intuitive and highly interactive large-screen visualization methods, making it difficult to meet the information visualization needs of modern management. Summary of the Invention

[0004] In view of this, the present invention proposes a method for building a digital twin visualization screen, which can solve the shortcomings of existing technologies, such as difficulty in quickly and accurately obtaining the required information, serious data silos, and relatively simple visualization display methods.

[0005] The technical solution of this invention is implemented as follows:

[0006] A method for building a digital twin visualization dashboard, specifically including:

[0007] Obtain the BIM model;

[0008] Extract family components from the BIM model and classify and store them according to classification criteria;

[0009] Establish the association between the background tag configuration library and the background family library;

[0010] Data is acquired and organized from the BIM model through the Revit interactive interface, and then written into the BIM model and the backend database respectively.

[0011] Import the BIM model into the IoT platform for lightweight conversion, and transfer the model data to the lightweight BIM model along with the GUID code.

[0012] Using IoT and BIM lightweight mobile terminals, IoT data is collected on-site from family components of the BIM lightweight model corresponding to smart devices, and the data is written to the background IoT database.

[0013] The backend IoT database uses GUID codes and device ID codes to associate model data and IoT data, thus integrating complete data.

[0014] BIM models and IoT platforms obtain complete data through the backend IoT database;

[0015] The BIM model is converted into a digital twin model supported by the UE engine, and the complete data is transferred to the digital twin model along with the GUID code and the device ID code;

[0016] Configure scenarios for the digital twin model through its interactive interface;

[0017] The IoT middleware platform interacts with the web-based visual front-end via GUID codes and device ID codes;

[0018] The digital twin model and the web visualization front end are linked through GUID code and device ID code to realize the construction of digital twin visualization screen.

[0019] As a further optional solution to the aforementioned method for building a digital twin visualization screen, the method also includes automatic updating of the digital twin model and data, specifically:

[0020] During the on-site IoT data acquisition process, if it is found that the on-site equipment does not match the model, the flag configuration corresponding to the family component is invoked and the flag is placed at the corresponding position;

[0021] IoT and BIM lightweight mobile devices acquire location data of marked configurations on lightweight models;

[0022] Add or delete information based on the on-site conditions of the equipment, collect IoT data, and write the tagged data and IoT data into the background IoT database;

[0023] The BIM model side reads the location data and marking information of the marked configuration in the database, as well as the device ID code in the IoT data, through the Revit interactive interface;

[0024] The Revit interactive interface automatically calls the corresponding family components in the background family library based on the association of the tag configuration, and executes the add or delete commands based on the configuration location data and tag information obtained from the database.

[0025] After completing the instructions to add or delete family components, the BIM model is used to acquire data and extract and organize the equipment IDs in the configuration, and write them into the BIM model and the background model database respectively.

[0026] The backend database then uses GUID and device ID codes to link the model data and IoT data, integrating the complete data.

[0027] Update the digital twin model and data based on complete data.

[0028] As a further optional solution to the aforementioned digital twin visualization screen construction method, the background family library is used for the storage and classification management of BIM model component families, including a standard family library and a project family library. The standard family library is a stable standard library formed by sorting out equipment requirements according to the functional indicators required by the Internet of Things platform and creating family components. The project family library is for a single project, extracting family components from the project to form a project family library.

[0029] As a further optional solution to the aforementioned digital twin visualization screen construction method, the background tag configuration library is used for the storage and classification management of tag configurations, including a standard tag configuration library and custom tag configurations. The standard tag configuration library is configured with different forms of tag configurations to form a stable standard tag configuration library according to the component classification requirements in the standard family library of the background family library. The custom tag configuration is used to customize the family component classification in the BIM model when there is a family component classification that does not correspond to the standard tag configuration.

[0030] As a further optional solution to the aforementioned digital twin visualization screen construction method, the backend database is used to store relevant data of family components in the IoT platform and BIM model, including a model database and an IoT database. The model database stores relevant data of family components in the BIM model, and the IoT database stores relevant data of IoT device products.

[0031] As a further alternative to the aforementioned digital twin visualization screen construction method, convolutional neural networks and natural language processing techniques are used to perform semantic parsing on the extracted static features of family components, and the semantic parsing results are used as a training set to input into a deep learning model for training.

[0032] As a further optional solution to the aforementioned digital twin visualization screen construction method, intelligent association between the background tag configuration library and the family library is carried out based on the organic combination of deep learning, reinforcement learning, and distributed architecture.

[0033] As a further optional solution to the aforementioned digital twin visualization screen construction method, on-site data is continuously monitored, automatically judged, and automatically updated based on visual marker enhancement technology, multimodal data matching model, and automatic repair and marker update technology.

[0034] A computing device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the digital twin visualization large screen construction method described above.

[0035] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the digital twin visualization screen construction method described above.

[0036] The beneficial effects of this invention are as follows: By extracting family components from the BIM model and classifying and storing them, effective organization and management of BIM model data are achieved; the association between the backend tag configuration library and the backend family library is established, enhancing data traceability and consistency; the BIM model is combined with IoT data, and precise association between model data and IoT data is achieved through GUID codes and device ID codes, ensuring data integrity and accuracy; the BIM model is transformed into a digital twin model using the UE engine, improving the model's realism and interactivity, enabling the visualization screen to display building information more intuitively; scene configuration of the digital twin model is achieved through the digital twin model's interactive interface, enhancing user customization capabilities and ease of operation; data interaction between the IoT middleware and the WEB visualization frontend, as well as the interaction between the digital twin model and the WEB visualization frontend, are also addressed. The interconnected EB visualization front-end enables real-time data updates and dynamic display, enhancing system usability and user experience. Importing BIM models into the IoT platform for lightweight transformation reduces model complexity and data volume, improving data processing efficiency and system response speed. Lightweight model data is transmitted to the lightweight BIM model via GUID codes, ensuring data integrity and consistency during transmission while reducing data processing difficulty and cost. IoT and lightweight BIM mobile terminals collect data from smart devices and write it to the backend IoT database, achieving automated data collection and storage. The backend IoT database uses GUID codes and device ID codes to link model data and IoT data, integrating complete data and providing strong support for intelligent analysis and decision-making. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1This is a flowchart illustrating the steps of a method for building a digital twin visualization screen, as provided in an embodiment of the present invention. Detailed Implementation

[0039] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. 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.

[0040] refer to Figure 1 A method for building a digital twin visualization dashboard, specifically including:

[0041] Obtain the BIM model;

[0042] Extract family components from the BIM model and classify and store them according to classification criteria;

[0043] Establish the association between the background tag configuration library and the background family library;

[0044] Data is acquired and organized from the BIM model through the Revit interactive interface, and then written into the BIM model and the backend database respectively.

[0045] Import the BIM model into the IoT platform for lightweight conversion, and transfer the model data to the lightweight BIM model along with the GUID code.

[0046] Using IoT and BIM lightweight mobile terminals, IoT data is collected on-site from family components of the BIM lightweight model corresponding to smart devices, and the data is written to the background IoT database.

[0047] The backend IoT database uses GUID codes and device ID codes to associate model data and IoT data, thus integrating complete data.

[0048] BIM models and IoT platforms obtain complete data through the backend IoT database;

[0049] The BIM model is converted into a digital twin model supported by the UE engine, and the complete data is transferred to the digital twin model along with the GUID code and the device ID code;

[0050] Configure scenarios for the digital twin model through its interactive interface;

[0051] The IoT middleware platform interacts with the web-based visual front-end via GUID codes and device ID codes;

[0052] The digital twin model and the web visualization front end are linked through GUID code and device ID code to realize the construction of digital twin visualization screen.

[0053] In this embodiment, by extracting and classifying family components from the BIM model, effective organization and management of BIM model data are achieved. The association between the backend tag configuration library and the backend family library is established, enhancing data traceability and consistency. The BIM model is combined with IoT data, and precise association between model data and IoT data is achieved through GUID codes and device ID codes, ensuring data integrity and accuracy. The UE engine is used to transform the BIM model into a digital twin model, improving the model's realism and interactivity, enabling the visualization screen to display building information more intuitively. Scene configuration of the digital twin model is achieved through the digital twin model's interactive interface, enhancing user customization capabilities and ease of operation. Data interaction between the IoT middleware and the WEB visualization frontend, as well as the interaction between the digital twin model and the WEB visualization frontend, are also addressed. The BIM visualization front-end enables real-time data updates and dynamic display, enhancing system usability and user experience. Importing the BIM model into the IoT platform for lightweight conversion reduces model complexity and data volume, improving data processing efficiency and system response speed. Lightweight model data is transmitted to the lightweight BIM model via GUID codes, ensuring data integrity and consistency during transmission while reducing data processing difficulty and cost. IoT and lightweight BIM mobile terminals collect data from smart devices and write it to the backend IoT database, achieving automated data collection and storage. The backend IoT database uses GUID codes and device ID codes to link model data and IoT data, integrating complete data and providing strong support for intelligent analysis and decision-making.

[0054] Preferably, the method further includes a digital twin model and automatic data updates, specifically:

[0055] During the on-site IoT data acquisition process, if it is found that the on-site equipment does not match the model, the flag configuration corresponding to the family component is invoked and the flag is placed at the corresponding position;

[0056] IoT and BIM lightweight mobile devices acquire location data of marked configurations on lightweight models;

[0057] Add or delete information based on the on-site conditions of the equipment, collect IoT data, and write the tagged data and IoT data into the background IoT database;

[0058] The BIM model side reads the location data and marking information of the marked configuration in the database, as well as the device ID code in the IoT data, through the Revit interactive interface;

[0059] The Revit interactive interface automatically calls the corresponding family components in the background family library based on the association of the tag configuration, and executes the add or delete commands based on the configuration location data and tag information obtained from the database.

[0060] After completing the instructions to add or delete family components, the BIM model is used to acquire data and extract and organize the equipment IDs in the configuration, and write them into the BIM model and the background model database respectively.

[0061] The backend database then uses GUID and device ID codes to link the model data and IoT data, integrating the complete data.

[0062] Update the digital twin model and data based on complete data.

[0063] In this embodiment, during on-site IoT data acquisition, once a discrepancy is detected between the on-site equipment and the model, the system can immediately call the tag configuration corresponding to the family component and place a tag at the corresponding location. This mechanism ensures timely detection and tagging of problems, providing accurate location information for subsequent data updates. By acquiring the location data of the tag configuration on the lightweight model through IoT and BIM lightweight mobile terminals, the system can accurately determine the location and range that needs to be updated, thereby improving the accuracy and efficiency of data updates. The system can automatically add or delete information based on the on-site equipment conditions and perform IoT data acquisition, writing the tag data and IoT data into the background IoT database. This automated process reduces manual intervention and improves the efficiency and accuracy of data updates. The Revit interactive terminal can automatically call the corresponding family component in the background family library based on the association of the tag configuration and, based on the group data obtained from the database... By acquiring location data and tagging information, the system executes add or delete commands, greatly simplifying the model update process and reducing operational difficulty. After completing the add or delete command for family components, the system acquires data from the BIM model and extracts and organizes the equipment IDs in the tagging configuration, writing them into both the BIM model and the backend model database. This step ensures data consistency between the BIM model and the backend database, avoiding data conflicts and redundancy. The backend database then uses GUID codes and equipment ID codes to link the model data and IoT data, integrating complete data. This mechanism ensures data integrity and accuracy, providing reliable data support for updating the digital twin model. Updating the digital twin model and data based on complete data allows the digital twin model to reflect the actual situation of the equipment on site in real time. This dynamic update mechanism improves the practicality and accuracy of the digital twin model.

[0064] Preferably, the background family library is used for the storage and classification management of BIM model component families, including a standard family library and a project family library. The standard family library is a stable standard library formed by sorting out equipment requirements according to the functional indicators required by the Internet of Things platform and creating family components. The project family library is for a single project, extracting family components from the project to form a project family library.

[0065] In this embodiment, the standard family library organizes device requirements based on the functional indicators needed by the IoT platform, and creates family components to form a stable standard library. This design ensures the standardization and normalization of family components, avoiding inconsistencies caused by different creators or project requirements. The standard family library unifies the expression of device information, improving data consistency and comparability. The project family library, on the other hand, is created by extracting family components from a single project. This design allows the project family library to flexibly adapt to the needs of different projects, avoiding inapplicability issues due to project differences. Furthermore, the project family library can be updated as projects progress. The library is expanded and updated to reflect changes in progress and requirements, ensuring its timeliness and accuracy. The backend library stores and categorizes family components, allowing users to quickly retrieve the required components, improving work efficiency. Categorization allows users to clearly understand the type, attributes, and uses of family components, avoiding misuse or omissions due to information confusion. The design of the standard and project family libraries facilitates smoother data integration between the BIM model and the IoT platform. The standard family library provides a unified data interface and expression method for the IoT platform, enabling it to accurately identify and parse equipment information in the BIM model.

[0066] Preferably, the background tag configuration library is used for the storage and classification management of tag configurations, including a standard tag configuration library and custom tag configurations. The standard tag configuration library is a stable standard tag configuration library with different forms of tag configurations set according to the component classification requirements in the standard family library of the background family library. The custom tag configuration is used to customize the family component classification in the BIM model when there is a family component classification that does not correspond to the standard tag configuration.

[0067] In this embodiment, the standard tag configuration library sets up different forms of tag configurations according to the component classification requirements in the standard family library of the background family library, forming a stable standard tag configuration library. This design ensures the standardization and consistency of tag configurations, enabling different projects or different personnel to follow the same specifications and standards when using them. Through the standard tag configuration library, the expression method and purpose of tags can be unified, improving data consistency and comparability. The introduction of custom tag configurations allows for the use of custom tags when family components in the BIM model have a classification of family components that do not correspond to the standard tag configuration. This design improves the flexibility and scalability of tag configurations, enabling tag configurations to adapt to different projects or different structures. To meet the specific needs of components, custom mark configurations can be created and modified according to actual requirements, ensuring the timeliness and accuracy of mark configurations. The backend mark configuration library stores and categorizes mark configurations, allowing users to quickly retrieve the required mark configurations, improving work efficiency. Through categorization management, users can clearly understand the type, form, and purpose of mark configurations, avoiding misuse or omissions caused by information confusion. The design of the standard mark configuration library and custom mark configurations makes data integration between the BIM model and the backend database smoother. The standard mark configuration library provides a unified data interface and expression method for the backend database, enabling the backend database to accurately identify and parse mark information in the BIM model.

[0068] Preferably, the background database is used to store relevant data of family components in the IoT platform and BIM model, including a model database and an IoT database. The model database stores relevant data of family components in the BIM model, and the IoT database stores relevant data of IoT device products.

[0069] In this embodiment, by dividing the backend database into a model database and an IoT database, separate data storage is achieved. The model database focuses on storing data related to family components in the BIM model, while the IoT database focuses on storing data related to IoT device products. This separation makes the data clearer and easier to manage and maintain. The separate design of the model database and the IoT database ensures the consistency and integrity of their respective data. Storing family component data in the BIM model and IoT device product data separately avoids data overlap and confusion, thereby ensuring data accuracy and reliability. The separate database design makes data access and processing more efficient, specifically for BIM models and IoT device products. To meet the different needs of different products, the storage structure and query performance of the model database and IoT database can be optimized separately, improving the speed and efficiency of data access. By storing data separately, data access permissions and security policies can be set more flexibly. For sensitive data in the BIM model, stricter security measures can be taken to protect it, while for data of IoT devices, appropriate access permissions can be set according to actual needs to ensure data security and privacy protection. Although the data is stored separately, the model database and IoT database can be integrated and work together through data interfaces or middleware. This design enables data between BIM models and IoT devices to be interconnected and interact.

[0070] Preferably, convolutional neural networks and natural language processing techniques are used to perform semantic parsing on the extracted static features of family components, and the semantic parsing results are used as the training set to input the deep learning model for training.

[0071] In this embodiment, by fusing convolutional neural networks (CNN) and natural language processing (NLP) techniques, static features of family components are extracted, and textual data such as device names and model descriptions are extracted for semantic parsing. The semantic parsing results are used as a training set to input into a deep learning model for training, thereby improving the matching efficiency and accuracy of the system. The specific steps are as follows:

[0072] Collect and import datasets: Collect images related to smart device family components based on the backend standard family library; collect images related to various labeled configurations based on the standard labeled configuration library; collect text data such as device names and model descriptions of family components based on classification criteria; import the datasets into image training sets and text training sets.

[0073] Data preprocessing: Pixel values ​​are standardized for the image training set data, and punctuation, word segmentation, and stop word removal are performed on the text training set data.

[0074] Model building involves extracting features from the data, classifying the data, and compiling the model.

[0075] Training and evaluating the model: Input the prepared training set data (including images and corresponding labels) and test set data (including images and corresponding labels) to train the model. The smaller the difference between the model's predicted values ​​and the true values, the higher the model's accuracy.

[0076] In addition, the specific steps for building the model are as follows:

[0077] Feature extraction: Static features of equipment such as geometric features, location information, and installation methods are extracted from images of equipment family components and marked configurations in the BIM model using convolutional neural networks. Natural language processing technology is used to perform semantic parsing on text data such as equipment names and model descriptions, and the embedded vectorization method (BERT model) is used to uniformly map them to a standardized data space.

[0078] Classification is achieved using softmax logistic regression.

[0079] Model compilation: IT compiles the model, selects the loss function, optimizer, metrics, etc., and analyzes and evaluates the model.

[0080] Preferably, the background tagging configuration library and the family library are intelligently associated based on the organic combination of deep learning, reinforcement learning and distributed architecture.

[0081] In this embodiment, through the organic combination of deep learning, reinforcement learning, and distributed architecture, a technological innovation from traditional static matching to dynamic adaptive association is achieved in the intelligent association between the tag configuration library and the family library. This avoids manual intervention and improves the efficiency of data matching and updating. The specific steps are as follows:

[0082] Extracting Project Model Family Data: Extracting family components from the project model, and extracting project model family data such as images and text data of the families according to the model requirements;

[0083] Data preprocessing: Combining automated data cleaning algorithms, the project model family data is denoised, outlier handled, and formatted. If the data missing rate exceeds the set threshold, an anomaly report is triggered and a re-collection is requested. If data is detected to be outside the reasonable range (by setting upper and lower limits), it is marked as invalid data and replaced or removed to ensure the integrity and consistency of the input data.

[0084] Importing data into the CNN+NLP model: Import the preprocessed data into the model through the program interface;

[0085] Setting an error threshold: The accuracy of the output results can be ensured by designing an error threshold for the system. However, sometimes the error threshold is too high, which will affect the system's operating efficiency. Therefore, it is necessary to select a reasonable error threshold to ensure the accuracy within a reasonable range while maximizing the system's operating efficiency.

[0086] Error determination: The Deep Q-Learning reinforcement learning algorithm is used to train the AI ​​system to learn device characteristics and historical matching rules. If the matching confidence is lower than the threshold (less than 80%), manual review and model parameter optimization are triggered. Families that meet the error threshold are automatically archived according to the classification criteria; families that do not meet the criteria are manually modified and archived. Finally, the identification, correction, classification, and storage of families are completed in the project family library.

[0087] Matching Decision: The CNN+NLP model performs label configuration matching decision on the project family library organized according to the classification criteria. If the decision condition is met, the system obtains the corresponding label configuration from the configuration library and automatically associates it. If the decision condition is not met, it means that the device type does not exist in the existing standard label library, triggering the custom label generation process. The system combines image recognition and pattern recognition technology to generate personalized labels for device family components, automatically store them in the configuration library, and automatically associate them.

[0088] Preferably, the field data is continuously monitored, automatically judged, and automatically updated based on visual marker enhancement technology, multimodal data matching model, and automatic repair and marker update technology.

[0089] In this embodiment, the visual marker enhancement technology uses on-site images captured by the device's camera and AR device to identify the device through deep learning image processing algorithms YOLO and Mask R-CNN. If the detected device does not match the BIM model marker, a marker repair process between the image and the model is triggered to dynamically associate the detected device marker with the marker in the BIM model.

[0090] Multimodal data matching model: By integrating real-time IoT data (device ID, operating status, etc.) with static BIM model data (location, geometric features, etc.), a multimodal association model is constructed. If there is a conflict between the matched static and dynamic data (such as the same device in different locations), a conflict resolution mechanism is triggered to achieve accurate matching between the tag configuration and the family library.

[0091] Automatic repair and marker update: The system can automatically detect inconsistencies between on-site equipment and the BIM model, predict equipment status through AI and dynamically generate or correct markers. If the inconsistency rate exceeds the normal range (e.g., more than 10%), a large-scale marker repair process is triggered. It also supports marker extension and custom markers in complex scenarios.

[0092] A computing device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the digital twin visualization large screen construction method described above.

[0093] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the digital twin visualization screen construction method described above.

[0094] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for constructing a digital twin visualization large screen, characterized in that, Specifically, it includes: Obtain the BIM model; Extract family components from the BIM model and classify and store them according to classification criteria; Establish the association between the background tag configuration library and the background family library; Data is acquired and organized from the BIM model through the Revit interactive interface, and then written into the BIM model and the backend database respectively. Import the BIM model into the IoT platform for lightweight conversion, and transfer the model data to the lightweight BIM model along with the GUID code. Using IoT and BIM lightweight mobile terminals, IoT data is collected on-site from family components of the BIM lightweight model corresponding to smart devices, and the data is written to the background IoT database. The backend IoT database uses GUID codes and device ID codes to associate model data and IoT data, thus integrating complete data. BIM models and IoT platforms obtain complete data through the backend IoT database; The BIM model is converted into a digital twin model supported by the UE engine, and the complete data is transferred to the digital twin model along with the GUID code and the device ID code; Configure scenarios for the digital twin model through its interactive interface; The IoT middleware platform interacts with the web-based visual front-end via GUID codes and device ID codes; The digital twin model and the web visualization front end are linked through GUID code and device ID code to realize the construction of a digital twin visualization screen; The method also includes a digital twin model and automatic data updates, specifically: During the on-site IoT data acquisition process, if it is found that the on-site equipment does not match the model, the flag configuration corresponding to the family component is invoked and the flag is placed at the corresponding position; IoT and BIM lightweight mobile devices acquire location data of marked configurations on lightweight models; Add or delete information based on the on-site conditions of the equipment, collect IoT data, and write the tagged data and IoT data into the background IoT database; The BIM model side reads the location data and marking information of the marked configuration and the device ID code in the IoT data from the background IoT database through the Revit interactive terminal; The Revit interactive interface automatically calls the corresponding family components in the background family library based on the association of the tag configuration, and executes the add or delete commands based on the configuration location data and tag information obtained from the database. After completing the instructions to add or delete family components, the BIM model is used to acquire data and extract and organize the equipment IDs in the configuration, and write them into the BIM model and model database respectively. The backend database then uses GUID and device ID codes to link the model data and IoT data, integrating the complete data. Update the digital twin model and data based on complete data.

2. The method for constructing a digital twin visualization large screen according to claim 1, characterized in that, The background family library is used for the storage and classification management of BIM model component families, including a standard family library and a project family library. The standard family library is a stable standard library formed by sorting out equipment requirements according to the functional indicators required by the Internet of Things platform and creating family components. The project family library is for a single project, extracting family components from the project to form a project family library.

3. The method for constructing a digital twin visualization large screen according to claim 2, characterized in that, The background tag configuration library is used for the storage and classification management of tag configurations. It includes a standard tag configuration library and custom tag configurations. The standard tag configuration library is a stable standard tag configuration library with different forms of tag configurations set according to the component classification requirements in the standard family library of the background family library. The custom tag configuration is used to customize the family component classification in the BIM model when there is a family component classification that does not correspond to the standard tag configuration.

4. The method for constructing a digital twin visualization large screen according to claim 3, characterized in that, The backend database is used to store relevant data of family components in the IoT platform and BIM model, including a model database and an IoT database. The model database stores relevant data of family components in the BIM model, and the IoT database stores relevant data of IoT device products.

5. The method for constructing a digital twin visualization large screen according to claim 4, characterized in that, Convolutional neural networks and natural language processing techniques are used to perform semantic parsing on the extracted static features of family components, and the semantic parsing results are used as the training set to train the deep learning model.

6. The method for constructing a digital twin visualization large screen according to claim 5, characterized in that, Based on the organic combination of deep learning, reinforcement learning, and distributed architecture, intelligent association is performed between the background tag configuration library and the family library.

7. The method for constructing a digital twin visualization large screen according to claim 6, characterized in that, Based on visual marker enhancement technology, multimodal data matching model, and automatic repair and marker update technology, the field data is continuously monitored, automatically judged, and automatically updated.

8. A computing device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the digital twin visualization large screen construction method according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the digital twin visualization screen construction method according to any one of claims 1-7.

Images