Low-code digital twin platform development method based on anchor point mechanism

Abstract

The invention discloses a low-code digital twin platform development method based on an anchor point mechanism, and the method comprises the steps: enabling an interaction point in a three-dimensional model to be dynamically associated with the data of the Internet of Things through introducing an anchor point abstraction mechanism, and achieving the condition-driven dynamic effect display in combination with a low-code configuration interface; according to the method, the problems of large code writing amount, difficulty in data and visualization integration and complexity in dynamic effect implementation in application development of a traditional digital twin platform are solved; a user configures anchor points through a graphical interface, the development efficiency is improved by more than 80%, and interaction codes do not need to be written; a dynamic effect engine arranged in the platform is provided with flicker, color change, light ring and other effect templates, condition triggering is supported through setting of a mapping rule, and the dynamic effect implementation cost is reduced; the ThingsBoard data and the three-dimensional model are seamlessly connected through an anchor point mechanism, and data-driven visualization is achieved.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and more specifically to a digital twin platform development method based on an anchor point mechanism. Background Technology

[0002] Current digital twin platforms suffer from the following pain points: 1. High development complexity: Traditional 3D visualization (such as Three.js) requires writing a large amount of code to implement model interaction and data binding, resulting in a high development threshold; 2. Disconnect between data and visualization: The integration of 3D models and IoT device data (such as ThingsBoard) relies on customized development and lacks standardized configuration methods; 3. Lack of low-code tools: Existing platforms make it difficult to directly define model behavior (such as dynamic effects) through a graphical interface, leading to low iteration efficiency; 4. Difficulty in implementing dynamic effects: Common digital twin effects such as flashing, color changes, and halo diffusion require manual coding and cannot be quickly adapted to business logic through configuration. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide a low-code digital twin platform development method based on the anchor point mechanism, which can solve the problems of large amount of code writing, difficulty in data and visualization integration, and complexity in the implementation of dynamic effects in traditional digital twin platform application development.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows.

[0005] The low-code digital twin platform development method based on the anchor point mechanism introduces an anchor point abstraction mechanism to dynamically associate interactive points in the 3D model with IoT data, and combines this with a low-code configuration interface to achieve condition-driven dynamic effect display. Specifically, it includes the following steps: S1. 3D Model Import and Parsing: Users import 3D model files into the 3D editor of the digital twin model. The platform automatically parses the model structure and extracts all Mesh objects and their hierarchical relationships. S2. Anchor point definition and type specification: In the design interface of the 3D editor, the user interactively selects the Mesh object in the model, marks it as an anchor point, and defines the anchor point type through the enumeration system. S3. Data source configuration and binding: In the platform's low-code configuration interface, users select a data source for the anchor point and bind the anchor point to the data source, generating mapping rules that trigger dynamic effect changes. S4. Real-time data association establishment and real-time synchronization of dynamic effects: Establish a real-time connection between anchor points and data sources, and realize dynamic effect display based on mapping rules and the cooperation of the dynamic effect engine; S5. Verification and Testing: The platform provides a simulated data mode, allowing users to input test values ​​to verify the anchor point's response capability.

[0006] The above-described low-code digital twin platform development method based on the anchor point mechanism includes the following steps in step S1: Model structure parsing. S11. File Format Recognition and Loader Selection: The system first automatically selects the corresponding Three.js loader based on the file extension. Each loader is responsible for converting the 3D model in binary or text format into a scene graph structure that can be recognized by Three.js. S12, Scene Graph Traversal and Mesh Extraction: After loading, the platform will receive a complete Scene object. At this time, the system will access each node in the scene graph through a recursive traversal algorithm. S13. Independent processing of Mesh objects; S14. Memory Management and Optimization: The system analyzes all extracted Meshes, merges similar materials, optimizes drawing calls, and establishes a spatial index structure to support efficient selection and collision detection.

[0007] In the above-mentioned low-code digital twin platform development method based on the anchor point mechanism, during the parsing process in step S1, each Mesh object in the model is assigned a unique identifier for subsequent anchor point definition.

[0008] In the above-described low-code digital twin platform development method based on the anchoring mechanism, in step S3, the data source supports multiple types: Static data: Directly input values ​​or expressions; Dynamic device data: Integrates with an IoT platform to subscribe to device telemetry data via API interface.

[0009] In the above-mentioned low-code digital twin platform development method based on the anchor point mechanism, in step S3, the user selects the device ID and data fields through a form, and the platform automatically generates data mapping rules.

[0010] In the above-described low-code digital twin platform development method based on the anchor point mechanism, the mapping rule generation operation in step S3 is as follows: S31. Conditional Expression Input: Users input logical expressions through a form, supporting comparison operators and logical operators; S32. Conditional Expression Conversion Output: The platform has a built-in lightweight rule engine that automatically parses conditional expressions and converts the input conditional expressions into executable functions. S33. Effect Parameter Configuration: Set the final output effect parameters according to the anchor point type and the set conditional expression; S34. Save the mapping rules automatically generated by the platform as structured data.

[0011] The above-mentioned low-code digital twin platform development method based on the anchor point mechanism includes the following operations in step S4: S41. Real-time data acquisition: The platform acquires real-time data through a subscription mechanism. When device data is updated, the data packet is pushed to the front end. S42. Mapping anchor points to data through key-value pairs: Associate anchor point IDs with data point IDs to form a mapping table; S43. Bind the rule engine to the data stream, and work with the platform's built-in dynamic effect engine to achieve the linkage between dynamic effect changes and data.

[0012] In the above-mentioned low-code digital twin platform development method based on the anchor point mechanism, in step S43, when real-time data arrives, the engine evaluates the conditions one by one. When the data changes meet the conditions, the platform triggers a callback function, locates the corresponding anchor point according to the mapping table, achieves high-performance rendering based on the shaders and animation mixers of the 3D editor in the dynamic effects engine, and applies dynamic effects.

[0013] Due to the adoption of the above technical solutions, the technical progress achieved by this invention is as follows.

[0014] This invention provides a low-code digital twin platform development method based on an anchor point mechanism. By introducing an anchor point abstraction mechanism, interactive points in a 3D model are dynamically associated with IoT data. Combined with a low-code configuration interface, condition-driven dynamic effects are displayed, solving the problems of large amounts of code writing, difficulty in integrating data and visualization, and complexity in implementing dynamic effects in traditional digital twin platform application development. Users can configure anchor points through a graphical interface, improving development efficiency by more than 80% without writing interactive code. The platform's built-in dynamic effect engine has effect templates such as flashing, color changes, and halos. Through configurable mapping rules, it supports triggering various dynamic effects based on conditions, reducing the cost of animation implementation. The anchor point mechanism seamlessly connects ThingsBoard data and 3D models, realizing data-driven visualization. Anchor point types and effects can be expanded through configuration to adapt to different industry needs (such as industry and smart cities). Attached Figure Description

[0015] Figure 1 This is a detailed flowchart of the present invention; Figure 2 This is a flowchart illustrating the specific operation of generating the mapping rules of this invention. Figure 3 This is a flowchart illustrating the real-time data association establishment and dynamic effect synchronization in step S4 of the present invention. Figure 4 This is a display image of the imported 3D model file on the platform in a specific embodiment of the present invention; Figure 5 This is a diagram illustrating the extracted Mesh objects and their hierarchical relationships in a specific embodiment of the present invention; Figure 6 This is an interface diagram of step S20 in a specific embodiment of the present invention; Figure 7 This is an interface diagram showing the setting of effect parameters in a specific embodiment of the present invention; Figure 8 This is a diagram showing the effect when the input temperature is 55°C during the verification and testing phase in a specific embodiment of the present invention. Figure 9 This is a diagram showing the effect when the input temperature is 50°C during the verification and testing phase in a specific embodiment of the present invention. Detailed Implementation

[0016] Low-code digital twin platform development methods based on anchor point mechanism, such as Figures 1 to 9 As shown, by introducing an anchor point abstraction mechanism, interactive points in the 3D model are dynamically associated with IoT data, and a condition-driven dynamic effect display is achieved by combining a low-code configuration interface. Specifically, the steps include: S1. 3D Model Import and Parsing: Users import 3D model files into the 3D editor of the digital twin model. The platform automatically parses the model structure and extracts all Mesh objects and their hierarchical relationships.

[0017] The model structure parsing operation is as follows: S11. File Format Recognition and Loader Selection: The system first automatically selects the corresponding Three.js loader based on the file extension (.glb, .gltf, .obj, .fbx, etc.). Each loader is responsible for converting binary or text-formatted 3D models into scene graph structures that Three.js can recognize.

[0018] S12, Scene Graph Traversal and Mesh Extraction: After loading, the platform will receive a complete Scene object. At this time, the system will access each node in the scene graph through a recursive traversal algorithm.

[0019] S13. Independent processing of Mesh objects: For each identified Mesh object, the platform performs the following operations: S131, Geometry Separation: Extract the geometry data of the Mesh. If multiple Meshes share the same geometry, the system will automatically create a copy.

[0020] S132, Material Independence: Separate material references to ensure that each Mesh has an independent material instance.

[0021] S133. Transformation Information Saving: Record local transformations (position, rotation, scaling) and world transformation matrices.

[0022] S134. Metadata Extraction: Obtain attributes such as name and user-defined data from the original model.

[0023] S14. Memory Management and Optimization: The system analyzes all extracted Meshes, merges similar materials, optimizes drawing calls, and establishes spatial index structures (such as BVH trees) to support efficient selection and collision detection.

[0024] The platform uses a multi-level index structure to maintain the hierarchical relationship of the model. The automatic construction process of the hierarchical relationship is as follows: a) Depth-first traversal: Starting from the root node of the scene, recursively traverse all child nodes; b. Relationship record: Record the parent node, child node, and sibling node of each node during the traversal process; c. Depth calculation: Calculate the depth level of each node in the tree structure; d. Transformation chain construction: Establishing the transformation inheritance relationship between parent and child nodes; e. Path generation: Generate a complete path from the root node to each node.

[0025] During the parsing process in step S1, each Mesh object in the model is assigned a unique identifier (such as a UUID) for subsequent anchor definition.

[0026] The platform employs a three-level identifier system to ensure the unique identification of each mesh: Level 1: Platform Global ID, which is a unique identifier at the system level. The format is "mesh_" plus a timestamp and a random string (e.g., mesh_1641234567890_abc123def). Its core feature is that it is globally unique. It is mainly used for database storage and external system references and can be permanently stored. Level 2: Three.js runtime UUID, which is an identifier used internally by the Three.js graphics engine at runtime, automatically generated by Three.js (e.g., f819c7c5-8b2f-4a1e-9c3d-6b5a8d7c9e0f). It is mainly used for real-time operation and rendering in 3D scenes on web pages; Level 3: Original model UUID. This identifier is obtained from the original file and retained when the model is imported (e.g., original_9876543210_xyz789). Its purpose is to retain a reference to the original model, which facilitates model version management and traceability.

[0027] The process of generating a unique identifier is as follows: Assigning an identifier: When a new 3D model is added to the system, the assignPlatformId method generates a global platform ID for it (format: mesh_standard_1641234567890_5_abc123def). This ID combines the model type, timestamp, counter, and random number to ensure global uniqueness. Create information profile: The system will create a detailed meshInfo profile for the model, which includes its IDs at all levels (platform ID, Three.js runtime UUID, raw UUID), the model's metadata (number of vertices, material type, bounding box, etc.), the calculated feature hash value, and its parent-child relationship in the scene; Registration and Indexing: The generated files will be stored in the main storage, and all reverse mapping tables and feature indexes will be updated to facilitate future searches by any dimension; Mounting references: To facilitate direct use in the code, the system will mount the platform ID and key information to the Three.js model object itself (through mesh.userData and custom methods).

[0028] S2. Anchor point definition and type specification: In the design interface of the 3D editor, the user interactively selects the Mesh object in the model, marks it as an anchor point, and defines the anchor point type through the enumeration system.

[0029] Anchor types include text, video, color, rotation, blinking, halo diffusion, and flowing light effects.

[0030] S3. Data Source Configuration and Binding: In the platform's low-code configuration interface, users select a data source for the anchor point and bind the anchor point to the data source, generating mapping rules that trigger dynamic effect changes.

[0031] In step S3, the data source supports multiple types: Static data: Directly input values ​​or expressions; Dynamic device data: Integrates with an IoT platform to subscribe to device telemetry data via API interface.

[0032] In step S3, the user selects the device ID and data field through a form, and the platform automatically generates data mapping rules. Specifically, the mapping rule generation process is as follows: S31. Conditional Expression Input: Users input logical expressions (such as temperature > 50℃) through a form, supporting comparison operators (>, <, =) and logical operators (AND, OR).

[0033] S32. Conditional Expression Conversion Output: The platform has a built-in lightweight rule engine that automatically parses conditional expressions and converts the input conditional expressions into executable functions.

[0034] S33. Effect Parameter Configuration: Set the effect parameters for the final output based on the anchor point type and the set conditional expression.

[0035] S34. Save the mapping rules automatically generated by the platform as structured data (such as JSON format).

[0036] S4. Real-time data association establishment and real-time synchronization of dynamic effects: Establish a real-time connection between anchor points and data sources, and realize dynamic effect display based on mapping rules and the cooperation of the dynamic effect engine.

[0037] S41. Real-time data acquisition: The platform acquires real-time data through a subscription mechanism. Specifically, the platform establishes a real-time connection with the data source through WebSocket or MQTT protocols. When the device data is updated, the data packet is pushed to the front end.

[0038] S42. Mapping anchor points to data through key-value pairs: Associate anchor point IDs with data point IDs to form a mapping table (e.g., if the anchor point ID is "pump bearing" and the data point ID is "sensor temperature", then obtain the "reading of the temperature sensor located on the pump bearing").

[0039] S43. Bind the rule engine to the real-time data stream and work with the platform's built-in dynamic effects engine to achieve linkage between dynamic effects changes and data: Specifically, when real-time data arrives, the engine evaluates the conditions one by one. When the data changes meet the conditions, the platform triggers a callback function, locates the corresponding anchor point according to the mapping table, and achieves high-performance rendering based on the shaders and animation mixers of the 3D editor in the dynamic effects engine, and applies dynamic effects.

[0040] In step S43, the triggered effect can be: The flickering effect is achieved by periodically changing the material's transparency. Halo Diffusion: Generates particle diffusion animation using a custom shader.

[0041] S5. Verification and Testing: The platform provides a simulated data mode, allowing users to input test values ​​to verify the anchor point's response capability.

[0042] This invention ensures accurate and efficient association between anchor points and datasets through model parsing, interactive configuration, and real-time communication technologies, avoiding the complexity of traditional manual coding. After binding with the data source, it generates mapping rules, forming a chain of "data → conditions → effects," making the condition rules for dynamic effects configurable. Complex business logic can be implemented without writing code, improving development efficiency.

[0043] The integrated architecture of the digital twin platform in this invention is as follows: Data layer: ThingsBoard handles device management, data acquisition, and alarms, forming the data foundation of the digital twin; Model layer: Three.js is responsible for 3D rendering, and the anchor point mechanism ensures that the model and data are accurately linked; Application layer: GoView low-code interface provides configuration tools, allowing users to define twin behaviors without coding.

[0044] The 3D model serves as a digital twin of physical entities (such as water pumps and tunnels). It is bound to the data of real devices through an anchoring mechanism. The platform collects IoT data through ThingsBoard and drives the virtual model's state update, forming a two-way mapping.

[0045] Users can interact with the digital twin platform through anchor points, such as clicking on the model to view real-time data and modifying parameters to control the equipment in reverse. The platform supports simulation mode, which can simulate the model behavior under different working conditions, demonstrating the "simulation optimization" value of digital twins.

[0046] The end-to-end architecture realizes a closed loop of "data-model-interaction", which conforms to the core feature of "virtual-real linkage" of digital twins and meets the full-element requirements of digital twins.

[0047] In addition, the platform integrates historical data analysis and machine learning algorithms to support trend prediction. It can not only display the current state of the model, but also simulate historical behavior through data backtracking to assist decision-making. For example, it can predict equipment failures based on temperature data and trigger early warning effects (such as flashing) in the 3D model.

[0048] The present invention will now be described in further detail with reference to specific embodiments.

[0049] The development method of a low-code digital twin platform based on the anchoring mechanism, taking the "water pump temperature monitoring" scenario as an example, includes the following steps: S10. 3D Model Import and Parsing: Users import the water pump 3D model file into the Three.js editor of the digital twin model. The platform automatically parses the model structure and extracts all Mesh objects and their hierarchical relationships. The 3D model displayed after importing into the platform is as follows: Figure 4 As shown, the extracted Mesh objects and their hierarchical relationships are as follows: Figure 5 As shown.

[0050] S20. Anchor Point Definition and Type Specification: In the Three.js editor's design interface, the user interactively selects the bearing in the pump model as a Mesh object and marks it as an anchor point. The anchor point type is color, and the anchor point information (such as position coordinates and type parameters) is saved as a JSON configuration and bound to the model's metadata. The interface for this step is as follows: Figure 6 As shown.

[0051] S30. Data Source Configuration and Binding: In the Vue component based on GoView, the user selects dynamic device data from an integrated IoT platform as the data source for the bearing mesh anchor point, subscribes to device telemetry data through the API interface, and binds the anchor point to the temperature sensor data in the data source to generate mapping rules that trigger dynamic effect changes.

[0052] Specifically, the mapping rule generation process is as follows: S301. Input conditional expression: The color is red when the temperature is >50°C and green when the temperature is ≤50°C.

[0053] S302: The platform has a built-in lightweight rule engine that automatically parses conditional expressions and converts the input conditional expressions into executable functions, such as converting temperature>50℃ into (telemetryValue)=>telemetryValue>50.

[0054] S303, Effect Parameter Configuration: Based on the anchor point type and the set conditional expression, set the final output effect parameters. The operation interface for setting effect parameters is as follows: Figure 7 As shown.

[0055] In this embodiment, when the temperature exceeds 50°C, in addition to setting the color to infrared, the frequency of the flashing anchor point is also set to 2Hz, and the duration is 15s.

[0056] S304. Save the mapping rules automatically generated by the platform as JSON data.

[0057] S40. Real-time data association establishment and dynamic effect real-time synchronization.

[0058] S401 Real-time data acquisition: The platform establishes a real-time connection with the data source through the WebSocket protocol. When the device data is updated, the data packet is pushed to the front end.

[0059] S402. Mapping anchor points to data through key-value pairs: Associate anchor point IDs with data point IDs to form a mapping table. Specifically, the anchor point ID is "pump bearing" and the data point ID is "sensor temperature". The actual purpose is to obtain the "reading of the temperature sensor located on the pump bearing".

[0060] S403 binds the rule engine to the data stream, and works with the platform's built-in dynamic effect engine to achieve the linkage between dynamic effect changes and data.

[0061] Specifically, when real-time data arrives, the engine evaluates each condition one by one. When the temperature value is greater than 50℃, the triggering condition of the mapping rule is met, the platform triggers the callback function, calls the corresponding Three.js effect function, locates the corresponding anchor point according to the mapping table, and achieves high-performance rendering based on the 3D editor's shader and animation mixer, and applies dynamic effects.

[0062] S50, Verification and Testing: When the input temperature is 55℃, the water pump bearing automatically turns red and flashes at a frequency of 2Hz for 15 seconds to enhance visual alertness. The effect is as follows: Figure 8 As shown; with an input temperature of 50℃, the water pump bearing turns green, and the effect is as follows: Figure 9 As shown.

Claims

1. A low-code digital twin platform development method based on anchor point mechanism, characterized in that, By introducing an anchor point abstraction mechanism, interactive points in the 3D model are dynamically associated with IoT data, and a condition-driven dynamic effect display is achieved by combining a low-code configuration interface. The specific steps include: S1. 3D Model Import and Parsing: Users import 3D model files into the 3D editor of the digital twin model. The platform automatically parses the model structure and extracts all Mesh objects and their hierarchical relationships. S2. Anchor point definition and type specification: In the design interface of the 3D editor, the user interactively selects the Mesh object in the model, marks it as an anchor point, and defines the anchor point type through the enumeration system. S3. Data source configuration and binding: In the platform's low-code configuration interface, users select a data source for the anchor point and bind the anchor point to the data source, generating mapping rules that trigger dynamic effect changes. S4. Real-time data association establishment and real-time synchronization of dynamic effects: Establish a real-time connection between anchor points and data sources, and realize dynamic effect display based on mapping rules and the cooperation of the dynamic effect engine; S5. Verification and Testing: The platform provides a simulated data mode, allowing users to input test values ​​to verify the anchor point's response capability.

2. The low-code digital twin platform development method based on anchor point mechanism according to claim 1, characterized in that, The model structure parsing operation in step S1 is as follows: S11. File Format Recognition and Loader Selection: The system first automatically selects the corresponding Three.js loader based on the file extension. Each loader is responsible for converting the 3D model in binary or text format into a scene graph structure that can be recognized by Three.js. S12, Scene Graph Traversal and Mesh Extraction: After loading, the platform will receive a complete Scene object. At this time, the system will access each node in the scene graph through a recursive traversal algorithm. S13. Independent processing of Mesh objects; S14. Memory Management and Optimization: The system analyzes all extracted Meshes, merges similar materials, optimizes drawing calls, and establishes a spatial index structure to support efficient selection and collision detection.

3. The low-code digital twin platform development method based on anchor point mechanism according to claim 1, characterized in that, During the parsing process in step S1, each Mesh object in the model is assigned a unique identifier for subsequent anchor point definition.

4. The low-code digital twin platform development method based on anchor point mechanism according to claim 1, characterized in that, In step S3, the data source supports multiple types: Static data: Directly input values ​​or expressions; Dynamic device data: Integrates with an IoT platform to subscribe to device telemetry data via API interface.

5. The low-code digital twin platform development method based on anchor point mechanism according to claim 1, characterized in that, In step S3, the user selects the device ID and data field through a form, and the platform automatically generates data mapping rules.

6. The low-code digital twin platform development method based on anchor point mechanism according to claim 5, characterized in that, In step S3, the mapping rule generation operation is as follows: S31. Conditional Expression Input: Users input logical expressions through a form, supporting comparison operators and logical operators; S32. Conditional Expression Conversion Output: The platform has a built-in lightweight rule engine that automatically parses conditional expressions and converts the input conditional expressions into executable functions. S33. Effect Parameter Configuration: Set the final output effect parameters according to the anchor point type and the set conditional expression; S34. Save the mapping rules automatically generated by the platform as structured data.

7. The low-code digital twin platform development method based on anchor point mechanism according to claim 1, characterized in that, Step S4 includes the following operations: S41. Real-time data acquisition: The platform acquires real-time data through a subscription mechanism. When device data is updated, the data packet is pushed to the front end. S42. Mapping anchor points to data through key-value pairs: Associate anchor point IDs with data point IDs to form a mapping table; S43. Bind the rule engine to the data stream, and work with the platform's built-in dynamic effect engine to achieve the linkage between dynamic effect changes and data.

8. The low-code digital twin platform development method based on anchor point mechanism according to claim 7, characterized in that, In step S43, when real-time data arrives, the engine evaluates the conditions one by one. When the data changes meet the conditions, the platform triggers a callback function, locates the corresponding anchor point according to the mapping table, achieves high-performance rendering based on the shaders and animation mixers of the 3D editor in the dynamic effects engine, and applies dynamic effects.

Images