A two-dimensional and three-dimensional linkage-based industrial control level configuration software system and method

Abstract

The application provides an industrial control level configuration software system and method based on two-dimensional and three-dimensional linkage, which is suitable for industrial automation, intelligent building and other types of engineering scenes. The system includes modular component deployment, point import and matching, drawing and model analysis, two-dimensional and three-dimensional linkage view management, device point confirmation, communication simulation test and mapping learning optimization and other functional modules. By importing two-dimensional CAD drawings and three-dimensional BIM models, automatic matching and visual binding of device components and point data are realized; and combined with AI assisted matching algorithm and template reuse mechanism, the configuration efficiency and accuracy are improved. At the same time, it supports joint debugging test and graphical feedback of the results, ensuring the authenticity and stability of the system mapping. The method can realize cross-platform configuration, intelligent component identification and engineering knowledge sedimentation, and has strong universality and expandability.

Description

Technical Field

[0001] This invention relates to the field of industrial configuration software technology, and in particular to an industrial control-level configuration software system and method based on two-dimensional and three-dimensional linkage. Background Technology

[0002] With the continuous advancement of industrial automation and digital transformation, 3D modeling technology is increasingly being applied in industries such as construction, power, and transportation, making 3D models a digital representation of real physical equipment and their spatial relationships. Integrating 2D drawings with 3D models, and combining real-time location data with configuration logic to achieve a "2D / 3D linkage" configuration system, has become an important direction for industry development.

[0003] However, existing industrial control configuration systems generally have the following shortcomings in the fusion of 2D and 3D data:

[0004] The data is severely fragmented: there is a lack of effective mapping mechanisms between 2D drawings, 3D models, and actual physical equipment. The configuration process often relies on manual identification and binding, which is inefficient and prone to errors. Secondly, there is a lack of intelligent assistance capabilities: equipment location matching largely depends on human experience, making it difficult to handle batch configuration and cross-system compatibility issues in large-scale scenarios. Therefore, we propose an industrial control-level configuration software system and method based on 2D and 3D linkage. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by providing an industrial control-level configuration software system and method based on two-dimensional and three-dimensional linkage, thereby solving the technical problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for operating an industrial control-level configuration software system based on two-dimensional and three-dimensional linkage includes the following steps:

[0008] S1: Based on the type of facilities and equipment and the communication protocol, select the corresponding modular components and deploy them to the server in a microservice manner. Set the component version information, deployment location, communication protocol parameters and start / stop status through the basic configuration interface.

[0009] S2: Import the location list data containing equipment model and coding information, establish the binding relationship between location data and components, call the AI-assisted algorithm to suggest matching for unsuccessful matching items and record the results of manual correction;

[0010] S3: Import 2D CAD drawings and 3D BIM model files, parse out the geometric data and attribute information of the components, and establish the mapping basis between the drawings and the model;

[0011] S4: Load 2D and 3D linked views in the visualization interface to display the initial position of the equipment in the model and the corresponding point data. It supports users to interactively adjust inaccurate positions or mapping relationships, including dragging and moving equipment, correcting numbers, and rebinding.

[0012] S5: Users check the binding information of each device in the 2D and 3D views to confirm whether the point data is consistent with the actual device. For unbound devices, point information can be added and configured from the component library.

[0013] S6: Based on the configuration results, call the communication protocol of the modular components to conduct communication integration tests with the actual device, send test commands and receive response data to verify whether the mapping and interaction between the virtual scene and the real device are accurate;

[0014] S7: Record the final mapping relationship of devices that have successfully completed the joint debugging and generate a device mapping list; for devices that have failed the joint debugging, the user corrects the errors and feeds them back to the AI ​​algorithm model to continuously optimize the accuracy of subsequent automatic matching.

[0015] The modular components in S1 support a protocol adaptive mechanism, which can automatically switch to a backup communication protocol or adjust parameters to retry when device communication fails, thereby improving the success rate of joint debugging.

[0016] The manual correction results recorded in S2 can be used as training samples for reinforcement learning models to improve the matching accuracy between location data and equipment models.

[0017] The 3D BIM model imported into S3 supports multiple formats, and the model components contain attribute identification information for matching point codes.

[0018] In S4, when a device is dragged in the 3D view, the system automatically updates the position label and device number in the 2D drawing, realizing two-way binding verification between the 2D and 3D views.

[0019] In S5, the point binding relationships configured by the user can be abstracted into structured template files, which can be exported and quickly reused in other projects.

[0020] The S6 system supports initiating parallel communication test tasks for multiple devices simultaneously and displays the test results graphically in a two-dimensional or three-dimensional interactive interface.

[0021] The mapping list file generated in S7 can interface with the facility asset management system or building lifecycle management platform to achieve integrated management of equipment throughout its entire lifecycle.

[0022] An industrial control-grade configuration software system based on 2D / 3D linkage, comprising:

[0023] The modular component deployment module is used to select the appropriate modular components according to the type of target facilities and equipment and communication protocol, and deploy them to the server in the form of microservices, and complete the configuration of component version, deployment location, communication parameters and start / stop status;

[0024] The site import and matching module is used to import a list of sites containing equipment model and serial number information, match corresponding modular components based on the component library, establish a binding relationship between sites and components, and call the AI ​​matching algorithm to provide suggestions for sites that fail to match and record the results of manual correction.

[0025] The drawing and model parsing module is used to import 2D CAD drawings and 3D BIM model files, parse component geometric information and attribute data, and build the mapping foundation between drawings and models.

[0026] The linked view and position adjustment module is used to display two-dimensional and three-dimensional linked views in the visualization interface, and supports users to drag and drop devices, correct numbers, rebind points, and adjust device positions.

[0027] The site confirmation and configuration module is used to verify the consistency between the site and the model device on a device-by-device basis, and allows users to manually add and configure site information for unbound devices;

[0028] The communication integration module is used to call the communication protocol to perform communication simulation tests with real devices, and to send test commands and receive response data to verify the accuracy of 2D and 3D mapping.

[0029] The relationship confirmation and learning optimization module records the mapping relationships of successfully integrated devices and generates a list. For failed items, the feedback data after user correction is used to train the AI ​​algorithm to improve the accuracy of subsequent automatic matching. The graphical visualization interface is used to display communication test results, changes in linked view operations, and system interaction feedback.

[0030] The beneficial effects of this invention are as follows: Through the deployment of modular components and an automatic point-of-use matching mechanism, this invention significantly reduces the tedious manual configuration work in traditional configuration systems, improving engineering deployment efficiency. It provides an efficient linkage mechanism between 2D CAD drawings, 3D BIM models, and real physical equipment, supporting drag-and-drop adjustment and synchronous updates of equipment positions, enhancing operational intuitiveness and accuracy, and facilitating operation and maintenance as well as visualization. The system has the ability to automatically adjust communication protocol parameters and switch to backup protocols, adapting to various industrial equipment communication environments and improving system stability and compatibility. By introducing artificial intelligence algorithms to intelligently match points and equipment models, and combining this with manually corrected samples for model optimization, the system becomes "smarter with use," possessing continuous learning and improvement capabilities. It supports the abstraction of user-configured binding relationships into structured templates, facilitating rapid reuse in other projects and promoting standardized deployment and knowledge accumulation. Communication testing with real equipment verifies the accuracy of 2D and 3D mapping, ensuring correct configuration before system launch and effectively reducing engineering debugging and maintenance costs. The equipment mapping list can seamlessly connect to asset management systems or platforms, connecting the data chain across design, construction, and operation and maintenance stages, promoting digital twins and intelligent operation and maintenance in industrial scenarios. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the steps and methods of the present invention. Detailed Implementation

[0032] 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, and 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.

[0033] Example 1: As Figure 1 As shown, this embodiment provides a method for operating an industrial control-level configuration software system based on two-dimensional and three-dimensional linkage, including the following steps:

[0034] S1: Modular component deployment and basic configuration: Based on the type of facility and equipment and the communication protocol, select the corresponding modular components and deploy them to the server in a microservice manner. Set the component version information, deployment location, communication protocol parameters and start / stop status through the basic configuration interface.

[0035] S2: Location Data Import and Matching: Import location list data containing information such as equipment model and code, match deployed modular components according to equipment type, establish binding relationship between location data and components, call AI-assisted algorithm to suggest matching for unsuccessful matching items and record manual correction results;

[0036] S3: Import and parsing of drawings and model files: Import 2D CAD drawings and 3D BIM model files, parse out the geometric data and attribute information of components, and establish the mapping basis between drawings and models;

[0037] S4: Linked 2D and 3D Views and Equipment Position Adjustment: Load linked 2D and 3D views in the visualization interface to display the initial position of the equipment in the model and the corresponding point data. Users can interactively adjust inaccurate positions or mapping relationships, including dragging and moving equipment, correcting numbers, and rebinding.

[0038] S5: Device location information confirmation and configuration: Users check the binding information of each device in the 2D and 3D views to confirm whether the location data is consistent with the actual device. For unbound devices, the location information can be added and configured from the component library.

[0039] S6: Joint Debugging Simulation Test and Communication Verification: Based on the configuration results, call the communication protocol of the modular components to conduct joint communication debugging tests with the actual device, send test commands and receive response data to verify whether the mapping and interaction between the virtual scene and the real device are accurate.

[0040] S7: Final confirmation and learning optimization of mapping relationships: Record the final mapping relationship of devices that have successfully completed joint debugging and generate a device mapping list; for devices that have failed joint debugging, the user corrects the data and feeds it back to the AI ​​algorithm model to continuously optimize the accuracy of subsequent automatic matching.

[0041] S1 includes the following sub-steps S110-S140:

[0042] S110: Device Protocol Identification and Component Selection

[0043] The system first reads the facility and equipment list file input or imported by the user, and extracts the field parameters, including the equipment type T. i Communication protocol P i Manufacturer M i wait;

[0044] Establish equipment demand vector V i ={T i ,P i M i}, by looking up a table or semantic matching, retrieve compatible modular components C from the module component library. i The matching rules are as follows:

[0045]

[0046] in: A collection of components; V CThis provides the protocol types and adaptation information supported by the component; Sim(·) is the compatibility scoring function, which can be a weighted cosine similarity or a decision tree scoring function based on the configuration template.

[0047] S120: Component Microservice Deployment to Server: After selecting a component, the system packages it as a container service (such as a Docker image) or deploys it to the specified server as a JAR file, service script, etc. k ;

[0048] The service deployment model can be expressed as: D(C j )→S k ,

[0049] Where: D(C j ) indicates a specific object or component C j To perform processing, scheduling, or decision-making;

[0050] →S k : Indicates that the object is assigned to a specific state, server, task unit, or processing module S. k ;

[0051] This indicates that among all possible candidate targets k, the one that makes expression L... k +R k The smallest k;

[0052] L k Indicates the current server load; R k It represents the cost of server resources (CPU, memory, port conflicts, etc.); and determines the optimal deployment location through a load balancing scheduling algorithm.

[0053] S130: Basic Parameter Configuration: In the basic configuration interface, users set the following parameters for the deployed components: Version Information V j : The version number supported by the component, such as v1.0.3; the deployment location IP. j :PORT j : The host address and port where the service runs; Communication protocol parameters: Communication mode (TCP / UDP); Encoding method (ASCII / Binary); Data frame structure (start bit, length, checksum);

[0054] Operating status flag δ j ∈{0,1}: Indicates whether the component is enabled.

[0055] The parameter setting table structure is as follows:

[0056] Parameters Data types Example value Version String "v1.2.1" IP Address String "192.168.1.5" Port Integer 502 Protocol Enum TCP DataFormat JSON Schema {...} EnabledFlag Boolean true

[0057] S140, Component Start-up and Operation Status Monitoring

[0058] The system supports the following component startup commands:

[0059] Start(C j ): Indicates that component C j Issue a "start" command. This component can be a software module, control unit, simulation block, subsystem, etc. (Logical Implication): Indicates that the "startup behavior" will lead to (or trigger) a definite result—creating and running an instance of the component; Runtime_Instance(C j ): Represents component C j A runtime instance has been successfully created, that is, it has changed from a static description to an executable object;

[0060] After startup, a running instance is generated and the service is registered; it also supports stop commands and health check APIs, periodically performs ping or heartbeat checks on the service status, records the response latency τ and the number of exceptions ∈, and triggers an alarm or automatically restarts the system when ∈ > θ.

[0061] Through the implementation of the above sub-steps, this invention realizes the automatic identification, dynamic deployment, structured configuration and intelligent monitoring of modular components, providing a solid and standardized foundation for subsequent point data import and scene construction. It has high reusability, platform independence and high scalability, effectively reducing the manual burden in the early stage of configuration engineering.

[0062] S2 includes the following sub-steps S210-S250:

[0063] S210: Structured Import of Point Data: Users upload a point list file through the system interface. The file includes device type (DeviceType), device code (DeviceCode), point name (PointName), and communication parameters (such as address, function code, etc.). The system validates the file's format and converts it into a structured data table: PointEntry. i ={T i C i ,P i A i}

[0064] Wherein: T i : The device type corresponding to the i-th point; C i Equipment number; P i Location function identifier; A i Communication parameters, such as register address, unit, precision, etc.

[0065] The system automatically generates a structured manifest:

[0066] Index <![CDATA[Device type T i > <![CDATA[Number C i > <![CDATA[Functional Point P i > <![CDATA[Communication parameter A i > 1 water pump WP001 Start-stop control Modbus0x0001 2 Temperature sensor TS007 Current temperature OPCTemp.Cur … … … … …

[0067] S220: Modular component pre-matching processing: The system pre-matches components based on the T value of each point. i Type: Queries whether there is a matching modular component used for protocol parsing and data communication for this type of device.

[0068] Function to determine matching relationship: Match(T) i )={M j |M j supports T i If a unique matching component M exists... j The system will automatically bind the PointEntry. i With component M j .

[0069] Match(T i ): indicates that for task T i Perform a matching operation, that is, find out which modules or resources can handle the task; M j : Indicates an available module, resource, node, or execution unit; {M j |M j supportsT i}: This is a set representing all sets that satisfy "support task T". i "A set of modules that meet the following conditions. The vertical bar "|" is the set construction symbol, representing "elements that satisfy the following conditions";

[0070] S230: AI-assisted matching algorithm for handling unmatched items: For points that cannot be directly matched, the system calls an AI-assisted matching algorithm to perform semantic, structural, and historical similarity analysis. Let the set of unmatched points be:

[0071] Define a set U, whose elements are all PointEntries that cannot be automatically matched to a module. i That is: perform Match(T) on task Ti. i When matching, an empty set is returned. These locations will serve as targets for AI recommendations.

[0072] For each pointEntry in set U i The following recommendation model is used:

[0073] in: This represents the optimal matching module recommended by the AI ​​model; arg max represents the module that maximizes the objective function from all candidate modules. It is a set of candidate modules; Sim AI (F i ,F j (), is a similarity function calculated by the AI ​​model, which measures the similarity of a point entry. i Feature F i With module M j Feature F j The similarity between them is considered; finally, the most similar module is selected as the recommended match.

[0074] F i : Represents the feature vector of a point; F j : Represents the capability vector of a component; Sim AI The similarity score calculated by the AI ​​model can be based on deep semantic networks or clustering; if Sim AI If the value is greater than θ, binding is recommended; otherwise, mark it as "pending manual processing".

[0075] S240: Manual Confirmation and Correction of Binding Relationship: AI matching suggestions will be presented on the front-end interface in the form of a "suggestion relationship graph". Users can confirm each AI recommendation result and perform the following operations: replace recommendation components; manually select other components; modify the communication parameters, units, and identifiers of points; each manual correction is recorded by the system as a reinforcement learning sample.

[0076] Sample = (PointEntry) i AIResult j HumanCorrection

[0077] PointEntry i : Represents an actual point entity, such as a sensor, actuator, or monitoring device; contains the input characteristic information of that point (such as type, location, function, etc.).

[0078] AIResult j : Indicates that the AI ​​model has this point PointEntry i The automatically provided prediction matching results, such as recommended modules or bound control logic, are the "initial answer" of the model's current output.

[0079] HumanCorrection refers to human intervention in AIResult. jThe corrected result represents a more reasonable and correct matching or processing method under human judgment; this value will be fed back to the AI ​​model as a supervisory signal for subsequent training. This sample will be fed back to the AI ​​model for the next round of model weight fine-tuning.

[0080] S250: Binding Relationship Archiving and Export: The final confirmed matching relationship will be saved in a structured file format, including but not limited to: point number, binding component ID, protocol name, parsing method, initial state, etc. It supports exporting to .json, .xml, or database formats for subsequent simulation integration and visualization. An example binding structure is as follows:

[0081] {

[0082] "PointCode":"WP001_START",

[0083] "ModuleID":"MODBUS_WATERPUMP_V1",

[0084] "Protocol":"Modbus-TCP",

[0085] "Address":"0x0001",

[0086] "BindTime":"2025-07-10T14:22:11"}

[0087] Through the above sub-steps, this method constructs a point binding process that combines structured data processing, AI intelligent recommendation, and manual closed-loop optimization, which significantly improves the data docking efficiency of large-scale heterogeneous devices, avoids repeated manual input errors, and has technical advantages such as self-adaptation, self-learning, transferability, and reusability.

[0088] S3 includes the following sub-steps S310-S350:

[0089] S310: 2D CAD Drawing File Import: Users can import CAD drawing files through the system's upload interface, supporting formats such as .dwg, .dxf, and .dgn. The system converts the files into a collection of structured geometric objects.

[0090] Each two-dimensional geometric component Includes the following information: geometric type (line, circle, polygon, etc.); vertex coordinate set V i ={(x1,y1),…,(x m ,y m Layer name, tile name, color, annotation text, and other attribute information.

[0091] The system temporarily stores the geometric set in the two-dimensional component parsing cache.

[0092] S320: 3D BIM Model File Import: The system supports importing BIM model file formats such as .rvt (Revit) and .ifc (Industry Foundation Classes) and extracting 3D component sets.

[0093]

[0094] Each three-dimensional component Includes: geometric structure (bounding box volume, polyhedral mesh, center point, etc.); attribute information (such as component name, system type, material, family type, instance ID, etc.); spatial location coordinates center point: P j =(x j ,y j ,z j ), used for projection and alignment.

[0095] S330: Unified Conversion of Component Attribute Structure: In order to establish a unified mapping mechanism, the system performs... and Each component in the code establishes a unified attribute representation vector:

[0096]

[0097] in: It is a two-dimensional point feature vector; T i It refers to the type of the point, such as temperature, pressure, switch, etc.; L i It refers to the spatial coordinates of a point on a two-dimensional drawing; N i It is the location name or number; A i These are the additional attributes of the location, such as communication protocol, electrical characteristics, and connection method;

[0098] It is a 3D component feature vector; T j It refers to the type of three-dimensional component, such as a fan, sensor, pump, etc.; M j It is the component model name or ID; P j It refers to the coordinates or geometric position of a component in three-dimensional space; A j It refers to the component's attribute information, such as material, interface, power requirements, etc.

[0099] S340: Initial Mapping Establishment between Drawings and Model Components: The system uses a feature matching algorithm to perform preliminary matching between 2D and 3D components, using the following matching function:

[0100]

[0101] The i-th two-dimensional point object;

[0102] The j-th 3D component object;

[0103] The feature vector of a two-dimensional point contains attributes such as type, location, and name;

[0104] The feature vector of a 3D component contains attributes such as type, model, and position;

[0105] The similarity calculation function between the two is represented by θ, which is calculated by cosine similarity, distance metric or AI model, and θ is the threshold (usually set to 0.7 to 0.85).

[0106] The system stores the matching results in a 2D and 3D component mapping table:

[0107] CAD Component ID BIM Component ID Matching similarity Confirm? g12_2D g87_3D 0.91 no

[0108] S350: User Confirmation of Mapping and Interactive Adjustment: The system visually displays the initial matching results. Users can manually confirm or drag and drop to correct the binding relationship between the component positions in the 2D drawings and the components in the 3D model. The corrected binding relationship will be used as the final configuration for subsequent S4 2D and 3D synchronous display and linkage operations.

[0109] This step establishes an attribute mapping relationship between 2D and 3D components through unified analysis and structural abstraction of CAD drawings and BIM models. This facilitates subsequent key functions such as visualized equipment positioning, point-to-point debugging, and view synchronization. The multi-dimensional mapping mechanism, employing attribute vector modeling, feature similarity calculation, and spatial projection alignment, significantly improves the accuracy of drawing-model integration and the system's intelligence level.

[0110] S4 includes the following sub-steps S410-S450:

[0111] S410: Linked View Loading and Initialization Display: The system calls CAD drawings and BIM model parsing data, loading them to form a synchronized 2D and 3D view interface; in the 3D view, based on the BIM component set... Render the scene model; in the 2D view, based on the set of geometric components. Simultaneous rendering of layout drawings; each device object D i Its initial position is determined by its center point P. i =(x i ,y i ,z iThe information is indicated and highlighted in the 3D scene; at the same time, each device number, point name, status and other information are displayed on the view in the form of labels.

[0112] S420: Linked Highlighting of Equipment Locations: The system automatically binds equipment components and location information to form a binding set. Equipment components With point p j Binding};

[0113] p represents the i-th 3D device component object; j This represents the j-th two-dimensional point object; This indicates a binding relationship, meaning the 3D component corresponds to a specific point; when a user clicks on a component... The system will automatically locate the corresponding component in the 2D drawing and highlight the bound point p. i Reverse click on 2D drawing components It will also be synchronized to the highlighted area in the 3D view. Achieve two-way linkage.

[0114] S430: Equipment Position Error Detection and Alert Mechanism: The system introduces a spatial position error detection mechanism. If the current position P of the equipment... i Its reference position on the drawing If the difference exceeds the threshold ∈, a message "Position deviation exceeds limit" will be displayed. The calculation formula is as follows: If Δ i If the value is greater than or equal to ∈, the system will prompt the user to perform position correction.

[0115] S440: Device dragging and dynamic updating: Users can drag the device's position in the 3D view with the mouse, and the system automatically updates the center point coordinates P of the component in real time. i This is simultaneously reflected in the bound point data; the coordinates of all dragged positions are updated to the database in real time.

[0116] Update(DeviceID i ,P i )

[0117] DeviceID i : Represents the unique identifier of the i-th device, which can be understood as the primary key of the device in the configuration system or database; P i This indicates the location feature information of the device in the current 2D drawing or 3D model, such as coordinates, viewpoint, attribute labels, etc. If the device position is moved and the original bound point is no longer compatible, the system can prompt whether to automatically unbind or perform a rebinding suggestion.

[0118] S450, Number Correction and Binding Relationship Adjustment: Users can click on the component label in the linkage interface to modify the equipment number (e.g., WP001→WP002); the system will update the equipment code field and update the relevant fields in all bound point data accordingly; if the equipment number changes and affects the uniqueness of the point, the system will perform a conflict check and provide candidate options.

[0119] Example interaction scenario: A user discovers that a pump device in the 3D scene is in the wrong position and moves it to the expected position by dragging it; at the same time, the system recognizes that its position number WP003 does not match the actual number, and the user clicks to change it to WP001; the system synchronously updates its position label and number in the 2D drawing, and the interaction is completed.

[0120] Through the above linkage adjustment mechanism, this invention achieves high-precision synchronous verification and interactive correction between three-dimensional equipment components and two-dimensional point annotations, greatly reducing problems such as point misalignment, duplicate numbering, and configuration conflicts caused by traditional static models, and has good human-computer interaction, configuration accuracy, and engineering deployment adaptability.

[0121] S5 includes the following sub-steps S510-S550:

[0122] S510: Loading Device and Location Binding Data: The system reads the current device and location binding mapping table from the database to form a device binding dataset.

[0123] Where: D i : The i-th device model component (including number, type, etc.); P j : Corresponding location data (such as address, function code, unit, etc.); if a device has not yet been bound to a location, it is marked as "not configured".

[0124] S520: Confirm binding relationships sequentially in the interactive interface: The user clicks on each device D in the 2D / 3D interactive interface. i The system pops up a device binding information window, displaying its currently bound point P. j Key information fields, such as:

[0125] Field Item value Equipment Number WP001 Equipment type water pump Location address 0x0001 Communication Protocol Modbus-TCP Data Units ON / OFF

[0126] Users need to manually confirm whether it matches the actual physical device; if it does not match, the modification process will begin.

[0127] S530: Consistency Verification and Anomaly Message: The system calculates the attribute consistency score between the bound point and the device feature, assuming the device feature vector is F. D The point configuration vector is F P Then we have:

[0128] ConsistencyScore (D) i ,P j ) = Sim(F D ,F P )

[0129] D i Let P represent the i-th device (such as an electricity meter, sensor, air conditioner, etc.). j This indicates the location of a point in the j-th drawing or model (such as a marker point in a 2D drawing, or a component node in a 3D model); F D Indicates device D i The feature vector (e.g., device type, label, function, topological location, etc.), F P Point P represents j The feature vectors (e.g., text labels, primitive names, spatial locations, etc.) are used, where Sim(·) can be a cosine similarity algorithm based on field weights or a fuzzy matching algorithm; when the matching degree is lower than the threshold θ c When the value is 0.6 (e.g., 0.6), the system pops up a message saying "May be a binding error".

[0130] S540: Creating and Binding Locations with Unbound Devices: For devices that are not currently bound, users can click the "Add Binding" button, and the system will automatically load the device type T. i and from the modular component library Recommended compatible location templates:

[0131] Users can choose from recommended locations or fill in new locations themselves. Fields include: communication protocol; function code or register address; acquisition cycle; data type and unit.

[0132] A new binding record is created after creation:

[0133] S550: Location configuration written to database and marked as complete: After confirming that the binding is correct, the system writes the final configuration record of the device and location to the database table, and marks the status field as "confirmed as complete".

[0134]

[0135]

[0136] Through the above steps, this invention realizes a step-by-step confirmation mechanism for device location mapping relationships and a flexible ability to add bindings, effectively ensuring the accuracy of location configuration and the consistency of system configuration, significantly reducing maintenance problems such as commissioning failures and false alarms caused by misbinding or omission of devices, and has good engineering adaptability and scalability.

[0137] S6 includes the following sub-steps S610-S650:

[0138] S610: Determine test targets and generate instruction templates: The system determines the test targets and generates instruction templates based on the configured point information table. Identify the set of devices currently available for joint testing.

[0139]

[0140] For each device D i The system calls its bound communication protocol module M k Generate a standard test instruction package:

[0141] Where: P j : The point to be tested; V test Test the expected value (e.g., control "on" or "off"); The message format is encoded according to a protocol (such as Modbus, OPC).

[0142] S620: Communication Link Establishment and Command Transmission: The system registers the address (IP address) through modular components. k ,PORT k Establish a communication link with the physical device and send commands in the following manner:

[0143] Send(Cmd i → D i The system will send timestamp T send The instructions are recorded in the log system as the basis for test reconciliation.

[0144] S630: Response Acquisition and Packet Parsing: After executing a command, the device returns a response data packet (Resp). i The system parses the protocol using a protocol component.

[0145] and the expected value V test Comparison: If Δ i <∈ is considered a correct response; otherwise, a communication error is indicated.

[0146] S640: Virtual State Update and Synchronization Comparison: The system returns values ​​from physical devices. Mapped into the virtual 3D scene, the corresponding device components are updated. Status attributes such as color, brightness, and angle: The synchronized status values ​​are then compared with the point statuses set in the user interface to confirm layer or icon consistency, which is used to intuitively judge the consistency of the simulation.

[0147] S650: Joint Debugging Test Result Recording and Diagnostic Output: The system will write the following data for each communication test into the test log:

[0148]

[0149] It also categorizes abnormal situations (communication failure, timeout, inconsistent response, etc.) and provides repair suggestions, such as: checking the network link; checking for incorrect protocol configuration; and checking whether the device is offline or has a delayed response.

[0150] Through the steps described above, the system can automatically complete the entire simulation verification process from virtual device control command generation to actual device response feedback to 3D scene state updates, achieving a closed-loop verification mechanism for consistency between virtual and real systems. This method effectively avoids problems such as incorrect point configuration, mismatched protocol parameters, and incorrect device mapping relationships, greatly improving debugging efficiency and system reliability.

[0151] S7 includes the following sub-steps S710-S750:

[0152] S710: Record successful mapping relationships during joint debugging: For device point combinations that return a "success" status during joint debugging tests, the system confirms their final mapping relationship as "valid binding" and generates a device mapping list table:

[0153]

[0154] This list contains the following field: Device Number D i Point code P j The system includes: 3D model component ID; communication protocol and component name; mapping timestamp and operator information. The system exports the inventory as a structured file (such as JSON, XML, or CSV) and writes it to the system database for subsequent maintenance.

[0155] S720: Marking Failed Devices and Error Classification: For combinations of devices that failed test (D k ,P l The system analyzes and categorizes the reasons for its failure:

[0156] D k This represents a device object, which could be an electrical device, a sensor, etc.; P l This indicates the point being attempted to be bound, which could be an annotation point on a 2D drawing or a component on a 3D model, where R... k This indicates the reason for the failure, which may include: communication timeout; point resolution failure; protocol incompatibility; binding error. The system provides suggestions based on the failure type, such as recommending component replacement, address modification, or connection checking.

[0157] S730: User Correction and Feedback Collection: Users correct failed devices, including: modifying device binding points; replacing communication components; adjusting data format, address, and other parameters; correction results are recorded as: in: Original erroneous feature vectors (such as location attributes, matching methods, etc.); Correct configuration features confirmed after manual correction.

[0158] S740: Feedback samples are used for AI model optimization training: The system will use all CorrectionSamples n The samples are fed into the AI-assisted matching module to continuously optimize its recommendation algorithm. Taking a deep neural network as an example, its loss function can be expressed as:

[0159] Where: N is the total number of samples (e.g., the number of matching pairs); The model predicts the matching label; y n The tags have been manually corrected and are now accurately bound. 2 This represents the square of the Euclidean norm (i.e., the square of the vector distance). As training samples accumulate, the model will gradually improve the matching accuracy of subsequent unbound points.

[0160] S750: Dynamically updated recommendation strategy and credibility score: The system updates the recommendation strategy and credibility score for each type of device T. i With modular component C j The recommendation accuracy is dynamically statistically analyzed to form a reliable matching score matrix:

[0161]

[0162] When a component's credibility score exceeds a set threshold (e.g., 0.9), the system will prioritize recommending this component in new projects. Through the above steps, this invention not only achieves final binding confirmation between devices and locations but also constructs a feedback learning mechanism based on user operation data. This enables the AI ​​model to "learn while using," resulting in stronger automatic adaptation capabilities, configuration accuracy, and deployment efficiency in future projects, thereby driving the continuous intelligent evolution of the system.

[0163] In this embodiment of the industrial control-level configuration software system, step S1 involves the deployment and configuration of modular components. To address the issue of device communication failures caused by protocol inconsistencies or improper parameter configurations, the modular components in this system are designed with a protocol adaptive mechanism, which possesses the following key functions:

[0164] Communication failure identification mechanism: When a modular component attempts to communicate with a target device, if it does not receive a response within the expected time or receives an error frame (such as a CRC check error, illegal function code, etc.), the system determines that the current communication has failed.

[0165] Automatic switching to backup protocols: After a communication failure is confirmed, the modular component automatically selects a backup communication protocol from a predefined list of protocol candidates and retryes communication. For example, if communication using the Modbus TCP protocol fails, the system can automatically switch to one of the following protocols: Modbus RTU, OPC UA, MQTT, etc., trying them one by one in sequence until communication is successful or the attempt list is exhausted.

[0166] Automatic retry of communication parameter combinations: If communication still fails after switching protocols, the system will enter an adaptive retry mode for communication parameters, automatically combining and retrying the following parameters: baud rate (e.g., 9600, 19200, 38400, etc.); parity check (e.g., None, Even, Odd); data bits (e.g., 7 bits, 8 bits); stop bits (1 bit, 2 bits); communication address or station number; frame timeout and response waiting time, etc. This combination process follows priority rules and limits the maximum number of attempts to avoid entering invalid loops.

[0167] Successful Configuration Recording and Optimization: After successfully establishing communication through protocol switching or parameter adjustment, the modular component records the current configuration combination as the "recommended configuration" for that device type. This recommended configuration will be cached in the system database. When connecting to the same model of device in the future, the system will prioritize the application of the successfully verified protocol and parameter configuration combination to improve deployment efficiency and communication success rate.

[0168] Through the aforementioned protocol adaptive mechanism, modular components possess the ability to automatically analyze, identify, and restore communication, completing joint debugging tasks without requiring users to manually modify protocol parameters. This mechanism significantly improves the success rate of initial device access, offering substantial advantages, especially in complex environments with heterogeneous devices. It can reduce labor costs, shorten configuration cycles, and enhance system stability and intelligence.

[0169] In step S2 of this embodiment, the system automatically matches the imported location list data with the equipment models in the system. However, considering the differences in equipment naming conventions, location coding rules, and attribute fields across different projects, the system allows users to manually correct failed or incorrect matches.

[0170] To further improve the accuracy of location-equipment model matching in subsequent projects, this invention provides a matching recommendation mechanism based on reinforcement learning optimization using manually corrected results. The key implementation methods of this mechanism are as follows: Manual correction of sample records: When the system matching results do not meet user expectations, the user can manually correct the erroneous items, such as: replacing the matching target; modifying equipment type identification; adjusting the location field structure (e.g., mapping "sensor number" to "equipment primary key"), etc.

[0171] The system records each correction operation as a "training sample" in a structured manner, with the sample structure as follows:

[0172]

[0173] Sample structure vectorization: The system transforms the above samples into feature vector pairs for training the matching algorithm model. For example: original point vector x i : Includes field names, device type, encoding structure, etc.; correct device model vector y i : Indicates the structural features of the correctly matched object.

[0174] Training Sample i =(x i ,y i )

[0175] Reinforcement learning optimization strategies for model training: The system feeds accumulated corrected samples into the matching model for learning and optimization, which can employ supervised learning (such as Multilayer Perceptron, MLP), reinforcement learning (such as Q-learning or policy gradient), etc. The goal is to optimize the model parameters so that the predicted matching objects tend to align with the user's actual choices.

[0176] Example of training loss function:

[0177] in: System prediction results; y i : Manually corrected labels.

[0178] Feedback and Continuous Optimization: As manually corrected samples accumulate, the system model will continuously learn user preferences and industry characteristics, enabling the matching strategy to evolve automatically. In subsequent projects, when importing location lists, the system can prioritize referencing historical experience samples to improve the accuracy of the initial automatic matching.

[0179] By using manually corrected results as training samples to input into the model for optimization, this embodiment constructs a location matching system with "learning ability," which has the following advantages: continuous self-optimization: the system becomes "smarter" the more it is used, and the matching results are more accurate; strong cross-project transferability: the model can be transferred between different projects and adapt to various industry data standards; significantly reduced manual costs: reducing repeated corrections and improving the success rate of first-time configuration.

[0180] In this embodiment, step S3 involves importing and parsing the 3D BIM model file. To enhance the system's applicability and compatibility, the BIM model supports the parsing and adaptation of various mainstream file formats, including but not limited to: Revit format (.rvt): widely used in architectural and MEP design; MicroStation format (.dgn): commonly used in municipal, transportation, and infrastructure design.

[0181] After importing the model, the system automatically extracts the geometric information and attribute identification information of each component, such as component number, equipment name, and system classification. This attribute information will be used to automatically match with the equipment codes in the point list, establishing a binding relationship between model components and point data, providing basic support for subsequent 2D and 3D linkage and simulation.

[0182] In step S4, the user can adjust the spatial position of the device by dragging in the 3D visualization view, for example, modifying the device's coordinates (x, y, z) in the building model. After the user completes the dragging, the system automatically triggers a synchronization mechanism to update the position annotation information in the 2D drawing associated with the device in real time. The synchronization process includes the following two key actions:

[0183] Position coordinate synchronization: This involves synchronizing the device's new coordinates (x, y) in the 3D model. ' ,y ' ,z ' Map the coordinates to the 2D drawing coordinate system (x2D, y2D) and update the corresponding drawing annotation points;

[0184] Number binding verification: Check whether the equipment number of the component in the 3D model is consistent with the associated annotation number in the 2D drawing. If they are inconsistent, the user will be automatically prompted to correct the binding.

[0185] This process enables the maintenance of the positional mapping relationship between equipment in 3D and 2D graphics, ensuring information consistency between the two during configuration and maintenance, achieving two-way binding verification and automatic updates between 2D and 3D graphics, and effectively improving the efficiency and accuracy of engineering configuration.

[0186] In step S5, after the user completes the binding relationship configuration between the device locations and model components, the system will abstract and extract this binding relationship into a structured template file, thus generating a reusable binding configuration data model. Specifically, this includes:

[0187] Structured abstract modeling: Convert key information such as equipment number, point identifier, model component ID, location information, and attribute fields into standardized data structures (such as XML, JSON, or custom structure definition language) to form a binding template file;

[0188] Template export mechanism: Users can use the export function provided by the system to save the above-mentioned binding relationship template file to their local machine or project resource library;

[0189] Template reuse mechanism: When users encounter similar equipment deployment scenarios in other projects, they can directly import existing binding template files, and the system will automatically complete the point matching and binding of devices with the same structure or similar devices, greatly improving configuration efficiency.

[0190] This mechanism has excellent cross-project migration capabilities, which helps to form standardized and modular knowledge accumulation and rapid deployment capabilities in engineering.

[0191] In step S6, the system supports initiating parallel communication test tasks to multiple devices simultaneously. Specifically:

[0192] Multi-device concurrent test scheduling mechanism: The system is based on an asynchronous communication architecture, which can send test commands to multiple target devices at the same time, collect their response data, and support parallel task scheduling optimization (such as thread pool, asynchronous queue, non-blocking IO, etc.).

[0193] Real-time collection and processing of test results: The response data of each device is collected and analyzed in real time to determine key indicators such as its communication status, protocol compatibility, and data integrity.

[0194] Graphical result display mechanism: The system displays the test results in a graphical way (e.g., color coding, icon status, progress animation, etc.) in a two-dimensional / three-dimensional visual interactive interface, realizing the visual perception of the test results;

[0195] 3D model linkage feedback: When a device fails to test, the corresponding component in the 3D model can be highlighted in different colors, display a prompt message, or flash a marker to help the user locate the problem.

[0196] The above technical solution improves the efficiency and intuitiveness of the system's communication configuration verification, and is suitable for the rapid commissioning and verification needs of industrial configuration systems.

[0197] In step S7, the mapping list file generated by the system can be used as a standardized data output interface file. This file records the final binding relationship, identification number, spatial location information, etc. of the device.

[0198] This mapping list file has good scalability and compatibility, and can be integrated with the following platform systems: Asset Management System: Enables access and sharing of information such as equipment asset ledger management, maintenance records, and operating status;

[0199] Building Information Modeling for Facility Management (BIM) platform: Imports mapping data through standard interface protocols (such as IFC, COBie, or custom APIs) to achieve synchronous linkage between equipment and BIM model during the operation and maintenance phase.

[0200] Through the above-mentioned docking mechanism, the binding results of equipment during the construction phase can be extended to the operation and maintenance management phase, realizing information connectivity, visual tracking and status monitoring of equipment throughout its entire life cycle, significantly improving operation and maintenance efficiency and management accuracy.

[0201] Example 2: This example provides an industrial control-level configuration software system based on two-dimensional and three-dimensional linkage, including:

[0202] The modular component deployment module is used to select the appropriate modular components according to the type of target facilities and equipment and communication protocol, and deploy them to the server in the form of microservices, and complete the configuration of component version, deployment location, communication parameters and start / stop status;

[0203] The site import and matching module is used to import a list of sites containing information such as equipment model and serial number, and match the corresponding modular components based on the component library to establish a binding relationship between sites and components. At the same time, it calls the AI ​​matching algorithm to provide suggestions for sites that fail to match and records the results of manual correction.

[0204] The drawing and model parsing module is used to import 2D CAD drawings and 3D BIM model files, parse component geometric information and attribute data, and build the mapping foundation between drawings and models.

[0205] The linked view and position adjustment module is used to display two-dimensional and three-dimensional linked views in the visualization interface, and supports users to drag and drop devices, correct numbers, rebind points, and adjust device positions.

[0206] The site confirmation and configuration module is used to verify the consistency between the site and the model device on a device-by-device basis, and allows users to manually add and configure site information for unbound devices;

[0207] The communication integration module is used to call the communication protocol to perform communication simulation tests with real devices, and to send test commands and receive response data to verify the accuracy of 2D and 3D mapping.

[0208] The relationship confirmation and learning optimization module is used to record the mapping relationship of successfully integrated devices and generate a list. For failed items, the feedback data after the user corrects them is used to train the AI ​​algorithm to improve the accuracy of subsequent automatic matching.

[0209] A graphical visualization interface is used to display communication test results, changes in linked view operations, and system interaction feedback.

[0210] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0211] 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.

[0212] In conclusion, 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 running an industrial control level configuration software system based on two-three dimensional linkage, characterized in that, Comprising the following steps: S1: According to the facility equipment type and communication protocol, select the corresponding modular component, deploy to the server side in a micro-service manner, set the version information, deployment location, communication protocol parameters and start-stop state of the component through the basic configuration interface; S2: Import the point list data containing device model and coding information, establish the binding relationship between point data and components, call the AI assisted algorithm to suggest matching for unsuccessful matching items and record the manual correction results; S3: Import two-dimensional CAD drawings and three-dimensional BIM model files, parse the geometric data and attribute information of the components, and establish the mapping basis between the drawings and the models; S4: Load the two-three dimensional linkage view in the visualization interface, show the initial position of the equipment in the model and the corresponding point data, support users to interactively adjust the inaccurate position or mapping relationship, including dragging and moving the equipment, correcting the number and rebinding; S5: The user checks the binding information of each device in the two-three dimensional view, confirms whether the point data is consistent with the real device, and can add and configure point information from the component library for the unbound device; S6: According to the configuration result, call the communication protocol of the modular component, and communicate with the actual device for joint debugging test, send test instructions and receive response data, to verify whether the mapping and interaction between the virtual scene and the real device are accurate; S7: Record the final mapping relationship of the successfully joint tested device and generate a device mapping list; for the joint tested device, the user corrects and feeds back to the AI algorithm model, continuously optimizes the accuracy of subsequent automatic matching; S7 includes the following sub-steps: S710: For the device point combination that returns a successful state in the joint test, the system confirms the final mapping relationship as valid binding, and generates a device mapping list table: ； The list contains: equipment number ; point code ; three-dimensional model component ID; communication protocol and component name; Mapping timestamp and operator information, the system exports the list as a structured file; S720: For the device combinations that failed the test the system analyzes the reason for their failure and classifies them: ; represents a device object; represents a point of attempted binding, represents a failure reason, the system provides suggestions based on the failure type; S730: the user corrects the failed device, including: modifying the device binding point; replacing the communication component; adjusting the data format and address parameters; and the correction result is recorded as: ; wherein: : the original error feature vector; : the correct configuration feature confirmed after manual correction; S740: The system will put all samples into the AI-assisted matching module for continuous optimization of its recommendation algorithm, which is a deep neural network, and the loss function is: ; Wherein: is the total number of samples; : The matching label predicted by the model; : The real binding label after manual correction; S750: The system dynamically counts the recommendation accuracy of each type of device with the modular components and forms a reliable matching score matrix: ； When the credibility score of a component is higher than the set threshold, the system will preferentially recommend this component in new projects.

2. The method according to claim 1, wherein, The modular component in S1 supports protocol adaptive mechanism, which can automatically switch to standby communication protocol or adjust parameters for retry when device communication fails, to improve the joint debugging success rate.

3. The method according to claim 2, wherein, The manual correction results recorded in S2 can be used as training samples for reinforcement learning model to improve the matching accuracy of point data and device model.

4. The method according to claim 3, wherein, The three-dimensional BIM model imported in S3 supports multiple formats, and the model components contain attribute identification information for matching point coding.

5. The method according to claim 4, wherein, When the device is dragged in the three-dimensional view in S4, the system automatically synchronously updates the position mark and equipment number in the two-dimensional drawing, realizing bidirectional binding verification between two and three dimensions.

6. The method according to claim 5, wherein, The point binding relationship configured by the user in S5 can be abstracted as a structured template file, which can be exported and quickly reused in other projects.

7. The method according to claim 6, wherein, In S6, the system supports initiating parallel communication test tasks for multiple devices at the same time, and displays the test results in a graphical way in the two-three dimensional linkage interface.

8. The method according to claim 7, wherein, The mapping list file generated in S7 can be interfaced with the facility asset management system or the building life cycle management platform to realize the whole life cycle linkage management of the equipment.

9. A two-three dimensional linkage-based industrial control level configuration software system, adopting the two-three dimensional linkage-based industrial control level configuration software system running method in any one of claims 1-8, characterized in that, Comprising: The modular component deployment module is configured to select corresponding modular components according to the type and communication protocol of the target facility equipment, and deploy the modular components to the server in the form of microservices, and complete configuration of component version, deployment location, communication parameters, and start-stop state. The point position import and matching module is configured to import a point position list containing equipment model and number information, match corresponding modular components based on a component library, establish a binding relationship between point positions and components, and call an AI matching algorithm to give suggestions for point positions that are not successfully matched and record manual correction results. The drawing model analysis module is configured to import two-dimensional CAD drawings and three-dimensional BIM model files, analyze component geometric information and attribute data, and build a mapping basis between drawings and models. The linkage view and position adjustment module is configured to display a two-three-dimensional linkage view in a visual interface, support user dragging of equipment, correction of numbers, and rebinding of point positions, and adjust equipment positions. The point position confirmation and configuration module is configured to verify the consistency of point positions and model equipment on a device-by-device basis, and allow users to manually add and configure point position information of unbound equipment. The communication joint debugging module is configured to call a communication protocol and a real equipment to perform communication simulation testing, send testing instructions, and receive response data, and verify two-three-dimensional mapping accuracy. The relationship confirmation and learning optimization module is configured to record mapping relationships of successfully joint debugged equipment and generate a list, and use feedback data for AI algorithm training after user correction of failed items to improve subsequent automatic matching accuracy. The graphical visualization interface is configured to display communication testing results, linkage view operation changes, and system interaction feedback.

Images