A method, system, device, and medium for visualizing scenarios based on digital twinning

Abstract

The application discloses a kind of based on digital twin visual scene method, system, equipment and medium, it is related to digital twin and visual modeling technical field, including collecting multi-format source data to carry out field standard mapping, the multi-format source data after field standard mapping is executed consistency check;The multi-format source data after consistency check is written into to be fused buffer, selects model component data in to-be-fused buffer and executes coordinate reference conversion, is associated with structured graphic information by building identification field after the model component data of conversion and structured graphic information are bound;Model component data and structured graphic information after association and binding are synchronously loaded into pre-defined template container, scene configuration file is generated according to container structure, and is registered to multi-end rendering service channel by calling publishing engine.The method is disclosed in the application by building affine transformation coordinate reference conversion scheme, the accurate alignment of multi-format model component under unified space coordinate system is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twins and visual modeling, and specifically to a visualization scene method, system, device and medium based on digital twins. Background Art

[0002] With the continuous development of information technology, the Internet of Things, and 3D visualization technologies, digital twins, a new modeling concept that integrates multi-source data, virtual-reality interaction, and dynamic simulation, have been widely applied in various fields, including urban management, industrial simulation, and building information modeling. The construction of digital twin scenarios typically requires the unified integration of large amounts of heterogeneous data and the visualization of this data in a 3D spatial environment, enabling holographic representation and dynamic monitoring of physical entities.

[0003] Existing technologies still lack uniformity in data collection and processing methods when building digital twin visualization scenarios. Multi-source data have significant differences in format, field structure, and semantic expression. There is a lack of efficient field mapping and data standardization mechanisms, resulting in low fusion efficiency and high manual participation.

[0004] There is a lack of semantic hierarchical binding mechanism between model components and business data. Three-dimensional models often exist only as geometric forms, and it is difficult to establish structured and retrievable associations with business forms, sensor data or image materials, which limits the interactive capabilities of the scene.

[0005] The scene construction process lacks universality and automation. Most existing systems rely on manual methods to complete model loading, layer configuration and attribute attachment. They lack a unified data structure and automated configuration process, resulting in low construction efficiency and poor adaptability.

[0006] The time dimension is insufficiently supported. Most visualization systems lack temporal binding and dynamic switching capabilities for component information, making historical backtracking and timeline analysis impossible. There is an urgent need to develop a visualization scenario construction method that supports semantic fusion and dynamic interaction for complex multi-source data scenarios. This method can improve the automation, interaction capabilities, and operational efficiency of scenario construction, thereby promoting the implementation of digital twin technology in various industries. Summary of the Invention

[0007] In view of the above-mentioned problems, the present invention is proposed.

[0008] Therefore, the technical problem solved by the present invention is: the existing digital twin visualization construction method has the following problems: the data fusion process relies on manual processing, there is a lack of structured binding mechanism between model components and graphic information, the temporal display capability between components is weak, and the automation and dynamic interaction capabilities are low. In addition, how to achieve spatial unified expression of multi-source model components while maintaining data consistency, and realize efficient semantic binding and dynamic mapping of time periods with graphic data based on component identification fields.

[0009] In order to solve the above technical problems, the present invention provides the following technical solution: collecting multi-format source data for field standard mapping, and performing consistency check on the multi-format source data after field standard mapping.

[0010] The multi-format source data after consistency verification is written into the buffer to be fused, the model component data is selected in the buffer to be fused to perform coordinate base conversion, and the converted model component data is associated and bound with the structured graphic information by constructing an identification field.

[0011] The associated and bound model component data and structured graphic information are synchronously loaded into the predefined template container, a scene configuration file is generated according to the container structure, and the publishing engine is called to register with the multi-terminal rendering service channel.

[0012] Constructing identification fields for associative binding includes adopting a coordinate reference conversion method based on affine transformation to achieve unified spatial expression of multi-source model components, and realizing integrated semantic binding of structured graphic information and model component data by constructing identification fields, including many-to-one and time period dynamic mapping binding mechanisms.

[0013] As a preferred solution of the digital twin-based visualization scene method described in the present invention, the acquisition of multi-format source data includes receiving structured and unstructured data files uploaded by users, writing the data files into the original data cache, and assigning a unique data number to each piece of data.

[0014] Perform format recognition operations on the received data files, call the built-in parser to parse the field structure according to data type, parse structured data into key-value pairs in row and column form, extract key fields from unstructured data through regular templates and perform paragraph semantic annotation through natural language processing, extract meta-field information and uniformly format it into a JSON data structure.

[0015] Data files include business ledgers in XLSX format, report documents in DOCX format, vector maps in SHAPEFILE format, building models in IFC format, and structured graphic information materials in PDF. The platform access module provides three paths: interface upload, WebService interface access, and IoT device port access.

[0016] Field standard mapping includes calling the platform's built-in field mapping rule table. The rule table stores the correspondence between the platform's standard fields and common business fields. Each standard field corresponds to multiple industry-common field names. The mapping relationship between the uploaded fields and the standard fields is determined through field name similarity matching and semantic keyword analysis.

[0017] When the field name cannot be matched directly, the platform adopts an auxiliary field identification strategy, using the data content type, unit, and value range characteristics to assist in determining the mapping relationship.

[0018] After the mapping is completed, the system automatically generates a field conversion log and a field original-mapping comparison table, and writes the standardized data after field unification into the structure field pool.

[0019] As a preferred solution of the digital twin-based visualization scene method described in the present invention, the consistency check includes verifying the data type of each data file, checking the field integrity of the data records, detecting whether all required fields are missing, automatically filling in the default values, and marking the records as abnormal records when the missing values ​​are serious and not entering the subsequent processing flow.

[0020] Data type verification includes checking the consistency of tables, vectors, texts, model components, files of different formats with the system supported format list, and refusing to accept incompatible file types.

[0021] As a preferred solution of the digital twin-based visualization scene method described in the present invention, the execution of coordinate reference conversion includes parsing the original coordinate system information of the selected model component data. When the model component data lacks a complete coordinate definition, the platform calls the default coordinate inference rule to estimate the coordinate system.

[0022] The original coordinate system information includes the projection definition information embedded in the model component data and the coordinate code manually marked by the user.

[0023] According to the defined relationship between the original coordinate system and the platform unified target coordinate system, an affine transformation matrix is ​​constructed and the coordinate point transformation is performed using a matrix expression.

[0024] The affine transformation matrix includes the rotation matrix, the scaling matrix and the translation vector.

[0025] When there is place name and address information in the model component data, perform forward and reverse parsing of the place name to parse the address, convert the address field into longitude and latitude coordinates, and reverse the longitude and latitude to the standard address structure.

[0026] After the conversion is completed, the position changes of the model component data before and after the conversion are displayed through the visual preview interface, and the user is allowed to fine-tune or manually calibrate the coordinate offset value. The model component data coordinates after manual confirmation are written into the scene construction data area.

[0027] As a preferred solution of the digital twin-based visualization scene method described in the present invention, the associative binding by constructing identification fields includes performing one-to-one and one-to-many binding operations on the converted model component data and structured graphic information.

[0028] The converted model component data includes the parsed and pre-processed model component data, and has a unique component identification field.

[0029] Structured graphic information includes business attribute forms, on-site photos, process documents, sensor metadata, and video stream source addresses. Structured data entries with clear primary key fields are extracted through the data import process.

[0030] The platform matches the converted model component data identification fields with the corresponding fields in the structured graphic information. When the field names are inconsistent, the field mapping table is called for mapping and matching.

[0031] Matching rules include three conditions: field synonyms, data type consistency, and structural context consistency.

[0032] When multiple structured graphic information records meet the same component matching conditions, the binding conflict judgment mechanism is executed and the main binding is performed in the order of sensor data > video image > image document > table data. The data that has not been mainly bound is written into the extended information pool as auxiliary binding.

[0033] The platform generates a component binding relationship table, which records the component ID, binding data type, data path, data source identifier and binding timestamp information, and writes the relationship table into the database.

[0034] When the structured graphic information contains a timestamp field, the binding relationship can be set as a dynamic mapping of time periods.

[0035] Setting the binding relationship as a dynamic mapping of time periods includes loading different graphic and text contents in different time periods on the same component, thereby forming a time semantic binding mechanism.

[0036] After all components are bound, a visual interface is provided to support component binding verification. Users can view the binding structured graphic information content by clicking on the model component. If there is an incorrect binding, it can be manually released and the binding source can be re-specified. All manual operations are recorded in the binding operation log table.

[0037] As a preferred solution of the digital twin-based visualization scene method described in the present invention, the synchronous loading into the predefined template container includes reading a data set of component identification, binding graphic content, business fields, and spatial positioning information, loading the component data into the template container according to the component group to which it belongs, and attaching the corresponding structured graphic information to the associated component in the form of a binding structure.

[0038] The geometric information of each component is written into the component display area, the bound business fields are attached to the component binding point in the form of configurable pop-up attributes, the structured graphic information is encoded into the image layer area, and configured as a switchable hot zone display item according to the associated timestamp.

[0039] When the information bound to a component has a time period attribute, a component-information index table is generated in chronological order, and the displayed graphic and text content is dynamically switched according to the current time status during scene operation.

[0040] Each time a component and structured graphic information is loaded, the spatial location, component number, data path and time range are recorded, and a loading index item is established to build a complete component visual reference structure.

[0041] Traverse all add-ins to perform data consistency checks. After the checks pass, all loaded results are written to the scene configuration list.

[0042] Consistency checks include location duplication detection, layer mapping integrity check, component reference uniqueness verification, and resource path validity verification.

[0043] After loading is completed, scene preview data is generated. Users can make personalized modifications during the preview phase, and all configuration results are written into the component configuration item table.

[0044] The preview data includes support for component style adjustment, layer display order configuration, and graphic and text interaction behavior definition.

[0045] As a preferred solution of the digital twin-based visualization scene method described in the present invention, the generation of the scene configuration file includes extracting the component space information, business binding fields, graphic resource paths, layer identifiers and interaction parameters loaded into the template container, organizing the content into a hierarchical structure based on the granularity of component units, and writing it into the scene configuration list in the form of key-value pairs.

[0046] Map all component numbers, generate a component index table, and build a quick access structure from component ID to layer data path, binding data path and component status attributes.

[0047] Read the viewing parameters, component visibility, hotspot display rules, plug-in activation options, and layer order settings adjusted by the user during the preview phase, and merge these configuration parameters into the scene configuration data set;

[0048] Generate a list of scene startup parameters, including the latitude and longitude coordinates of the initial positioning perspective, scene zoom level, lighting rendering parameters, background base map selection, component loading order priority, and interaction event binding path.

[0049] Construct a complete scene description structure, which is expressed in JSON object structure format.

[0050] Save the scene configuration structure in JSON object format as a configuration file named SceneManifest.json. The file contains the absolute paths of all components and resources, business attribute reference paths, real-time data subscription addresses, and the start and end ranges of the time period dynamic binding logic.

[0051] After the scene configuration file is saved, it is written into the file directory index structure and a unique identifier is registered in the platform. After the registration is completed, the scene packaging operation is triggered, and the configuration file together with the component resources, structured graphic information, and plug-in scripts are packaged into a scene deployment package.

[0052] Another object of the present invention is to provide a visualization scene system based on digital twins, which can realize the integrated semantic fusion of model components and structured graphic information through a graphic binding mechanism based on component identification fields and an affine coordinate transformation strategy, thereby solving the problem of the separation of model data and business data and the inability to establish a computable association relationship in the current digital twin visualization construction technology.

[0053] As a preferred solution of the digital twin-based visualization scene system described in the present invention, it includes: an acquisition mapping verification module, a benchmark conversion and association binding module, and a synchronous loading and generation of scene configuration file module.

[0054] The acquisition mapping verification module is used to acquire multi-format source data for field standard mapping, and perform consistency verification on the multi-format source data after field standard mapping.

[0055] The reference conversion and association binding module is used to write the multi-format source data after consistency verification into the buffer to be fused, select model component data in the buffer to be fused to perform coordinate reference conversion, and associate and bind the converted model component data with structured graphic information by constructing an identification field.

[0056] The synchronous loading and scene configuration file generation module is used to synchronously load the associated and bound model component data and structured graphic information into the predefined template container, generate the scene configuration file according to the container structure, and call the publishing engine to register with the multi-terminal rendering service channel.

[0057] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a step of a visualization scene method based on digital twins.

[0058] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a digital twin-based scene visualization method.

[0059] Beneficial effects of the present invention: The digital twin-based visualization scene method provided by the present invention achieves precise alignment of multi-format model components in a unified spatial coordinate system by constructing an affine transformation coordinate base conversion scheme, effectively improving the accuracy of spatial data fusion. Through a structured graphic information binding scheme driven by component identification fields, an integrated semantic relationship between the model and business data is established, significantly enhancing the query configurability and time dimension expression capabilities of the scene. The present invention has achieved better results in heterogeneous data fusion, component semantic expression, and dynamic scene configuration. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0061] Figure 1 An overall flow chart of a digital twin-based scene visualization method provided for the first embodiment of the present invention.

[0062] Figure 2 This is a flowchart of an operation of a digital twin-based scene visualization method provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0063] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0064] Example 1, with reference to Figure 1-2 , as an embodiment of the present invention, provides a visualization scene method based on digital twins, including:

[0065] S1: Collect multi-format source data for field standard mapping, and perform consistency check on the multi-format source data after field standard mapping.

[0066] Receive structured and unstructured data files uploaded by users, write the data files into the original data buffer, and assign a unique data number to each data.

[0067] Perform format recognition operations on the received data files, call the built-in parser to parse the field structure according to data type, parse structured data into key-value pairs in row and column form, extract key fields from unstructured data through regular templates and perform paragraph semantic annotation through natural language processing, extract meta-field information and uniformly format it into a JSON data structure.

[0068] A preferred solution for unified formatting into JSON data structure is:

[0069] To uniformly represent and process structured and unstructured data in subsequent processes, the platform needs to encapsulate the data in a unified data structure format after completing standard field mapping. To this end, the following JSON structure standard representation scheme is proposed as the preferred structure to enhance data parsability and cross-module call versatility.

[0070] ;

[0071] in, Indicates the standardized structured data records, Indicates the The first field names, express The corresponding field value.

[0072] Data files include business ledgers in XLSX format, report documents in DOCX format, vector maps in SHAPEFILE format, building models in IFC format, and structured graphic information materials in PDF. The platform access module provides three paths: interface upload, WebService interface access, and IoT device port access.

[0073] Field standard mapping includes calling the platform's built-in field mapping rule table. The rule table stores the correspondence between the platform's standard fields and common business fields. Each standard field corresponds to multiple industry-common field names. The mapping relationship between the uploaded fields and the standard fields is determined through field name similarity matching and semantic keyword analysis.

[0074] When the field name cannot be matched directly, the platform adopts an auxiliary field identification strategy, using the data content type, unit, and value range characteristics to assist in determining the mapping relationship.

[0075] After the mapping is completed, the system automatically generates a field conversion log and a field original-mapping comparison table, and writes the standardized data after field unification into the structure field pool.

[0076] Perform data type verification on each data file, perform field integrity check on data records, detect whether all required fields are missing, automatically fill in default values, and mark the record as an abnormal record if there are serious missing values ​​and temporarily prevent it from entering the subsequent processing flow.

[0077] Data type verification includes checking the consistency of tables, vectors, texts, model components, files of different formats with the system supported format list, and refusing to accept incompatible file types.

[0078] Furthermore, for structured tables and vector data, the primary key uniqueness verification subroutine is called to quickly compare all primary key field values ​​by establishing a primary key hash table, find duplicate records and generate duplicate data reports. All duplicate primary key records must be reviewed by the administrator and the handling method must be decided.

[0079] A preferred solution for quickly comparing all primary key field values ​​by establishing a primary key hash table is:

[0080] During primary key uniqueness verification, the platform uses a hash table structure to construct primary key mappings to enable rapid indexing and duplicate detection for large datasets. The following formula is a preferred representation of this hash mechanism, which helps to promptly detect duplicate records during the data import phase.

[0081]

[0082] in, Represents a primary key hash index table, Indicates the The identification field of a component, express and The primary key field value is repeated, Indicates that these two records are not the same.

[0083] For spatial data and model component data, the coordinate field consistency verification program is used to confirm that all coordinate fields use the platform's unified coordinate system, and the coordinate conversion process is automatically triggered for inconsistent coordinate data.

[0084] If time series data is involved, call the time sequence verification submodule to check whether the timestamp of each record meets the monotonically increasing requirements, and compare it with the time interval of the relevant business master data to check for abnormal time periods or discontinuous records.

[0085] After the verification is completed, the data records that pass all consistency rules will be written into the structured data pool, and an exception report file will be generated for the data that fails the consistency verification. The abnormal data will be archived separately for manual verification. All verification operation processes and results will be recorded in the system verification log table.

[0086] S2: Write the multi-format source data after consistency verification into the buffer to be fused, select the model component data in the buffer to be fused to perform coordinate reference conversion, and associate and bind the converted model component data with the structured graphic information by constructing an identification field.

[0087] The original coordinate system information of the selected model component data is parsed. When the model component data lacks a complete coordinate definition, the platform calls the default coordinate inference rule to estimate the coordinate system.

[0088] The original coordinate system information includes the projection definition information embedded in the model component data and the coordinate code manually marked by the user.

[0089] According to the defined relationship between the original coordinate system and the platform unified target coordinate system, an affine transformation matrix is ​​constructed and the coordinate point transformation is performed using a matrix expression.

[0090] A preferred solution for coordinate point conversion is:

[0091] To ensure that heterogeneous model components are spatially aligned to a unified coordinate system, an affine transformation is performed on the original coordinate information. The following matrix expression is the preferred affine transformation scheme adopted, covering the rotation, scaling, and translation processes.

[0092] ;

[0093] in, Represents the original coordinate point, Represents the coordinate point after transformation, represents the affine transformation matrix, represents the rotation matrix, represents the scaling matrix, Represents the translation vector.

[0094] The affine transformation matrix includes the rotation matrix, the scaling matrix and the translation vector.

[0095] When there is place name and address information in the model component data, perform forward and reverse parsing of the place name to parse the address, convert the address field into longitude and latitude coordinates, and reverse the longitude and latitude to the standard address structure.

[0096] A preferred solution for reversing longitude and latitude into a standard address structure is:

[0097] When matching component fields with graphic information fields, the platform adopts a three-dimensional consistency judgment strategy, including field name synonymy, data type consistency, and structural context consistency.

[0098] ;

[0099] in, Indicates the The data address field of each model component, , Indicates the The longitude and latitude coordinates obtained by parsing the address fields, represents the forward parsing function of place names, Represents the reverse parsing function of place names.

[0100] After the conversion is completed, the position changes of the model component data before and after the conversion are displayed through the visual preview interface, and the user is allowed to fine-tune or manually calibrate the coordinate offset value. The model component data coordinates after manual confirmation are written into the scene construction data area.

[0101] A preferred solution for fine-tuning is:

[0102] Considering the possible deviation of the transformed coordinates, a manual visualization fine-tuning mechanism is provided to allow users to perform secondary corrections through the offset vector.

[0103] ;

[0104] in, Indicates the final confirmed coordinates, which will be written into the scene construction data area. Indicates the user's control of the Manually fine-tune the offset vector of each component.

[0105] Perform one-to-one and one-to-many binding operations on the converted model component data and structured graphic information.

[0106] The converted model component data includes the parsed and pre-processed model component data, and has a unique component identification field.

[0107] Structured graphic information includes business attribute forms, on-site photos, process documents, sensor metadata, and video stream source addresses. Structured data entries with clear primary key fields are extracted through the data import process.

[0108] The platform matches the converted model component data identification fields with the corresponding fields in the structured graphic information. When the field names are inconsistent, the field mapping table is called for mapping and matching.

[0109] A preferred solution for calling the field mapping table for mapping matching is:

[0110] When matching model component fields with graphic information fields, the platform adopts the triple rules of field synonymy, data type consistency, and context structure consistency.

[0111] ;

[0112] in, Represents the model component field, Indicates the structured graphic information field. Indicates whether the field name is a synonym. Indicates the data type of the field, Represents the structure context of a field.

[0113] Matching rules include three conditions: field synonyms, data type consistency, and structural context consistency.

[0114] When multiple structured graphic information records meet the same component matching conditions, the binding conflict judgment mechanism is executed and the main binding is performed in the order of sensor data > video image > image document > table data, and other data is written into the extended information pool as auxiliary binding.

[0115] A preferred solution for executing the binding conflict judgment mechanism is:

[0116] If there are multiple graphic data binding candidates for the same component, a scoring mechanism is built based on the data type, and a scoring function is used to determine the primary binding object.

[0117] ;

[0118] in, Indicates a certain graphic data record to be bound. Indicates whether the image and text data contains sensor data. Indicates whether to include video stream information. Indicates whether it contains picture / image documents. Indicates whether it contains tabular or structured numerical data, 、 、 、 Indicates the weighted coefficients of the four types of information in the score, which are used for priority sorting. Indicates data record The binding priority score value.

[0119] The platform generates a component binding relationship table, which records the component ID, binding data type, data path, data source identifier and binding timestamp information, and writes the relationship table into the database.

[0120] When the structured graphic information contains a timestamp field, the binding relationship can be set as a dynamic mapping of time periods.

[0121] Setting the binding relationship as a dynamic mapping of time periods includes loading different graphic and text contents in different time periods on the same component, thereby forming a time semantic binding mechanism.

[0122] A preferred solution for setting the binding relationship as a dynamic mapping of time periods is:

[0123] It supports dynamic mapping of components and graphic content in different time periods, introduces a binding structure based on time intervals, and realizes time-series content switching.

[0124] ;

[0125] in, Representation component In time Corresponding to the bound data content, Represents any point in time, Indicates the start time of binding data. Indicates the end time of the binding data.

[0126] After all components are bound, a visual interface is provided to support component binding verification. Users can view the binding structured graphic information content by clicking on the model component. If there is an incorrect binding, it can be manually released and the binding source can be re-specified. All manual operations are recorded in the binding operation log table.

[0127] S3: Synchronously load the associated and bound model component data and structured graphic information into the predefined template container, generate a scene configuration file based on the container structure, and call the publishing engine to register with the multi-terminal rendering service channel.

[0128] Read the data set of component identification, binding graphic content, business fields, and spatial positioning information, load the component data into the template container according to the component group to which it belongs, and attach the corresponding structured graphic information to the associated component in the form of a binding structure.

[0129] The geometric information of each component is written into the component display area, the bound business fields are attached to the component binding point in the form of configurable pop-up attributes, the structured graphic information is encoded into the image layer area, and configured as a switchable hot zone display item according to the associated timestamp.

[0130] When the information bound to a component has a time period attribute, a component-information index table is generated in chronological order, and the displayed graphic and text content is dynamically switched according to the current time status during scene operation.

[0131] Each time a component and structured graphic information is loaded, its spatial position, component number, data path and time range are recorded, and a loading index item is established to build a complete component visual reference structure.

[0132] A preferred solution for building a complete component visual reference structure is:

[0133] During the process of building the scene configuration, the component path, status and interaction parameters are recorded to build a reference index, and the platform defines a triple structure to support scene loading.

[0134] ;

[0135] in, Representation component The resource index item, Indicates the component resource path, Indicates the component status. Represents the component interaction plug-in identifier and additional properties.

[0136] Traverse all add-ins to perform data consistency checks. After the checks pass, all loaded results are written to the scene configuration list.

[0137] Consistency checks include location duplication detection, layer mapping integrity check, component reference uniqueness verification, and resource path validity verification.

[0138] After loading is completed, scene preview data is generated, allowing users to make personalized modifications during the preview stage, and all configuration results are written into the component configuration item table.

[0139] The preview data includes support for component style adjustment, layer display order configuration, and graphic and text interaction behavior definition.

[0140] Extract the component space information, business binding fields, graphic resource paths, layer identifiers, and interaction parameters loaded into the template container, organize the content into a hierarchical structure based on the component unit granularity, and write it into the scene configuration list in the form of key-value pairs.

[0141] Map all component numbers, generate a component index table, and build a quick access structure from component ID to layer data path, binding data path and component status attributes.

[0142] Read the viewing parameters, component visibility, hotspot display rules, plug-in activation options, and layer order settings adjusted by the user during the preview phase, and merge these configuration parameters into the scene configuration data set.

[0143] Generate a list of scene startup parameters, including the latitude and longitude coordinates of the initial positioning perspective, scene zoom level, lighting rendering parameters, background base map selection, component loading order priority, and interaction event binding path.

[0144] Construct a complete scene description structure, which is expressed in JSON object structure format.

[0145] A preferred solution for constructing a complete scene description structure is:

[0146] After the scene is built, all components and their indexes are encapsulated into a unified configuration structure for cross-platform deployment and parsing.

[0147] ;

[0148] in, Represents the scene main configuration structure, Indicates the unique number of the scene. Represents component interaction event mapping, Indicates the unique number of the scene. Represents scene rendering parameters.

[0149] Save the scene configuration structure in JSON object format as a configuration file named "SceneManifest.json". The file contains the absolute paths of all components and resources, business attribute reference paths, real-time data subscription addresses, and the start and end ranges of the time period dynamic binding logic.

[0150] After the scene configuration file is saved, it is written into the file directory index structure and a unique identifier is registered in the platform. After the registration is completed, the scene packaging operation is triggered, and the configuration file together with the component resources, structured graphic information, and plug-in scripts are packaged into a scene deployment package.

[0151] Example 2 is an embodiment of the present invention, which provides a visualization scene system based on digital twins, including an acquisition mapping verification module, a benchmark conversion and association binding module, and a synchronous loading and generation of scene configuration file module.

[0152] The acquisition mapping verification module is used to acquire multi-format source data for field standard mapping and perform consistency verification on the multi-format source data after field standard mapping.

[0153] It should also be noted that the data tags and component IDs generated by the acquisition mapping verification module are used as input for coordinate transformation and field matching by the reference transformation and association binding module.

[0154] The coordinate conversion is based on the original coordinate field and unit judgment results output by the acquisition mapping verification module.

[0155] The image and text binding field matching uses the field standard mapping comparison table in the collection mapping verification module as the matching criteria.

[0156] The datum conversion and association binding module is used to write the multi-format source data after consistency verification into the buffer to be fused, select the model component data in the buffer to be fused to perform coordinate datum conversion, and associate and bind the converted model component data with the structured graphic information by constructing an identification field.

[0157] It should also be noted that the reference conversion and association binding module in the binding relationship structure as a synchronous load and generate scene configuration file module text loading scheduling basis.

[0158] All components ID, spatial position, time period, text path in the reference conversion and association binding module has been defined complete, synchronous load and generate scene configuration file synchronous load and generate scene configuration file module directly quoted.

[0159] Component state and layer hanging priority is written by the reference conversion and association binding module output.

[0160] Synchronous load and generate scene configuration file module to reference conversion and association binding module output binding information generated interactive index.

[0161] Synchronous load and generate scene configuration file module reads the reference conversion and association binding module in the time period structure, form dynamic rendering scheduling strategy.

[0162] Synchronous load and generate scene configuration file module for the association after the model component data and structured text information synchronization to the pre-defined template container, according to the container structure to generate scene configuration file, call publishing engine registered to the multi-terminal rendering service channel.

[0163] It should also be noted that the synchronous load and generate scene configuration file module component configuration table in the field for display, interactive pop-up window called data field, directly from the acquisition mapping verification module module of the unified field structure pool.

Claims

1. A visualization scene method based on digital twins, characterized in that: include: Collect multi-format source data for field standard mapping, and perform consistency check on the multi-format source data after field standard mapping; Writing the multi-format source data after consistency verification into the buffer to be fused, selecting the model component data in the buffer to be fused to perform coordinate base conversion, and associating and binding the converted model component data with the structured graphic information by constructing an identification field; The association binding by constructing the identification field includes: Perform one-to-one or one-to-many binding operations on the converted model component data and structured graphic information; The converted model component data includes the parsed and pre-processed model component data, and has a unique component identification field; Structured graphic information includes business attribute forms, on-site photos, process documents, sensor metadata, and video stream source addresses. Structured data entries with clear primary key fields are extracted through the data import process. The platform matches the converted model component data identification fields with the corresponding fields in the structured graphic information. If the field names are inconsistent, the field mapping table is called for mapping and matching. Matching rules include three conditions: field synonyms, data type consistency, and structural context consistency; When multiple structured graphic information records meet the same component matching conditions, the binding conflict judgment mechanism is executed in the order of sensor data > video image > image document > table data, and the primary binding is performed. The data that has not been primarily bound is written into the extended information pool as auxiliary binding; The platform generates a component binding relationship table that records the component ID, binding data type, data path, data source identifier, and binding timestamp information, and writes the relationship table into the database; When the structured graphic information contains a timestamp field, the binding relationship can be set as a dynamic mapping of time periods; Setting the binding relationship as a dynamic mapping of time periods includes loading different graphic and text contents in different time periods on the same component to form a time semantic binding mechanism; After all components are bound, a visual interface is provided to support component binding verification. Users can view the binding structured graphic information by clicking on the model component. If there is an incorrect binding, they can manually release it and re-specify the binding source. All manual operations are recorded in the binding operation log table. Synchronously load the associated and bound model component data and structured graphic information into the predefined template container, generate a scene configuration file based on the container structure, and call the publishing engine to register with the multi-terminal rendering service channel; Constructing identification fields for associative binding includes adopting a coordinate reference conversion method based on affine transformation to achieve unified spatial expression of multi-source model components, and realizing integrated semantic binding of structured graphic information and model component data by constructing identification fields, including many-to-one and time period dynamic mapping binding mechanisms.

2. The digital twin-based scene visualization method according to claim 1, characterized in that: The acquisition of multi-format source data include, Receive structured and unstructured data files uploaded by users, write the data files into the original data buffer, and assign a unique data number to each data; Perform format recognition on the received data files, call the built-in parser to parse the field structure according to data type, parse structured data into key-value pairs in row and column format, extract key fields from unstructured data through regular templates and perform paragraph semantic annotation through natural language processing, extract meta-field information and uniformly format it into a JSON data structure; Data files include business records in XLSX format, report documents in DOCX format, vector maps in SHAPEFILE format, building models in IFC format, and structured graphic information materials in PDF format. The platform access module provides three paths: interface upload, WebService interface access, and IoT device port access. Field standard mapping involves calling the platform's built-in field mapping rule table, which stores the correspondence between platform standard fields and common business fields. Each standard field corresponds to multiple industry-common field names. The mapping relationship between uploaded fields and standard fields is determined through field name similarity matching and semantic keyword analysis. When the field name cannot be directly matched, the platform adopts an auxiliary field identification strategy, using the data content type, unit, and value range characteristics to assist in determining the mapping relationship; After the mapping is completed, the system automatically generates a field conversion log and a field original-mapping comparison table, and writes the standardized data after field unification into the structure field pool.

3. The digital twin-based scene visualization method according to claim 2, characterized in that: The performing consistency check includes: Verify the data type of each data file, perform field integrity checks on data records, detect whether all required fields are missing, automatically fill in default values, and mark records as abnormal records if there are serious missing values ​​and temporarily prevent them from entering the subsequent processing flow; Data type verification includes checking the consistency of tables, vectors, texts, model components, files of different formats with the system supported format list, and refusing to accept incompatible file types.

4. The digital twin-based scene visualization method according to claim 3, characterized in that: The performing of the coordinate base transformation includes: Analyze the original coordinate system information of the selected model component data. When the model component data lacks a complete coordinate definition, the platform uses the default coordinate inference rules to estimate the coordinate system. The original coordinate system information includes the projection definition information embedded in the model component data and the coordinate code manually marked by the user; According to the definition relationship between the original coordinate system and the platform unified target coordinate system, the affine transformation matrix is ​​constructed and the coordinate point transformation is performed using the matrix expression; Affine transformation matrices include rotation matrices, scaling matrices, and translation vectors; When the model component data contains place name and address information, perform forward and reverse parsing of the place name to parse the address, convert the address field into longitude and latitude coordinates, and reverse the longitude and latitude to the standard address structure; After the conversion is completed, the position changes of the model component data before and after the conversion are displayed through the visual preview interface, and the user is allowed to fine-tune or manually calibrate the coordinate offset value. The model component data coordinates after manual confirmation are written into the scene construction data area.

5. The digital twin-based scene visualization method according to claim 4, characterized in that: The synchronous loading into the predefined template container includes: Read the data set of component identification, binding graphic content, business fields, and spatial positioning information, load the component data into the template container according to the component group to which it belongs, and attach the corresponding structured graphic information to the associated component in the form of a binding structure; Write the geometric information of each component into the component display area, attach the bound business fields to the component binding point in the form of configurable pop-up attributes, encode the structured graphic information into the image layer area, and configure it as a switchable hot zone display item based on the associated timestamp; When the information bound to a component has a time period attribute, a component-information index table is generated in chronological order, and the displayed graphic and text content is dynamically switched according to the current time status during the scene operation; Each time a component and structured graphic information is loaded, the spatial location, component number, data path and time range are recorded, and a loading index item is established to build a complete component visual reference structure; Traverse all add-ins to perform data consistency checks. After the checks pass, write all loaded results into the scene configuration list; Consistency checks include location duplication detection, layer mapping integrity check, component reference uniqueness check, and resource path validity verification; After loading is completed, scene preview data is generated. Users can make personalized modifications during the preview phase, and all configuration results are written into the component configuration item table. The preview data includes support for component style adjustment, layer display order configuration, and graphic and text interaction behavior definition.

6. The digital twin-based scene visualization method according to claim 5, characterized in that: The generating of the scenario configuration file comprises: Extract component space information, business binding fields, graphic resource paths, layer identifiers, and interaction parameters loaded into the template container, organize the content into a hierarchical structure based on component units, and write it into the scene configuration list in the form of key-value pairs; Map all component numbers, generate a component index table, and build a quick access structure from component ID to layer data path, binding data path, and component status attributes; Read the viewing angle parameters, component visibility, hotspot display rules, plug-in activation options, and layer order settings adjusted by the user during the preview phase, and merge these configuration parameters into the scene configuration data set; Generate a list of scene startup parameters, including the latitude and longitude coordinates of the initial positioning perspective, scene zoom level, lighting rendering parameters, background base map selection, component loading order priority, and interaction event binding path; Construct a complete scene description structure, which is expressed in JSON object structure format; Save the scene configuration structure in JSON object format as a configuration file named SceneManifest.json. The file contains the absolute paths of all components and resources, business attribute reference paths, real-time data subscription addresses, and the start and end ranges of the time period dynamic binding logic; After the scene configuration file is saved, it is written into the file directory index structure and a unique identifier is registered in the platform. After the registration is completed, the scene packaging operation is triggered, and the configuration file together with the component resources, structured graphic information, and plug-in scripts are packaged into a scene deployment package.

7. A digital twin-based visualization scene system, characterized by: It includes acquisition mapping verification module, benchmark conversion and association binding module, synchronous loading and generation of scene configuration file module; The acquisition mapping verification module is used to acquire multi-format source data for field standard mapping and perform consistency verification on the multi-format source data after field standard mapping; The reference conversion and association binding module is used to write the multi-format source data after consistency verification into the buffer to be fused, select the model component data in the buffer to be fused to perform coordinate reference conversion, and associate and bind the converted model component data with the structured graphic information by constructing an identification field; The association binding by constructing the identification field includes: Perform one-to-one or one-to-many binding operations on the converted model component data and structured graphic information; The converted model component data includes the parsed and pre-processed model component data, and has a unique component identification field; Structured graphic information includes business attribute forms, on-site photos, process documents, sensor metadata, and video stream source addresses. Structured data entries with clear primary key fields are extracted through the data import process. The platform matches the converted model component data identification fields with the corresponding fields in the structured graphic information. If the field names are inconsistent, the field mapping table is called for mapping and matching. Matching rules include three conditions: field synonyms, data type consistency, and structural context consistency; When multiple structured graphic information records meet the same component matching conditions, the binding conflict judgment mechanism is executed in the order of sensor data > video image > image document > table data, and the primary binding is performed. The data that has not been primarily bound is written into the extended information pool as auxiliary binding; The platform generates a component binding relationship table, which records the component ID, binding data type, data path, data source identifier and binding timestamp information, and writes the relationship table into the database; When the structured graphic information contains a timestamp field, the binding relationship can be set as a dynamic mapping of time periods; Setting the binding relationship as a dynamic mapping of time periods includes loading different graphic and text contents in different time periods on the same component to form a time semantic binding mechanism; After all components are bound, a visual interface is provided to support component binding verification. Users can view the binding structured graphic information by clicking on the model component. If there is an incorrect binding, they can manually release it and re-specify the binding source. All manual operations are recorded in the binding operation log table. The synchronous loading and scene configuration file generation module is used to synchronously load the associated and bound model component data and structured graphic information into the predefined template container, generate the scene configuration file according to the container structure, and call the publishing engine to register with the multi-terminal rendering service channel.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the digital twin-based scene visualization method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the digital twin-based scene visualization method according to any one of claims 1 to 6 are implemented.

Images