Digital twin integrated system

Abstract

The invention belongs to the technical field of data twinning, and discloses a digital twinning integrated system which comprises the following steps: acquiring multi-source heterogeneous data, and performing compression processing, fusion optimization and standard integration processing on the multi-source heterogeneous data to obtain a corresponding standardized digital twinning set; constructing a corresponding high-precision digital twinning scene based on the obtained standardized digital twinning data set; performing scene compression and scene optimization processing on the high-precision digital twinning scene to obtain a corresponding lightweight digital twinning scene packet; building a corresponding digital twinning platform based on the lightweight digital twinning scene packet; carrying out diversified service release and authority management on the digital twinborn platform to obtain a corresponding standardized digital twinborn service set; performing real-time monitoring and performance analysis on a digital twinborn service process based on the standardized digital twinborn service set, and performing feedback adjustment on a corresponding digital twinborn platform based on the standardized digital twinborn service set; the application efficiency and the application range of the digital twinning service are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of digital twin technology, and more specifically, to a digital twin integrated system. Background Art

[0002] With the continuous advancement of smart city construction, digital twin technology, as a key means to achieve the deep integration of the physical world and the digital world, is being used more and more widely. Digital twin technology builds digital models of physical entities to achieve real-time monitoring, analysis and prediction of the entity status, providing new technical support for urban planning, management and decision-making.

[0003] However, the current digital twin technology still faces many challenges and limitations in its practical application. First, urban spatial data presents multi-source heterogeneous characteristics, and these data formats are diverse and standards vary, making it difficult to effectively integrate them. Second, traditional three-dimensional scene construction methods often find it difficult to balance model accuracy and system performance, resulting in the construction of digital twin scenes that are either insufficiently accurate or overloaded. In addition, existing digital twin systems usually have high requirements for terminal hardware, especially strong dependence on GPUs, which limits their deployment and application on low-configuration terminal devices.

[0004] In terms of data processing, existing technologies mostly use simple data compression and integration methods, which are difficult to handle massive heterogeneous data, resulting in data redundancy and low retrieval efficiency. In addition, they lack effective data association mechanisms and are difficult to reflect the complex relationships between urban elements. In terms of scene construction, existing technologies focus more on the construction of static geometric models, lack effective simulation of dynamic characteristics and time series changes, and are difficult to truly reflect the characteristics of urban space evolution over time. In terms of system performance, existing digital twin platforms often face problems such as slow loading and delayed response, especially when dealing with large-scale scenes, the user experience is poor.

[0005] In terms of service release, existing technologies lack unified service standards and interface specifications, making it difficult to support diversified application needs. In addition, they lack effective permission management and security control mechanisms, posing data security risks. In terms of system optimization, existing technologies mostly use static configuration methods, lack real-time monitoring of operating status and dynamic adjustment capabilities, and are difficult to adapt to changing usage scenarios and load requirements.

[0006] In view of this, the present invention proposes a digital twin integrated system to solve the above problems. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: A digital twin integrated system, comprising: A data acquisition module is used to acquire multi-source heterogeneous data of the target city space, and to compress and integrate the acquired multi-source heterogeneous data to obtain a corresponding preliminary digital twin database, and to perform standard integration processing on the acquired preliminary digital twin database to obtain a corresponding standardized digital twin set; the multi-source heterogeneous data includes spatial data and time series data; The scene construction module constructs multi-dimensional spatiotemporal scenes based on the obtained standardized digital twin data set to obtain the corresponding high-quality visual scenes, and performs attribute mapping and scene verification on them to obtain the corresponding high-precision digital twin scenes; The lightweight publishing module builds a data stream distribution scheme based on the corresponding high-precision digital twin scene, and obtains the corresponding dynamic scene loading framework based on it; at the same time, it builds the corresponding high-performance rendering configuration based on the high-precision digital twin scene, and uses the high-performance rendering configuration and dynamic scene loading framework to package the scene as a whole and configure the publishing parameters to obtain the corresponding lightweight digital twin scene package; A cross-platform adaptation module, used to perform multi-terminal compatibility processing and GPU-free rendering optimization on the lightweight digital twin scene package to obtain a corresponding digital twin platform; The service management module is used to publish diversified services and manage permissions on the digital twin platform to obtain the corresponding standardized digital twin service set; Run the monitoring module to conduct real-time monitoring and performance analysis of the corresponding digital twin platform based on the standardized digital twin service set, generate the corresponding platform adjustment strategy, and make feedback adjustments to the corresponding digital twin platform based on it.

[0008] Furthermore, the multi-source heterogeneous data includes spatial data and time series data; the spatial data includes BIM model data, GIS map data, point cloud data, and image data; and the time series data includes IoT device data and sensor data.

[0009] Furthermore, the process of obtaining the standardized digital twin dataset includes: Decompose the BIM model data and GIS map data in the corresponding spatial data hierarchically, obtain the hierarchical structure corresponding to the spatial entities in the corresponding urban space, and obtain the subordinate relationship between different spatial entities; and build the corresponding initial data structure tree based on it; Perform data redundancy analysis on the constructed initial data structure tree to obtain a streamlined initial data structure tree, and extract spatial features of corresponding spatial entities based on the streamlined initial data structure tree; and select a corresponding spatial index algorithm based on it, and index and construct spatial entities in the urban space based on the selected spatial index algorithm to obtain corresponding efficient index data; Constructing a three-dimensional urban model; and simplifying the constructed three-dimensional urban model; Set different observation distances and angles, and generate LOD models with different levels of detail by combining the simplified urban 3D model. Organize the generated LOD models to obtain a corresponding multi-level detail model set. At the same time, establish a transition mapping relationship between LOD models with different levels of detail. Performing texture compression and material optimization on the obtained multi-level detail model set to obtain a corresponding lightweight surface model; Perform spatiotemporal alignment and semantic annotation on the collected time series data to obtain corresponding multi-source fusion data; Construct a mapping rule set; and based on the mapping rule set, perform urban space holographic mapping and data association analysis on the corresponding multi-source fusion data to construct a corresponding urban spatiotemporal association network; The lightweight surface model is used as the basic framework of urban space, the urban spatiotemporal correlation network is used as the driving mechanism of urban dynamics, and a preliminary digital twin database is constructed by combining the pre-established correlation mapping mechanism between spatial data and time series data; The metadata of the preliminary digital twin database obtained is normalized to obtain a standardized digital twin dataset.

[0010] Furthermore, the process of constructing a high-quality visual scene includes: Based on the standardized digital twin dataset, the overall scene in the target city space is partitioned and divided into boundaries to obtain the corresponding scene partitioning scheme. Based on the scheme, the scene elements in the corresponding overall scene are classified and attributed to obtain the corresponding scene element library. Based on the streamlined initial data structure tree, the topological relationship between each scene element in the corresponding scene element library is obtained, and the initial scene skeleton is constructed in combination with the obtained scene element library. The constructed initial scene skeleton is then material mapped and illuminated to obtain the corresponding high-quality visual scene.

[0011] Furthermore, the process of acquiring a high-precision digital twin scene includes: Configure the physical properties and define the dynamic characteristics of the corresponding high-quality visual scene to obtain the corresponding scene model with physical characteristics; and perform dynamic behavior simulation and state transition rule setting based on the scene model with physical characteristics to obtain the corresponding intelligent interaction scene; Perform historical time series data mapping and future state prediction on the obtained intelligent interaction scenarios to obtain corresponding multi-temporal scenario sequences; The obtained multi-phase scene sequence is verified for spatiotemporal continuity and data consistency is maintained to obtain the corresponding high-precision digital twin scene.

[0012] Furthermore, the process of obtaining the lightweight digital twin scene package includes: Design on-demand loading strategies and streaming plans for the corresponding high-precision digital twin scenarios, obtain the corresponding hierarchical loading priorities and progressive transmission strategies, and build the corresponding data stream distribution solutions based on them; Based on the data stream distribution scheme, the corresponding scene data is hierarchically stored and cached to optimize, and a corresponding hierarchical data storage structure is obtained; and based on the hierarchical data storage structure, the viewpoint-related details are progressively loaded and the near and far levels are adaptively controlled to obtain corresponding dynamic loading control parameters, and non-visible area data is delayed loaded and automatically unloaded according to the dynamic loading control parameters to obtain a corresponding dynamic scene loading framework; Extract the corresponding scene data to obtain the corresponding scene geometry data and scene texture data, and perform mesh simplification and topology optimization on the obtained scene geometry data to obtain the corresponding lightweight geometry model; at the same time, perform adaptive compression and resolution adjustment on the obtained scene texture data to obtain the corresponding optimized texture set; Based on lightweight geometric models and optimized texture sets, rendering pipeline optimization and shader simplification are performed to obtain the corresponding high-performance rendering configuration. The overall scene packaging and release parameter configuration are performed through high-performance rendering configuration and dynamic scene loading framework to obtain the corresponding lightweight digital twin scene package.

[0013] Furthermore, the construction process of the digital twin platform includes: Obtain the hardware configuration and performance parameters of various terminal devices, and build the corresponding terminal capability classification table based on them, and obtain the corresponding device capabilities based on them; Based on the terminal capability classification table, adaptive allocation and priority scheduling of rendering resources are performed to obtain a corresponding resource scheduling strategy; based on the obtained resource scheduling strategy, software rendering algorithm optimization and CPU computing efficiency improvement are performed to obtain a corresponding efficient soft rendering engine; and based on the efficient soft rendering engine, multi-threaded parallel processing and task allocation optimization are performed to obtain a corresponding parallel rendering framework; Based on the parallel rendering framework, scene features are simplified and key information is retained to obtain a corresponding terminal adaptation scene model, and the terminal adaptation scene model is adapted to multi-platform interface layout and interactive mode conversion to obtain a corresponding cross-platform interactive interface; the obtained cross-platform interactive interface is program packaged and interface standardized to obtain a corresponding initial twin platform, and the obtained initial twin platform is tested for environmental compatibility and performance optimization to obtain a corresponding digital twin platform.

[0014] Furthermore, the construction process of the standardized digital twin service set includes: Perform two-dimensional and three-dimensional map service configuration and Internet map service integration on the service interface of the corresponding digital twin platform to obtain a corresponding multi-map service set, and perform service registration and directory management on the multi-map service set to obtain a corresponding service resource directory; Based on the service resource directory, user authority level design and access control rule definition are performed to obtain a corresponding service access control matrix, and a corresponding security service access mechanism is constructed according to the service access control matrix; based on the security service access mechanism, service quality monitoring and load balancing configuration are performed on the digital twin platform to obtain a corresponding service quality assurance plan, and service publishing and subscription management are performed through the service quality assurance plan to obtain a corresponding standardized digital twin service set.

[0015] Furthermore, the process of generating a corresponding platform adjustment strategy and performing feedback adjustment on the corresponding digital twin platform based on the strategy includes: Monitor the operation data involved in the data twin service process of the corresponding digital twin platform in real time to obtain the corresponding performance data flow and abnormal event set; Based on the abnormal event set and in combination with the historical operation data of the digital twin platform, a performance bottleneck analysis and resource usage evaluation are performed to obtain a corresponding performance bottleneck report; Obtain the user interaction behavior and scene loading performance of the target user during the corresponding digital twin service process, and generate corresponding user experience evaluation results based on them; Based on the user experience evaluation results and performance bottleneck reports, a multi-objective optimization problem is constructed and solved, and a corresponding comprehensive optimization plan is constructed based on the solution results; based on the comprehensive optimization plan, parameter adjustment instructions and resource allocation strategy updates are generated to obtain a corresponding platform adjustment plan; based on the platform adjustment plan, the constructed data twin platform is optimized.

[0016] Technical effects and advantages of a digital twin integrated system of the present invention: 1. Through the efficient fusion of multi-source heterogeneous data, lightweight processing and high-precision construction of multi-dimensional space-time scenes, compact management of urban spatial data and realistic simulation of dynamic scenes are achieved, providing accurate data support and visualization foundation for digital twins, improving data processing efficiency and scene realism; 2. Through cross-platform compatibility optimization, GPU-free rendering and standardized service management, the terminal hardware threshold is lowered and the application scenarios are broadened; combined with real-time monitoring and dynamic optimization strategies, the stability, security and long-term efficient operation of the digital twin platform are ensured, and the technology is widely implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a digital twin integrated system of the present invention; Figure 2 It is a schematic diagram of a digital twin integration method of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Example 1 See also Figure 1 As shown, the digital twin integrated system described in this embodiment includes: A data acquisition module is used to acquire multi-source heterogeneous data of the target city space, and to compress and integrate the acquired multi-source heterogeneous data to obtain a corresponding preliminary digital twin database, and to perform standard integration processing on the acquired preliminary digital twin database to obtain a corresponding standardized digital twin set; the multi-source heterogeneous data includes spatial data and time series data; The scene construction module constructs multi-dimensional spatiotemporal scenes based on the obtained standardized digital twin data set to obtain the corresponding high-quality visual scenes, and performs attribute mapping and scene verification on them to obtain the corresponding high-precision digital twin scenes; The lightweight publishing module builds a data stream distribution scheme based on the corresponding high-precision digital twin scene, and obtains the corresponding dynamic scene loading framework based on it; at the same time, it builds the corresponding high-performance rendering configuration based on the high-precision digital twin scene, and uses the high-performance rendering configuration and dynamic scene loading framework to package the scene as a whole and configure the publishing parameters to obtain the corresponding lightweight digital twin scene package; A cross-platform adaptation module, used to perform multi-terminal compatibility processing and GPU-free rendering optimization on the lightweight digital twin scene package to obtain a corresponding digital twin platform; The service management module is used to publish diversified services and manage permissions on the digital twin platform to obtain the corresponding standardized digital twin service set; Run the monitoring module to conduct real-time monitoring and performance analysis of the corresponding digital twin platform based on the standardized digital twin service set, generate the corresponding platform adjustment strategy, and make feedback adjustments to the corresponding digital twin platform based on it.

[0020] It should be further explained that, in the specific implementation process, the process of acquiring multi-source heterogeneous data includes: Setting up a collection unit, and deploying the collection unit to a pre-set data collection point in the target urban space. The data collection point can be set at a road traffic node inside or outside a building in the target urban space, around a public facility, and at a pre-set environmental monitoring point or other location that can be used to obtain urban space data; Based on the acquisition unit, multi-source data is collected on the target city space to obtain corresponding multi-source heterogeneous data, and the data is uploaded to the cloud for storage. The multi-source heterogeneous data includes spatial data such as BIM model data, GIS map data, point cloud data, image data, and time series data such as IoT device data and sensor data.

[0021] It should be further explained that, in the specific implementation process, the process of obtaining the preliminary digital twin database includes: The BIM model data and GIS map data in the corresponding spatial data are hierarchically decomposed to obtain the hierarchical structure corresponding to the spatial entities in the corresponding urban space, and the subordinate relationship between different spatial entities; the subordinate relationship includes adjacent relationship, connection relationship and inclusion relationship; for example, the BIM model data can be decomposed into a hierarchical structure such as project, building, floor, room and component; the GIS map data can be decomposed into a hierarchical structure such as region, plot, road and facility; Based on the obtained hierarchical structure and subordinate relationship, a corresponding initial data structure tree is constructed, wherein the initial data structure tree is represented by a directed graph, and is composed of a number of tree nodes, node edges and edge attributes, wherein the tree nodes represent spatial entities, the node table represents the subordinate relationship between spatial entities, and the edge attributes include the relationship type, spatial distance and connection strength between spatial entities, etc.; Performing data redundancy analysis on the constructed initial data structure tree to obtain a streamlined initial data structure tree, wherein the data redundancy analysis refers to identifying and removing duplicate data, similar attribute data and low-value information in the constructed initial data structure tree; so as to reduce redundancy and improve access efficiency; for example: for the constructed initial data structure tree, identifying component elements in the spatial entity with a geometric expression repetition rate exceeding 90%, marking them as redundant geometry; identifying edge attributes with a repetition rate exceeding 85%, marking them as redundant attributes; identifying data fields with information entropy below a preset threshold, marking them as low-value information; based on the redundant marking results, optimizing the initial data structure tree, removing duplicate geometric elements, merging similar attribute data, filtering low-value information, and obtaining streamlined data expression; Based on the simplified initial data structure tree, the spatial features of the corresponding spatial entity are extracted, and the spatial features include the geometric center coordinates, the outer frame boundary and the geometric shape description of the spatial entity; and the corresponding spatial index algorithm is selected in combination with the pre-set query requirements; for example, for dense spatial data in highly urbanized areas, the R-tree index algorithm is used; for sparse spatial data in the outer areas of the city, the quadtree index algorithm is used; for complex three-dimensional scenes, the octree index algorithm is used; Then, based on the selected spatial index algorithm, the spatial entities in the urban space are indexed and constructed to obtain corresponding efficient index data, which includes the location information, range information and hierarchical relationship of the spatial entities, so as to realize efficient organization and rapid retrieval of urban space related data; Based on the collected multi-source heterogeneous data, a three-dimensional urban model corresponding to the target urban space is constructed, wherein the construction process of the corresponding three-dimensional urban model is a prior art and will not be elaborated in detail in the present invention; The urban 3D model is simplified by using a mesh reduction algorithm and combining the obtained simplified protection priorities; Set different observation distances and observation angles, and generate LOD models with different levels of detail by combining the simplified urban 3D model. The LOD model includes LOD 0 (minimal outline), LOD 1 (Basic Blocks), LOD 2 (Main Features), LOD 3 (Detailed Structure) and LOD 4 (full details); the completeness of the geometric information retained by the LOD model at different levels of detail is LOD 0 <LOD 1 <LOD 2 <LOD 3 <LOD 4 ; Then, a corresponding multi-level detail model set is constructed based on the obtained LOD models of different detail levels; at the same time, a transition mapping relationship between LOD models of different detail levels is established to achieve smooth switching between different LOD models; The obtained multi-level detail model set is texture compressed and material optimized to obtain the corresponding lightweight surface model; texture compression includes three steps: resolution downsampling, image compression and texture atlas integration; material optimization clusters and simplifies material parameters to merge known material sets into simplified material sets, reducing the number of material types while maintaining visual expressiveness; for example, materials with similar colors (color difference △E<5) and similar reflection characteristics are merged into a shared material; for materials used in large areas, atlas technology (such as TextureAtlas) is used to merge multiple small textures into a large texture atlas to reduce rendering state switching; The collected IoT device data, sensor data and other time series data are subjected to spatiotemporal alignment and semantic annotation to obtain corresponding multi-source fusion data; the spatiotemporal alignment process includes two main steps: time standardization and spatial registration; time standardization refers to unifying data with different time sampling frequencies into the same time reference system, such as converting hourly, daily and monthly data into a unified time granularity; spatial registration refers to converting spatial data in different coordinate systems into a unified geographic coordinate system to ensure the consistency of spatial position; semantic annotation adds standardized descriptive information to time series data based on urban spatial entities, including data type, unit, numerical range and semantic association, etc.; Then, a mapping rule set between the physical space and the digital space is constructed, and the mapping rule set includes geometric mapping rules, attribute mapping rules and behavior mapping rules; wherein the geometric mapping rule is used to define how the physical entity is expressed in the digital space; the attribute mapping rule is used to define how the physical attribute is converted into the digital attribute; the behavior mapping rule is used to define how the physical behavior is simulated into the digital behavior; it should be further explained that the construction process of the mapping rule set is a prior art, and the present invention will not elaborate on it in detail; Based on the mapping rule set, the obtained multi-source fusion data is subjected to urban space holographic mapping and data association analysis to construct a corresponding urban space-time association network, which is composed of network nodes and network edges. The network nodes represent urban events or states, and the network edges represent the influence relationship between events or states. Among them, urban space holographic mapping refers to mapping the urban space from the physical space to the digital space based on the mapping rule set; data association analysis refers to identifying the association relationship between different data through association rule mining; The lightweight surface model is used as the basic framework of urban space, the urban spatiotemporal correlation network is used as the driving mechanism of urban dynamics, and a preliminary digital twin database is constructed by combining the pre-established correlation mapping mechanism between spatial data and time series data; It should be further explained that, in the specific implementation process, the association mapping mechanism refers to a mapping relationship matrix used to describe the bidirectional association relationship between spatial data and time series data; for example: for each spatial entity, its corresponding time series data stream is associated; for each time series data point, its affected spatial range is associated; and a many-to-many mapping relationship matrix is ​​established based on it.

[0022] It should be further explained that, in the specific implementation process, the process of obtaining the standardized digital twin set includes: The metadata of the preliminary digital twin database obtained is normalized to obtain a standardized digital twin data set; metadata normalization defines a unified data description framework, which includes basic information such as data source, collection time, accuracy level, and update frequency.

[0023] It should be further explained that, in the specific implementation process, the construction process of high-quality visual scenes includes: Based on the standardized digital twin data set, the overall scene in the target city space is partitioned and the boundary is divided to obtain the corresponding scene partition scheme, which includes main partition information, sub-partition information, partition boundary description and the association relationship between partitions; wherein, the scene partition process needs to consider indicators such as data density, functional relevance and load balance; and the scene partition refers to dividing the overall scene corresponding to the target city space into multiple sub-areas by adopting a spatial partitioning algorithm; the spatial partitioning algorithm adopted in the present invention is a grid partitioning algorithm; Based on the scene partitioning scheme, the scene elements in the corresponding overall scene are classified and attributed to obtain the corresponding scene element library; element classification refers to classifying the various types of scene elements in the sub-area, and configuring the corresponding attribute parameters for the classified scene elements, forming a rich scene element library; for example, from the time dimension, the scene element types can be divided into static elements (such as buildings, roads, terrain, municipal facilities, vegetation and fixed facilities) and dynamic elements (such as transportation, pedestrians, environmental factors, temporary facilities and urban events); the attribute configuration refers to configuring the corresponding attribute parameters for the classified scene elements; for example, the attribute parameters required to be configured for the building elements in the static elements include building height, number of floors, building age, structure type, usage function, energy consumption characteristics, etc.; the attribute parameters required to be configured for the vehicle elements in the dynamic elements include: vehicle type, speed range, turning rules, stay time and other attributes; Further, the topological relationship between each scene element in the corresponding scene element library is obtained based on the simplified initial data structure tree, and the topological relationship includes spatial connection relationship (such as adjacent, intersecting, containing, etc.), functional connection relationship (such as power supply, water supply, transportation, etc.) and logical connection relationship (such as subordinate, control, influence, etc.); Then, an initial scene skeleton is constructed based on the obtained topological relationship and scene element library. The initial scene skeleton is represented by an attribute graph model. The nodes in the attribute graph model represent scene elements, and the edges represent the relationships between scene elements. The node attributes include element ID, type, geometric features, and attribute set; the edge attributes include relationship type, directionality, etc. After the initial scene skeleton is constructed, material mapping and lighting optimization are performed on the constructed initial scene skeleton to obtain a corresponding high-quality visual scene; wherein, the material mapping refers to accurately simulating the optical characteristics of various materials by applying physical based rendering technology, so that the material performance simulated in the corresponding initial scene skeleton is more consistent with the actual material performance; the lighting optimization is to enhance the light and shadow effects of the initial scene skeleton by applying global illumination, ambient occlusion and other technologies.

[0024] It should be further explained that, in the specific implementation process, the acquisition process of high-precision digital twin scenes includes: Physical property configuration and dynamic property definition are performed on high-quality visual scenes to obtain scene models with physical properties; wherein the physical property configuration sets physical parameters of high-quality visual scenes based on a pre-built physical property database, and the physical parameters include mass, density, elasticity, friction coefficient, etc.; the physical property database contains reference data sets such as material density table, friction coefficient table, elasticity coefficient table, etc.; usually prepared in advance based on industry literature; dynamic property definition is used to set behavior rules of scene elements that change over time; for example, the law of change of traffic flow over time; Based on the scene model with physical characteristics, dynamic behavior simulation and state transition rule setting are performed to obtain the corresponding intelligent interaction scene; the dynamic behavior simulation is based on the physical engine for motion calculation and collision detection; the state transition rule is used to define the state changes of scene elements under different conditions; for example, the red, yellow, and green state transitions of traffic lights; The obtained intelligent interaction scenarios are mapped with historical time series data and predicted for future states to obtain corresponding multi-temporal scenario sequences; historical time series data mapping refers to associating historical multi-source heterogeneous data with scenario elements; for example, historical traffic flow data is mapped to the road network, or historical energy consumption data is mapped to buildings; future state prediction predicts the possible future evolution of the scenario through a time series model; for example, a long short-term memory network (LSTM) is used to predict traffic flow: The obtained multi-phase scene sequence is subjected to spatiotemporal continuity verification and data consistency maintenance to obtain the corresponding high-precision digital twin scene; spatiotemporal continuity verification is used to ensure the smooth transition of the scene in the time dimension; data consistency maintenance is used to handle data conflicts and differences caused by different data sources; In this embodiment, the multi-dimensional space-time scene construction process adopts a layered construction strategy, first building a basic spatial framework, then adding dynamic behaviors and time series changes, and finally performing overall optimization and verification to ensure the accuracy and realism of the scene.

[0025] It should be further explained that, in the specific implementation process, the process of obtaining the lightweight digital twin scene package includes: Based on the high-precision digital twin scene, on-demand loading strategy design and streaming planning are carried out; the on-demand loading strategy refers to determining the hierarchical loading priority based on the user's visual needs. The process of obtaining the hierarchical loading priority includes: Acquire the user's historical viewpoint information, which includes parameters such as the user's viewpoint position, viewpoint direction, viewpoint angle, and moving speed, and construct a corresponding viewpoint heat map based on the historical viewpoint information. The viewpoint heat map can be used to reflect the distribution of the user's viewpoint movement trajectory; Furthermore, based on the viewpoint distribution map, the viewpoint area concerned by the corresponding user is obtained, and based on the viewpoint distribution map, the initial loading priority of the scene data in the corresponding high-precision digital twin scene is determined; for example, the current visible area is set as the first priority, and the expected visible area is set as the second priority; the expected visible area refers to the visible area that is predicted to enter the user's field of view based on the viewpoint movement trajectory and combined with the current visible area; Then, the semantic relationship of each scene element in the corresponding high-precision digital twin scene is obtained for identification, and based on it, the corresponding initial loading priority is further subdivided to obtain the corresponding hierarchical loading priority; for example, the second highest priority is assigned to key facilities with interactive functions (such as traffic control equipment and important public buildings) to ensure that these elements are loaded in time; the second lowest priority is assigned to purely decorative elements (such as decorative trees and landscape sketches), and loading can be delayed; The streaming transmission planning refers to packaging the scene data in the high-precision digital twin scene in blocks of fixed size based on a pre-planned streaming transmission mechanism, and designing a progressive data transmission strategy based on network conditions and user experience requirements; wherein each data block contains a metadata header and actual scene data content; the metadata header is used to describe the block content, dependencies, and display requirements; Furthermore, a corresponding data stream distribution scheme is constructed based on the obtained hierarchical loading priority and progressive transmission strategy; Based on the data stream distribution scheme, the corresponding scene data is hierarchically stored and cached to optimize, and a corresponding hierarchical data storage structure is obtained; wherein the analytical data storage structure adopts a pyramid structure; the hierarchical storage distributes the scene data to storage media of different levels, for example: obtaining the data importance of the corresponding scene data, and obtaining the user's access frequency to different scene data based on historical visual information; further, the storage structure of the corresponding scene data is divided based on the data importance and access frequency, and hierarchical storage is performed based on the storage structure, and the storage structure includes four levels of core data, high-frequency data, regular data and low-frequency data; for example, the scene skeleton can be divided into core data, and a solid-state storage strategy can be adopted to ensure fast startup; The cache optimization selects a corresponding cache method according to the storage structure of the corresponding scene data and the user behavior. For example, for low-frequency data, the low-frequency detail data is loaded only after the user stays in a specific area for longer than the expected time. Based on the hierarchical data storage structure, progressive loading of viewpoint-related details and adaptive control of near and far levels are performed to obtain corresponding dynamic loading control parameters, and delayed loading and automatic unloading of non-visible area data are performed according to the dynamic loading control parameters to obtain a corresponding dynamic scene loading framework; wherein the dynamic loading control parameters include LOD switching distance and memory occupancy upper limit, etc.; the dynamic scene loading framework is used to implement real-time memory management and resource scheduling to ensure stable operating performance.

[0026] Among them, the progressive loading of viewpoint-related details refers to dynamically adjusting the detail level of the loaded content according to the user's viewpoint position and movement trend; for example, according to a pre-set level selection function, and in combination with the actual viewpoint distance, a suitable detail level is selected; the mathematical formula of the level selection function is: ; In the formula, represents the model precision factor, Indicates the benchmark accuracy, usually 1.0; and Respectively represent the reference distance and the actual observation distance; Represents the visibility factor, which is related to the importance of scene elements in the field of view. The higher the model precision factor, the higher the level of detail selected. The near-far layer adaptive control dynamically adjusts the LOD model accuracy according to the visual distance; for example, the required LOD model accuracy is obtained based on a pre-set adaptive control function; wherein the mathematical expression formula of the adaptive control function is: ; In the formula, LOD i is the level of detail of the selected LOD model; i=0, 1, 2, 3, 4; Indicates the viewpoint angle; Indicates the radius of the circumscribed circle of the corresponding spatial entity; dynamically switches LOD models at different levels of detail based on progressive loading of viewpoint-related details and adaptive control of near and far levels; Lazy loading of non-visible areas uses a multi-level priority queue to sort loading tasks according to the visibility and importance of spatial entities; automatic unloading monitors memory usage, and when memory pressure increases, unloads non-essential data in reverse order of priority; for example, the frustum culling technology is used to identify the current non-visible area; for occluded objects, the occlusion culling technology is applied to avoid loading invisible parts; for area data that has not been visible for a long time, the unloading mechanism is triggered to release memory; low-precision proxy data of non-visible areas is retained to facilitate rapid restoration of display.

[0027] Data extraction is performed on the obtained scene data to obtain corresponding scene geometry data and scene texture data, and mesh simplification and topology optimization are performed on the obtained scene geometry data to obtain a corresponding lightweight geometry model. Mesh simplification refers to reducing the model complexity by reducing the number of vertices and facets; for example, the edge with the least impact is selected for collapse by iteratively calculating the collapse cost through the edge collapse algorithm until the target simplification rate is reached; topology optimization adjusts the connection relationship of the mesh to improve rendering efficiency; At the same time, the obtained scene texture data is adaptively compressed and the resolution is adjusted to obtain the corresponding optimized texture set; adaptive compression selects different compression algorithms and compression rates according to the characteristics of texture content; for example, a low compression rate is used to retain details in high-frequency detail areas, and a high compression rate is used to reduce the amount of data in flat areas; resolution adjustment allocates appropriate texture resolutions to different objects according to the importance and visual impact of the objects, for example, high-resolution textures are used for important buildings, and low-resolution textures are used for ordinary vegetation, thereby reducing memory usage while ensuring visual effects; Based on lightweight geometric models and optimized texture sets, rendering pipeline optimization and shader simplification are performed to obtain corresponding high-performance rendering configurations; rendering pipeline optimization includes batch merging, draw call reduction, and state sorting to reduce API call overhead; shader simplification reduces shader complexity through algorithm optimization and precision adjustment; for example, for distant objects, a simplified lighting model and low-precision calculations are used; for nearby important objects, a complete physical lighting model and high-precision calculations are used; The corresponding lightweight digital twin scene package is obtained by packaging the scene as a whole and configuring the release parameters through high-performance rendering configuration and dynamic scene loading framework; the scene as a whole packaging organizes and compresses the dynamic scene loading framework, lightweight geometric model, optimized texture set and high-performance rendering configuration in a unified manner; the release parameter configuration sets the scene loading strategy, rendering options and interaction parameters to ensure the best performance experience on different devices; for example, for high-performance PC devices, high-quality rendering options are configured; for mobile devices, power-saving optimization options are configured. The resulting lightweight digital twin scene package has high compression ratio, low memory usage and fast loading characteristics, and is suitable for deployment on various terminals.

[0028] It should be further explained that in the specific implementation process, the construction process of the digital twin platform includes: Obtain the hardware configuration and performance parameters of various types of terminal devices, and build a corresponding terminal capability grading table based on them. The terminal capability grading table can be used to grade terminal devices according to indicators such as CPU performance, memory capacity, network bandwidth, etc., and obtain corresponding device capabilities. The device capabilities include device scores and device levels; the terminal devices include but are not limited to mobile terminals and computer terminals; Based on the terminal capability classification table, the corresponding lightweight digital twin scene package is adaptively allocated and prioritized for rendering resources to obtain a corresponding resource scheduling strategy; the resource scheduling strategy refers to dynamically adjusting the scene complexity and loading content according to the device level; wherein, the adaptive allocation of rendering resources refers to dynamically adjusting the rendering resolution, texture quality, geometric accuracy and special effects level of the corresponding terminal device according to the device capability; for example, for high-scoring devices, 4K rendering resolution and high-quality textures are allocated; for low-scoring devices, 720p rendering resolution and compressed textures are used; the priority scheduling is to reasonably allocate computing resources according to the importance of scene elements and device capabilities; for example, on low-performance devices, the rendering quality of scene elements under the main vision is prioritized, and the level of detail of secondary scene elements is reduced; Based on the resource scheduling strategy, the software rendering algorithm is optimized and the CPU computing efficiency is improved for the high-performance rendering configuration in the corresponding lightweight digital twin scene to obtain the corresponding high-efficiency soft rendering engine; the high-efficiency soft rendering engine can be used to realize functions such as rasterization, texture mapping and basic lighting; the software rendering algorithm optimization refers to reducing the amount of calculation by reducing the calculation accuracy and other methods; the CPU computing efficiency improvement is to improve the CPU rendering performance of the terminal device by optimizing the SIMD instruction set and other methods; The efficient soft rendering engine is subjected to multi-threaded parallel processing and task allocation optimization to obtain a corresponding parallel rendering framework; the parallel rendering framework fully utilizes the multi-core CPU resources of the terminal device by adopting task segmentation and thread pool technology; multi-threaded parallel processing is used to decompose the rendering tasks generated by the efficient soft rendering engine into independent work units, and assign them to multiple CPU cores for parallel execution; for example, a tile partitioning strategy is adopted to divide the screen into multiple tiles, and each thread is responsible for rendering one or more tiles. Task allocation optimization improves multi-core utilization and realizes dynamic load balancing through work stealing algorithm and load balancing technology; Based on the parallel rendering framework, the lightweight digital twin scene is simplified in scene features and key information is retained to obtain a corresponding terminal adaptation scene model, and the terminal adaptation scene model is adapted to the multi-platform interface layout and the interaction mode is converted to obtain a corresponding cross-platform interaction interface; wherein, the terminal adaptation scene model is used to dynamically adjust the geometric complexity and texture details according to the device capabilities; the cross-platform interaction interface supports multiple input methods such as touch, mouse and keyboard, and gestures to achieve consistency and smoothness of operation; it should be further explained that scene feature simplification refers to identifying and extracting key features of the scene, removing visual redundancy, and reducing scene complexity; for example, simplifying a building into a representation of an outline plus key feature points, retaining recognizability while greatly reducing the amount of data; key information retention is used to ensure that the functionality and interactivity of the scene are not affected; for example, even if the visual performance is simplified, the attribute data and interaction functions of the building are still retained; The obtained cross-platform interactive interface is packaged and interface standardized to obtain the corresponding initial twin platform, and the obtained initial twin platform is tested for environmental compatibility and performance optimized to obtain the corresponding digital twin platform; the program packaging is used to integrate the lightweight digital twin scene package, efficient rendering engine and cross-platform interactive interface into the initial twin platform, and the interface standardization is used to perform compatibility processing on the service interface of the initial twin platform so that it can adapt to different terminal devices and service platforms (that is, the service interface after interface standardization can support multiple API interface protocols to cope with the API interface differences of different terminal devices or service platforms); performance optimization refers to optimizing the compatibility issues tested during the environmental compatibility test to ensure that it can achieve the expected performance on different terminal devices and service platforms. The service platform includes GIS software platform and BIM software platform, etc. It should be further explained that in the specific implementation process, the construction process of the standardized digital twin service set includes: Perform two-dimensional and three-dimensional map service configuration and Internet map service integration on the service interface of the corresponding digital twin platform to obtain a corresponding multi-map service set, and perform service registration and directory management on the multi-map service set to obtain a corresponding service resource directory; wherein the multi-map service set provides diversified map services such as vectors, images, terrains, and three-dimensional models; the service resource directory adopts classification organization and metadata description to support service discovery and combined calling.

[0029] It should be further explained that the configuration of 2D and 3D map services includes the configuration and release of base map services, tile services and scene services; for example, the release of multi-level tile map services that comply with the WMTS standard supports efficient map browsing. Internet map service integration connects to mainstream online map services, such as Amap, Baidu Map, and Tiandi Map, to expand data coverage; Service registration refers to recording the published service information into the service registry by using existing service registration centers such as Eureka or Consul. The service information includes the service name, access address, interface description and version information. Directory management organizes the hierarchical relationship and dependency relationship between services to form a service resource directory tree for easy browsing and retrieval of services.

[0030] Based on the service resource directory, user authority level design and access control rule definition are performed, and a corresponding service access control matrix is ​​constructed based thereon. According to the service access control matrix, service call authentication and authorization process are implemented to obtain a corresponding security service access mechanism; the service access control matrix is ​​used to define the authority relationship between user roles and resource operations; the security service access mechanism refers to user access verification through the use of token authentication and encrypted communication to ensure the security of data and services.

[0031] Based on the secure service access mechanism, the digital twin platform is monitored for service quality and load balancing is configured to obtain a corresponding service quality assurance plan, and the service quality assurance plan is used to manage service publishing and subscription of the corresponding digital twin platform to build a corresponding standardized digital twin service set; wherein the service quality assurance plan includes performance monitoring, concurrency control and fault recovery strategy; the standardized digital twin service set provides a unified service catalog, registration, discovery and calling mechanism to achieve flexible combination and convenient access of digital twin capabilities; Furthermore, the digital twin platform can provide digital twin services to target users based on the constructed standardized digital twin service set; and enable the digital twin platform to have reliable digital twin service capabilities.

[0032] It should be further explained that, in the specific implementation process, the process of generating the corresponding platform adjustment strategy and making feedback adjustments to the corresponding digital twin platform based on it includes: Monitor the operation data involved in the data twin service process of the corresponding digital twin platform in real time to obtain the corresponding performance data stream, which includes key indicators such as CPU occupancy, memory usage, response time and throughput; Based on the performance data stream, anomaly detection and threshold alarm are performed on the digital twin platform to obtain a corresponding abnormal event set; the abnormal event set includes performance fluctuations and abnormal behaviors in the digital twin platform that exceed preset indicator thresholds; Furthermore, based on the abnormal event set and in combination with the historical operation data of the digital twin platform, a performance bottleneck analysis and resource usage evaluation are performed to obtain a corresponding performance bottleneck report; the performance bottleneck analysis identification is used to obtain performance constraints in the digital twin service process, for example, based on the abnormal event set and historical operation data, performance constraints are analyzed (such as by analyzing the usage of resources such as CPU, memory, network and disk, identifying the resources that first reach the bottleneck); resource usage evaluation is used to analyze the efficiency of resource usage and the rationality of allocation; Obtain the user interaction behavior of the target user during the corresponding digital twin service process, which includes user behavior data such as user operation frequency, dwell time, and interaction path; and simultaneously obtain the scene loading performance of the digital twin platform during the corresponding digital twin service process, which includes indicators such as scene initialization time, frame rate stability, and interaction response delay; Generate corresponding user experience evaluation results based on user interaction behavior and scene loading performance; user experience evaluation results can be used to quantify operation fluency, loading waiting time and interactive responsiveness; Furthermore, a multi-objective optimization problem is constructed based on the user experience evaluation result and the performance bottleneck report, and the multi-objective optimization problem includes multiple objectives such as load balancing performance, data accuracy and user experience; the constructed multi-objective optimization problem is solved, and a corresponding comprehensive optimization solution is constructed based on the solution result; wherein, the particle swarm optimization algorithm is used in the solution process of the multi-objective optimization problem, and the particle swarm optimization algorithm is a prior art, and the present invention will not elaborate on it in detail; Furthermore, based on the comprehensive optimization plan, parameter adjustment instructions and resource allocation strategy updates are generated to obtain the corresponding platform adjustment plan; the parameter adjustment instructions are used to guide the adjustment process of each parameter in the data twin platform; the resource allocation strategy update is used to optimize the allocation method of computing and storage resources; Furthermore, based on the platform adjustment plan, the corresponding compression processing, fusion optimization and the corresponding hierarchical loading priority process are dynamically adjusted to achieve continuous optimization of the data twin platform.

[0033] The digital twin integration system described in the embodiment of the present application realizes the compact expression and unified management of massive map data by efficiently compressing and fusion-optimizing multi-source heterogeneous data; creates a digital twin scene with high precision and time evolution through multi-dimensional spatiotemporal scene construction and dynamic update; greatly reduces the amount of data transmission through extreme compression and optimization processing, and realizes efficient and smooth scene loading; greatly reduces the terminal hardware requirements through multi-terminal compatibility processing and GPU-free rendering optimization; meets the diverse application needs through diversified service publishing and permission management; realizes the continuous optimization of the performance of the digital twin platform through real-time monitoring and performance analysis, and ensures long-term stable and efficient operation. This system significantly improves the efficiency and accuracy of digital twin technology, greatly broadens the boundaries of application scenarios, and opens up a new path for the widespread application of digital twin technology.

[0034] Example 2 See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, which provides a digital twin integration method, including: Step 1: Obtain multi-source heterogeneous data of the target city space, compress and optimize the obtained multi-source heterogeneous data, obtain the corresponding preliminary digital twin database, perform standard integration processing on the obtained preliminary digital twin database, and obtain the corresponding standardized digital twin set; the multi-source heterogeneous data includes spatial data and time series data; Step 2: Based on the obtained standardized digital twin data set, a multi-dimensional spatiotemporal scene is constructed to obtain the corresponding high-quality visual scene, and attribute mapping and scene verification are performed to obtain the corresponding high-precision digital twin scene; Step 3: Build a data stream distribution solution based on the corresponding high-precision digital twin scene, and obtain the corresponding dynamic scene loading framework based on it; Step 4: Build the corresponding high-performance rendering configuration based on the high-precision digital twin scene, and use the high-performance rendering configuration and dynamic scene loading framework to package the scene as a whole and configure the release parameters to obtain the corresponding lightweight digital twin scene package; Step 5: Perform multi-terminal compatibility processing and GPU-free rendering optimization on the lightweight digital twin scene package to obtain a corresponding digital twin platform; Step 6: Publish diversified services and manage permissions on the digital twin platform to obtain the corresponding standardized digital twin service set; Step 7: Based on the standardized digital twin service set, conduct real-time monitoring and performance analysis of the corresponding digital twin platform, generate the corresponding platform adjustment strategy, and make feedback adjustments to the corresponding digital twin platform based on it.

[0035] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for a person skilled in the art to modify the technical solutions described in the aforementioned embodiments or to replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0036] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0037] In the description of the present invention, it should be understood that the terms "first", "second", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0038] In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0039] In the description of the present invention, "several" means one or more than one, and "a large number" means two or more than two.

[0040] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0041] The formulas in this manual are dimensionless and calculated numerically. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the corresponding recent real situation. The preset parameters and thresholds in the formulas are set by technicians in this field according to actual conditions.

[0042] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

Claims

1. A digital twin integrated system, characterized in that: include: A data acquisition module is used to acquire multi-source heterogeneous data of the target city space, and to compress and integrate the acquired multi-source heterogeneous data to obtain a corresponding preliminary digital twin database, and to perform standard integration processing on the acquired preliminary digital twin database to obtain a corresponding standardized digital twin set; the multi-source heterogeneous data includes spatial data and time series data; The scene construction module constructs multi-dimensional spatiotemporal scenes based on the obtained standardized digital twin data set to obtain the corresponding high-quality visual scenes, and performs attribute mapping and scene verification on them to obtain the corresponding high-precision digital twin scenes; A lightweight publishing module builds a data stream distribution solution based on the corresponding high-precision digital twin scene, and obtains the corresponding dynamic scene loading framework based on it; At the same time, the corresponding high-performance rendering configuration is built based on the high-precision digital twin scene, and the overall scene packaging and release parameter configuration are performed through the high-performance rendering configuration and dynamic scene loading framework to obtain the corresponding lightweight digital twin scene package; A cross-platform adaptation module, used to perform multi-terminal compatibility processing and GPU-free rendering optimization on the lightweight digital twin scene package to obtain a corresponding digital twin platform; The service management module is used to publish diversified services and manage permissions on the digital twin platform to obtain the corresponding standardized digital twin service set; Run the monitoring module to conduct real-time monitoring and performance analysis of the corresponding digital twin platform based on the standardized digital twin service set, generate the corresponding platform adjustment strategy, and make feedback adjustments to the corresponding digital twin platform based on it.

2. A digital twin integrated system according to claim 1, characterized in that: The multi-source heterogeneous data includes spatial data and time series data; the spatial data includes BIM model data, GIS map data, point cloud data, and image data; the time series data includes IoT device data and sensor data.

3. A digital twin integrated system according to claim 2, characterized in that: The process of obtaining a standardized digital twin dataset includes: Decompose the BIM model data and GIS map data in the corresponding spatial data hierarchically, obtain the hierarchical structure corresponding to the spatial entities in the corresponding urban space, and obtain the subordinate relationship between different spatial entities; and build the corresponding initial data structure tree based on it; Perform data redundancy analysis on the constructed initial data structure tree to obtain a streamlined initial data structure tree, and extract spatial features of corresponding spatial entities based on the streamlined initial data structure tree; and select a corresponding spatial index algorithm based on it, and index and construct spatial entities in the urban space based on the selected spatial index algorithm to obtain corresponding efficient index data; Constructing a three-dimensional urban model; and simplifying the constructed three-dimensional urban model; Set different observation distances and angles, and generate LOD models with different levels of detail by combining the simplified urban 3D model. Organize the generated LOD models to obtain a corresponding multi-level detail model set. At the same time, establish a transition mapping relationship between LOD models with different levels of detail. Performing texture compression and material optimization on the obtained multi-level detail model set to obtain a corresponding lightweight surface model; Perform spatiotemporal alignment and semantic annotation on the collected time series data to obtain corresponding multi-source fusion data; Construct a mapping rule set; and based on the mapping rule set, perform urban space holographic mapping and data association analysis on the corresponding multi-source fusion data to construct a corresponding urban spatiotemporal association network; The lightweight surface model is used as the basic framework of urban space, the urban spatiotemporal correlation network is used as the driving mechanism of urban dynamics, and a preliminary digital twin database is constructed by combining the pre-established correlation mapping mechanism between spatial data and time series data; The metadata of the preliminary digital twin database obtained is normalized to obtain a standardized digital twin dataset.

4. A digital twin integrated system according to claim 3, characterized in that: The process of building a high-quality visual scene includes: Based on the standardized digital twin dataset, the overall scene in the target city space is partitioned and divided into boundaries to obtain the corresponding scene partitioning scheme. Based on the scheme, the scene elements in the corresponding overall scene are classified and attributed to obtain the corresponding scene element library. Based on the streamlined initial data structure tree, the topological relationship between each scene element in the corresponding scene element library is obtained, and the initial scene skeleton is constructed in combination with the obtained scene element library. The constructed initial scene skeleton is then material mapped and illuminated to obtain the corresponding high-quality visual scene.

5. A digital twin integrated system according to claim 4, characterized in that: The process of acquiring a high-precision digital twin scene includes: Configure the physical properties and define the dynamic characteristics of the corresponding high-quality visual scene to obtain the corresponding scene model with physical characteristics; and perform dynamic behavior simulation and state transition rule setting based on the scene model with physical characteristics to obtain the corresponding intelligent interaction scene; Perform historical time series data mapping and future state prediction on the obtained intelligent interaction scenarios to obtain corresponding multi-temporal scenario sequences; The obtained multi-phase scene sequence is verified for spatiotemporal continuity and data consistency is maintained to obtain the corresponding high-precision digital twin scene.

6. A digital twin integrated system according to claim 5, characterized in that: The acquisition process of the dynamic scene loading framework includes: Based on the corresponding high-precision digital twin scenarios, on-demand loading strategy design and streaming transmission planning are carried out to obtain the corresponding hierarchical loading priority and progressive transmission strategy, and based on them, the corresponding data stream distribution solution is built; Based on the data stream distribution solution, the corresponding scene data is hierarchically stored and cached to optimize, and a corresponding hierarchical data storage structure is obtained; and based on the hierarchical data storage structure, the scene data is progressively loaded with viewpoint-related details and adaptively controlled with far and near levels to obtain corresponding dynamic loading control parameters, and according to the dynamic loading control parameters, non-visible area data is delayed loaded and automatically unloaded to obtain a corresponding dynamic scene loading framework.

7. A digital twin integrated system according to claim 6, characterized in that: The process of obtaining the lightweight digital twin scene package includes: Extract the corresponding scene data to obtain the corresponding scene geometry data and scene texture data, and perform mesh simplification and topology optimization on the obtained scene geometry data to obtain the corresponding lightweight geometry model; at the same time, perform adaptive compression and resolution adjustment on the obtained scene texture data to obtain the corresponding optimized texture set; Based on lightweight geometric models and optimized texture sets, rendering pipeline optimization and shader simplification are performed to obtain the corresponding high-performance rendering configuration. The overall scene packaging and release parameter configuration are performed through high-performance rendering configuration and dynamic scene loading framework to obtain the corresponding lightweight digital twin scene package.

8. A digital twin integrated system according to claim 7, characterized in that: The process of building a digital twin platform includes: Obtain the hardware configuration and performance parameters of various terminal devices, and build the corresponding terminal capability classification table based on them, and obtain the corresponding device capabilities based on them; Based on the terminal capability classification table, adaptive allocation and priority scheduling of rendering resources are performed to obtain a corresponding resource scheduling strategy; based on the obtained resource scheduling strategy, software rendering algorithm optimization and CPU computing efficiency improvement are performed to obtain a corresponding efficient soft rendering engine; and based on the efficient soft rendering engine, multi-threaded parallel processing and task allocation optimization are performed to obtain a corresponding parallel rendering framework; Based on the parallel rendering framework, scene features are simplified and key information is retained to obtain a corresponding terminal adaptation scene model, and the terminal adaptation scene model is adapted to multi-platform interface layout and interactive mode conversion to obtain a corresponding cross-platform interactive interface; the obtained cross-platform interactive interface is program packaged and interface standardized to obtain a corresponding initial twin platform, and the obtained initial twin platform is tested for environmental compatibility and performance optimization to obtain a corresponding digital twin platform.

9. A digital twin integrated system according to claim 8, characterized in that: The process of building a standardized digital twin service set includes: Perform two-dimensional and three-dimensional map service configuration and Internet map service integration on the service interface of the corresponding digital twin platform to obtain a corresponding multi-map service set, and perform service registration and directory management on the multi-map service set to obtain a corresponding service resource directory; Based on the service resource directory, user authority level design and access control rule definition are performed to obtain a corresponding service access control matrix, and a corresponding security service access mechanism is constructed according to the service access control matrix; based on the security service access mechanism, service quality monitoring and load balancing configuration are performed on the digital twin platform to obtain a corresponding service quality assurance plan, and service publishing and subscription management are performed through the service quality assurance plan to obtain a corresponding standardized digital twin service set.

10. A digital twin integrated system according to claim 9, characterized in that: The process of generating the corresponding platform adjustment strategy and making feedback adjustments to the corresponding digital twin platform based on it includes: Monitor the operation data involved in the data twin service process of the corresponding digital twin platform in real time to obtain the corresponding performance data flow and abnormal event set; Based on the abnormal event set and in combination with the historical operation data of the digital twin platform, a performance bottleneck analysis and resource usage evaluation are performed to obtain a corresponding performance bottleneck report; Obtain the user interaction behavior and scene loading performance of the target user during the corresponding digital twin service process, and generate corresponding user experience evaluation results based on them; Based on the user experience evaluation results and performance bottleneck reports, a multi-objective optimization problem is constructed and solved, and a corresponding comprehensive optimization plan is constructed based on the solution results; based on the comprehensive optimization plan, parameter adjustment instructions and resource allocation strategy updates are generated to obtain a corresponding platform adjustment plan; based on the platform adjustment plan, the constructed data twin platform is optimized.

Images