Simulation deduction data multi-scale visualization method and system based on dynamic linkage

Abstract

This application relates to the field of data visualization and simulation technology, and provides a method and system for multi-scale visualization of simulation data based on dynamic linkage. By constructing a multi-layer semantic graph, the attributes, behaviors, and relationships of simulation objects are structurally expressed, thus clearly defining the layers of simulation scenario data and avoiding difficulties in multi-scale data fusion due to a lack of semantic relationship definitions. Semantic differential analysis generates differential update instructions, enabling efficient tracking of data changes during simulation, precise location of change nodes, and reduced computational load. Furthermore, the differential update instructions drive a dynamic linkage synchronization mechanism, achieving cross-scale semantic incremental synchronization and resolving the asynchrony in cross-layer data updates caused by a lack of incremental semantic drive.

Description

Technical Field

[0001] This application relates to the field of data visualization and simulation technology, and in particular to a multi-scale visualization method and system for simulation data based on dynamic linkage. Background Technology

[0002] Currently, in fields such as intelligent simulation and deduction, complex system modeling, and emergency decision support, the multi-dimensional, multi-temporal, and multi-semantic characteristics of data are becoming increasingly prominent. As simulation systems expand in scale, single-scale or static data display methods are insufficient to meet the needs of global understanding of complex scenarios. Traditional simulation visualization systems mostly employ timed full refreshes, which, due to the lack of incremental semantic drivers, leads to asynchronous data updates across different layers. To achieve highly consistent visualization and dynamic semantic linkage of complex simulation data, a visualization method capable of synchronous updates and multi-scale dynamic display across multiple semantic structures is urgently needed. Summary of the Invention

[0003] Therefore, it is necessary to provide a method and system for multi-scale visualization of simulation and deduction data based on dynamic linkage to address the above-mentioned technical problems.

[0004] A multi-scale visualization method for simulation and inference data based on dynamic linkage includes the following steps: Acquire simulation scenario data, extract the attributes, behaviors and relationships of simulation objects and convert them into corresponding semantic units, assign a unique identifier to each type of semantic unit, and construct a multi-layer semantic model. Based on the dependencies between semantic units in the multi-layer semantic model and the pre-defined semantic mapping rules between layers, a multi-layer semantic graph is constructed with semantic units as nodes and semantic dependencies between nodes as edges. Obtain snapshots of multi-layer semantic graphs at continuous time points in the simulation, perform semantic differential analysis based on the obtained snapshots, and generate differential update instructions. Driven by differential update instructions, a dynamic linkage synchronization mechanism is triggered to propagate and update the state of the multi-layer semantic graph, resulting in an updated multi-layer semantic graph. The dynamic linkage synchronization mechanism includes a differential triggering module and a state propagation module. The differential triggering module generates a propagation task queue according to the differential update instructions. The state propagation module accesses the semantic mapping table sequentially according to the propagation task queue. A multi-scale visualization dataset is generated based on the updated multi-layer semantic graph and mapped to the visualization interface according to the preset multi-scale view layout. Real-time monitoring of the dynamic linkage synchronization process and visualization process, and adaptive adjustment of rendering priority and dynamic linkage frequency based on monitoring data.

[0005] A multi-scale visualization system for simulation and deduction data based on dynamic linkage, comprising: The semantic model building module is used to acquire simulation scenario data, extract the attributes, behaviors and relationships of simulation objects and convert them into corresponding semantic units, and assign a unique identifier to each type of semantic unit to build a multi-layer semantic model. The semantic graph construction module is used to construct a multi-layer semantic graph based on the dependencies between semantic units in the multi-layer semantic layer model and the preset semantic mapping rules between layers, with semantic units as nodes and semantic dependencies between nodes as edges; The update instruction generation module is used to obtain multi-layer semantic graph snapshots at continuous time points in the simulation, perform semantic differential analysis based on the obtained multi-layer semantic graph snapshots, and generate differential update instructions. The semantic graph update module is used to trigger a dynamic linkage synchronization mechanism to perform state propagation and update of the multi-layer semantic graph, driven by differential update instructions, to obtain the updated multi-layer semantic graph. The dynamic linkage synchronization mechanism includes a differential trigger module and a state propagation module. The differential trigger module generates a propagation task queue according to the differential update instructions. The state propagation module accesses the semantic mapping table sequentially according to the propagation task queue. The visualization module is used to generate multi-scale visualization datasets based on the updated multi-layer semantic graph and map them to the visualization interface according to the preset multi-scale view layout. The monitoring and update module is used to monitor the dynamic linkage synchronization process and visualization process in real time, and adaptively adjust the rendering priority and dynamic linkage frequency based on the monitoring data.

[0006] The aforementioned multi-scale visualization method and system for simulation data based on dynamic linkage constructs a multi-layer semantic graph to structurally represent the attributes, behaviors, and relationships of simulation objects. This ensures clear layering of simulation scenario data and avoids difficulties in multi-scale data fusion caused by a lack of semantic relationship definitions. Semantic differential analysis generates differential update instructions, enabling efficient tracking of data changes during simulation, precise location of change nodes, and reduced computational load. Furthermore, the differential update instructions drive a dynamic linkage synchronization mechanism, achieving cross-scale incremental semantic synchronization and resolving cross-layer data update asynchrony caused by a lack of incremental semantic drivers. Attached Figure Description

[0007] Figure 1 This is a flowchart illustrating a multi-scale visualization method for simulation data based on dynamic linkage in one embodiment. Figure 2 This is a block diagram of a multi-scale visualization system for simulation data based on dynamic linkage, as shown in one embodiment. Detailed Implementation

[0008] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0009] The multi-scale visualization method for simulation simulation data based on dynamic linkage provided in this application can be applied to simulation visualization systems.

[0010] In one embodiment, such as Figure 1 As shown, a multi-scale visualization method for simulation and inference data based on dynamic linkage is provided, including the following steps: Step 201: Obtain simulation scenario data, extract the attributes, behaviors and relationships of simulation objects and convert them into corresponding semantic units, assign a unique identifier to each type of semantic unit, and construct a multi-layer semantic model.

[0011] Step 202: Based on the dependency relationships between semantic units in the multi-layer semantic layer model and the preset semantic mapping rules between layers, construct a multi-layer semantic graph with semantic units as nodes and semantic dependencies between nodes as edges.

[0012] Step 203: Obtain snapshots of the multi-layer semantic graph at continuous time points in the simulation, perform semantic differential analysis based on the obtained snapshots of the multi-layer semantic graph, and generate differential update instructions.

[0013] Step 204: Driven by the differential update instruction, a dynamic linkage synchronization mechanism is triggered to propagate and update the state of the multi-layer semantic graph, resulting in an updated multi-layer semantic graph. The dynamic linkage synchronization mechanism includes a differential trigger module and a state propagation module. The differential trigger module generates a propagation task queue according to the differential update instruction; the state propagation module accesses the semantic mapping table sequentially according to the propagation task queue.

[0014] Step 205: Generate a multi-scale visualization dataset based on the updated multi-layer semantic graph, and map it to the visualization interface according to the preset multi-scale view layout.

[0015] Step 206: Monitor the dynamic linkage synchronization process and visualization process in real time, and adaptively adjust the rendering priority and dynamic linkage frequency based on the monitoring data.

[0016] In the aforementioned multi-scale visualization method for simulation data based on dynamic linkage, a multi-layer semantic graph is constructed to structurally represent the attributes, behaviors, and relationships of simulation objects. This ensures clear layering of simulation scenario data and avoids difficulties in multi-scale data fusion due to a lack of semantic relationship definitions. Semantic differential analysis generates differential update instructions, enabling efficient tracking of data changes during simulation, precise location of change nodes, and reduced computational load. Furthermore, the differential update instructions drive a dynamic linkage synchronization mechanism, achieving cross-scale incremental semantic synchronization and resolving cross-layer data update asynchrony caused by a lack of incremental semantic drivers.

[0017] In one embodiment, step 201 involves acquiring simulation scenario data, extracting the attributes, behaviors, and relationships of the simulation objects and converting them into corresponding semantic units, assigning a unique identifier to each semantic unit, and constructing a multi-layer semantic model, including: Acquire simulation scenario data, including static attribute data, behavior log data, and status parameters of the simulation objects; Feature decomposition is performed on the simulation scenario data. The static attributes of objects are encapsulated into attribute semantic units in the form of key-value pairs, and the behavior logs are encapsulated into behavior semantic units in the form of event sequences. Relationship semantic units are established based on the interaction relationships between simulation objects. Each semantic unit is assigned a unique identifier, and all semantic units are divided into multiple semantic layers according to their level of abstraction. Each semantic layer maintains a logical connection through semantic dependency mapping, resulting in a multi-layer semantic layer model.

[0018] In this embodiment, feature decomposition and encapsulation of simulation scenario data into semantic units provide an operable semantic foundation for subsequent semantic graph construction and differential analysis. By assigning a unique identifier and hierarchical division to each type of semantic unit, unified data expression across objects and time dimensions is achieved, forming the starting layer for semantic computation.

[0019] It should be noted that static attribute data represents the inherent characteristics of the simulation object at the initial moment, such as geometric dimensions, physical parameters, logical categories, etc.; behavior log data records the behavioral events that occur on the simulation timeline; and state parameter sequences are used to reflect the continuous or discrete quantities of the object's state as it changes over time.

[0020] Specifically, in one embodiment, feature decomposition is performed on the simulation scenario data, and the set of simulation objects is set as follows: ,in Let N represent the i-th simulation object, and N represent the number of simulation objects. The data consists of three parts: ;in For attribute data collection, A collection of behavioral logs. This is a set of state parameters.

[0021] from Extract static attribute items and encapsulate them into attribute semantic units using key-value pairs:

[0022] Where, k b Indicates the attribute name, v b This represents the corresponding numerical or enumerated parameter. , These are the attribute name and the set of attribute values, respectively. (Encapsulated) Static semantic information used to describe an object.

[0023] Behavior log This represents the sequence of action events of an object on the timeline.

[0024] Among them, e i,p (t p ) represents an object In time The p-th event that occurs. These events are encapsulated into behavioral semantic units in chronological order: Semantic units are used to express the behavioral logic of an object in the deduction process.

[0025] The interaction relationships between objects are represented by a relation matrix. express:

[0026] in, Representation Object and There is an interactive relationship. This indicates that no relation exists. Relational semantic units are generated based on the relation matrix: ;in Encapsulate functions for relationships to unify the representation format of different types of relationships (such as causality, dependency, and spatial association).

[0027] Assign a unique identifier to each semantic unit:

[0028] in, Indicates the semantic unit type, Represents time stamps. All semantic units are divided into multiple semantic layers according to their level of abstraction. ;in Represents the underlying object layer. The global abstraction layer is represented by a semantic layer. Each semantic layer maintains logical connections through semantic dependency mapping, resulting in a multi-layer semantic layer model. The semantic parser reads the semantic unit index table of each layer to achieve traceability of semantic access paths between layers.

[0029] Multi-layer semantic layer model Represented as: .

[0030] In one embodiment, the set of semantic units is set as follows: Define hierarchical functions based on the type and scope of semantic units. ;in Represents a hierarchical mapping function. Indicates the first A semantic unit, Semantic layer The hierarchical division is based on the dependency depth and scope of semantic units. Lower-level semantics correspond to the object level, and higher-level semantics correspond to the aggregation or global level. Step 202: Based on the dependency relationships between semantic units in the multi-layer semantic model and the preset semantic mapping rules between levels, a multi-layer semantic graph is constructed with semantic units as nodes and semantic dependencies between nodes as edges, including: Based on the dependency relationship between semantic units in the multi-layer semantic model, the lower-level semantics correspond to the object layer, and the higher-level semantics correspond to the global abstraction layer, and the semantic units are assigned to the corresponding semantic layers. Representing semantic units as graph nodes: ; in, This is a semantic unit encoding function used to convert different types of semantic units into a unified node data format; For a set of nodes; For the first i One node; For the first i One semantic unit; Assign a hierarchy number to each node and generate mapping pointers Mapping pointers are used to represent semantic correspondences between layers; Using semantic dependencies between nodes as edges:

[0031]

[0032] in, This is an edge set used to describe the semantic dependency network between nodes; For the first j One node; for and The semantic influence weights between them; This is a function representing the degree of semantic dependency between attributes; This is a function representing the degree of semantic dependency between behaviors; This is a function of the semantic dependency between relations; The weight coefficient corresponding to the attribute; These are the weighting coefficients corresponding to the behaviors; These are the weight coefficients corresponding to the relationships; A multi-layer semantic graph is constructed using semantic units as nodes and semantic dependencies between nodes as edges. ,in, This is a semantic layer set.

[0033] In this embodiment, semantic units are converted into semantic graph structures, enabling semantic nodes, their levels, weights, and mapping rules to be uniformly described within the same structural system. Furthermore, the multi-layered semantic graph preserves dependency chains between nodes at the same level and contains mapping paths between different levels, providing a searchable topological foundation for subsequent semantic difference calculations and state propagation.

[0034] In one embodiment, step 202, based on the dependencies between semantic units in the multi-layer semantic layer model and the preset semantic mapping rules between layers, constructs a multi-layer semantic graph with semantic units as nodes and semantic dependencies between nodes as edges, and further includes: Establish an inter-level index table based on mapped pointers:

[0035] in, This is the index between a node in the k-th semantic layer and a node in the (k+1)-th semantic layer. For the k-th semantic layer i One node; For the (k+1)th semantic layer j One node; Generate an in-layer topology table based on edge sets:

[0036] in, This is the topology table for the k-th semantic layer, recording the semantic dependencies between nodes.

[0037] In this embodiment, during the semantic graph construction phase, an inter-layer index table is constructed to represent the mapping relationship between semantic nodes in different layers, providing a hierarchical correspondence basis for subsequent semantic differential propagation; an intra-layer topology table is constructed to define the dependency chains of nodes in the same layer. The intra-layer topology table is arranged in weight order to ensure the determinism of dependency calculation and updates. The inter-layer index table and the intra-layer topology table serve as index structures for path lookup, supporting the precise location of semantic update paths in differential calculation.

[0038] In one embodiment, the preset semantic mapping rule between levels is:

[0039] in, This is the set of nodes in the k-th semantic layer; This is the node set of the (k+1)th semantic layer; This refers to the semantic mapping rules between the k-th semantic layer and the (k+1)-th semantic layer. When multiple lower-level nodes correspond to the same higher-level node, calculate the aggregation mapping weight:

[0040] in, For the (k+1)th semantic layer j The aggregate weight of each node reflects the sum of the semantic strengths of its subordinate nodes; for and The semantic influence weight between them.

[0041] This indicates how the semantics of lower-level nodes are mapped to higher-level abstractions.

[0042] In one embodiment, step 203 involves acquiring snapshots of the multi-layer semantic graph at continuous time points in the simulation, performing semantic differential analysis based on the acquired snapshots, and generating differential update instructions, including: Obtain snapshots of multi-layer semantic graphs at consecutive time points in the simulation:

[0043] in, For time A multi-layered semantic graph; For time A multi-layered semantic graph; For time The set of nodes; For time The set of nodes; For time edge set; For time edge set; For time The semantic layer set; For time The semantic layer set; Will and Semantic difference analysis was performed on the three dimensions of attributes, behavior, and relationship to obtain attribute difference tables, behavior difference tables, and relationship difference tables. The node identifiers in the attribute difference table, behavior difference table, and relation difference table are aggregated to generate a difference update instruction.

[0044] In this embodiment, semantic differential analysis is performed on the multi-layer semantic graph snapshots at continuous time points in the simulation, considering attributes, behaviors, and relationships. Changes in these three categories are hierarchically identified, forming structured differential data records for unified management of semantic change information. This allows for subsequent semantic updates and synchronous propagation in a structured manner. A unique node identifier is used to associate the differential content, ensuring the semantic update path is traceable and the computation is reproducible, avoiding inconsistencies in the simulation state. The generation of the differential update instruction set provides input for the subsequent semantic synchronization module, supporting consistency maintenance between semantic layers and local update mechanisms.

[0045] Specifically, in one embodiment, each node in a multi-layer semantic graph snapshot It consists of semantic attribute vectors, behavioral event vectors, and a relational chain structure: ;in Represents a set of semantic attributes; Represents a set of behavioral events; This represents the relationship chain associated with this node.

[0046] Attribute difference analysis, for the same node identifier Calculate the change in attribute: Subtraction represents the set difference operation, and the result includes the addition and deletion of items:

[0047] Where 'a' represents an element in the semantic attribute set. The results are recorded in an attribute difference table. ; ;in Indicates the category of attribute change. This indicates the corresponding level number.

[0048] Behavioral difference analysis, the behavioral event difference is calculated as follows: This result is used to identify changes in node behavior during simulation, such as state triggering and event completion. Behavioral Difference Table ;in For behavior change categories, This is an indicator of the intensity or priority of behavioral change.

[0049] Relationship chain difference analysis, for any node The set of relationships at different times is and The relation difference is defined as: ;when When this occurs, it indicates that the node's connectivity in the semantic graph has changed. The system records the difference and associates it with the changed set of dependent nodes, forming a relation difference table. ;in Indicates the type of relationship change. This represents the set of related nodes that the node depends on.

[0050] Aggregation is performed based on the node identifiers in the difference table to generate a unified differential update instruction set. The definition of the differential update instruction set is as follows: ; in It is a difference aggregation function; The update instruction for the i-th node contains the joint result of three types of differences. W i =< , , , >;among them A unique identifier for a node; For change type; Numbering the corresponding level; This sets the parameters to be changed, including the specific differential content and the indices of the dependent nodes. The output is a differential update instruction set. This is used by the subsequent semantic synchronization propagation module to perform semantic state updates.

[0051] In one embodiment, step 204, driven by a differential update instruction, triggers a dynamic linkage synchronization mechanism to propagate and update the state of the multi-layer semantic graph, resulting in an updated multi-layer semantic graph, including: The differential triggering module parses the differential update instruction to obtain the timestamp, node identifier, and dependent node set, and generates a propagation task queue based on the timestamp. The state propagation module reads task items from the propagation task queue and, based on the node level number and the preset semantic mapping rules between levels, retrieves them from the inter-level index table. The corresponding mapping pointer is obtained to locate the position of the target node in the semantic layer and determine the update path. The state of the multi-layer semantic graph is propagated and updated according to the update path. After the node state data is updated, the results are aggregated to the higher layer, and then the inter-layer state merging is completed according to the semantic consistency constraint rules. The merged high-level state values ​​are read by the lower-level nodes and reversed for synchronization updates, outputting the updated multi-layer semantic graph. .

[0052] In this embodiment, inter-layer indexing is used to implement path mapping, and inter-layer state merging is completed according to semantic consistency constraint rules, ensuring the orderliness of the semantic update process and cross-layer consistency. The updated semantic graph structure provides a semantically unified input data foundation for subsequent multi-scale visualization mapping.

[0053] Specifically, in one embodiment, the differential update instruction set is set as follows: ; in, Represents a unique identifier for a semantic node. Indicates the instruction timestamp. Indicates the difference type, Indicates the difference content. This represents the set of dependent nodes. The differential triggering module parses the differential update instruction to obtain the timestamp, node identifier, and set of dependent nodes, and generates a propagation task queue based on the timestamps. :

[0054] Where, q g To propagate the first in the task queue g one element, g =1, ..., m, recording task numbers, target node identifiers, target level numbers, and a list of dependent nodes, used for scheduling by the state propagation module. The state propagation module reads task items from the propagation task queue and, based on the node level numbers and preset semantic mapping rules between levels, retrieves them from the inter-level index table. The corresponding mapping pointer is obtained to locate the target node's position in the semantic layer and determine the update path. Let the node state be: ;in A semantic vector representing the attributes of a node. Represents a behavioral semantic vector. This represents a semantic description of the relationship. When a node... Received differential update command At that time, execute the state update function: ;in This is a difference-driven update function used to correct the current state of a node based on the difference values. The updated node state is then updated according to the mapping function. The request propagates upwards, forming a higher-level state update request. The propagation path is based on the inter-layer index table. Confirmed. When a higher-level node receives state updates from multiple lower-level nodes, it performs an inter-layer state merging operation based on semantic consistency constraints. Let the merged state of the higher-level node be represented as: ;in This represents a semantic merge function used to aggregate the states of multiple source nodes. The semantic merge function takes dependency chain priority as a parameter and is defined as follows: ;in This is the dependency chain priority weight coefficient, used to control the proportion of influence of updates from different nodes on the merge result. After the merge is complete, the state of the higher-level nodes... It is updated, and the new semantic value serves as the reference input for the lower-level nodes.

[0055] After inter-layer propagation is complete, the lower-layer node reads the updated higher-layer state value based on the mapping pointer. The system executes the reverse synchronization function: ;in This represents a semantic synchronization function used to ensure consistent updates of attribute values ​​and semantic states between upper and lower layers. After the synchronization function completes its execution, the state of the lower-level nodes is corrected, and the intra-layer topology table is updated. The node dependency chain state is then updated to ensure the effectiveness of subsequent differential propagation paths.

[0056] In one embodiment, step 204, driven by a differential update instruction, triggers a dynamic linkage synchronization mechanism to propagate and update the state of the multi-layer semantic graph, obtaining the updated multi-layer semantic graph, and further includes: The state propagation module sets a synchronization lock control flag on the access path. When a node enters the update state, the flag is set to 1, indicating that the node is in a locked state; after the update is completed, the flag is set to 0, indicating that the node can be accessed again.

[0057] The set of flag bits is defined as follows: .

[0058] In this embodiment, by setting a synchronization lock control flag on the access path, it is ensured that the multi-source differential instructions will not cause the node to be accessed repeatedly or overwritten during the propagation process.

[0059] In one embodiment, step 205 involves generating a multi-scale visualization dataset based on the updated multi-layer semantic graph and mapping it to the visualization interface according to a preset multi-scale view layout. Specifically, this involves reading the updated multi-layer semantic graph. The system sequentially parses the nodes and relational structures in each layer of the semantic layer. Each node... Includes attribute set: ; in, This represents the c-th attribute item, including coordinate parameters, hierarchy labels, rendering parameters, and semantic identifier information. During the parsing process, the system uses the hierarchy labels... A hierarchical index table is established to map node data at different levels to the corresponding rendering cache. Based on the parsing results, a multi-scale visualization dataset is constructed in memory. Defined as: ;in Indicates the corresponding semantic layer A collection of visualized data. Multi-scale visualized datasets. Each semantic layer corresponds to a data rendering cache; the data rendering cache stores the coordinate data, hierarchy labels, and rendering attributes of the semantic nodes in that layer. Includes node data cache area Its structure is as follows: ;in This refers to the three-dimensional coordinate data of the node; These are hierarchical tags used to distinguish the semantic layers to which they belong; This is a set of rendering attributes, including node color, transparency, shape parameters, and identifier references. The data generation function is defined as follows: ;in This represents the cache build function, which is used to extract coordinate information and rendering parameters from node attributes and organize them into a rendering cache structure.

[0060] Define the mapping function as: ;in The first part representing the visual interface Each view area. This function performs spatial mapping based on hierarchy labels and view layout parameters, including coordinate projection matrices. With scaling matrix Mapping operations are expressed as: ;in These are the mapped screen coordinates. This is a translation matrix used to adjust the coordinate position; This is the scaling matrix, used to adjust the scaling ratio of the coordinates; These are the original coordinate points used to locate the node's display position within the view area. The system establishes an independent rendering cache for each semantic layer and dynamically loads it through the visualization engine. There is a one-to-one correspondence between the rendering cache and the view area, and this mapping is managed by the view index table.

[0061] In one embodiment, the visualization interface employs a layered rendering pipeline architecture.

[0062] In this embodiment, a layered rendering pipeline architecture is adopted. The visualization data of each semantic layer is processed through an independent pipeline, and the semantic structure at different scales is rendered in layers to ensure that the same node is displayed consistently in multiple views.

[0063] Specifically, in one embodiment, the global pipeline is responsible for processing the upper semantic layer. The data from the global abstraction layer is used to generate a summary situational view; the local pipeline is responsible for processing the lower semantic layer. to The data is used to generate local detail views. The rendering pipeline is defined as follows: Global rendering pipeline processing node set: Local rendering pipeline processing node set: ;in This indicates the number of parent levels contained in the global view.

[0064] To ensure consistent representation of cross-layer semantic nodes across different views, a semantic channel is established between the rendering pipelines. The semantic channel is defined as follows: ; in This indicates that the same semantic node is in the first position. Layer and First Visualize the correspondences in the dataset using layers based on semantic identifiers. Establishing data synchronization between pipelines ensures that identical nodes maintain a consistent visual appearance across multiple view layers. The synchronization function for semantic channels is represented as: ; in and The nodes are respectively in the th Layer and First The set of rendering properties of a layer These are cross-layer synchronization parameters used to define consistency rules for color, transparency, and shape.

[0065] In this embodiment, the semantic channel transmits node rendering attributes between different levels, ensuring that the semantic associations of cross-scale views remain synchronized at the visual level, thus providing a foundation for the visual analysis of subsequent simulation results.

[0066] In one embodiment, step 206 involves real-time monitoring of the dynamic linkage synchronization process and the visualization process, and adaptively adjusting the rendering priority and dynamic linkage frequency based on the monitoring data. Specifically, this is done with a fixed time step. The running status is sampled to form a time series dataset: ;in Indicates time Number of semantic differences triggered internally; Length of the rendering task queue; and These represent the current resource utilization rates of the processor and graphics processing unit, respectively. Within each sampling period, these parameters are written to the monitoring buffer and the semantic difference density is calculated. ;in This represents the differential triggering intensity per unit time, used to reflect the frequency of semantic layer updates. A preset semantic differential triggering threshold is also provided. When detected When this occurs, it is considered to be in a high-frequency semantic update state, requiring the scheduling module to adjust the priority. The judgment condition is expressed as follows: ;in This indicates that the system has entered the scheduling and adjustment mode. This indicates that the current scheduling state will be maintained. The judgment result and the sampled data are packaged into a feedback signal and sent to the scheduling module, forming a feedback input vector: .

[0067] Based on feedback signals, the scheduling module adjusts the refresh priority parameters of each task in the task scheduling table. The task scheduling table is defined as follows: ; in Indicates the task identifier; Assign a semantic layer number to the task; Priority for task refresh; Sets the task refresh rate. When... At that time, according to the semantic hierarchy weight function: Task priorities are redistributed. The new priority calculation formula is: ; where semantic layer number The smaller the value, the lower the level and the higher its weight. The smaller the value, the lower the refresh priority.

[0068] The rules for adjusting the task refresh rate are defined as follows: ;in Number the threshold layer; This is a frequency scaling factor used to reduce the refresh rate of lower semantic layers, thereby reducing the rendering computational load. (Adjusted scheduling table) It is written back to the system scheduling cache and takes effect in the next rendering cycle.

[0069] After the scheduling parameters are adjusted, task distribution is performed according to the new scheduling table. The linkage frequency control function is defined as follows: ;in, Indicates the first Layered tasks in time The linkage cycle length is determined. The trigger interval of each task layer is updated according to this function, thereby achieving dynamic frequency control based on semantic hierarchy.

[0070] At the end of each cycle, the feedback log is updated, and the current differential strength, number of task executions, and average response time are recorded in the system status table for reference in subsequent cycles.

[0071] In this embodiment, a real-time monitoring and feedback mechanism is introduced into the semantic linkage and rendering update stages. Task priority and refresh frequency are dynamically adjusted based on differential strength and resource load, thereby achieving adaptive linkage control between semantic levels and ensuring visualization continuity and system response stability in multi-layered semantic dynamic update scenarios. The input-output relationships between modules are clear, and the parameters are fully defined, enabling stable operation of dynamic feedback closed-loop control.

[0072] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0073] In one embodiment, such as Figure 2 As shown, a multi-scale visualization system for simulation and deduction data based on dynamic linkage is provided, including: The semantic model construction module 901 is used to acquire simulation scenario data, extract the attributes, behaviors and relationships of simulation objects and convert them into corresponding semantic units, and assign a unique identifier to each type of semantic unit to build a multi-layer semantic model.

[0074] The semantic graph construction module 902 is used to construct a multi-layer semantic graph based on the dependency relationships between semantic units in the multi-layer semantic layer model and the preset semantic mapping rules between layers, with semantic units as nodes and semantic dependencies between nodes as edges.

[0075] The update instruction generation module 903 is used to obtain multi-layer semantic graph snapshots at continuous time points in the simulation, perform semantic differential analysis based on the obtained multi-layer semantic graph snapshots, and generate differential update instructions.

[0076] The semantic graph update module 904 is used to trigger a dynamic linkage synchronization mechanism to perform state propagation and update of the multi-layer semantic graph, driven by differential update instructions, so as to obtain the updated multi-layer semantic graph. The dynamic linkage synchronization mechanism includes a differential trigger module and a state propagation module. The differential trigger module generates a propagation task queue according to the differential update instructions; the state propagation module accesses the semantic mapping table sequentially according to the propagation task queue.

[0077] The visualization module 905 is used to generate a multi-scale visualization dataset based on the updated multi-layer semantic graph and map it to the visualization interface according to the preset multi-scale view layout.

[0078] The monitoring and update module 906 is used to monitor the dynamic linkage synchronization process and visualization process in real time, and adaptively adjust the rendering priority and dynamic linkage frequency based on the monitoring data.

[0079] Specific limitations regarding the multi-scale visualization system for simulation data based on dynamic linkage can be found in the limitations of the multi-scale visualization method for simulation data based on dynamic linkage mentioned above, and will not be repeated here. Each module in the aforementioned multi-scale visualization system for simulation data based on dynamic linkage can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0080] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0081] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A dynamic linkage-based simulation deduction data multi-scale visualization method, characterized in that, Includes the following steps: Acquire simulation scenario data, extract the attributes, behaviors and relationships of simulation objects and convert them into corresponding semantic units, assign a unique identifier to each type of semantic unit, and construct a multi-layer semantic model. Based on the dependencies between semantic units in the multi-layer semantic model and the pre-defined semantic mapping rules between layers, a multi-layer semantic graph is constructed with semantic units as nodes and semantic dependencies between nodes as edges. Obtain snapshots of multi-layer semantic graphs at continuous time points in the simulation, perform semantic differential analysis based on the obtained snapshots, and generate differential update instructions. Driven by differential update instructions, a dynamic linkage synchronization mechanism is triggered to propagate and update the state of the multi-layer semantic graph, resulting in an updated multi-layer semantic graph. The dynamic linkage synchronization mechanism includes a differential triggering module and a state propagation module. The differential triggering module generates a propagation task queue according to the differential update instructions. The state propagation module accesses the semantic mapping table sequentially according to the propagation task queue. A multi-scale visualization dataset is generated based on the updated multi-layer semantic graph and mapped to the visualization interface according to the preset multi-scale view layout. Real-time monitoring of the dynamic linkage synchronization process and visualization process, and adaptive adjustment of rendering priority and dynamic linkage frequency based on monitoring data.

2. The dynamic-link-based simulation deduction data multi-scale visualization method according to claim 1, characterized in that, Acquire simulation scenario data, extract the attributes, behaviors, and relationships of simulation objects and transform them into corresponding semantic units, assign a unique identifier to each semantic unit, and construct a multi-layer semantic model, including: Acquire simulation scenario data, including static attribute data, behavior log data, and status parameters of the simulation objects; The simulation scenario data is decomposed into features. The static attributes of the objects are encapsulated into attribute semantic units in the form of key-value pairs. The behavior logs are encapsulated into behavior semantic units in the form of event sequences. Relationship semantic units are established based on the interaction relationships between the simulation objects. Each semantic unit is assigned a unique identifier, and all semantic units are divided into multiple semantic layers according to their level of abstraction. Each semantic layer maintains a logical connection through semantic dependency mapping, resulting in a multi-layer semantic layer model.

3. The dynamic-link-based simulation-reasoning-data multiscale visualization method according to claim 1, wherein, Based on the dependencies between semantic units in the multi-layer semantic model and the pre-defined semantic mapping rules between layers, a multi-layer semantic graph is constructed with semantic units as nodes and semantic dependencies between nodes as edges, including: Based on the dependency relationship between semantic units in the multi-layer semantic model, the lower-level semantics correspond to the object layer, and the higher-level semantics correspond to the global abstraction layer, and the semantic units are assigned to the corresponding semantic layers. Representing semantic units as graph nodes: wherein, is a semantic unit encoding function for converting different types of semantic units into a uniform node data format; is a node set; is a first i node; is a first i semantic unit; assigning a hierarchical number to each node and generating a mapping pointer , the mapping pointer being used to represent the inter-layer semantic correspondence relationship; Using semantic dependencies between nodes as edges: wherein, is a set of edges; is a set of nodes; j is a set of nodes; is a semantic influence weight between and ; is a dependency degree function between semantics of attributes; is a dependency degree function between semantics of behaviors; is a dependency degree function between semantics of relations; is a weight coefficient corresponding to attributes; is a weight coefficient corresponding to behaviors; is a weight coefficient corresponding to relations; A multi-layer semantic graph is constructed with semantic units as nodes and semantic dependencies between nodes as edges wherein, is a set of semantic layers.

4. The dynamic-link-based simulation-reasoning-data multiscale visualization method according to claim 3, characterized in that, Based on the dependencies between semantic units in the multi-layer semantic model and the pre-defined semantic mapping rules between layers, a multi-layer semantic graph is constructed with semantic units as nodes and semantic dependencies between nodes as edges. This also includes: Establish an inter-level index table based on mapped pointers: in, This is the index between a node in the k-th semantic layer and a node in the (k+1)-th semantic layer. For the k-th semantic layer i One node; For the (k+1)th semantic layer j One node; Generate an in-layer topology table based on edge sets: wherein, is a topology table for the k-th semantic layer.

5. The multi-scale visualization method for simulation and deduction data based on dynamic linkage according to claim 2, characterized in that, Obtain snapshots of multi-layer semantic graphs at continuous time points in the simulation, perform semantic differential analysis based on the obtained snapshots, and generate differential update instructions, including: Obtain snapshots of multi-layer semantic graphs at consecutive time points in the simulation: in, For time A multi-layered semantic graph; For time A multi-layered semantic graph; For time The set of nodes; For time The set of nodes; For time edge set; For time edge set; For time The semantic layer set; For time The semantic layer set; Will and Semantic difference analysis was performed on the three dimensions of attributes, behavior, and relationship to obtain attribute difference tables, behavior difference tables, and relationship difference tables. The node identifiers in the attribute difference table, behavior difference table, and relation difference table are aggregated to generate a difference update instruction.

6. The multi-scale visualization method for simulation and deduction data based on dynamic linkage according to claim 5, characterized in that, Driven by differential update instructions, a dynamic linkage synchronization mechanism is triggered to propagate and update the state of the multi-layer semantic graph, resulting in an updated multi-layer semantic graph, including: The differential triggering module parses the differential update instruction to obtain the timestamp, node identifier, and dependent node set, and generates a propagation task queue based on the timestamp. The state propagation module reads task items from the propagation task queue and, based on the node level number and the preset semantic mapping rules between levels, retrieves them from the inter-level index table. The corresponding mapping pointer is obtained to locate the position of the target node in the semantic layer and determine the update path. The state of the multi-layer semantic graph is propagated and updated according to the update path. After the node state data is updated, the results are aggregated to the higher layer, and then the inter-layer state merging is completed according to the semantic consistency constraint rules. The merged high-level state value is read by the lower-level nodes and a reverse synchronization update is performed, outputting the updated multi-layer semantic graph.

7. The multi-scale visualization method for simulation and inference data based on dynamic linkage according to claim 6, characterized in that, Driven by differential update instructions, a dynamic linkage synchronization mechanism is triggered to propagate and update the state of the multi-layer semantic graph, resulting in an updated multi-layer semantic graph, which also includes: The state propagation module sets a synchronization lock control flag on the access path. When a node enters the update state, the flag is set to 1, indicating that the node is in a locked state; after the update is completed, the flag is set to 0, indicating that the node can be accessed again.

8. The multi-scale visualization method for simulation and deduction data based on dynamic linkage according to claim 1, characterized in that, The visualization interface uses a layered rendering pipeline architecture.

9. The multi-scale visualization method for simulation and deduction data based on dynamic linkage according to claim 1, characterized in that, The preset semantic mapping rules between levels are as follows: in, This is the set of nodes in the k-th semantic layer; This is the node set of the (k+1)th semantic layer; This refers to the semantic mapping rules between the k-th semantic layer and the (k+1)-th semantic layer. When multiple lower-level nodes correspond to the same higher-level node, calculate the aggregation mapping weight: in, For the (k+1)th semantic layer j The aggregate weight of each node; for and The semantic influence weight between them.

10. A multi-scale visualization system for simulation and deduction data based on dynamic linkage, characterized in that, include: The semantic model building module is used to acquire simulation scenario data, extract the attributes, behaviors and relationships of simulation objects and convert them into corresponding semantic units, and assign a unique identifier to each type of semantic unit to build a multi-layer semantic model. The semantic graph construction module is used to construct a multi-layer semantic graph based on the dependencies between semantic units in the multi-layer semantic layer model and the preset semantic mapping rules between layers, with semantic units as nodes and semantic dependencies between nodes as edges; The update instruction generation module is used to obtain multi-layer semantic graph snapshots at continuous time points in the simulation, perform semantic differential analysis based on the obtained multi-layer semantic graph snapshots, and generate differential update instructions. The semantic graph update module is used to trigger a dynamic linkage synchronization mechanism to perform state propagation and update of the multi-layer semantic graph, driven by differential update instructions, to obtain the updated multi-layer semantic graph. The dynamic linkage synchronization mechanism includes a differential trigger module and a state propagation module. The differential trigger module generates a propagation task queue according to the differential update instructions. The state propagation module accesses the semantic mapping table sequentially according to the propagation task queue. The visualization module is used to generate multi-scale visualization datasets based on the updated multi-layer semantic graph and map them to the visualization interface according to the preset multi-scale view layout. The monitoring and update module is used to monitor the dynamic linkage synchronization process and visualization process in real time, and adaptively adjust the rendering priority and dynamic linkage frequency based on the monitoring data.

Images