Simulation integration system and method based on a graph architecture

Abstract

The application provides a simulation integrated system and method based on a graph architecture, wherein entities participating in simulation are represented by using a knowledge graph, and a situation subgraph representing a simulation scene and a current state of the entities in the simulation scene is generated. The situation subgraph constructed based on the knowledge graph structure can accommodate entities of different granularities. After uniform time stamp timing, according to simulation task requirements, relevant entities participating in simulation are simulated according to the attributes of the current entities, and the calculation results are synchronously updated to the situation subgraph. Because the granularity of the entity models in the situation subgraph is different, different granularities of entities can be constructed through the relationship between the entity models, and the entities can be quickly adjusted and generated according to simulation task requirements, thereby solving the problem that the current simulation system needs to have a uniform model granularity and it is difficult to adjust the model.

Description

Technical Field

[0001] This invention relates to the field of system simulation, and in particular to a simulation integration system and method based on a graph architecture. Background Technology

[0002] Currently, personnel face significant challenges in training due to the high cost of equipment and the difficulty in organizing and scheduling personnel. With the development of computer systems and real-world simulation technologies, the use of simulation systems for training has become increasingly important. Currently, simulation systems primarily use the LVC (Low Voltage Conversion) simulation framework as their foundation, requiring all entity models to adhere to a unified interface and possess a consistent model granularity. If the model granularity needs adjustment, all entity models in the overall simulation system must be adjusted uniformly, resulting in poor scalability for different business simulation requirements. Secondly, existing simulation systems cannot be used interchangeably for simulations of different granularities because the interfaces for different granularity models are different. Furthermore, modeling an entity in a simulation system includes physical models, data models, and computational models, which are interdependent and coupled, hindering rapid adjustment and reconstruction of models to meet specific needs. Summary of the Invention

[0003] The technical problem to be solved by this invention is how to overcome the shortcomings of the prior art, and a simulation integration system and method based on graph architecture is proposed.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A simulation integration system based on a graph architecture includes a basic knowledge graph, an input module, a situational subgraph, a spatiotemporal unified module, and a computation module; The aforementioned basic knowledge graph is a general knowledge graph for the domain to be simulated. The input module is used to obtain simulation task requirements. The situational subgraph is used to extract subgraphs related to the equipment and loads involved in the simulation task from the basic knowledge graph, instantiate the nodes in the subgraph to form entities according to the simulation task requirements, assign attributes to each entity, and construct specific interaction relationships for each entity based on the connection relationships in the subgraph. The spatiotemporal unification module is used to provide unified timestamps for each entity in the situational subgraph. The calculation module has a calculation model for each entity in the situational subgraph. According to the simulation task requirements, it calculates the simulation task for entity pairs with task-related relationships using the corresponding calculation model and displays the calculation results as simulation results.

[0005] Furthermore, the calculation module stores the calculated results of the entity pairs with task associations in the corresponding entities of the situation subgraph according to the simulation task requirements, and updates the corresponding entity results of the situation subgraph.

[0006] Furthermore, when the calculation module performs calculations at any simulation time point, it extracts the entity data of the current time in the situation sub-graph for calculation.

[0007] This invention also provides a simulation integration method based on a graph architecture, comprising the following steps: Step 1: Construct a general-purpose basic knowledge graph based on the simulation domain; Step 2: Obtain simulation task requirements; Step 3: Based on the equipment and payloads involved in the simulation task requirements, extract the knowledge graph subgraphs corresponding to the equipment and payloads from the basic knowledge graph; Step 4: Instantiate the nodes in the knowledge graph subgraph according to the equipment and load to form entities, and assign attributes to each entity. Based on the relationship between the nodes in the knowledge graph subgraph, construct the connection relationship for the equipment entity and the load entity to form the situation subgraph in the simulation task requirements. Step 5: Set a unified timestamp for each entity in the situation subgraph. When the simulation task starts, each entity related to the task calculates its simulation state based on its own calculation model and the entity attributes at this time, and outputs the simulation results.

[0008] Furthermore, in step 5, each calculation model saves the calculation results on the corresponding entities and updates the corresponding entity results in the situation subgraph.

[0009] Furthermore, when assigning unified timestamps to all entities in the situational subgraph, speedup ratios can be used to accelerate the progress of the simulation task, depending on the needs of the simulation task.

[0010] Furthermore, when the computational model performs simulation calculations for any simulation point in time, the entity data at the current moment in the situational sub-graph is extracted for calculation.

[0011] By adopting the above technical solution, the present invention has the following beneficial effects: This invention provides a simulation integration system and method based on a knowledge graph architecture. By representing entities participating in the simulation using a knowledge graph, a situational subgraph representing the simulation scenario and the current state of the entities participating in the simulation task is generated. This invention uses a knowledge graph-based simulation inference structure to generate the situational subgraph, which can flexibly construct entity models of different granularities according to simulation business needs. Entity attributes of different granularities can be accessed on demand through the relationships between entity models. After unified timestamp synchronization, simulation calculations are performed on the relevant entities participating in the simulation based on their current attributes, and the calculation results are synchronously updated to the situational subgraph. Because the entity models in the situational subgraph have different granularities, entities of different granularities can be constructed through the relationships between entity models, and entities can be quickly adjusted and generated according to simulation task needs, solving the problems of current simulation systems requiring a unified model granularity and the difficulty of adjusting models. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of the structure of the present invention; Figure 2 This is a system flowchart of the present invention. Detailed Implementation

[0013] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] The following provides a specific embodiment of a simulation integration system based on a graph architecture according to the present invention, such as... Figure 1 As shown, it includes a basic knowledge graph, an input module, a situational subgraph, a spatiotemporal unified module, and a computation module; the basic knowledge graph is a general knowledge graph for the domain to be simulated.

[0015] The basic knowledge graph in this embodiment mainly includes prior knowledge of the equipment, providing a knowledge base for generating the situational subgraph of the mission. For the simulated equipment platform, each equipment platform is treated as an entity, such as a satellite platform or a ground mobile platform; for the payloads mounted on the equipment platform, each payload is treated as an entity, such as a visible light camera on a satellite platform; there are relationships between equipment platform entities and payload entities, with equipment platform entities carrying payload entities; there are relationships between equipment platform entities and entities of the same type, with equipment platform entities carrying other equipment platform entities; there are relationships between payload entities, with payload entities depending on other payload entities; and there are relationships between payload entities and equipment platforms, with payload entities depending on equipment platform entities.

[0016] The input module is used to obtain the simulation task requirements.

[0017] The situational subgraph is used to extract subgraphs related to the equipment and payloads participating in the simulation task from the basic knowledge graph. Nodes in the subgraph are instantiated into entities according to the simulation task requirements, attributes are assigned to each entity, and specific interaction relationships are constructed for each entity based on the connections in the subgraph. The basic knowledge graph provides some basic knowledge relationships within the domain to be simulated. To quickly construct the knowledge graph needed for the simulation task, only a portion of the knowledge graph related to the equipment and payloads of the simulation task needs to be extracted from the basic knowledge graph and instantiated into entities. The instantiated situational subgraph, which is the situational subgraph oriented towards the simulation scenario, mainly includes the equipment entities and payloads participating in the simulation scenario. Equipment is constructed according to the simulation's predetermined requirements, and entity information is saved and updated.

[0018] For each piece of equipment constructed, each piece of equipment is considered an entity. The main attributes of each equipment entity include equipment category, equipment model, equipment name, and timestamp. For each simulated equipment platform, each equipment platform is considered an entity. For each payload mounted on an equipment platform, each payload is considered an entity. Relationships exist between equipment platform entities and equipment entities; an equipment platform entity belongs to an equipment entity. Relationships also exist between payload entities and equipment entities. Specifically, an equipment platform entity carries a payload entity. Furthermore, an equipment platform entity has relationships with entities of the same type; an equipment platform entity carries another equipment platform entity. Relationships exist between payload entities; a payload entity depends on another payload entity. A payload entity also depends on an equipment platform entity. Payload entities interact with equipment entities; this is expressed as a payload on one piece of equipment being able to interact with another piece of equipment, such as a visible light satellite payload being able to photograph a ground target. The main attributes of these relationships include interaction type (strike, reconnaissance, etc.) and interaction value.

[0019] The spatiotemporal unification module is used to provide unified timestamps for all entities in the situational awareness subgraph. For simulation purposes, each entity in the situational awareness subgraph needs a unified time node when performing calculations according to the simulation task. This ensures that each entity has a unified time basis when using its own computational model, synchronizing the simulation results. The spatiotemporal unification module can also set the start time, end time, and speedup ratio of the simulation task. After the simulation task begins, it maintains the current simulation time and speedup ratio and broadcasts them.

[0020] The calculation module has a calculation model for each entity in the situational subgraph. According to the simulation task requirements, it calculates the simulation task for entity pairs with task-related relationships using the corresponding calculation model and displays the calculation results as simulation results.

[0021] In this embodiment, since the situational subgraph constructed based on the knowledge graph structure can accommodate entities of different granularities, after unified timestamp synchronization, the relevant entities participating in the simulation perform simulation calculations according to the current entity's attributes, and the calculation results are synchronously updated to the situational subgraph. Because the entity models in the situational subgraph have different granularities, entities of different granularities can be constructed through the relationships between entity models, and entities can be quickly adjusted and generated according to the simulation task requirements, solving the problem that current simulation systems need to have a unified model granularity and that adjusting models is difficult. Furthermore, since the calculation models are independent, the calculation model of a specific entity can be adjusted individually.

[0022] In this embodiment, the calculation module stores the calculated results of entity pairs with task associations in the corresponding entities of the situation subgraph according to the simulation task requirements, and updates the corresponding entity results of the situation subgraph.

[0023] In this embodiment, when the calculation module performs calculations at any simulation time point, it extracts the entity data of the current time in the situation sub-graph for calculation.

[0024] This invention also provides a simulation integration method based on a graph architecture, such as... Figure 2 As shown, it includes the following steps: Step 1: Construct a general basic knowledge graph based on the simulation domain; the basic knowledge graph in this embodiment mainly includes equipment prior knowledge, which provides a knowledge base for the generation of the mission situation subgraph.

[0025] Step 2: Obtain simulation task requirements.

[0026] Step 3: Based on the equipment and payloads involved in the simulation task requirements, extract the corresponding knowledge graph subgraphs from the basic knowledge graph. Since the basic knowledge graph is quite large, only the subgraphs related to the simulation task requirements need to be extracted. These subgraphs will reveal the dependencies between the equipment and payloads, laying the foundation for entity instantiation.

[0027] Step 4: Instantiate the nodes in the knowledge graph subgraph based on the equipment and load to form entities, and assign attributes to each entity. Based on the relationships between nodes in the knowledge graph subgraph, construct connection relationships for the equipment entities and load entities to form a situation subgraph in the simulation task requirements. In this embodiment, only the part of the knowledge graph related to the simulation task is extracted from the basic knowledge graph for instantiation. The instantiated situation subgraph, which is the situation subgraph oriented towards the simulation scenario, can store simulation entity information of different granularities, mainly including equipment entities participating in the simulation scenario. Equipment is constructed according to the simulation scenario requirements, and entity information is saved and updated. When constructing simulation entities, the entity body is constructed first, and then the corresponding load capacity entity is selected according to the simulation simulation requirements. The association relationship between the body and the load is constructed by constructing dependency relationships and mounting relationships, and the entity model is formed as needed. The construction method of the situation subgraph is as follows: First, based on the simulation task requirements, obtain entity information and establish the entity entity; Then, based on the capability requirements of the simulation task entity, obtain the corresponding equipment platform or payload; Retrieve the knowledge graph; if this entity has dependencies associated with it, retrieve the dependent entities. Set detailed equipment values ​​according to the simulation task requirements; Once the equipment entities are constructed, the interaction relationships between them are established according to the simulation task requirements.

[0028] Step 5: Set a unified timestamp for each entity in the situation subgraph. When the simulation task starts, each entity related to the task calculates its simulation state based on its own calculation model and the entity attributes at this time, and outputs the simulation results.

[0029] In this embodiment, each computational model saves the computation results on the corresponding entities and updates the corresponding entity results of the situation subgraph.

[0030] In this embodiment, while providing unified timestamps for each entity in the situational subgraph, the speedup ratio can also be used to accelerate the progress of the simulation task as needed.

[0031] In this embodiment, when the calculation model performs simulation calculations for any simulation time point, the entity data at the current time in the situation sub-graph is extracted for calculation.

[0032] This application utilizes a knowledge graph-based situational awareness subgraph that can accommodate entities of varying granularities. After unified timestamp synchronization, relevant entities participating in the simulation perform simulation calculations based on their current attributes, and the results are synchronously updated in the situational awareness subgraph. Because the entity models in the situational awareness subgraph have different granularities, entities of different granularities can be constructed through the relationships between them. Entities can be quickly adjusted and generated according to simulation task requirements, solving the problem of current simulation systems requiring a unified model granularity and facing difficulties in model adjustment. Furthermore, since the computational models are independent, the computational model of a single entity can be adjusted individually.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A simulation integration system based on a graph architecture, characterized in that, It includes a basic knowledge graph, an input module, a situational subgraph, a spatiotemporal unification module, and a computation module; The aforementioned basic knowledge graph is a general knowledge graph for the domain to be simulated. The input module is used to obtain simulation task requirements. The situational subgraph is used to extract subgraphs related to the equipment and loads involved in the simulation task from the basic knowledge graph, instantiate the nodes in the subgraph to form entities according to the simulation task requirements, assign attributes to each entity, and construct specific interaction relationships for each entity based on the connection relationships in the subgraph. The spatiotemporal unification module is used to provide unified timestamps for each entity in the situational subgraph. The calculation module has a calculation model for each entity in the situation subgraph. According to the simulation task requirements, it calculates the simulation task for entity pairs with task-related relationships using the corresponding calculation model and displays the calculation results as simulation results. The calculation module stores the calculated results of entity pairs with task associations in the corresponding entities of the situation subgraph according to the simulation task requirements, and updates the corresponding entity results of the situation subgraph. When the calculation module performs calculations at any simulation time point, it extracts the entity data of the current time in the situation subgraph for calculation.

2. A simulation integration method based on a graph architecture, characterized in that, Includes the following steps: Step 1: Construct a general-purpose basic knowledge graph based on the simulation domain; Step 2: Obtain simulation task requirements; Step 3: Based on the equipment and payloads involved in the simulation task requirements, extract the knowledge graph subgraphs corresponding to the equipment and payloads from the basic knowledge graph; Step 4: Instantiate the nodes in the knowledge graph subgraph according to the equipment and load to form entities, and assign attributes to each entity. Based on the relationship between the nodes in the knowledge graph subgraph, construct the connection relationship for the equipment entity and the load entity to form the situation subgraph in the simulation task requirements. Step 5: Set a unified timestamp for each entity in the situation subgraph. When the simulation task starts, each entity related to the task calculates its simulation state according to its own calculation model and the entity attributes at this time, and outputs the simulation results. Each computational model stores the computation results on the corresponding entities and updates the corresponding entity results in the situation subgraph; When the computational model performs simulation calculations for any simulation point in time, the entity data at the current moment in the situational sub-graph is extracted for calculation.

3. The simulation integration method according to claim 2, characterized in that, When assigning unified timestamps to all entities in the situational subgraph, speedup ratios can also be used to accelerate the progress of the simulation task, depending on the needs of the simulation task.

Images