Airport operation situation prediction method and system based on digital twinning

Abstract

The invention relates to the technical field of airport information processing, and provides an airport operation situation prediction method and system based on digital twinning, and the method comprises the steps: deploying a twinning space; wherein the twin space is used for representing a real airport geographic scene; loading at least one digital twin instance corresponding to the coordinate information and the entity object in the twin space; wherein the coordinate information is position information of the digital twin instance corresponding to the entity object; in response to a target scheduling instruction of the target digital twinning instance, executing hierarchical scheduling deduction of the target digital twinning instance in the twinning space to determine operation situation information of the target digital twinning instance in the real airport geographic scene; wherein the hierarchical scheduling deduction comprises a deduction space based on risk grading.

Description

Technical Field

[0001] The present invention relates to the technical field of airport information processing, and particularly relates to a method and system for predicting the operation situation of an airport based on digital twin. Background Art

[0002] Digital twin is a scenario simulation technology that constructs physical entities in a virtual space to achieve digital transformation with precise mapping. Through two-way real-time interaction, under the drive of data, it realizes the full life cycle management of the simulation scenario, simulates the operation mechanism of scenario devices, and realizes the global control and local control of the simulation scenario.

[0003] In terms of airport operation situation analysis, the existing technology realizes the dispatching control of the airport through big data technology. When digital twin technology is applied in the airport operation process, it is mainly used for monitoring the equipment status and analyzing the rationality in the airport dispatching process. In terms of control accuracy, it can only judge whether there are dispatching conflicts through the monitoring equipment and sensing equipment set in the airport after the plane descends or during the process of the plane moving its position.

[0004] Moreover, in terms of the accuracy of the existing digital twin technology, it can only determine the position information of different planes, equipment or fixed facilities on the airport map to realize the global digital twin supervision. The computing power will be allocated to each device, plane and fixed facility on the airport map, and only global supervision can be realized. There will be accuracy errors when singly supervising a single plane, equipment or fixed device. Therefore, it is impossible to realize the dispatching analysis under the spatio-temporal reference of the scenario and the plane, equipment or fixed facilities. Therefore, it is impossible to judge whether there are risks according to the dispatching instructions of the plane, as well as the risk level or accident level that the risk situation may cause. Summary of the Invention

[0005] The present application proposes a method for predicting the operation situation of an airport based on digital twin. The present application constructs a multi-level deduction by embedding the risk dimension into the geographical deduction space through the dynamic mapping of twin instances and the risk grading deduction space, and outputs the micro situation through instance-level deduction to realize the prediction of the airport operation status with high accuracy.

[0006] In a first aspect, the present application proposes a method for predicting the operation situation of an airport based on digital twin, including: Deploying a twin space; wherein, the twin space is used to represent the real airport geographical scenario; Loading at least one digital twin instance corresponding to coordinate information and an entity object in the twin space; wherein, the coordinate information is the position information of the digital twin instance corresponding to the entity object; In response to the target scheduling instruction of the target digital twin instance, perform hierarchical scheduling deduction of the target digital twin instance in the twin space to determine the operation situation information of the target digital twin instance in the real airport geographical scenario; wherein, the hierarchical scheduling deduction includes a deduction space based on risk classification.

[0007] During the operation of the airport in this application, the spatial positions of multiple entity objects can be synchronized through the twin space, and virtual deduction under dynamic risk constraints can be carried out to determine the airport operation situation information, reducing the conflict probability and delay time of airport operation.

[0008] Combined with the first aspect, computing nodes and geographical topology nodes are deployed inside the twin space; wherein, the computing nodes are used to process the first data, and the first data is the non-pre-scheduled dynamic data representing entity objects in the locally processed scheduling data; the geographical topology nodes are used to determine the dynamic boundary of the scheduling data according to the first data and the entity modeling data; wherein, the entity modeling data is the scene entity in the real airport geographical information.

[0009] This application processes local dynamic data through computing nodes to prevent data delay, and sets the boundaries of entity objects through the geographical topology results to prevent the disconnection between virtual deduction and real scene space constraints.

[0010] Combined with the first aspect, the loading of at least one digital twin instance corresponding to coordinate information and entity objects in the twin space includes: Determine at least one entity object to be loaded in the real airport geographical scenario, and determine the entity dependency relationship of at least one entity object to be loaded in the real airport geographical scenario; wherein, the entity dependency relationship includes a first dependency relationship and a second dependency relationship, the first dependency relationship is used to represent the geographical location information of the entity object to be loaded, and the second dependency relationship represents the three-dimensional layout information of the entity object to be loaded; Construct a composite coordinate system in the geographical scenario and the functional area to which at least one entity object to be loaded belongs according to the entity dependency relationship; Load the twin coordinate information of at least one entity object to be loaded in the twin space according to the first dependency relationship and the composite coordinate system; Load the three-dimensional layout information corresponding to the twin coordinate information in the twin space according to the second dependency relationship and the composite coordinate system to form at least one digital twin instance.

[0011] During the loading process of the digital twin instance in this application, combining a single coordinate system and two-dimensional position mapping can prevent incomplete structural restoration of three-dimensional entity objects in the virtual space and position deviation of entity objects in different functional areas due to inconsistent coordinates.

[0012] In combination with the first aspect, loading at least one digital twin instance corresponding to coordinate information and an entity object in the twin space further includes: Determine the scheduling attribute of the entity object according to the position information of the entity object in the real airport geographical scene; wherein, the scheduling attribute includes a fixed attribute and a movable attribute; Divide the scheduling instruction set of the entity object according to the scheduling attribute; Configure a synchronization response mechanism for the digital twin instance in the twin space according to the scheduling instruction set; wherein, the synchronization response mechanism configures an operable instruction set that the digital twin instance can respond to according to the scheduling attribute and the real airport geographical scene.

[0013] This application can prevent the confusion of instructions for fixed entities and movable entities. Through the instruction set and the synchronization response mechanism, the twin instance only responds to its operable instructions.

[0014] In combination with the first aspect, the target scheduling instruction is the first instruction in the operable instruction set; wherein, the first instruction is a target instruction without instruction conflict between the target digital twin instance and another digital twin instance in the real-time twin space. When there is a scheduling behavior in the target digital twin instance, a new operation for the current scheduling behavior is generated.

[0015] This application associates instruction selection with real-time conflict detection, prevents the selection of static instructions from causing conflicts in the real scenario, and combines new operations to prevent the inability to adapt to real-time changes during the instruction execution process.

[0016] In combination with the first aspect, the hierarchical scheduling deduction includes: Respond to the deduction data of the target digital twin instance executing the target scheduling instruction, and obtain the first deduction sample data; Determine the first deduction space for generating spatio-temporal annotations in the twin space according to the first deduction sample data; Execute scheduling conflict determination through the first deduction space in response to the second scheduling instruction synchronously executed during the time period of each spatio-temporal annotation in the twin space; Wherein, when there is no scheduling conflict, output the stable operation situation information of the real airport geographical scene; When there is a scheduling conflict, determine the conflict object, form a second deduction space based on the conflict object, and the second deduction space is used to form a third deduction space according to the risk level.

[0017] This application solves the conflict detection of multi-instruction parallelism and the unclear spatio-temporal relationship through spatio-temporal annotations, and prevents the inability to accurately locate local risks after global conflict detection through hierarchical deduction spaces.

[0018] In combination with the first aspect, the third deduction space is used to replace the scheduling operation of the target digital twin instance in the twin space with the scheduling operation of the conflicting object, determine the scheduling conflict scenario, and the first conflict behavior representing the scheduling conflict behavior in the twin space.

[0019] Through the replacement operation, this application actively simulates the conflict scenario, which can solve the problem of locating the root cause of the conflict and specifically converting the conflict behavior into a visual operation in the twin space.

[0020] In combination with the first aspect, the second deduction space is further used to determine a conflict instruction according to the conflicting object, and configure a risk level corresponding to the instruction source based on the instruction source of the conflict instruction; wherein, the risk level is matched with the corresponding risk items and control items in the real airport geographical scenario. The risk items are used to determine the ontological risk value of the conflicting object, and the control items are used to determine the diffusion risk value of the conflicting object according to the ontological risk value, and use the diffusion risk value as the target risk value for determining the risk level.

[0021] This application can solve the problems of the homogenization of conflict risks caused by instructions from different sources and only focusing on local conflict scenarios while ignoring object characteristics and global impacts.

[0022] In combination with the first aspect, the operation situation information includes the potential risk vector of the target digital twin instance in the real airport geographical scenario, and through the potential risk vector; wherein, the potential risk vector is the target risk value existing in the hierarchical scheduling deduction of the second digital twin instance caused by the scheduling behavior of the target digital twin instance under the target scheduling instruction.

[0023] During the evolution process, this application can quantify risks.

[0024] In the second aspect, this application proposes an airport operation situation prediction system based on digital twins, including: Space deployment module: used to deploy the twin space; wherein, the twin space is used to represent the real airport geographical scenario; Instance configuration module: used to load at least one digital twin instance corresponding to coordinate information and entity objects in the twin space; wherein, the coordinate information is the position information of the digital twin instance corresponding to the entity object. Deduction and prediction module: used to respond to the target scheduling instruction of the target digital twin instance, perform hierarchical scheduling deduction of the target digital twin instance in the twin space to determine the operation situation information of the target digital twin instance in the real airport geographical scenario; wherein, the hierarchical scheduling deduction includes a deduction space based on risk classification.

[0025] During the operation of the airport, this application can achieve the spatial position synchronization of multiple entity objects through the digital twin space, conduct virtual deduction under dynamic risk constraints, determine the airport operation situation information, and reduce the conflict probability and delay time of airport operation.

[0026] Other features and advantages of the present invention will be described in the following specification, and part of them will become obvious from the specification, or be understood by implementing the present invention. The objectives and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the written specification and the drawings.

[0027] The technical solution of the present invention will be further described in detail below through the drawings and embodiments. Description of the Drawings

[0028] The drawings are used to provide further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention.

[0029] In the drawings: Figure 1 is the architecture diagram of the airport digital twin system in the embodiment of the present invention; Figure 2 is the architecture diagram of the traditional airport digital twin system; Figure 3 is the method flow diagram of the airport operation situation prediction method based on digital twin in the embodiment of the present invention; Figure 4 is the process diagram of loading digital twin instances in the digital twin space in the embodiment of the present invention; Figure 5 is the processing diagram of the digital twin instances by the digital twin processing layer in the embodiment of the present invention; Figure 6 is the scheduling management diagram of the digital twin instances by the digital twin processing layer in the embodiment of the present invention; Figure 7 is the hierarchical deduction process diagram of the digital twin instances by the digital twin processing layer in the embodiment of the present invention. Detailed Embodiment

[0030] The following describes the preferred embodiments of the present invention with reference to the drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0031] In the existing airport digital twin architecture, refer to Figure 2 , mainly for different scenarios of the airport, monitor the airport, collect flight data of aircraft and aircraft position scheduling data, as well as ground handling control data, to realize the construction of the digital twin space of the overall airport.

[0032] It should be understood that the key to building a digital twin space lies in the digital mapping of different physical objects in the airport. In the existing technology, the airport tower can only collect airplanes, ground crew, special vehicles, etc. based on the monitoring devices and sensing devices inside the airport to achieve position simulation. However, aircraft monitoring and ground service monitoring are two different independent data - end monitoring, without the same spatio - temporal reference. Therefore, there are element conflicts between different control systems, and conflict prediction and control cannot be achieved.

[0033] In addition, referring to Figure 2 , traditional digital twins can only monitor the whole physical object. The position information of its location can only be monitored through point - by - point twins. During the monitoring process, the monitored conflicts can only be specific conflicts such as using the same runway at the same moment with the same arrival and departure positions between conflicting objects. Because the accuracy of traditional twin systems is insufficient, it is impossible to monitor and predict the unexpected conflicts that may occur between airplanes and special vehicles, airplanes and ground crew, special vehicles and ground crew, and between the geographical environment of physical entities and airplanes under dispatching instructions, nor can it judge the risk status caused by different dispatching instructions.

[0034] Based on the above problems, the present application provides the following solutions; Embodiment 1: Referring to Figure 3 , the present application proposes a method for predicting the operation situation of an airport based on digital twins, including: Step S100: Deploy a twin space; where the twin space is used to represent the real - world airport geographical scene, and its function is to construct a virtual mapping environment corresponding to the real airport; The twin space is a digital twin model. According to the geographical information of the real airport, it constitutes a virtual three - dimensional space based on GIS + BIM, integrating static scene data such as terrain, runways, and terminals. The twin space is a synchronous space for the operation situation environment for the loading and deduction of digital twin instances.

[0035] Step S101: Load at least one digital twin instance corresponding to coordinate information and an entity object into the twin space; where the coordinate information is the position information of the digital twin instance corresponding to the entity object; the digital twin instance is to map the real - world airport entity object into the twin space; The coordinate information is the specific information of the entity object. By modeling entity objects such as aircraft, ground support vehicles, and baggage systems as independent twins, real-time binding of spatial coordinates and the three-dimensional structure of the entity object, as well as real-time behavior, generates digital twin instances. The dynamic and static entity objects are precisely located in terms of space, time, occupied space, etc. in the virtual space, enabling spatio-temporal synchronous distribution simulation of entities in different systems but in the same regional space. The digital twin instance has corresponding attribute information, which includes coordinates, speed, fuel quantity, etc., through message queues, etc. The position information of the digital twin instance enables the twin space to reflect the distribution state of entity objects in the real airport in real time.

[0036] Step S102: In response to the target scheduling instruction of the target digital twin instance, perform hierarchical scheduling deduction of the target digital twin instance in the twin space to determine the operation situation information of the target digital twin instance in the real airport geographical scene; among them, the hierarchical scheduling deduction includes a deduction space based on risk classification. The hierarchical scheduling deduction is to predict the real operation situation through virtual deduction, which can reduce the calculation cost of undifferentiated deduction in the full scene and improve the pertinence of the prediction result at the same time.

[0037] Through the target scheduling instruction, the scheduling behavior of any digital twin entity is deduced, and the deduction sub-space is divided. The deduction sub-space is in different risk types and degrees of risk, and simulation deduction can be carried out based on different deduction sub-spaces. Simulation events are injected into the sub-space, and the behavior path of the twin is calculated through a preset rule engine. The risk classification is mainly set based on the specific deduction sub-space.

[0038] Finally, structured data such as airport traffic heat maps, conflict point warnings, and resource scheduling suggestions are generated through the deduction results, extending the digital twin technology from industrial scenarios to airport geographical scenarios, and solving the technical problem of spatial position synchronization of multiple entity objects; it can significantly reduce the probability of operation conflicts and delay time at the airport.

[0039] Embodiment 2: Refer to Figure 4 and Figure 1 In the twin space of the present application, a computing node 2010 and a geographical topology node 2020 are deployed, which belong to a distributed computing unit and are deployed in the server of the twin space; the computing node 2010 and the geographical topology node 2020 belong to the twin processing layer 20. The computing node 2010 and the geographical topology node 2020 have node division functions for improving the parallel processing ability of the twin space.

[0040] First, entity modeling is performed, such as step S1001: Obtain entity modeling data. The computing node 2010 will first determine the specific modeling data of the physical entity to be modeled through the geographical topology node 2020; Among them, the computing node 2010 is used to process the first data, which is the non-pre-scheduled dynamic data representing entity objects in the locally processed scheduling data (not the data generated by operations that should be performed in the regular plan); the scheduling data includes pre-scheduled data and non-pre-scheduled data, and the pre-scheduled data is the data in the plan such as the planned taxi time and fixed route, For example, in step S1002: According to the entity modeling data, determine the non-pre-scheduled dynamic data of the modeling entity in the local scheduling data to generate the first data. The computing node 2010 processes the non-pre-scheduled dynamic data in the local scheduling data, which can quickly respond to emergencies in the real airport. Since it does not involve cloud data processing and external servers, there is no delay.

[0041] The non-pre-scheduled dynamic data is the non-pre-scheduled dynamic data generated in the event of an emergency during the operation of the airport. The non-pre-scheduled dynamic data includes sudden equipment failures or meteorological and terrain damages, etc. Its function is to provide sample data in the prediction process; The geographical topology node 2020 is used to determine the dynamic boundary of the scheduling data according to the first data and the entity modeling data; among them, the entity modeling data is the scene entity in the real airport geographical information (the scene entity includes geographical entity, equipment entity and building entity). For example, in step S1003: Determine the scene entity of the modeling entity and the dynamic boundary of the scene entity in the twin space. The geographical topology node 2020 performs entity modeling according to the non-pre-scheduled dynamic data that appears in the real airport environment, calculates the influence range through the spatial topology relationship, and can determine the deduced subspace range according to the non-pre-scheduled dynamic data, limiting the deduced range to prevent the deduced result from conflicting with the real space.

[0042] Embodiment 3: Refer to Figure 5 and Figure 1 , this application loads at least one digital twin instance corresponding to coordinate information and entity objects in the twin space, including: The instance configuration module 202 determines the loaded entity object by receiving the data of the scene fixed entity and the scene moving entity of the space deployment module 201, and then executes instance deployment through the space deployment module 201. Determine at least one entity object to be loaded in the real airport geographical scene, and determine the entity dependency relationship of at least one entity object to be loaded in the real airport geographical scene; among them, the entity dependency relationship includes a first dependency relationship and a second dependency relationship. The first dependency relationship is used to represent the geographical location information of the entity object to be loaded, and the second dependency relationship represents the three-dimensional layout information of the entity object to be loaded; The first dependency relationship is a dependency relationship constructed for the coordinate data of the entity object corresponding to the geographical location information. The second dependency relationship realizes high-precision measurement between different entities through the relative spatial relationship between entities and the height hierarchy in the three-dimensional space, such as the vertical distance between the taxiway and the runway. The combination of the two realizes the unique topological constraint of the entity in the three-dimensional space, and neither can be missing. In terms of the accuracy of the twin model, it can achieve the determination of spatial distribution misalignment at the centimeter level; According to the entity dependency relationship, construct a composite coordinate system in the geographical scene and the functional area to which at least one entity object to be loaded belongs; The composite coordinate system is a spatio-temporal reference system that integrates the geographical scene (runway area, apron area) and the entity dependency relationship (aircraft taxiing function), and synchronizes the absolute position and relative layout of the equipment.

[0043] According to the first dependency relationship and the composite coordinate system, load the twin coordinate information of at least one entity object to be loaded in the twin space; the twin coordinate information accurately maps the plane position of the entity object in the real scene to the twin space, so that the plane position of the digital twin instance is exactly the same as that of the real entity object.

[0044] Loading the twin coordinate information based on the first dependency relationship can realize the spatial anchor point configuration of the entity object; According to the second dependency relationship and the composite coordinate system, load the three-dimensional layout information corresponding to the twin coordinate information in the twin space to form at least one digital twin instance. The three-dimensional layout information maps the three-dimensional structure of the entity object in the real scene to the twin space, so that the three-dimensional structure can also be mapped to the twin space, improving the restoration degree of the real scene corresponding to the digital twin instance in the twin space. It is to construct the spatial relationship network of the entity object and realize the full-dimensional mapping from the physical entity to the digital twin.

[0045] This application associates the entity dependency relationship with the subdivided plane and three-dimensional structures to prevent the incomplete restoration of the three-dimensional entity object in the twin space, and combines the composite coordinate system to integrate multi-source coordinate benchmarks, enabling entity objects with different functions to be mapped when the coordinates are not unified without position deviation.

[0046] Embodiment 4: Refer to Figure 6 and Figure 1 , this application loads at least one digital twin instance corresponding to the coordinate information and the entity object in the twin space, and further includes: According to the position information of the entity object in the real airport geographical scene, determine the scheduling attribute of the entity object; wherein, the scheduling attribute includes a fixed attribute and a movable attribute; The location information of entity objects in the real airport geographical scenario, including whether the entity object is a long-term fixed entity or an entity that moves depending on a power system, is classified according to the scheduling attribute (fixed device or mobile device). Under the attribute classification, it is possible to prevent incorrect instructions for object entities in the twin space and improve the pertinence of instructions.

[0047] Entity objects with fixed attributes belong to immovable entity objects such as jet bridges, tower control towers, and some fixed devices. Entity objects with movable attributes are movable objects such as airplanes, ground service vehicles, and engineers. Through the scheduling attribute, first, the accuracy of scheduling instructions can be verified. Second, the belonging location of each entity object and the movable boundary can be determined. According to the scheduling attribute, divide the scheduling instruction set of entity objects. The scheduling instruction set is the exclusive control instruction for each type of entity object. Fixed-attribute objects only respond to status query instructions, and movable-attribute objects need to respond to path planning instructions. It can filter out irrelevant instructions and eliminate illegal operations. At the same time, it can also prevent redundancy of the instruction set to a certain extent and reduce the computational load.

[0048] Configure the synchronous response mechanism of the digital twin instance in the twin space according to the scheduling instruction set. Among them, the synchronous response mechanism configures the operable instruction set that the digital twin instance can respond to according to the scheduling attribute and the real airport geographical scenario.

[0049] The synchronous response mechanism sets dynamic response rules in the twin space (for example: when the synchronous response mechanism is based on the MQTT protocol, when the real device receives an instruction, the twin instance synchronously triggers the deduction of the same instruction). It restricts the range of operable instructions according to the scheduling attribute, combines the real geographical scenario to constrain the validity of the instructions, realizes the dynamic contraction / expansion of the instruction set with the entity position, reduces the response delay from the second level to the millisecond level, and can ensure that the twin instance only responds to its operable instructions. In the twin space, there will also be no distortion of the deduction result due to irrelevant instructions.

[0050] Embodiment 5: Refer to Figure 6 and Figure 1 In this application, the target scheduling instruction is the first instruction in the operable instruction set. Among them, the first instruction is the target instruction without instruction conflict between the target digital twin instance and another digital twin instance in the real-time twin space. When there is a scheduling behavior in the target digital twin instance, a new operation for the current scheduling behavior is generated.

[0051] The first instruction is a valid instruction pre-screened in the operable instruction set. The first instruction can conform to the non-instruction conflict with another twin instance in the real-time twin space. When the target instruction is a scheduling behavior, new operations for the current behavior will be generated to achieve the unity of instruction safety and resource optimization. The operable instruction set is an exclusive instruction set configured based on scheduling attributes. The first instruction is an instruction preferentially selected from the exclusive instruction set to ensure the matching of the target scheduling instruction with the scheduling attributes of the digital twin instance. In actual implementation, if two aircraft use the same taxiway, it belongs to an instruction conflict. By detecting the instruction targets of both in real time (such as taxiway occupancy time, path), it is judged whether there is a conflict (such as time overlap or space overlap), and only the non-conflicting instructions are selected as the first instruction. When the target digital twin instance executes a scheduling behavior (such as an aircraft starting to taxi), the original instruction may need to be adjusted due to real-time scenario changes (such as sudden obstacles). The new operation is a dynamic supplement to the original instruction. By obtaining the deduction feedback in the twin space in real time (such as collision warning), the generation of new operations is triggered to prevent static instructions from not adapting to scenario changes.

[0052] Embodiment 6: Refer to Figure 7 and Figure 1 , the twin space of the present application can perform hierarchical scheduling deduction, including: In response to the deduction data of the target digital twin instance executing the target scheduling instruction, obtain the first deduction sample data; When the target digital twin instance (aircraft) executes the target instruction (taxi to any parking area), deduction data will be generated. The first deduction sample data is the deduction process data based on the target instance executing the target instruction, and an initial deduction data set can be generated, which can ensure that the deduction process is based on the dynamic data of real scheduling behavior and prevent deduction deviation due to data deficiency.

[0053] According to the first deduction sample data, determine the first deduction space for generating space-time annotations in the twin space; The first deduction space is a virtual deduction environment with space-time marks constructed in the twin space according to the deduction data; the space-time annotation is to add tags in the time dimension and space dimension to the first deduction sample data. The first deduction space is a virtual environment constructed based on the marked tags, simulating the overlapping situation of multiple scheduling instructions in time and space, making the space-time association visible, showing the execution trajectories of multiple instructions, and providing a clear analysis scenario for conflict determination.

[0054] Execute scheduling conflict determination through the second scheduling instruction synchronously executed by the first deduction space in response to the time period marked by each space-time in the twin space; the second scheduling instruction is the scheduling instruction of other digital twin instances in the same time period. Through the space-time annotation of the first deduction space, synchronously simulate the execution process of these instructions, detect whether there are conflicts such as time overlap (such as occupying the same taxiway at the same time) or space overlap (such as path crossing). The conflict determination is multi-instruction parallel conflict detection, avoiding missed conflict judgments caused by sequential execution detection.

[0055] Among them, when there is no scheduling conflict, output the stable operation situation information of the real airport geographical scene; if all the second scheduling instructions have no overlap with the target scheduling instruction in space-time, output the stable operation situation information of the real airport.

[0056] When there is a scheduling conflict, determine the conflict object, and form a second deduction space based on the conflict object. The second deduction space is used to form a third deduction space according to the risk level.

[0057] The hierarchical conflict in this application: Locate the conflict object (such as conflict aircraft A and ground service vehicle B), and then construct a second deduction space (conflict object exclusive sandbox). Finally, generate a third deduction space according to the risk level (such as simulating the collision consequences of a high-risk scenario). In the hierarchical conflict handling mechanism, without conflict, the stable operation situation information can be directly output. In the case of conflict, conflict handling is carried out according to the subspace. The subspace is highly correlated with the risk level and risk type, and high-precision deduction under low computing power can be achieved.

[0058] Embodiment 7: Refer to Figure 7 , the third deduction space of this application is used to replace the scheduling operation of the target digital twin instance in the twin space with the scheduling operation of the conflict object, determine the scheduling conflict scenario, and the first conflict behavior representing the scheduling conflict behavior in the twin space.

[0059] In the second deduction space, the conflict object has been identified (such as target digital twin instance A and conflict object B). The scheduling operation replacement mechanism of this application replaces the scheduling operation of the target digital twin instance with the scheduling operation of the conflict object in the third deduction space. For example: replace the landing instruction of aircraft A with the movement instruction of ground service vehicle B. Through the simulation of the replaced operation, determine the scheduling conflict scenario. Finally, dynamically generate the first conflict behavior in the twin space. The focus of this application is to actively construct the conflict scenario and deduce the micro behavior, locate the root cause and reason of the conflict. The scheduling conflict scenario refers to the specific space-time conditions when the conflict occurs. The third deduction space can determine detailed conflict scenario information by recording the execution process of the replaced scheduling operation and combining with space-time annotation, and transform the abstract conflict into observable and analyzable behavior.

[0060] Example 8: Refer to Figure 7 , the second deduction space of this application is also used to determine a conflict instruction according to a conflict object, and configure a risk level corresponding to the instruction source based on the instruction source of the conflict instruction; wherein, the risk level is matched with corresponding risk items and control items in the real airport geographical scenario, the risk items are used to determine the ontological risk value of the conflict object, and the control items are used to determine the diffusion risk value of the conflict object according to the ontological risk value, and use the diffusion risk value as the target risk value for determining the risk level.

[0061] Determining a conflict instruction according to a conflict object can realize the traceability of the conflict instruction. Furthermore, through the traceability analysis of the instruction, the risk level can be determined, the ontological risk value of the conflict object can be calculated, and then the diffusion risk value can be deduced according to the ontological risk value. Finally, the diffusion risk value is used as the target risk value to drive the update of the risk level. By analyzing the historical scheduling records and real-time deduction data of the conflict object, the instruction directly causing the conflict can be located. The instruction source refers to the issuing entity (such as the ground scheduling system, the tower control system) or type of the conflict instruction. The authorities and influence scopes of different instruction sources are different. Therefore, different risk levels need to be configured for them to achieve differential risk assessment. The risk items are risk factors related to the inherent attributes of the conflict object, and the ontological risk value is the quantification of the risk factors, avoiding only focusing on the conflict scenario and ignoring the object characteristics. The object characteristics may lead to small risk events and cause large risk consequences. The control items are factors affecting risk diffusion in the airport operation environment (such as the traffic density at the conflict location, the availability of the alternate taxiway, the response time of the rescue resources). The diffusion risk value is the quantification of the factors affecting risk diffusion, and the target risk value is the diffusion risk value, that is, the final basis for determining the risk level after integrating the ontological risk value and the control items.

[0062] Example 9: Refer to Figure 7 , the operation situation information of this application includes the potential risk vector of the target digital twin instance in the real airport geographical scenario, and through the potential risk vector; wherein, the potential risk vector is the target risk value existing in the hierarchical scheduling deduction of the second digital twin instance caused by the scheduling behavior of the target digital twin instance under the target scheduling instruction.

[0063] The operation situation information is a comprehensive representation of the airport operation state; The potential risk vector is to quantify the risk impact of the target instance behavior on the associated entity, representing the potential risk of the target digital twin instance; The risk conduction mechanism is that the scheduling behavior of the target instance causes the target risk value to occur in the deduction of the second digital twin instance (associated entity) (such as flight delays causing subsequent flight chain delays), preventing only focusing on the direct conflict object and ignoring the indirect risk of the target scheduling behavior to other instances.

[0064] Example 10: Referring to Figure 1 , this application proposes an airport operation situation prediction system based on digital twin. The system of this application includes a tower dispatching layer 10, a twin processing layer 20, and a data acquisition layer 30. The data acquisition layer 30 is used to collect object instances and real airport geographical scene data and upload them to the twin processing layer 20. The tower dispatching layer 10 is used to receive the hierarchical deduction data from the twin processing layer 20 and send scheduling instructions to the twin processing layer 20 to assist in real-time deduction.

[0065] This application is configured within the twin processing layer 20, which is the core processing module of the digital twin airport operation situation prediction system, and specifically includes: Spatial deployment module 201: Used to deploy the twin space; among them, the twin space is used to represent the real airport geographical scene, and its function is to construct a virtual mapping environment corresponding to the real airport; The twin space is a digital twin model. According to the geographical information of the real airport, it constitutes a virtual three-dimensional space based on GIS + BIM, integrating static scene data such as terrain, runways, and terminals. The twin space is a synchronous space for the operation situation environment built for the loading and deduction of digital twin instances.

[0066] Instance configuration module 202: Used to load at least one digital twin instance corresponding to coordinate information and an entity object in the twin space; among them, the coordinate information is the position information of the digital twin instance corresponding to the entity object; the digital twin instance is to map the real airport entity object to the twin space; The coordinate information is the specific information of the entity object. By modeling entity objects such as airplanes, ground service vehicles, and baggage systems as independent twins, real-time binding of the spatial coordinates and three-dimensional structure of the entity object, as well as real-time behavior, generates digital twin instances, realizing precise positioning of dynamic and static entity objects in terms of space, time, occupied space, etc. in the virtual space, enabling entity objects in different systems but in the same regional space to achieve spatio-temporal synchronous distribution simulation. The digital twin instance has corresponding attribute information, and the attribute information includes coordinates, speed, fuel quantity, etc., through message queues, etc. The position information of the digital twin instance enables the twin space to reflect the distribution state of entity objects in the real airport in real time.

[0067] Deduction and prediction module 203: Used to respond to the target scheduling instruction of the target digital twin instance and perform hierarchical scheduling deduction of the target digital twin instance in the twin space to determine the operation situation information of the target digital twin instance in the real airport geographical scene; among them, the hierarchical scheduling deduction includes a deduction space based on risk classification. The hierarchical scheduling deduction predicts the real operation situation through virtual deduction, which can reduce the calculation cost of undifferentiated deduction of the full scene and improve the pertinence of the prediction result at the same time.

[0068] Through the target scheduling instruction, the scheduling behavior of any digital twin entity is deduced, the deduction subspace is divided, and the simulation deduction can be carried out based on different deduction subspaces for different risk types and degrees. Simulation events are injected into the subspace, and the behavior path of the twin is calculated through the preset rule engine. The risk grading is mainly set based on the specific deduction subspace.

[0069] Finally, structured data such as airport traffic heat maps, conflict point warnings, and resource scheduling suggestions are generated through the deduction results, extending the digital twin technology from industrial scenarios to airport geographical scenarios, and solving the technical problem of spatial position synchronization of multiple entity objects; it can significantly reduce the probability of operation conflicts and delay time at the airport. Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and modifications.

Claims

1. A method for predicting the operation status of an airport based on digital twins, characterized in that, Including: Deploying a twin space; wherein, the twin space is used to represent the real airport geographical scenario; Loading at least one digital twin instance corresponding to coordinate information and an entity object in the twin space; wherein, the coordinate information is the location information of the digital twin instance corresponding to the entity object; In response to a target scheduling instruction of a target digital twin instance, performing hierarchical scheduling deduction of the target digital twin instance in the twin space to determine the operation situation information of the target digital twin instance in the real airport geographical scenario; wherein, the hierarchical scheduling deduction includes a deduction space based on risk classification.

2. The method for predicting the operation situation of an airport based on digital twin according to claim 1, wherein, A computing node and a geographical topology node are deployed inside the twin space; wherein, the computing node is used to process first data, and the first data is non-pre-scheduled dynamic data representing entity objects in the locally processed scheduling data; the geographical topology node is used to determine the dynamic boundary of the scheduling data according to the first data and entity modeling data; wherein, the entity modeling data is a scene entity in the real airport geographical information.

3. The method for predicting the operation situation of an airport based on digital twin according to claim 1, wherein, The loading of at least one digital twin instance corresponding to coordinate information and an entity object in the twin space includes: Determining at least one entity object to be loaded in the real airport geographical scenario, and determining the entity dependency relationship of at least one entity object to be loaded in the real airport geographical scenario; wherein, the entity dependency relationship includes a first dependency relationship and a second dependency relationship, the first dependency relationship is used to represent the geographical location information of the entity object to be loaded, and the second dependency relationship represents the three-dimensional layout information of the entity object to be loaded; Constructing a composite coordinate system in the geographical scenario and the functional area to which at least one entity object to be loaded belongs according to the entity dependency relationship; Loading the twin coordinate information of at least one entity object to be loaded in the twin space according to the first dependency relationship and the composite coordinate system; Loading the three-dimensional layout information corresponding to the twin coordinate information in the twin space according to the second dependency relationship and the composite coordinate system to form at least one digital twin instance.

4. The method for predicting the operation situation of an airport based on digital twin according to claim 3, wherein The loading of at least one digital twin instance corresponding to coordinate information and an entity object in the twin space further includes: Determining the scheduling attribute of the entity object according to the location information of the entity object in the real airport geographical scenario; wherein, the scheduling attribute includes a fixed attribute and a movable attribute; Dividing the scheduling instruction set of the entity object according to the scheduling attribute; Configuring a synchronous response mechanism for the digital twin instance in the twin space according to the scheduling instruction set; wherein, the synchronous response mechanism configures an operable instruction set that the digital twin instance can respond to according to the scheduling attribute and the real airport geographical scenario.

5. The method for predicting the operation situation of an airport based on digital twin according to claim 4, wherein, The target scheduling instruction is the first instruction in the operable instruction set; wherein, the first instruction is a target instruction without instruction conflict between the target digital twin instance and another digital twin instance in the real-time twin space, and when there is a scheduling behavior of the target digital twin instance, a new operation of the current scheduling behavior is generated.

6. The method for predicting the operation situation of an airport based on digital twin according to claim 1, wherein The hierarchical scheduling deduction includes: In response to the deduction data of the target digital twin instance executing the target scheduling instruction, obtaining first deduction sample data; Determining a first deduction space for generating spatio-temporal annotations in the twin space according to the first deduction sample data; Execute scheduling conflict determination through the second scheduling instruction synchronously executed by the first deduction space in response to the time period marked by each space-time in the twin space; Among them, when there is no scheduling conflict, output the stable operation situation information of the real airport geographical scenario; When there is a scheduling conflict, determine the conflict object, form a second deduction space based on the conflict object, and the second deduction space is used to form a third deduction space according to the risk level.

7. The method for predicting the operation situation of an airport based on digital twin according to claim 6, characterized in that, The third deduction space is used to replace the scheduling operation of the target digital twin instance in the twin space with the scheduling operation of the conflict object, determine the scheduling conflict scenario, and the first conflict behavior representing the scheduling conflict behavior in the twin space.

8. The method for predicting the operation situation of an airport based on digital twin according to claim 6, characterized in that, The second deduction space is also used to determine the conflict instruction according to the conflict object, and configure the risk level corresponding to the instruction source based on the instruction source of the conflict instruction; among them, the risk level is matched with the corresponding risk items and control items in the real airport geographical scenario, the risk item is used to determine the ontology risk value of the conflict object, and the control item is used to determine the diffusion risk value of the conflict object according to the ontology risk value, and use the diffusion risk value as the target risk value for determining the risk level.

9. The method for predicting the operation situation of an airport based on digital twin according to claim 8, wherein, The operation situation information includes the potential risk vector of the target digital twin instance in the real airport geographical scenario, and passes through the potential risk vector; among them, the potential risk vector is the target risk value existing in the hierarchical scheduling deduction of the second digital twin instance caused by the scheduling behavior of the target digital twin instance under the target scheduling instruction.

10. An airport operation situation prediction system based on digital twin, characterized in that, Include: Space deployment module: used to deploy the twin space; among them, the twin space is used to represent the real airport geographical scenario; Instance configuration module: used to load at least one digital twin instance corresponding to the coordinate information and the entity object in the twin space; among them, the coordinate information is the position information of the digital twin instance corresponding to the entity object. Deduction prediction module: used to respond to the target scheduling instruction of the target digital twin instance, execute the hierarchical scheduling deduction of the target digital twin instance in the twin space, so as to determine the operation situation information of the target digital twin instance in the real airport geographical scenario; among them, the hierarchical scheduling deduction includes a deduction space based on risk classification.

Images