Flight operation guarantee intelligent collaborative intervention method based on DT and complex network

Abstract

The invention discloses a flight operation guarantee intelligent collaborative intervention method based on DT and a complex network, and the method comprises the following steps: 1, building a flight ground operation guarantee digital twinborn model based on a data driving method, and achieving the virtual-real synchronization; step 2, establishing a flight ground operation guarantee complex network model; and step 3, flight operation guarantee intelligent collaborative intervention based on the digital twin model and the complex network model. Compared with the prior art, the method has the positive effects that the method has great value and practical significance for improving the traffic capacity of the airport and enhancing resource guarantee. Along with continuous development of the aviation transportation industry and complication of flight guarantee work, the method has wide application prospects, can be applied to various large airports, facilitates optimization of flight guarantee work procedures and improvement of operation efficiency, and meanwhile plays a positive role in promoting modernization management and intelligent development of the aviation guarantee industry.

Description

Technical Field

[0001] The present invention relates to an intelligent collaborative intervention method for flight operation support based on DT (digital twin technology) and complex networks. Background Art

[0002] The core business of airport transportation production organizations is all centered around flight operation support. Facing the full-process business scenario of flight support, the design and implementation of collaborative perception and intelligent decision-making systems that combine digital twins with complex system science has a guiding and driving role in promoting the construction of smart airports. From the perspective of soft science, the construction of smart civil aviation is not only about the renewal of digital and intelligent equipment and technological transformation, but also about the transformation and innovation of behavior patterns, decision-making methods and decision-making mechanisms, involving management processes, business process optimization and reengineering, and other aspects.

[0003] In recent years, with the rapid development and application of technologies such as the Internet of Things, cloud computing, edge computing, big data, and artificial intelligence, highly reliable digital twin intelligent management systems with the characteristics of fast response speed, strong flexible reconstruction capabilities, and process visualization have become a research hotspot. Digital twins use digital twins of real physical systems to achieve real-time graphical visualization of complex system operation data, scenario-based, and deep human-machine-object interaction.

[0004] Airport flight operation support is a complex system engineering project. There are intricate connections between various operations, which interact, influence and restrict each other, forming a highly complex causal chain coupling system. Therefore, the key to improving flight regularity and the efficiency of departure flight release lies in strengthening the synergy between the various elements of the flight support process. The larger the fleet size, the greater the turnover of passengers, baggage, cargo and mail, and the more important the collaborative work. With the accelerated recovery of the industry, the contradiction between the rapid growth of flight volume and the insufficient resource guarantee capacity is still prominent. It is urgent to use digital means to solve the management process, business process optimization and reengineering problems in civil aviation management and operation, as well as the improvement and innovation of decision-making methods and decision-making mechanisms, guide the development of underlying technologies and their application in civil aviation, and support the construction of smart airports.

[0005] Traditional empirical management is mainly based on manual guidance and scheduling, which makes on-site management difficult and easily leads to crowded and disordered sites and huge resource consumption. Bottleneck problems include the inability to accurately identify key processes that require intervention, intervention / control is not timely, intervention / control means are single, and there is insufficient motivation for business integration and innovation. Based on the data visualization of digital twin technology, the present invention uses complex network theory and methods to solve the problem of collaborative management of airport flight support. In response to the needs of flight delay prevention and control, with the research on process resource relationships as the main line, the modeling of complex systems for flight operation support and the research on their dynamic behaviors as the focus, and collaborative optimization control as the goal, it is committed to promoting the research and development of intelligent collaborative control technology and equipment for airport flight operation support, in line with the attributes of "demand traction, breaking bottlenecks". Summary of the invention

[0006] In order to overcome the above shortcomings of the prior art, the present invention proposes an intelligent collaborative intervention method for flight operation support based on DT and complex networks. The present invention is based on data visualization of digital twin technology, uses complex network theory and methods, and builds a collaborative perception and intelligent decision-making system for the full-process business scenario of flight support. The research on the fusion application of digital twin technology and complex networks is conducive to promoting industry data fusion sharing and value mining, and is of great significance to enhancing the airport's global perception capabilities, improving flight support collaborative services and lean management levels.

[0007] The technical solution adopted by the present invention to solve the technical problem is: an intelligent collaborative intervention method for flight operation guarantee based on DT and complex network, comprising the following steps:

[0008] Step 1: Establish a digital twin model of flight ground operation support based on a data-driven approach to achieve virtual-real synchronization;

[0009] Step 2: Establish a complex network model for flight ground operation support;

[0010] Step 3: Intelligent collaborative intervention for flight operation support based on digital twin models and complex network models.

[0011] Compared with the prior art, the present invention has the following positive effects:

[0012] Aiming at the needs of airport's global collaborative operation and full-process passenger intelligent service, the present invention establishes a digital twin model and collaborative optimization algorithm for flight ground operation support based on a data-driven method. Multi-source heterogeneous data and twin information processing methods such as digital twin identification information, spatial location information, process status, resource status information, flight status information, etc. are proposed. From the perspective of business logic, four types of information models, namely behavior data, rule data, object data and strategy data, are defined, and a digital twin virtual-real synchronization mechanism for flight ground operation support with data acquisition and drive as the core is established. Under the dependent network architecture of flight support collaborative control, the Unreal Engine (UE) and Python integrated development tools are used, combined with information model application, to solve intelligent business problems such as key process identification, key path length detection, collaborative control, etc. in the full-process operation scenario, and the global situation awareness, prediction, decision-making and control are realized through the fusion application of digital twin technology (DT) and complex network. Finally, taking the departure passenger guidance service in the terminal of a hub airport in the southwest region as an example, a simulation experiment is carried out in the developed flight operation support three-dimensional visualization decision-making system. The results show that the critical task identification and collaborative intervention scheme enables passengers expected to be delayed to complete boarding smoothly, shortening the passenger service guarantee time by an average of about 3′17″.

[0013] As a carrier of innovative information technology, the digital twin airport is a typical implementation path for smart airports. By modeling and simulating the operating status of physical systems, it realizes global situational awareness, information collaboration and intelligent prediction, effectively empowers applications at all levels of the airport, including the decision-making, management, operation and operational levels, and continuously guides the continuous improvement of airport security, operational efficiency and service experience. The present invention has great value and practical significance for improving airport traffic capacity and enhancing resource security. With the continuous development of the air transportation industry and the complexity of flight support operations, the method of the present invention has broad application prospects and can be applied to various large airports, which is conducive to optimizing flight support operation processes and improving operational efficiency. At the same time, it also plays a positive role in promoting the modernization and intelligent development of the aviation support industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will now be described by way of example with reference to the accompanying drawings, in which:

[0015] Figure 1 Develop process flow charts for mission-critical discovery and collaborative intervention. DETAILED DESCRIPTION

[0016] An intelligent collaborative intervention method for flight operation support based on DT and complex networks, including the following contents: 1 Construction of digital twin system for flight ground operation support

[0017] 1.1 Functional architecture of digital twin system

[0018] The flight operation support three-dimensional visualization decision-making system consists of a digital twin visualization platform and a flight operation support collaborative decision-making subsystem. Among them, the technical architecture of the digital twin visualization platform includes four main levels: data acquisition and storage layer, modeling and simulation layer, visualization and interaction layer, and analysis and optimization layer. The flight operation support collaborative decision-making subsystem is divided into three application levels: flight support process network analysis and modeling subsystem (Service process network Analysis and Modeling S / S, referred to as SPN S / S), service resource network analysis and modeling subsystem (Service resources network Analysis and Modeling S / S, referred to as SRN S / S), and adaptive scheduling resource management system (Adaptive Scheduling Resource-based system, referred to as ASS) under the dependent network framework formed by the coupling of dual networks. The digital twin and collaborative decision-making subsystem realize real-time data synchronization and bidirectional mapping through UE5 and Python data interfaces. The flight support global digital twin provides comprehensive and accurate data support for complex network optimization and control algorithms. The target optimization strategy based on the complex network is fed back to the physical world in real time through the information model to affect its behavior.

[0019] 1.2 Virtual-Real Synchronization Architecture of Digital Twin System

[0020] Based on the five-dimensional modeling framework of digital twins, the present invention constructs a digital twin model of flight ground operation support and a virtual-reality synchronization mechanism from three dimensions: physical layer, digital layer, and information fusion layer, providing better support for the operation and management of the physical world.

[0021] The physical layer integrates various data and information of the physical system corresponding to the airport's full-area intelligent perception, mainly including ground support service processes and status information of aircraft, passengers, baggage and cargo, flight ground operation support facility equipment resource information (including operation status information), flight information and geographic space information, etc. It collects and transmits physical world data through multi-source information fusion scene monitoring, multi-modal intelligent perception, radio frequency identification technology (RFID), 3D visual recognition and positioning and other technical means to realize the encapsulation of multi-type and multi-scale physical data.

[0022] The digital layer is a digital mapping of the physical layer, mainly including mathematical models and multi-source models of production factors, state transition models, and three-dimensional data models of virtual 3D space, etc. By receiving unified and standardized information data from the information fusion layer, the physical attributes, behaviors, rules and models are associated and integrated, and the status of each production factor in reality is mapped, thereby achieving global perception, situation analysis and collaborative decision-making of the airport operation system. For example, the three-dimensional model of the terminal is both a visual presentation of the building structure, space environment, transportation and service resources, and a monitoring method for resource occupancy and operation status.

[0023] The information fusion layer is the driver for data interaction between the physical layer and the digital layer. Through communication network protocols such as OPC UA and TCP / IP, the multi-source heterogeneous data collected from the underlying devices of the physical layer are standardized and classified. Under the premise of ensuring data consistency and reliability, the real-time production data such as the movement of objects and resource status in the real world are transmitted to the virtual world to drive the movement or status update of the virtual model. The content in the virtual world such as the digital layer prediction data or control signals can also be fed back to the real world to achieve intervention in the flight support process and adaptive dynamic scheduling of resources. Digital twins are digital versions of "clones" of the real world, and their key feature is the dynamic simulation of physical objects. The information fusion layer continuously interacts with the physical layer and the digital layer, and establishes a digital twin synchronization mechanism with data collection and drive as the core from four aspects: geometric physics, attribute information, behavioral logic, and constraint rules.

[0024] 1.3 Digital Twin Data Transmission and Drive

[0025] Data is the core element of the digital twin system. It originates from physical entities, virtual models and service systems, and is re-integrated into various parts after fusion processing to promote the operation of the entire system. After the virtual model is established, it needs to be given "power" to achieve virtual-real synchronization of digital twins, and its core technology is data transmission and drive. The system uses communication network protocols such as OPC UA and TCP / IP to realize the collection, transmission and preprocessing of multi-source heterogeneous data. The real-time data after cleaning and modeling is stored in the MySQL database (the demonstration system stores it in the form of global variables and synchronizes airport data through TCP transmission control protocol), and is collected regularly to update the data queue. The client connects to the database through the data API interface, reads and writes data and subscribes to it by browsing the address space, synchronizes the ground support service process and status information related to aircraft, passengers and baggage, facility equipment resource information, flight information and geospatial information to the twin system, drives the movement or status update of the virtual model, and feeds back the predicted data and control signals to the physical system, ultimately realizing the intervention in the flight support process and adaptive dynamic scheduling of resources.

[0026] 1.4 Establishment of digital twin information model

[0027] Information collection is the basis of digital twins. After fusion processing, information data is passed to the digital layer to create an accurate model of the system or drive the movement and status update of the virtual model.

[0028] 1.4.1 Classification and processing of multi-source information

[0029] The multi-source heterogeneous information that needs to be collected includes digital twin identification information, spatial location information, process status information, resource status information, and flight dynamic information.

[0030] The global identification of digital twins can give physical objects a unique digital identity code, enabling rapid indexing, positioning, and loading of related information of objects in the asset database. The global identification objects involved in the digital twin system for flight ground operations support include site facilities, equipment, aircraft, passengers, and luggage. Its information sources include OID identification system, GSI identification system, flight information system, barcode, QR code, RFID tag, face recognition technology, multimodal feature extraction and recognition technology, video motion target tracking technology, etc.

[0031] Spatial location information includes geographic information data, indoor map data and activity target tracking and positioning data. Geographic information data maps the physical space structure and relationship to the digital world, reflecting the temporal and spatial characteristics and dynamic changes of the real world, such as the terminal building configuration and the topological structure of the airport road network. Indoor map data focuses on describing the composition, distribution and mutual relationship of the internal space elements of the building, such as the lobby, check-in area, security inspection area, etc. Activity target tracking and positioning data provides the location and movement trajectory of passengers, luggage, aircraft, vehicles, etc., supporting the business system to locate, monitor, dispatch and supervise them safely.

[0032] The flight support service process refers to the order and method of providing various services to meet the needs of flight safety operation support. The operational services associated with aircraft, passengers, baggage and cargo are arranged and combined in a certain time and space order to form a network structure with specific service functions. The process status specifically refers to the completion progress or status of the node task, such as the target completion time, refueling amount and other information. The flight support service network has obvious time-varying characteristics. The network topology structure and node process status change with time and environment. It is necessary to track the entire chain of support processes, obtain real-time status information of the process, and predict the network evolution trend. The flight support process status information and collection method are shown in Table 1.

[0033] Table 1 Flight support process data collection

[0034]

[0035] Resource status information mainly includes facility and equipment type, quantity, capacity, usage priority, open / closed status, current service efficiency, VIP attributes, maintenance records, task plans, operators and other material information. The main channels for obtaining resource status information are as follows: 1) Intelligent sensing monitoring: equipment status monitoring is achieved through the integration of single-chip microcomputers and sensors; 2) Image recognition monitoring: through intelligent image processing and analysis to identify equipment usage status, automatically count resource status information such as queue length / number of people and equipment load in specific scenarios; 3) Other business systems, such as the airport collaborative decision-making system A-CDM, departure system DCS, security inspection system, aircraft stand allocation system, etc. The data sharing platform can obtain the occupancy and allocation plan of resources such as aircraft stands and boarding gates in real time to enhance resource coordination.

[0036] The twin system flight dynamics information mainly includes 45 ground support nodes involved in the flight from takeoff to landing, from completing the stopover to taking off again, and also includes supplementary information such as delays, cancellations, calculated takeoff time, and estimated off-block time. Most of the information can be obtained through the A-CDM system, and some information can be transmitted through external interfaces such as the AMAN and DMAN systems for flight arrival and departure sequencing and the VariFlight platform.

[0037] 1.4.2 Establishment of information model

[0038] Data modeling is a core technology in the field of digital twins, which helps to deeply understand and optimize physical systems and provide support for decision-making and planning. The information model is an abstract description of the content, format, interaction sequence and participating roles of information exchange.

[0039] 1. Behavior Data (BED)

[0040] The main body of digital twins is a data-driven model built for physical entities and behavioral logic. Therefore, behavioral logic data is essentially the specific form and carrier of process status information.

[0041] In Table 1, the behavior data model of any process Process is

[0042]

[0043] In the formula, Name Ps is the process name; Endurance0 Ps is the average total service time; Endurance1 Ps Because the service time is still needed; is the estimated start time; is the actual start time; is the estimated end time; is the actual end time; Capacity Ps For custom items, such as refueling amount, etc.; PreProcess Ps PostProcess Ps For the next work; Standard Process Ps Indicates whether it is a standard process; InterveneYN Ps Indicates whether the process can be intervened; MatchResource_ServTime Ps The service resources and corresponding service time matched to the process, if InterveneYN Ps =0, the item is "empty"; if InterveneYN Ps =1, the string is: resource 1 name-service time; resource 2 name-service time...

[0044] 2. Rule Data Model (RUD)

[0045] The essence of a process is rules and specifications. Business processes are the rules and specifications for conducting business. Once the flight safety operation guarantee process is solidified, its management and business processes are also solidified, and the corresponding rules and specifications are also determined. Subsequent management and business instances running on the process must be executed according to the established process. The standardization of management and business processes has significantly improved the efficiency of flight guarantee.

[0046] Taking passenger services in the terminal as an example, the passenger flow can be generally divided into the following categories: international departure, international arrival, domestic departure, domestic arrival, transfer and stopover, etc. Domestic passengers have simple procedures and occupy less time in the terminal, but the flow is large. International passengers need to go through passport and quarantine procedures, and they also have more luggage and stay in the terminal for a long time. Passengers on transfer flights are divided into international to international, international to domestic, domestic to domestic and domestic to international, and special flow routes need to be arranged. Stopover flights only make a short stop at the airport, and passengers are usually guided to the waiting area to rest and prepare to board the plane again, without luggage transfer problems.

[0047] Although there are individual differences in the passenger service process at the terminal, the departure service process is similar for passenger groups with the same travel nature, category, arrival mode, and selected service mode. In order to simplify the rule data model of the digital twin system, this paper designs 9 standard service processes (see Table 2) based on the passenger arrival mode Origin, luggage check-in status LuggageYN, and online check-in status OnlineCheckinYN, and defines the order and method of each service, and builds a standard process library.

[0048] Table 2 Standard service process for departing passengers

[0049]

[0050]

[0051] In Table 2, the rule data model of any procedure is

[0052] RUD={Name Pd , Process1, Process2, Process3,…, Process n} (2)

[0053] In the formula, Name Pd Process name; n Represents the nth service process in the sequence.

[0054] During the initialization phase of the digital twin system, the process rule data of objects (aircraft, passengers, luggage, and cargo) are automatically acquired by calling the terminal passenger service standard process library.

[0055] 3. Object Data Model (OBD)

[0056] The primary task of digital twin application is to create digital twin models of application objects, involving static and dynamic twin object data models of aircraft, passengers, baggage, cargo, and site facilities and equipment. Static data models include color, attributes, shape, size, etc. On this basis, dynamic data models introduce the time dimension, so that digital models can be automatically updated and adjusted over time, reflecting changes in the spatial position, process status, and behavior rules of application objects.

[0057] Taking passenger service at the terminal as an example, the object data model of the passenger's real-time status information is:

[0058]

[0059] In the formula, Name Pg Passenger name; TimeStamp Pg is the timestamp; Flight Pg Flight number; SchProcedure Pg CtProcess is the standard process to be executed (name); Pg For the current process (name); CtSrvResource Pg Service resources matched to the current process; CtProcessDuration PgStandard service time for the current process; CtProcessStartTime Pg The start time of the current process; CtProcessRemainingTime Pg The remaining time of the current process; CtProcessPFTime Pg Schedule completion time for the current process; NextProcess Pg For the next process (name); Process n Indicates the nth service process (name) being executed / to be executed.

[0060] Object data model for real-time status information of check-in counters CheckinCounters

[0061]

[0062] In the formula, Name CC Check-in counter number; StandarTime CC Standard service time; CtStandarTime CC The current standard service time (including the current waiting time in the queue); ManualCheckinYN CC Indicates whether it is a manual check-in counter; SpecialCounterYN CC Indicates whether it is a special counter (VIP counter); CounterOpenYN CC Indicates whether the counter is open for use.

[0063] 4. Strategy Data (STD)

[0064] Identifying the critical paths that affect the normal operation of flights and proposing early intervention strategies are important scenarios for the intelligent application of digital twin systems. The intervention strategy requires changing specific field data in the object data model. For example, the strategy model for passenger process intervention is

[0065] STD={OBD Pg ,InterventionProcess Pg} (5)

[0066] Where, OBD Pg Object data model for passenger status information; InterventionProcess Pg The process for which intervention is recommended, in the format of ProcessName-MatchResource_ServTime;…

[0067] 5. Entity Relationship (ER) Model

[0068] ER={BED,RUD,OBD,STD}.

[0069] 2. Construction of multi-layer complex network model for flight ground operation support

[0070] Flight support operations / services (Process) are interrelated and coordinated to form a complex spatiotemporal network relationship. The flight support process network (SPN network) is defined as G A , whose expression is

[0071]

[0072] Where V A = {v Aij |i∈N A} is the network G A The node set of node v Aij represents the jth job or service provided to object i, E A ={e Aij |e Aij =(v Ai ,v Aj ),i∈N A ,j∈N A ; i≠j} is the network G A Edge set, directed edge e Aij Represents the spatiotemporal relationship between two nodes, i.e. the subsequent job can only start after the previous job is completed.

[0073] There are also correlations between service resources. For example, there is substitutability and priority of use between similar resources, and the flight support service process determines the order of use of different types of resources. The resource system can also be described by a network and presents a typical community structure, in which there are more connections between similar nodes, while the connections between different types of nodes are relatively few or regarded as edge weights of ∞. Define the service resource network (SRN network) as G B , whose expression is

[0074] G B =(V B ,E B ) (7)

[0075] Where V B = {v Bi |i∈N B} is the network G B The node set of node v Bi Represents the i-th resource used in providing a certain type of service; EB ={e Bij |e Bij =(v Bi ,v Bj ),i∈N B ,j∈N B ; i≠j} is the network G B Connect edge set, edge e Bij Represents the existence of cooperation or substitution relationship between resources.

[0076] Flight safety operation guarantee requires finding a balance between resource allocation and process optimization. Resources should match service needs. A With G B The nodes with interdependent relationships are connected by inter-layer coupling with dependent edges to establish a flight support collaborative control dependency network G, which is expressed as

[0077] G=(G A ,G B ,E AB ) (8)

[0078] In the formula, E AB ={e AiBj |e AiBj =(v Ai ,v Bj ),Ai∈N A ,Bj∈N B} is the edge set of dependent edges, and the dependent edge e AiBj Represents the use of resources v Bj Provides a job or service v to object i Ai , G A The number of nodes is much larger than G B , the inter-layer coupling mode is one-to-one or many-to-one connection, node v Bj The out-degree (number of coupling edges) indicates the use of resource B j The number of objects for which the service is provided.

[0079] 3 Intelligent collaborative intervention for flight operation support based on DT and CN

[0080] Taking the regularity of departing flights at a hub airport in the southwest region as an example, we conducted a test on the fusion application of digital twin technology and complex networks, developed a three-dimensional visualization decision-making system for flight operation guarantee, and verified its application in delay prediction, key process identification, and cross-subject multi-resource collaborative intervention capabilities. The system adopts a C / S architecture and realizes two-way communication through TCP. Maya is responsible for the creation of the 3D model of the airport terminal, UE5 is the server, responsible for the creation and rendering of the three-dimensional scene; Python is the client, responsible for writing algorithm programs, and providing decision-making and control support for flight safety operation guarantee through complex network analysis and calculation.

[0081] Based on the principle and method of identifying key processes in flight ground operation support, under the dependent network architecture of flight support collaborative control, the graph theory problem of mathematical modeling can be solved by using Python. The NetworkX function can be used to represent and operate the weighted directed graph (G). The two main functions and functions used to calculate the critical path and path length are shown in Table 3.

[0082] Table 3 Functions for calculating critical path and path length

[0083]

[0084] G (NetworkX graph) is a directed acyclic graph, weight (str, optional) is the edge weight.

[0085] dag_longest_path() returns a list of vertices in the longest path of G, that is, a list of nodes in the critical path.

[0086] dag_longest_path_length() returns the weighted length of the edges that make up the longest path in G, that is, the length of the critical path.

[0087] Taking the departing passenger service guarantee in the terminal of a hub airport in the southwest region as an example, 10 flights, passengers and resource data of the airport on a certain day were randomly selected as experimental samples, and the UE system read and wrote data for monitoring, analysis and simulation. In the passenger guarantee service process in the terminal, there are three key processes for intervention: check-in, security check, and from security check to the waiting area. VIP channels are set up in the check-in area and the security check area, and electric vehicles are arranged behind the security check area. Under the dependent network architecture of flight guarantee collaborative control, Python's NetworkX library is used to predict delays and identify key tasks through the critical path method, and the relevant prediction, process intervention and resource scheduling strategy information are fed back to the UE system in real time. The UE system implements collaborative intervention, and the intervention method is to change the service resources matched by a process in the subsequent process. The process and results are presented through the digital twin visualization platform.

[0088] The steps for discovering key tasks and implementing intelligent collaborative intervention for flight operation support are as follows:

[0089] Step 1: Use Python to quickly generate the dependent network topology structure of flight support collaborative control and initialize the status data of network nodes based on the twin information model;

[0090] Step 2 uses the TCP data interface to monitor and receive the dynamic data of the twin object sent by UE5, and updates the flight support collaborative control dependent network data in real time;

[0091] Step 3: Substitute the G and weight corresponding to flight i into the function dag_longest_path_length() to return the critical path length (duration). Based on the current time, if the critical path length is less than or equal to the scheduled departure time of flight i, return to Step 2. Otherwise, proceed to the next step;

[0092] Step 4 Substitute the G and weight corresponding to flight i into the function dag_longest_path() to return the list of processes on the critical path. Ps and MatchResource_ServTimes Ps Comprehensively judge the key processes that can be intervened;

[0093] Step 5: Pass the policy information such as passengers and processes that need intervention to the UE system;

[0094] Step 6 When UE5 receives the intervention instruction, it will guide the passenger to the corresponding VIP service: VIP check-in counter, VIP security check channel, or take a battery car to the waiting area according to the number of interventions InterveneCount and the number of key processes that can be intervened in the passenger's subsequent processes, combined with the passenger's current security service stage;

[0095] After completing the process intervention in Step 7, UE5 immediately sends the real-time status information of passengers and resource status information to Python;

[0096] Step 8 returns to Step 2, and Python determines whether the next process intervention is required.

[0097] The specific implementation process is as follows Figure 1 As shown, EDT is the estimated departure time (estimated_boarding_time), and SDT is the planned departure time (plan_boarding_time). The intelligent collaborative intervention method for flight operation guarantee based on DT and CN proposed in the present invention has wide applicability and can cover all research objects involved in the entire flight operation guarantee process, including passengers, baggage, aircraft, etc. Passenger guidance service is an important part of the airport flight operation guarantee process. Therefore, the present invention selects the service guarantee for departing passengers in the terminal as a specific example for intelligent intervention experiments. Through the simulation intervention experiment, a total of 10 passengers were intervened. By comparing the estimated boarding time before and after the intervention, it was found that the overall time was advanced by 1′7″~5′44″, and the average was 3′17″. All passengers boarded the plane smoothly, and the results showed that the intervention was effective.

Claims

1. An intelligent collaborative intervention method for flight operation support based on DT and complex networks, characterized by: The steps include: Step 1: Establish a digital twin model of flight ground operation support based on a data-driven approach to achieve virtual-real synchronization; Step 2: Establish a complex network model for flight ground operation support; Step 3: Intelligent collaborative intervention for flight operation support based on digital twin models and complex network models.

2. The flight operation support intelligent collaborative intervention method based on DT and complex network according to claim 1 is characterized by: In step 1, a digital twin model of flight ground operation support is constructed from three dimensions: physical layer, digital layer, and information fusion layer, where: (1) The physical layer integrates various data and information of the physical system corresponding to the airport's global intelligent perception, mainly including ground support service processes and status information of aircraft, passengers, baggage, and cargo, flight ground operation support facility equipment resource information and operation status information, flight information, and geographic space information; (2) The digital layer is a digital mapping of the physical layer, which mainly includes mathematical models and multi-source models of production factors, state transition models, and three-dimensional data models of virtual 3D space. By receiving information data in a unified and standardized manner from the information fusion layer, the physical attributes, behaviors, rules, and models are associated and integrated to map the status of each production factor in reality. (3) The information fusion layer is the driver that realizes the data interaction between the physical layer and the digital layer. Through the communication network protocol, the multi-source heterogeneous data collected from the underlying devices of the physical layer are standardized and classified. Under the premise of ensuring data consistency and reliability, the real-time production data in the real world is transmitted to the virtual world to drive the movement or status update of the virtual model. At the same time, the digital layer prediction data or control signal is fed back to the real world to realize the continuous interaction between the physical layer and the digital layer. From the four aspects of geometric physics, attribute information, behavioral logic and constraint rules, a digital twin synchronization mechanism with data collection and drive as the core is established.

3. The flight operation support intelligent collaborative intervention method based on DT and complex network according to claim 2 is characterized by: The multi-source heterogeneous data includes global identification information, spatial location information, process status information, resource status information and flight dynamic information of the digital twin, where: 1) The global identification of digital twins gives physical objects a unique digital identity code, enabling rapid indexing, positioning, and loading of related information of objects in the asset database; global identification objects include site facilities, equipment, aircraft, passengers, and luggage; 2) Spatial location information includes geographic information data, indoor map data and activity target tracking and positioning data, among which: geographic information data maps the physical space structure and relationship to the digital world, reflecting the temporal and spatial characteristics and dynamic changes of the real world; indoor map data focuses on describing the composition, distribution and mutual relationship of the internal space elements of the building; activity target tracking and positioning data provides the location and movement trajectory of passengers, luggage, aircraft and vehicles, supporting the business system to locate, monitor, dispatch and supervise them safely; 3) Process status information refers to the completion progress or status of each node task of the operational service associated with aircraft, passengers, baggage and cargo in the flight support service process; 4) Resource status information mainly includes facility and equipment type, quantity, capacity, usage priority, open and closed status, current service efficiency, VIP attributes, maintenance records, task plans, operators and other material information; 5) Flight dynamic information mainly includes the ground support nodes involved in the flight from take-off to landing, from completing the stopover to taking off again, and also includes supplementary information such as delays, cancellations, calculated take-off time, and estimated off-block time.

4. The flight operation support intelligent collaborative intervention method based on DT and complex network according to claim 2 is characterized by: The method for establishing the complex network model of flight ground operation support described in step 2 is: (1) Build a flight support process network G A G A =(V A ,E A ) Where V A = {v Aij |i∈N A } is the network G A The node set of node v Aij represents the jth job or service provided to object i, E A ={e Aij |e Aij =(v Ai ,v Aj ),i∈N A ,j∈N A ; i≠j} is the network G A The edge set of the directed edge e Aij Represents the spatiotemporal relationship between two nodes; (2) Building a service resource network G B G B =(V B ,E B ) Where V B = {v Bi |i∈N B } is the network G B The node set of node v Bi represents the i-th resource used in providing a certain type of service; E B ={e Bij |e Bij =(v Bi ,v Bj ),i∈N B ,j∈N B ; i≠j} is the network G B The edge set of Bij There is a cooperative or substitution relationship between representative resources; (3) Constructing the flight support collaborative control dependent network G G=(G A ,G B ,E AB ) In the formula, E AB ={e AiBj |e AiBj =(v Ai ,v Bj ),Ai∈N A ,Bj∈N B } is the edge set of the dependent edges of network G, and the dependent edge e AiBj Represents the use of resources v Bj Provides a job or service v to object i Ai , G A The number of nodes is much larger than G B , the inter-layer coupling mode is one-to-one or many-to-one connection, node v Bj The number of coupling edges indicates the use of resource B j The number of objects for which the service is provided.

5. The flight operation support intelligent collaborative intervention method based on DT and complex network according to claim 4 is characterized by: The intelligent collaborative intervention method for flight operation support based on the digital twin model and the complex network model described in step 3 includes the following steps: The first step is to build a flight support collaborative control interdependent network G and initialize the status data of network nodes based on the digital twin information model; Step 2: Through the TCP data interface, monitor and receive the dynamic data of the twin object sent by the UE, and update the flight support collaborative control dependent network data in real time; Step 3: Calculate the critical path duration and determine whether the critical path duration is less than or equal to the scheduled departure time of the flight: if so, return to step 2; if not, proceed to step 4; Step 4: Calculate the critical path and its process list, then determine the critical processes that can be intervened based on the behavior data model attributes, and pass the passengers and process strategy information that need to be intervened to the UE; Step 5: When the UE receives the intervention instruction, it will guide the passenger to the corresponding VIP service based on the number of interventions and the number of key processes that can be intervened in the passenger's subsequent processes, combined with the passenger's current security service stage; Step 6. Return to step 2 after completing a process intervention.

6. The flight operation support intelligent collaborative intervention method based on DT and complex network according to claim 5 is characterized by: Functions for calculating critical paths and critical path durations include: 1) The function dag_longest_path(G) is used to return a list of key path nodes whose value is G; 2) A function dag_longest_path_length(G) is used to return the value of the critical path duration G.

7. The intelligent collaborative intervention method for flight operation support based on DT and complex network according to claim 5 is characterized by: The digital twin information model includes: a behavior data model, a rule data model, an object data model and a strategy model.

8. The flight operation support intelligent collaborative intervention method based on DT and complex network according to claim 7 is characterized by: The behavioral data model of any process is In the formula, Name Ps is the process name; Endurance0 Ps is the average total service time; Endurance1 Ps Because service time is still needed; is the estimated start time; is the actual start time; is the estimated end time; is the actual end time; Capacity Ps For custom items; PreProcess Ps PostProcess Ps For the next work; Standard Process Ps Indicates whether it is a standard process; InterveneYN Ps Indicates whether the process can be intervened; MatchResource_ServTime Ps The service resources and corresponding service time matched to the process.

9. The flight operation support intelligent collaborative intervention method based on DT and complex network according to claim 7 is characterized by: The rule data model for any process is RUD={Name Pd ,Process1,Process2,Process3,…,Process n } In the formula, Name Pd Is the process name; Process n Represents the nth service process in the sequence.

10. The flight operation support intelligent collaborative intervention method based on DT and complex network according to claim 7, characterized in that: The strategic model for passenger process intervention is: STD={OBD Pg ,InterventionProcess Pg } In the formula, InterventionProcess Pg Object data model OBD of passenger status information for the process of suggesting intervention Pg for: In the formula, Name Pg Passenger name; TimeStamp Pg is the timestamp; Flight Pg Flight number; SchProcedure Pg CtProcess is the standard process to be executed; Pg For the current process; CtSrvResource Pg Service resources matched to the current process; CtProcessDuration Pg Standard service time for the current process; CtProcessStartTime Pg The start time of the current process; CtProcessRemainingTime Pg The remaining time of the current process; CtProcessPFTime Pg Schedule completion time for the current process; NextProcess Pg For the next process; Process n Indicates the nth service process in sequence that is being executed or is to be executed.

Images