A flight plan robustness evaluation system and method

Abstract

A flight plan robustness evaluation system and method. The system comprises a network structure server, a simulation driving server, a first client, a second client, a third client and a fourth client connected with each other through a network. The flight plan robustness evaluation system and method provided by the application apply the dynamic traffic distribution concept to the air traffic network operation, realize the dynamic evaluation of the flight plan robustness, and help the civil aviation operation monitoring and management personnel to fine-tune the flight plan in advance when the robustness evaluation is low, so as to improve the robustness of the flight plan in the implementation process and avoid the occurrence of large-area flight delay.

Description

Technical Field

[0001] This invention belongs to the field of air traffic management technology, and specifically relates to a flight plan robustness evaluation system and method. The evaluation results can be used by air traffic management departments to optimize flight plans and formulate congestion mitigation strategies, as well as by airlines to prepare flight plans. Background Technology

[0002] The robustness of flight schedules significantly impacts the safety and efficiency of air traffic operations. Currently, flight schedules are primarily developed by individual airlines and submitted to the Civil Aviation Administration of China (CAAC) for approval. However, the approval process mainly focuses on maximizing airport capacity utilization, neglecting to consider the robustness of nationwide flight schedules from a network perspective. This results in poor resilience to external uncertainties. When unforeseen factors such as weather, military activities, or traffic congestion cause delays to some scheduled flights, a butterfly effect can easily occur, leading to widespread flight delays and severely threatening the safety and efficiency of air traffic operations. Therefore, effectively assessing the robustness of flight schedules to guide their scientific development is crucial for improving the effectiveness of air traffic operations in my country.

[0003] With the steady development of the air transport industry, my country's civil aviation airports completed 11.6505 million aircraft takeoffs and landings in 2019, a 5.2% increase over the previous year and approximately 2.4 times the 4.841 million takeoffs and landings in 2009. The International Air Transport Association (IATA), in its "Air Passenger Traffic Forecast for the Next 20 Years" report, points out that China will replace the United States as the largest aviation market around 2029. In this environment, the effective execution of flight schedules is more susceptible to external factors, potentially leading to widespread flight delays. Improving the robustness of flight schedules presents both an opportunity and a challenge for the development of civil aviation industries and airlines worldwide.

[0004] Current research on the robustness of flight schedules mainly involves constructing a transportation network with airports as nodes and connections between airports as edges (if there are flights between two airports, there is a connection between them). Complex network theory is then applied to calculate the topological characteristics of the transportation network to assess the robustness of the flight schedule. However, this approach fails to consider the functional characteristics of the transportation network, resulting in poor evaluation results and a lack of practicality. Therefore, it is crucial to consider both the structure and function of the air traffic network when evaluating the robustness of flight schedules. This will improve the effectiveness and scientific rigor of the evaluation, guide the scientific planning of flight schedules, and significantly enhance the safety and efficiency of air traffic operations.

[0005] Secondly, traditional network robustness studies mostly involve directly removing the attacked node or edge from the network. However, in reality, the attacked node or edge may still retain some transportation functions. Therefore, directly removing the node or edge does not reflect the actual operating conditions.

[0006] In addition, existing studies mostly focus on the robustness of flight schedules for individual airlines, and no research has yet been found on the robustness of national flight schedules.

[0007] Given the current state of flight schedule robustness assessment, there is a lack of an effective and practical flight schedule robustness assessment system and method. Summary of the Invention

[0008] To address the aforementioned problems, the present invention aims to provide a flight schedule robustness assessment system and method.

[0009] To achieve the above objectives, the flight plan robustness evaluation system provided by the present invention includes a network structure server, a simulation driver server, a first client, a second client, a third client, and a fourth client that are interconnected via a network.

[0010] The network structure server is equipped with an airspace navigation database, which is used to provide clients with navigation data services including airports, navigation beacons, waypoints, air routes, restricted areas, danger zones, restricted areas, obstacles, and sectors.

[0011] The simulation driver server is equipped with a simulation driver database, which is used to provide flight dynamics model and flight motion model data services to the client.

[0012] The first client has a flight planning subsystem installed, which is used to generate flight plans that conform to a certain distribution pattern;

[0013] The second client has an air traffic network modeling subsystem installed, which is used to build an air traffic network structure model;

[0014] The third client is equipped with a flight display and interaction subsystem, which is used to display and interact with the air traffic network environment generated by the air traffic network modeling subsystem and the evaluation results generated by the flight plan robustness evaluation subsystem.

[0015] The fourth client is equipped with a flight schedule robustness evaluation subsystem, which includes a flight reassignment module, a network structure robustness evaluation module, a traffic function robustness evaluation module, and a comprehensive flight schedule robustness evaluation module.

[0016] The evaluation method using the flight schedule robustness evaluation system provided by the present invention includes the following steps performed in sequence:

[0017] Step 1) Select airspace navigation data, including airports, navigation stations, waypoints, air routes, restricted areas, danger zones, restricted areas, obstacles, and sectors, from the airspace navigation database of the network structure server; pre-select the airports to be attacked and random attack events, including severe weather and aircraft malfunctions, involved in random attacks.

[0018] Step 2) Based on the actual flight operation distribution pattern of the evaluated airspace, compile an original flight plan that conforms to the actual distribution pattern in the flight planning subsystem of the first client.

[0019] Step 3) Input the airspace navigation data selected in Step 1) and the original flight plan compiled in Step 2) into the air traffic network modeling subsystem in the second client. Based on the original flight plan, construct the air traffic network structure model with airports as nodes and lines between airports as edges.

[0020] Step 4) Input the airspace navigation data selected in Step 1), the original flight plan compiled in Step 2), and the air traffic network structure model established in Step 3) into the flight display and interaction subsystem in the third client. Call the flight motion model and flight dynamic model in the simulation driving database in the simulation driving server 2, set the simulation time, perform the original flight plan operation simulation before the attack, obtain the flight plan simulation results before the attack, and display them on the flight display and interaction subsystem.

[0021] Step 5) Based on the flight plan simulation results obtained in Step 4) above, record the simulation data before the attack, including the actual departure time and actual arrival time of all flights, and input this simulation data before the attack into the flight plan robustness evaluation subsystem in the fourth client.

[0022] Step 6) The flight reassignment module of the flight plan robustness evaluation subsystem in the fourth client formulates the flight takeoff reassignment strategy for the attacked airport. Based on the attacked airport pre-selected in Step 1) and the random attack event, the airport takeoff capacity of the attacked airport is adjusted. Based on the flight reassignment strategy, the overcapacity takeoff flights of the attacked airport are reassigned from the perspective of actual operation to generate a flight reassignment plan, thereby adjusting the original flight plan.

[0023] Step 7) Input the airspace navigation data selected in Step 1) and the flight reassignment plan generated in Step 6) into the flight display and interaction subsystem in the third client, call the flight motion model and flight dynamic model in the simulation driver database in the simulation driver server 2, set the simulation time, simulate the flight reassignment plan after the attack, obtain the simulation results of the flight reassignment plan after the attack, and display them on the flight display and interaction subsystem.

[0024] Step 8) Based on the simulation results of the flight reassignment plan after the attack obtained in Step 7), record the simulation data after the attack, including the actual departure time and actual arrival time of all flights, and input these simulation data after the attack into the flight plan robustness evaluation subsystem in the fourth client.

[0025] Step 9) In the network structure robustness evaluation module of the flight plan robustness evaluation subsystem of the fourth client, the robustness of the flight plan before and after the attack is calculated using the simulation data before the attack obtained in Step 5) and the simulation data after the attack obtained in Step 8).

[0026] Step 10) In the traffic function robustness evaluation module of the flight plan robustness evaluation subsystem of the fourth client, based on the flight reallocation plan obtained in Step 6), calculate the traffic function robustness indicators including flight delay rate, airport cluster saturation change rate and flight cancellation rate.

[0027] Step 11) In the flight plan robustness comprehensive evaluation module, the global weighted network efficiency change rate η obtained in step 10) above is used. E,t Flight delay rate η D,t Airport group saturation change rate η M,t and flight cancellation rate η C,t Normalization is performed to obtain the normalized global weighted network efficiency change rate η' E,t， Normalized flight delay rate η' D,t Normalized airport cluster saturation change rate η' M,t and normalized flight cancellation rate η' C,t Four indicators were used, and then these four normalized indicators were weighted using an expert scoring method to obtain a comprehensive evaluation value η. t It is used to comprehensively measure the robustness of flight plans;

[0028] Step 12) The flight display and interaction subsystem in the third client will use the global weighted network efficiency change rate η obtained in step 10) above. E,t Flight delay rate η D,t Airport group saturation change rate η M,t and flight cancellation rate η C,t and the comprehensive evaluation value η obtained in step 11). t The display shows the change in the robustness of the air traffic network over time.

[0029] In step 6), the flight takeoff reassignment strategy involves the assumptions, forms, and methods of flight takeoff reassignment.

[0030] 1) The assumptions for the flight takeoff reallocation are as follows:

[0031] ① Severe incidents only attack nodes in the air traffic network, without considering the attack situation on the edges of the air traffic network;

[0032] ②The road and rail transportation between adjacent airports is well-developed and convenient;

[0033] ③The airlines have sufficient aircraft and crew capacity at various airports;

[0034] ④ Assuming sufficient route capacity, only the limitations of airport takeoff capacity on flight operations are considered;

[0035] ⑤ Without considering the mutual influence between departing and arriving flights, the airport has sufficient landing capacity;

[0036] ⑥ Assume that after the airport is attacked, its functions will not be completely lost and a certain amount of capacity will still remain.

[0037] ⑦ Flights can only be delayed by a maximum of one hour.

[0038] 2) There are three main forms of flight takeoff redistribution: flight delay, flight transfer, and flight cancellation.

[0039] ① Flight Delay: If the airport's takeoff capacity decreases after an attack, and the number of planned takeoff flights in the subsequent time period is less than the airport's takeoff capacity, then, without affecting the normal takeoff of planned flights in the next time period, the excess takeoff flights during the attack period will be delayed until the next time period.

[0040] ② Flight diversion and takeoff: If the attacked airport belongs to a certain airport group, and the passengers and cargo of the affected flights at that airport can be transferred in advance by road or rail to other transferable airports in the airport group that are not yet at full capacity, then the transferable airports will perform the transportation tasks for the affected flights at the attacked airport during the original planned time period.

[0041] ③ Flight cancellation: If, after an airport is attacked, the number of planned flights in the subsequent time period has reached the airport's takeoff capacity, and there are no other airports nearby that can be transferred to receive the affected flights, or the takeoff capacity of the airports that can be transferred has reached saturation, then the flight will be cancelled.

[0042] 3) Flight takeoff reallocation method: When the flight demand of the original flight plan exceeds the takeoff capacity of the attacked airport after the landing, the excess takeoff flights are first considered to be delayed to the next time period; until the takeoff capacity of the attacked airport in the next time period is reached, if there are still some flights that have not been allocated in the previous time period, these flights are arranged to take off from other transferable airports in the airport group to which the attacked airport belongs; until the takeoff capacity of other transferable airports is reached, if there are still some flights that have not been allocated in the previous time period at the attacked airport, these flights are canceled.

[0043] In step 9), the specific method for calculating the robustness of flight plans before and after the attack using the pre-attack simulation data obtained in step 5) and the post-attack simulation data obtained in step 8) is as follows:

[0044] 9.1) Calculate the shortest path d between any pair of airports (i,j) in the air traffic network based on the Floyd algorithm. ij ;

[0045] 9.2) Based on the flight schedule, calculate the number of flights q on the shortest path between any airport pair (i,j). ij ;

[0046] 9.3) Based on the above shortest path d between airport pairs (i,j), ij and flight volume q ij Calculate the weighted network efficiency of the airport pair:

[0047]

[0048] When airport pairs (i,j) are not connected, the shortest path d ij When the network approaches infinity, the weighted network efficiency ε is used. ij =0 is used to characterize the connectivity of airport pair (i,j);

[0049] 9.4) Calculate the average of the weighted network efficiencies between all the above airport pairs to characterize the global weighted network efficiency of the entire air traffic network:

[0050]

[0051] Where n is the total number of airports in the air traffic network, and the number of airport pairs is n(n-1);

[0052] 9.5) Considering the dynamic nature of flight schedules, the rate of change of the globally weighted network efficiency at different time periods is further calculated as an indicator of the robustness of the topology:

[0053]

[0054] Where E0 is the globally weighted network efficiency of the flight plan before the attack, E a The globally weighted network efficiency of flight schedules after an attack.

[0055] In step 10), the specific method for calculating the robustness indicators of transportation function, including flight delay rate, airport cluster saturation change rate, and flight cancellation rate, based on the flight reallocation plan obtained in step 6) is as follows:

[0056] 10.1) Calculate the flight delay rate after a flight schedule attack;

[0057] Based on the simulation results of the third client's flight reassignment plan after the attack, obtained in step 7), the flight delay rate is calculated:

[0058]

[0059] Among them, D i,t Let x be the number of flight delays at airport i during time period t. i,t η represents the number of planned departure flights at airport i during time period t. D,t Let be the flight delay rate for time period t, and n be the total number of airports;

[0060] 10.2) Rate of change of computer field cluster saturation;

[0061] Based on the simulation results of the third client's flight reallocation plan after the attack obtained in step 7), the airport group saturation change rate is calculated:

[0062]

[0063] Among them, b j Let NA be the takeoff capacity of airport j. In equation (5), the denominator is the sum of the takeoff capacities of all airport groups in the air traffic network, and the numerator is the number of flights taking off from other airports. i Let be the number of airports in the airport group to which airport i belongs. If airport i does not belong to any airport group, then the number of airports is NA. i =0;M i,t η is the number of flights that depart from airport i within the airport cluster and are transferred to other available airports during time period t; M,t Let be the rate of change of airport cluster saturation during time period t;

[0064] 10.3) Calculate the flight cancellation rate;

[0065] Based on the simulation results of the flight reallocation plan after the attack from a third client, the flight cancellation rate was calculated:

[0066]

[0067] Among them, C i,tThe number of flight cancellations at airport i during time period t; η C,t Let t be the flight cancellation rate for time period t.

[0068] In step 11), the comprehensive evaluation value η t The calculation formula is:

[0069] η t =1-aη′ D,t -bη′ M,t -cη′ C,t -dη′ E,t (7)

[0070] Where a, b, c, and d are the weights of the four normalized indicators, respectively.

[0071] Advantages of this invention:

[0072] 1. Current research on flight schedule robustness evaluation mainly analyzes the impact of attacks on the static structure of air traffic networks, neglecting the dynamic characteristics of the traffic system, such as dynamic traffic flow, i.e., the redistribution of flights after an attack. In actual operation, flights at failed nodes in the air traffic network may be transferred to other nodes and will not completely disappear. Therefore, this application considers the redistribution of departing flights caused by attacks and proposes three forms of flight redistribution for evaluation, thus making the robustness evaluation more comprehensive.

[0073] 2. Traditional flight scheduling robustness studies often directly remove attacked nodes or edges from the network to analyze its efficiency. However, in flight scheduling networks, attacked nodes or edges often do not completely fail, making direct removal unreasonable. For example, in an airport network, severe weather at an airport does not necessarily mean the airport is completely closed; only its takeoff and landing capacity decreases. Therefore, this application changes the impact of an attack on nodes to an adjustment of airport takeoff capacity, making the attack pattern more reasonable and the robustness evaluation more realistic.

[0074] 3. Existing studies on flight schedule robustness primarily focus on the robustness of real-time traffic networks or airline flight schedules, with virtually no research on the robustness of national or regional overall flight schedules. Air traffic arises from the simultaneous operation of flights by multiple airlines, and flights from different companies influence each other. Furthermore, low robustness of the traffic network during the planning phase significantly impacts its robustness during the operational phase. Therefore, this application takes a national flight schedule preparation perspective, enabling a holistic assessment of flight schedule robustness and providing new insights into flight schedule robustness research.

[0075] The system and method of this invention apply the concept of dynamic traffic allocation to the operation of air traffic networks, realizing the dynamic evaluation of flight plan robustness. When the robustness evaluation is low, it helps to guide civil aviation operation monitoring and management personnel to make fine adjustments to the flight plan in advance, so as to improve the robustness of the flight plan during implementation and avoid large-scale flight delays. Attached Figure Description

[0076] Figure 1 This is a block diagram of the flight schedule robustness evaluation system provided by the present invention;

[0077] Figure 2 This is a schematic diagram of an airport cluster considered in an embodiment of the present invention.

[0078] Figure 3 This is a flowchart of the flight schedule robustness evaluation method provided by the present invention.

[0079] Figure 4 This is a robustness variation diagram of flight schedules on a certain day in an embodiment of the present invention. Detailed Implementation

[0080] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific examples described herein are merely for explaining the invention and are not intended to limit the invention.

[0081] like Figure 1 As shown, the flight plan robustness evaluation system provided by the present invention includes a network structure server 1, a simulation driving server 2, a first client 3, a second client 4, a third client 5, and a fourth client 6 that are interconnected by a network.

[0082] The network structure server 1 is equipped with an airspace navigation database, which is used to provide navigation data services to clients, including airports, navigation stations, waypoints, air routes, restricted areas, danger zones, restricted areas, obstacles, and sectors.

[0083] The simulation driver server 2 is equipped with a simulation driver database, which is used to provide flight dynamics model and flight motion model data services to the client.

[0084] The first client 3 is equipped with a flight planning subsystem, which is used to generate flight plans that conform to a certain distribution pattern;

[0085] The second client 4 is equipped with an air traffic network modeling subsystem, which is used to establish an air traffic network structure model;

[0086] The third client 5 is equipped with a flight display and interaction subsystem, which is used to display and interact with the air traffic network environment generated by the air traffic network modeling subsystem and the evaluation results generated by the flight plan robustness evaluation subsystem.

[0087] The fourth client 6 is equipped with a flight schedule robustness evaluation subsystem, which includes a flight reassignment module, a network structure robustness evaluation module, a traffic function robustness evaluation module, and a flight schedule robustness comprehensive evaluation module.

[0088] like Figure 3 As shown, the evaluation method using the flight schedule robustness evaluation system provided by the present invention includes the following steps performed in sequence:

[0089] Step 1) Select airspace navigation data, including airports, navigation stations, waypoints, air routes, restricted areas, danger zones, restricted areas, obstacles, and sectors, from the airspace navigation database of the network structure server 1; pre-select the airports to be attacked and random attack events, including severe weather and aircraft malfunctions, involved in random attacks.

[0090] Step 2) Based on the actual flight operation distribution pattern of the evaluated airspace, compile an original flight plan that conforms to the actual distribution pattern in the flight planning subsystem of the first client 3;

[0091] Step 3) Input the airspace navigation data selected in Step 1) and the original flight plan compiled in Step 2) into the air traffic network modeling subsystem in the second client 4. Based on the original flight plan, construct the air traffic network structure model with airports as nodes and lines between airports as edges (if there are planned flights between two airports, there is a connection between the two airports; otherwise, there is no connection).

[0092] Step 4) Input the airspace navigation data selected in Step 1), the original flight plan compiled in Step 2), and the air traffic network structure model established in Step 3) into the flight display and interaction subsystem in the third client 5, call the flight motion model and flight dynamic model in the simulation driving database in the simulation driving server 2, set the simulation time, perform the original flight plan operation simulation before the attack, obtain the flight plan simulation results before the attack, and display them on the flight display and interaction subsystem;

[0093] Step 5) Based on the flight plan simulation results obtained in Step 4) above, record the simulation data before the attack, including the actual departure time and actual arrival time of all flights, and input this simulation data before the attack into the flight plan robustness evaluation subsystem in the fourth client 6.

[0094] Step 6) The flight reassignment module of the flight plan robustness evaluation subsystem in the fourth client 6 formulates the flight takeoff reassignment strategy for the attacked airport. Based on the attacked airport and random attack events pre-selected in Step 1), the airport takeoff capacity of the attacked airport is adjusted. Based on the flight reassignment strategy, the overcapacity takeoff flights of the attacked airport are reassigned from the perspective of actual operation to generate a flight reassignment plan, thereby adjusting the original flight plan.

[0095] The flight takeoff reassignment strategy involves the assumptions, forms, and methods of flight takeoff reassignment.

[0096] 1) The assumptions for the flight takeoff reallocation are as follows:

[0097] ① Severe incidents only attack nodes (airports) in the air traffic network, without considering the attack situation of the edges (air routes) of the air traffic network;

[0098] ②The road and rail transportation between adjacent airports is well-developed and convenient;

[0099] ③The airlines have sufficient aircraft and crew capacity at various airports;

[0100] ④ Assuming sufficient route capacity, only the limitations of airport takeoff capacity on flight operations are considered;

[0101] ⑤ Without considering the mutual influence between departing and arriving flights, the airport has sufficient landing capacity;

[0102] ⑥ Assume that after the airport is attacked, its functions will not be completely lost and a certain amount of capacity will still remain.

[0103] ⑦ Flights can only be delayed by a maximum of one hour.

[0104] 2) There are three main forms of flight takeoff redistribution: flight delay, flight transfer, and flight cancellation.

[0105] ① Flight Delay: If the airport's takeoff capacity decreases after an attack, and the number of planned takeoff flights in the subsequent time period is less than the airport's takeoff capacity, then, without affecting the normal takeoff of planned flights in the next time period, the excess takeoff flights during the attack period will be delayed until the next time period.

[0106] ② Flight diversion and takeoff: If the attacked airport belongs to a certain airport group, and the passengers and cargo of the affected flights at that airport can be transferred in advance by road or rail transportation (such as high-speed rail) to other transferable airports in the airport group whose capacity is not saturated, then the transferable airports will perform the transportation tasks of the affected flights at the attacked airport during the original planned time period.

[0107] ③ Flight cancellation: If, after an airport is attacked, the number of planned flights in the subsequent time period has reached the airport's takeoff capacity, and there are no other airports nearby that can be transferred to receive the affected flights, or the takeoff capacity of the airports that can be transferred has reached saturation, then the flight will be cancelled.

[0108] 3) Flight takeoff reallocation method: When the flight demand of the original flight plan exceeds the takeoff capacity of the attacked airport after the landing, the excess takeoff flights are first considered to be delayed to the next time period; until the takeoff capacity of the attacked airport in the next time period is reached, if there are still some flights that have not been allocated in the previous time period, these flights are arranged to take off from other transferable airports in the airport group to which the attacked airport belongs; until the takeoff capacity of other transferable airports is reached, if there are still some flights that have not been allocated in the previous time period at the attacked airport, these flights are canceled.

[0109] Step 7) Input the airspace navigation data selected in Step 1) and the flight reassignment plan generated in Step 6) into the flight display and interaction subsystem in the third client 5, call the flight motion model and flight dynamic model in the simulation driver database in the simulation driver server 2, set the simulation time, simulate the flight reassignment plan after the attack, obtain the simulation results of the flight reassignment plan after the attack, and display them on the flight display and interaction subsystem.

[0110] Step 8) Based on the simulation results of the flight reassignment plan after the attack obtained in Step 7), record the simulation data after the attack, including the actual departure time and actual arrival time of all flights, and input these simulation data after the attack into the flight plan robustness evaluation subsystem in the fourth client 6.

[0111] Step 9) In the network structure robustness evaluation module of the flight plan robustness evaluation subsystem of the fourth client 6, the robustness of the flight plan before and after the attack is calculated using the simulation data before the attack obtained in step 5) and the simulation data after the attack obtained in step 8).

[0112] The specific method is as follows:

[0113] 9.1) Calculate the shortest path d between any pair of airports (i,j) in the air traffic network based on the Floyd algorithm. ij ;

[0114] 9.2) Based on the flight schedule, calculate the number of flights q on the shortest path between any pair of airports (i,j). ij ;

[0115] 9.3) Based on the above shortest path d between airport pairs (i,j), ij and flight volume q ijCalculate the weighted network efficiency of the airport pair:

[0116]

[0117] When airport pairs (i,j) are not connected, the shortest path d ij When the network approaches infinity, the weighted network efficiency ε is used. ij =0 is used to characterize the connectivity of airport pair (i,j);

[0118] 9.4) Calculate the average of the weighted network efficiencies between all the above airport pairs to characterize the global weighted network efficiency of the entire air traffic network:

[0119]

[0120] Where n is the total number of airports in the air traffic network, and the number of airport pairs is n(n-1);

[0121] 9.5) Considering the dynamic nature of flight schedules, the rate of change of the globally weighted network efficiency at different time periods is further calculated as an indicator of the robustness of the topology:

[0122]

[0123] Where E0 is the globally weighted network efficiency of the flight plan before the attack, E a Globally weighted network efficiency for flight schedules after an attack;

[0124] The aforementioned rate of change in global weighted network efficiency reflects the impact of an attack on flight schedules on node connectivity efficiency. The smaller the value of this indicator, the greater the decline in global weighted network efficiency, and the worse the flight schedules' ability to resist attacks, i.e., the worse their robustness.

[0125] Step 10) In the traffic function robustness evaluation module of the flight plan robustness evaluation subsystem of the fourth client 6, based on the flight reallocation plan obtained in Step 6), calculate the traffic function robustness indicators including flight delay rate, airport group saturation change rate and flight cancellation rate.

[0126] The specific method is as follows:

[0127] 10.1) Calculate the flight delay rate after a flight schedule attack. The more flights delayed after an airport is attacked, the worse the robustness of the flight schedule. Therefore, the flight delay rate can be used as one of the indicators of the robustness of transportation functions.

[0128] Based on the simulation results of the flight reassignment plan after the attack on the third client 5 obtained in step 7), the flight delay rate is calculated:

[0129]

[0130] Among them, D i,t Let x be the number of flight delays at airport i during time period t. i,t η represents the number of planned departure flights at airport i during time period t. D,t Let t be the flight delay rate for time period t, and n be the total number of airports.

[0131] 10.2) The rate of change in the saturation of a computer cluster; the overall departure capacity of an airport cluster can, to some extent, measure the robustness of flight schedules. Compared to flight cancellations or delays, although diverting flights reduces losses to some extent, transferring passengers to other available airports incurs costs and increases operational pressure on those airports. Therefore, the greater the rate of change in the saturation of an airport cluster before and after an attack, the worse the local robustness of that airport cluster.

[0132] Based on the simulation results of the flight reassignment plan after the attack on the third client 5 obtained in step 7), the airport group saturation change rate is calculated:

[0133]

[0134] Among them, b j Let NA be the takeoff capacity of airport j. In equation (5), the denominator is the sum of the takeoff capacities of all airport groups in the air traffic network, and the numerator is the number of flights taking off from other airports. i Let be the number of airports in the airport group to which airport i belongs. If airport i does not belong to any airport group, then the number of airports is NA. i =0;M i,t η is the number of flights that depart from airport i within the airport cluster and are transferred to other available airports during time period t; M,t Let be the rate of change of airport cluster saturation during time period t.

[0135] 10.3) Calculate the flight cancellation rate; after a flight schedule is attacked, the higher the flight cancellation rate, the greater the traffic loss during that period, and the worse the robustness of the flight schedule. Based on the simulation results of the flight reallocation plan after the attack by the third client 5, the flight cancellation rate is calculated:

[0136]

[0137] Among them, C i,t The number of flight cancellations at airport i during time period t; η C,t Let t be the flight cancellation rate for time period t.

[0138] Step 11) In the flight plan robustness comprehensive evaluation module, the global weighted network efficiency change rate η obtained in step 10) above is used. E,t Flight delay rate η D,t Airport group saturation change rate ηM,t and flight cancellation rate η C,t Normalization is performed to obtain the normalized global weighted network efficiency change rate η' E,t， Normalized flight delay rate η' D,t Normalized airport cluster saturation change rate η' M,t and normalized flight cancellation rate η' C,t Four indicators were used, and then these four normalized indicators were weighted using an expert scoring method to obtain a comprehensive evaluation value η. t This is used to comprehensively measure the robustness of flight schedules. The comprehensive evaluation value η is... t The calculation formula is:

[0139] η t =1-aη′ D,t -bη′ M,t -cη′ C,t -dη′ E,t (7)

[0140] Wherein, a, b, c, and d are the weights of the four normalized indicators; the weights used in this invention are a = 0.1, b = 0.15, c = 0.25, and d = 0.5.

[0141] After the flight schedule was attacked, the overall evaluation value η t The larger the value, the less impact a deliberate attack would have on flight schedules during that period, and the more robust the flight schedules would be.

[0142] Step 12) The flight display and interaction subsystem in the third client 5 will use the global weighted network efficiency change rate η obtained in step 10) above. E,t Flight delay rate η D,t Airport group saturation change rate η M,t and flight cancellation rate η C,t and the comprehensive evaluation value η obtained in step 11). t The display shows the change in the robustness of the air traffic network over time.

[0143] Here are some application examples of the flight schedule robustness evaluation system and method provided by the present invention:

[0144] (1) In the first client 3, a simulation environment is set up, and the attacked airport is selected. In this embodiment, five high-capacity airports across the country are selected, namely ZBAA, ZGGG, ZPPP, ZUUU, and ZUCK. The random attack event is thunderstorm, such as... Figure 2 As shown;

[0145] (2) Input my country's flight schedule for January 2019 into the second client 4 to generate the flight schedule required for the simulation before the attack;

[0146] (3) Simulate the flight schedule before the attack on the third client 5, count the take-off and landing times of each flight, and store them on the fourth client 6.

[0147] (4) In response to the attack airport and attack event set by the first client 3, the flight reassignment module of the flight plan robustness evaluation subsystem of the fourth client 6 reassigns the flight and generates the flight reassignment plan after the attack.

[0148] Based on the selected random attack event, the takeoff capacity of the attacked airport was adjusted, and the adjustment results are shown in Table 1:

[0149] Table 1. Takeoff Capacity of the Attacked Airport Before and After the Attack

[0150]

[0151] The flight redistribution results for each airport in each time period were calculated based on the takeoff capacity after the attack, as shown in Table 2-4.

[0152] Table 2 Flight Delays at Different Airports During Different Time Periods

[0153]

[0154] Table 3 Flight transfer volume at each airport during different time periods

[0155]

[0156] Table 4. Flight cancellations at various airports during different time periods

[0157]

[0158] (5) Using the network structure robustness evaluation module in the flight plan robustness evaluation subsystem of the fourth client 6, the global weighted network efficiency change rate of the flight plan in each time period before the computer field attack is shown in Table 5.

[0159] Table 5. Change rate of global weighted network efficiency at different time periods

[0160]

[0161] Using the traffic function robustness evaluation module, robustness indicators for traffic function were calculated, as shown in Table 6.

[0162] Table 6 Robustness Indicators of Flight Schedule Transportation Function

[0163]

[0164] (6) The robustness of the flight plan is comprehensively evaluated using the flight plan robustness evaluation module in the flight plan robustness evaluation subsystem of the fourth client 6, as shown in Table 7.

[0165] Table 7 Robustness Comprehensive Evaluation

[0166]

[0167]

[0168] (7) Display the robustness change curve of the flight schedule for a certain day in the flight display and interaction subsystem of the third client 5, such as Figure 4 As shown in the figure, this graph reflects the robustness of flight schedules at different times. It can be seen that the robustness of flight schedules is worst between 6:00 and 7:00, while the robustness is relatively high before 5:00. The robustness after 8:00 shows slight fluctuations.

Claims

1. A flight schedule robustness evaluation system, characterized in that: The flight plan robustness evaluation system includes a network structure server (1), a simulation driver server (2), a first client (3), a second client (4), a third client (5), and a fourth client (6) that are interconnected by a network. The network structure server (1) is equipped with an airspace navigation database, which is used to provide navigation data services to clients, including airports, navigation stations, waypoints, air routes, restricted areas, danger zones, restricted areas, obstacles, and sectors. The simulation driver server (2) is equipped with a simulation driver database, which is used to provide flight dynamics model and flight motion model data services to the client; The first client (3) is equipped with a flight planning subsystem, which is used to generate flight plans that conform to a certain distribution pattern; The second client (4) is equipped with an air traffic network modeling subsystem, which is used to establish an air traffic network structure model; The third client (5) is equipped with a flight display and interaction subsystem, which is used to display and interact with the air traffic network environment generated by the air traffic network modeling subsystem and the evaluation results generated by the flight plan robustness evaluation subsystem. The selected airspace navigation data, the original flight plan and the air traffic network structure model are input into the flight display and interaction subsystem in the third client (5), the flight motion model and flight dynamic model in the simulation driving database in the simulation driving server (2) are called, the simulation time is set, the original flight plan before the attack is simulated, the flight plan simulation results before the attack are obtained and displayed on the flight display and interaction subsystem. The fourth client (6) is equipped with a flight plan robustness evaluation subsystem, including a flight reassignment module, a network structure robustness evaluation module, a traffic function robustness evaluation module and a flight plan robustness comprehensive evaluation module; Based on the flight plan simulation results before the attack, record the simulation data before the attack, including the actual departure time and actual arrival time of all flights, and input these simulation data before the attack into the flight plan robustness evaluation subsystem in the fourth client (6). The flight redistribution module of the flight plan robustness evaluation subsystem in the fourth client (6) formulates the flight take-off redistribution strategy for the attacked airport. Based on the pre-selected attacked airport and random attack events, the airport take-off capacity of the attacked airport is adjusted. Based on the flight redistribution strategy, the overcapacity take-off flights of the attacked airport are redistributed from the perspective of actual operation to generate a flight redistribution plan, thereby adjusting the original flight plan. The selected airspace navigation data and the generated flight reassignment plan are input into the flight display and interaction subsystem in the third client (5), the flight motion model and flight dynamic model in the simulation driving database in the simulation driving server (2) are called, the simulation time is set, the flight reassignment plan after the attack is simulated, the simulation result of the flight reassignment plan after the attack is obtained and displayed on the flight display and interaction subsystem; Based on the obtained simulation results of the flight reassignment plan after the attack, the simulation data after the attack, including the actual departure time and actual arrival time of all flights, is recorded, and these simulation data after the attack are input into the flight plan robustness evaluation subsystem in the fourth client (6). In the network structure robustness evaluation module of the flight plan robustness evaluation subsystem of the fourth client (6), the robustness of the flight plan before and after the attack is calculated using the simulation data before and after the attack, and the global weighted network efficiency change rate is calculated. Calculate the average of the weighted network efficiencies between all the aforementioned airport pairs to characterize the global weighted network efficiency of the entire air traffic network: （2）； in, The total number of airports in the air traffic network, and the number of airport pairs. indivual; Considering the dynamic nature of flight schedules, the rate of change of the globally weighted network efficiency at different time periods is further calculated as an indicator of the robustness of the topology: （3）； in, To optimize the global weighted network efficiency of flight schedules prior to the attack. Globally weighted network efficiency for flight schedules after an attack; In the traffic function robustness evaluation module of the flight plan robustness evaluation subsystem of the fourth client (6), traffic function robustness indicators, including flight delay rate, airport group saturation change rate and flight cancellation rate, are calculated according to the flight reassignment plan. Based on the simulation results of the flight reassignment plan after the attack obtained from the third client (5), the airport group saturation change rate was calculated: （5）； Among them, b j Let NA be the takeoff capacity of airport j. In equation (5), the denominator is the sum of the takeoff capacities of all airport groups in the air traffic network, and the numerator is the number of flights taking off from other airports. i Let be the number of airports in the airport group to which airport i belongs. If airport i does not belong to any airport group, then the number of airports is NA. i = 0; M i,t η represents the number of flights that depart from airport i within the airport cluster and are transferred to other available airports during time period t; M,t Let be the rate of change of airport cluster saturation during time period t; In the flight plan robustness comprehensive evaluation module, the aforementioned global weighted network efficiency change rate is used. Flight delay rate η D,t Airport group saturation change rate η M,t and flight cancellation rate η C,t Normalization is performed to obtain the normalized global weighted network efficiency change rate η' E,t Normalized flight delay rate η' D,t Normalized airport cluster saturation change rate η' M,t and normalized flight cancellation rate η' C,t Four indicators were used, and then these four normalized indicators were weighted using an expert scoring method to obtain a comprehensive evaluation value η. t .

2. An evaluation method using the flight schedule robustness evaluation system of claim 1, characterized in that: The evaluation method includes the following steps performed in sequence: Step 1) Select airspace navigation data, including airports, navigation stations, waypoints, air routes, restricted areas, danger zones, restricted areas, obstacles and sectors, in the airspace navigation database of the network structure server (1); pre-select the airports to be attacked and random attack events, including bad weather and aircraft malfunctions, involved in random attacks. Step 2) Based on the actual flight operation distribution pattern of the evaluated airspace, compile an original flight plan that conforms to the actual distribution pattern in the flight planning subsystem of the first client (3); Step 3) Input the airspace navigation data selected in Step 1) and the original flight plan compiled in Step 2) into the air traffic network modeling subsystem in the second client (4). Based on the original flight plan, construct the air traffic network structure model with airports as nodes and the lines between airports as edges. Step 4) Input the airspace navigation data selected in Step 1), the original flight plan compiled in Step 2), and the air traffic network structure model established in Step 3) into the flight display and interaction subsystem in the third client (5), call the flight motion model and flight dynamic model in the simulation driving database in the simulation driving server (2), set the simulation time, perform the original flight plan operation simulation before the attack, obtain the flight plan simulation results before the attack, and display them on the flight display and interaction subsystem; Step 5) Based on the flight plan simulation results obtained in Step 4) above, record the simulation data before the attack, including the actual departure time and actual arrival time of all flights, and input these simulation data before the attack into the flight plan robustness evaluation subsystem in the fourth client (6). Step 6) The flight redistribution module of the flight plan robustness evaluation subsystem in the fourth client (6) formulates the flight take-off redistribution strategy for the attacked airport. Based on the attacked airport and random attack events pre-selected in Step 1), the airport take-off capacity of the attacked airport is adjusted. Based on the flight redistribution strategy, the overcapacity take-off flights of the attacked airport are redistributed from the perspective of actual operation to generate a flight redistribution plan, thereby adjusting the original flight plan. Step 7) Input the airspace navigation data selected in Step 1) and the flight reassignment plan generated in Step 6) into the flight display and interaction subsystem in the third client (5), call the flight motion model and flight dynamic model in the simulation driving database in the simulation driving server (2), set the simulation time, simulate the flight reassignment plan after the attack, obtain the simulation result of the flight reassignment plan after the attack, and display it on the flight display and interaction subsystem; Step 8) Based on the simulation results of the flight reassignment plan after the attack obtained in Step 7), record the simulation data after the attack, including the actual departure time and actual arrival time of all flights, and input these simulation data after the attack into the flight plan robustness evaluation subsystem in the fourth client (6). Step 9) In the network structure robustness evaluation module of the flight plan robustness evaluation subsystem of the fourth client (6), the robustness of the flight plan before and after the attack is calculated using the simulation data before the attack obtained in step 5) and the simulation data after the attack obtained in step 8), and the global weighted network efficiency change rate is calculated. Step 10) In the traffic function robustness evaluation module of the flight plan robustness evaluation subsystem of the fourth client (6), the traffic function robustness indicators, including flight delay rate, airport group saturation change rate and flight cancellation rate, are calculated based on the flight reassignment plan obtained in Step 6) above. Step 11) In the flight plan robustness comprehensive evaluation module, the obtained global weighted network efficiency change rate will be used. Flight delay rate η D,t Airport group saturation change rate η M,t and flight cancellation rate η C,t Normalization is performed to obtain the normalized global weighted network efficiency change rate η' E,t Normalized flight delay rate η' D,t Normalized airport cluster saturation change rate η' M,t and normalized flight cancellation rate η' C,t Four indicators were used, and then these four normalized indicators were weighted using an expert scoring method to obtain a comprehensive evaluation value η. t ; Step 12) The global weighted network efficiency change rate obtained in the flight display and interaction subsystem of the third client (5) Flight delay rate η D,t Airport group saturation change rate η M,t and flight cancellation rate η C,t and the comprehensive evaluation value η obtained in step 11). t The display shows the change in the robustness of the air traffic network over time.

3. The evaluation method for the flight schedule robustness evaluation system as described in claim 2, characterized in that: In step 6), the flight takeoff reassignment strategy involves the assumptions, forms, and methods of flight takeoff reassignment. 1) The assumptions for the flight takeoff reallocation are as follows: ① Severe incidents only attack nodes in the air traffic network, without considering the attack situation on the edges of the air traffic network; ②The road and rail transportation between adjacent airports is well-developed and convenient; ③ Airlines have sufficient aircraft and crew capacity at various airports; ④ Assuming sufficient route capacity, only the limitations of airport takeoff capacity on flight operations are considered; ⑤ Without considering the mutual influence between departing and arriving flights, the airport has sufficient landing capacity; ⑥ Assume that after the airport is attacked, its functions will not be completely lost and it will still have a certain capacity. ⑦ Flights can only be delayed by a maximum of one hour. 2) There are three main forms of flight takeoff reassignment: flight delay, flight transfer, and flight cancellation; ① Flight Delay: If the airport's takeoff capacity decreases after an attack, and the number of planned takeoff flights in the subsequent time period is less than the airport's takeoff capacity, then, without affecting the normal takeoff of planned flights in the next time period, the excess takeoff flights during the attack period will be delayed until the next time period. ② Flight diversion and takeoff: If the attacked airport belongs to a certain airport group, and the passengers and cargo of the affected flights at that airport can be transferred in advance by road or rail to other transferable airports in the airport group that are not yet at full capacity, then the transferable airports will perform the transportation tasks for the affected flights at the attacked airport during the original planned time period. ③ Flight cancellation: If, after an airport is attacked, the number of planned flights in the subsequent time period has reached the airport's takeoff capacity, and there are no other airports nearby that can be transferred to receive the affected flights, or the takeoff capacity of the airports that can be transferred has reached saturation, then the flight will be cancelled. 3) Flight takeoff reallocation method: When the flight demand in the original flight plan exceeds the takeoff capacity of the attacked airport after landing, the first consideration is to delay the excess takeoff flights to the next time slot; Until the next time period reaches the airport takeoff capacity of the attacked airport, if there are still some flights that have not been allocated in the previous time period, these flights will be arranged to take off from other transferable airports in the airport group to which the airport belongs. Until the takeoff capacity of other available airports is reached, if there are still some flights that have not been allocated at the attacked airport in the previous period, those flights will be canceled.

4. The evaluation method for the flight schedule robustness evaluation system as described in claim 2, characterized in that: In step 9), the specific method for calculating the robustness of flight plans before and after the attack using the pre-attack simulation data obtained in step 5) and the post-attack simulation data obtained in step 8) is as follows: 9.1) Calculate the shortest path d between any pair of airports (i,j) in the air traffic network based on the Floyd algorithm. ij ; 9.2) Based on the flight schedule, calculate the number of flights q on the shortest path between any pair of airports (i,j). ij ; 9.3) Based on the above, the shortest path d between airport pairs (i,j) ij and flight volume q ij Calculate the weighted network efficiency of the airport pair: （1） ； When airport pairs (i,j) are not connected, the shortest path d ij When the network approaches infinity, the weighted network efficiency ε is used. ij =0 is used to characterize the connectivity of airport pair (i,j); 9.4) Calculate the average of the weighted network efficiencies between all the above airport pairs to characterize the global weighted network efficiency of the entire air traffic network: （2）； in, The total number of airports in the air traffic network, and the number of airport pairs. indivual; 9.5) Considering the dynamic nature of flight schedules, the rate of change of the globally weighted network efficiency at different time periods is further calculated as an indicator of the robustness of the topology: （3）； in, To optimize the global weighted network efficiency of flight schedules prior to the attack. The globally weighted network efficiency of flight schedules after an attack.

5. The evaluation method for the flight schedule robustness evaluation system according to claim 2, characterized in that: In step 10), the specific method for calculating the traffic function robustness indicators, including flight delay rate, airport cluster saturation change rate, and flight cancellation rate, based on the flight reallocation plan obtained in step 6) is as follows: 10.1) Calculate the flight delay rate after a flight schedule attack; Based on the simulation results of the flight reassignment plan after the attack obtained by the third client (5) in step 7), the flight delay rate is calculated: （4）； Among them, D i,t Let x be the number of flight delays at airport i during time period t. i,t η represents the number of planned departure flights at airport i during time period t. D,t Let be the flight delay rate for time period t, and n be the total number of airports; 10.2) Rate of change of computer field cluster saturation; Based on the simulation results of the flight reassignment plan after the attack obtained by the third client (5) in step 7), the airport group saturation change rate is calculated: （5）； Among them, b j Let NA be the takeoff capacity of airport j. In equation (5), the denominator is the sum of the takeoff capacities of all airport groups in the air traffic network, and the numerator is the number of flights taking off from other airports. i Let be the number of airports in the airport group to which airport i belongs. If airport i does not belong to any airport group, then the number of airports is NA. i = 0; M i,t η represents the number of flights that depart from airport i within the airport cluster and are transferred to other available airports during time period t; M,t Let be the rate of change of airport cluster saturation during time period t; 10.3) Calculate the flight cancellation rate; Based on the simulation results of the flight reallocation plan after the attack by the third client (5), the flight cancellation rate was calculated: （6）； Among them, C i,t η represents the number of flight cancellations at airport i during time period t; C,t Let t be the flight cancellation rate for time period t.

6. The evaluation method for the flight schedule robustness evaluation system according to claim 2, characterized in that: In step 11), the comprehensive evaluation value η t The calculation formula is: （7）； Where a, b, c, and d are the weights of the four normalized indicators, respectively.

Images