Urban air traffic vertiport siting method

Abstract

The application discloses a kind of urban air traffic vertical take-off and landing airport site selection method, it is related to urban low-altitude traffic facility site selection planning technical field, ground traffic travel history data is mapped to the several discretization units in target city spatial grid after, obtain demand point set;Demand point set is carried out clustering analysis based on clustering algorithm, generate candidate take-off and landing airport point set;Based on candidate take-off and landing airport point and its type, demand point set and ground traffic travel history data, define the basic data set required for vertical take-off and landing airport site selection optimization model and model constraint condition including the hierarchical dependency constraint between different types of candidate take-off and landing airport point;Multi-objective evolutionary algorithm is used to solve vertical take-off and landing airport site selection optimization model, and the optimal layout scheme is output.Introducing facility function heterogeneity modeling and hierarchical dependency constraint in multi-objective site selection model makes the site selection scheme more in line with actual operation design.

Description

Technical Field

[0001] This invention relates to the field of urban low-altitude transportation facility site selection and planning technology, specifically to a method for selecting the site of a vertical take-off and landing airport for urban air transportation. Background Technology

[0002] With the acceleration of urbanization and the intensification of ground traffic congestion, urban air transportation using electric vertical takeoff and landing (VTOL) aircraft for passenger and cargo transport has become an emerging mode of travel. As the physical nodes of the urban air transport network, the site selection, planning, and network layout of VTOL airports directly affect the operational efficiency and feasibility of the entire system. Research has found that existing VTOL airport site selection and planning technologies mainly suffer from the following technical bottlenecks and shortcomings:

[0003] First, existing technical solutions typically treat vertical takeoff and landing (VTOL) airports in different locations as homogeneous facilities, ignoring the heterogeneity of their functions. In actual operational scenarios, VTOL airports of different sizes and locations exhibit significant differences in energy supply (such as charging / battery swapping), maintenance support, and daily turnaround capacity. Existing technical models often assume that all nodes have the same service capabilities, leading to a mismatch between the planned facility layout and actual operational needs. For example, planning high-intensity maintenance tasks at small sites lacking maintenance capabilities will render them physically impossible to execute.

[0004] Second, existing technologies lack a collaborative design for the connectivity of hierarchical networks at vertical takeoff and landing (VTOL) airports. Current research largely focuses on the single issue of site selection (i.e., determining location), neglecting the issue of route connectivity between facilities (i.e., network design). Because there are objective hierarchical dependencies between VTOL airports of different functional levels (i.e., lower-level facilities often rely on higher-level facilities for operation, maintenance, and scheduling), ignoring this hierarchical constraint in the site selection model can lead to the planning of "island facilities" or "dead-end routes" in the network. For example, if a terminal station is not connected to a higher-level node with supporting capabilities, the continuous operation of aircraft cannot be guaranteed.

[0005] In summary, due to the discrepancy between existing technical solutions and actual physical operational constraints, neglecting differences in facility types and hierarchical dependencies during the site selection planning and network layout design of vertical take-off and landing (VTOL) airports can lead to physically unconnectable or operationally unsustainable layouts. Therefore, there is an urgent need for an optimization method for the site selection of VTOL airports in urban air traffic to generate scientifically sound planning schemes that meet actual operational conditions. Summary of the Invention

[0006] The purpose of this invention is to provide a method for selecting vertical take-off and landing (VTOL) airports in urban air traffic. By combining the types of VTOL airports and the hierarchical dependency constraints between different types of candidate take-off and landing airports, the method introduces facility function heterogeneity modeling and hierarchical dependency constraints into the multi-objective site selection model, making the obtained site selection scheme more in line with actual operational requirements.

[0007] To achieve the above objectives, this application proposes the following solution:

[0008] On the one hand, the present invention provides a method for site selection of vertical take-off and landing airports for urban air traffic, specifically including the following steps:

[0009] S1. Obtain historical ground transportation data and urban geographic information data of the target city. Based on the urban geographic information data, divide the spatial grid of the target city into several discretized units, and map the historical ground transportation data to the corresponding discretized units to obtain the demand point set.

[0010] S2. Based on the clustering algorithm, perform cluster analysis on the demand point set, select several candidate take-off and landing airport points, and generate a candidate take-off and landing airport point set;

[0011] S3. Construct a vertical take-off and landing airport site selection optimization model. Based on the technical service capabilities of the candidate take-off and landing airport sites, determine the types of candidate take-off and landing airport sites. Based on the set of candidate take-off and landing airport sites, the types of candidate take-off and landing airport sites, the demand point set, and historical ground traffic data, define the basic data set required for the vertical take-off and landing airport site selection optimization model, as well as the model constraints, including the hierarchical dependency constraints between different types of candidate take-off and landing airport sites.

[0012] S4. Use a multi-objective evolutionary algorithm to solve the vertical take-off and landing airport site selection optimization model and output the optimal layout scheme.

[0013] The concept of this invention is as follows:

[0014] Current urban air traffic location models typically treat vertical takeoff and landing (VTOL) airports as functionally homogeneous nodes, neglecting the physical differences in charging infrastructure, maintenance capabilities, and daily turnover capacity between facilities of different levels. This homogeneous assumption leads to planned facility layouts that fail to meet the energy supply and maintenance needs of actual operations, making some facilities "islands" unable to operate independently and resulting in wasted infrastructure construction. Furthermore, route planning often focuses on determining the location of individual facilities, ignoring the design of the route network topology between facilities. Especially when introducing facilities of different functional levels, the lack of clear hierarchical connection constraints (i.e., the dependency relationship between lower-level facilities and higher-level facilities) can result in logically disconnected networks, failing to form an effective urban air traffic service system.

[0015] Therefore, in response to the problems existing in the prior art, such as the physical infeasibility of the planning scheme due to the neglect of facility functional heterogeneity, and the lack of route connectivity due to the disconnect between site selection and network design, this application proposes the following solutions:

[0016] 1. Construct a hierarchical network architecture that conforms to physical operational constraints: By defining vertical take-off and landing airports (such as hubs, bases, and stations) with different functional levels and establishing a strict hierarchical dependency constraint mechanism, ensure that the planned network is physically feasible in terms of energy, maintenance, and scheduling.

[0017] 2. Achieve coordinated optimization of site selection and route network: Integrate facility site selection decisions and route connectivity decisions into the same optimization framework, generate corresponding route connection schemes while determining facility locations, and ensure the overall connectivity and service efficiency of the network.

[0018] 3. Provides an efficient two-stage solution strategy: Through a two-stage framework of "clustering pre-screening of candidate points + multi-objective optimization solution", the computational complexity is reduced while ensuring the quality of the solution, and a scientific planning scheme that balances service coverage and construction cost is quickly generated, providing data support for the construction of urban air traffic infrastructure.

[0019] In some specific implementation schemes, the process of obtaining the demand point set is as follows:

[0020] S11. Perform data cleaning on historical ground transportation travel data, remove data with abnormal travel routes and trajectories, and obtain a valid travel set;

[0021] S12. Based on urban geographic information data, the space of the target city is gridded and divided into several discrete units.

[0022] S13. Map the coordinates of the starting point and ending point of the valid trip set to the corresponding discretized units, perform spatial aggregation on the travel data in each discretized unit, obtain the weight of each discretized unit, treat each discretized unit as a demand point, and finally form a weighted demand point set.

[0023] In some specific implementation schemes, the process of obtaining the candidate take-off and landing airport point set is as follows:

[0024] S21. Use a clustering algorithm to perform cluster analysis on the demand point set, and iteratively calculate the cluster evaluation index under different numbers of cluster centers within a preset range;

[0025] S22. By plotting the curve of the change of cluster evaluation index, identify the abrupt change point of the curve, and determine the number of cluster centers corresponding to the abrupt change point as the optimal number of candidate facilities.

[0026] S23. Select high-density demand points from the demand point set as candidate take-off and landing airport points, and select the demand points with the best number of candidate facilities from the high-density demand points and store them in the candidate take-off and landing airport point set.

[0027] In some specific implementations, the basic dataset includes a set of candidate takeoff and landing airports. Itinerary Collection ( The set of valid trips obtained after data cleaning in step S11; and the set of candidate take-off and landing airport types. The set of administrative divisions Z, and the construction of the vertical take-off and landing airport site selection optimization model also includes:

[0028] The objective function for constructing a vertical takeoff and landing (VTOL) airport site selection optimization model includes a first objective function for maximizing the number of air segments served and a second objective function for minimizing the facility construction costs for the system operator. The first objective function Z1 is:

[0029]

[0030] in, Indicates itinerary Should I select from the candidate takeoff and landing airports? To candidate take-off and landing airports The air flight segment;

[0031] The second objective function Z2 is:

[0032]

[0033] in, The type is The construction cost of candidate take-off and landing airport sites, Indicate whether to include candidate take-off and landing airports Type construction as .

[0034] In some specific implementation schemes, the model constraints include:

[0035] Uniqueness constraint: Ensure that each candidate take-off and landing airport point can only be assigned one type in space;

[0036]

[0037] Administrative division constraints: Within each administrative division of the target city, the number of high-level node candidate take-off and landing airport sites is limited.

[0038]

[0039] in, Indicates administrative region z The set of candidate take-off and landing airports within the country. Indicates the first Whether each candidate take-off and landing airport site is to be developed into a high-level node hub ;

[0040] Hierarchical constraints: There are hierarchical dependencies between different types of candidate take-off and landing airports;

[0041] Consistency and symmetry constraints of flight routes: Flight routes between candidate take-off and landing airports must meet the requirement of two-way traffic.

[0042] Trip selection constraints: Simulate the entire process of real urban air transportation services to ensure that the system's allocation of travel demand conforms to intermodal transport logic and does not exceed network capacity;

[0043] Maximum daily processing capacity constraints for each candidate takeoff and landing airport:

[0044]

[0045] in, Indicates the air flight segment y :journey For candidate take-off and landing airports Arrival at candidate take-off and landing airports ; Indicates the air flight segment y :journey For candidate take-off and landing airports Arrival at candidate take-off and landing airports ; Indicates the first The types are Candidate take-off and landing airports, Representation type The maximum daily processing capacity of candidate take-off and landing airports.

[0046] In some specific implementation schemes, the candidate take-off and landing airport points include at least high-level nodes with the highest level of technical service capabilities, medium-level nodes with medium-level technical service capabilities, and low-level nodes with basic level technical service capabilities, with the following hierarchical dependency relationship:

[0047] Each selected mid-level node must have a direct flight path to at least one selected high-level node, defined as:

[0048]

[0049] in, Indicates the first The candidate take-off and landing airports and the first Whether to open air routes between the candidate take-off and landing airports. Indicates the first Whether the candidate take-off and landing airport sites are designated as high-level nodes. hub , Indicates the first Whether each candidate take-off and landing airport site is to be developed into a medium-level node base ;

[0050] Each low-level node, acting as a terminal access facility, must have a direct flight path with at least one high-level or mid-level node, defined as:

[0051]

[0052] in, Indicates the first Whether each candidate take-off and landing airport site is constructed as a low-level node. .

[0053] In some specific implementation schemes, the basic dataset includes a set of ground transportation modes. The simulation of a real-world urban air service process includes: departure ground transfer, air flight, and destination ground transfer. Travel modes include purely ground transportation and ground-air-ground intermodal transport. Trip selection constraints include:

[0054] 1) Each traveler may only choose between purely ground transportation and ground-air-ground intermodal transportation:

[0055]

[0056] in, Indicates itinerary Whether to use purely ground transportation;

[0057] 2) Aerial flight segment Candidate take-off and landing airports and routes between It is enabled:

[0058]

[0059] 3) When selecting the ground-air-ground intermodal transport mode, both the departure and arrival airports must be operational:

[0060]

[0061]

[0062] in, and These represent the candidate take-off and landing airports. and Whether it is built as a type The takeoff and landing airports, and the candidate takeoff and landing airports. and They serve as departure and arrival airports respectively in ground-air-ground intermodal transport;

[0063] 4) The ground connection method and the air flight method should be consistent:

[0064]

[0065]

[0066] 5) Each traveler may only choose one combination of candidate departure and arrival airports:

[0067]

[0068]

[0069] in, Indicates itinerary Whether to use ground transportation From the point of origin to the candidate take-off and landing airport point , Indicates itinerary Whether to use ground transportation From candidate take-off and landing airports Depart for your destination.

[0070] In some specific implementation schemes, the process of outputting the optimal layout scheme in step S4 is as follows:

[0071] S41. Generate an initial population using multi-base integer encoding based on the number of candidate take-off and landing airports;

[0072] S42. Check whether the initial population meets the model constraints, and perform constraint feasibility repair on the initial population. After each new individual is generated, execute the repair operator to force the model constraints to be met, including hub number repair and hierarchical constraint repair, to obtain a feasible population.

[0073] S43. A greedy heuristic strategy is used to decode the feasible population, simulate the actual operation process, and calculate the fitness of each individual in the feasible population based on the objective function value.

[0074] S44. Based on the fitness of each individual, execute an evolutionary loop to iteratively solve for the optimal addressing scheme; when the iteration reaches the preset termination condition, output a Pareto optimal solution set containing multiple candidate schemes;

[0075] S45. Based on the preset preference parameters, select a candidate solution from the Pareto optimal solution set as the optimal layout solution and output it.

[0076] In some specific implementation schemes, step S43 is performed as follows:

[0077] Each site selection scheme is treated as an individual in the feasible population. Based on the layout of candidate take-off and landing airports in the individual, and according to the maximum physical range limit of the electric vertical take-off and landing aircraft, a set of all feasible routes is generated.

[0078] For each trip in the trip set, calculate the generalized cost of each trip under different travel modes, including pure ground transportation and air intermodal transportation.

[0079] Compare the generalized costs of each trip under pure ground transportation and air intermodal transportation modes. If the generalized cost of air intermodal transportation is lower, and there are feasible air flight segments and the remaining physical capacity of the candidate take-off and landing airports meets the demand, then the trip is allocated to air intermodal transportation, and the daily processing capacity of the corresponding origin and destination candidate take-off and landing airports is deducted. If the generalized cost of pure ground transportation is lower, then the trip remains a pure ground transportation trip.

[0080] The objective function value is obtained by summing the number of trips corresponding to the served air segments and the cost of building each candidate take-off and landing airport. The individual fitness is then calculated based on the objective function value.

[0081] In some specific implementation plans, the generalized cost of air intermodal transport includes the time cost. With monetary cost :

[0082]

[0083]

[0084] Among them, the time cost of ground transfer from departure. and monetary costs It depends on the ground transportation method selected from the departure point to the takeoff point and the candidate takeoff and landing airports where the takeoff point is located;

[0085] Time cost of air travel The monetary cost of air travel depends on the flight path distance between candidate airports, transfer time, and takeoff and landing physical maneuver time. It depends on the route distance and the unit flight operating cost parameters;

[0086] Destination ground connection time cost and monetary costs It depends on the vertical takeoff and landing airport point at the time of departure and the ground transportation connection selected when leaving the candidate takeoff and landing airport point.

[0087] The advantages of this invention over the prior art are as follows:

[0088] 1. This invention improves the practical operational feasibility and physical connectivity of vertical takeoff and landing (VTOL) airport networks: By introducing facility functional heterogeneity modeling and hierarchical dependency constraints into the site selection model, it overcomes the shortcomings of traditional site selection models that treat facilities as homogeneous nodes and are difficult to support actual operations. By constructing a hierarchical network architecture that conforms to physical operational constraints, it defines different functional levels of VTOL airports (such as hubs, bases, and stations) and their differences in charging, maintenance, and turnaround capabilities, and mandates that lower-level facilities must establish route connections with higher-level facilities (hierarchical constraints). This ensures that the planned network is physically feasible in terms of energy supply and operation and maintenance support, avoiding the emergence of "isolated facilities" or "dead-end routes" that cannot maintain independent operation, allowing the site selection scheme to directly guide actual infrastructure construction.

[0089] 2. Significantly reduces the computational complexity of large-scale site selection optimization problems and improves solution efficiency. This invention designs a two-stage planning framework that integrates spatial clustering pre-screening and multi-objective optimization algorithms, effectively solving the computational bottleneck under large-scale demand points at the city level. The first stage uses spatial clustering methods (such as K-means++) to screen high-potential candidate points from massive discrete demand units, greatly reducing the solution space dimension of the subsequent optimization model; the second stage performs multi-objective optimization only on the candidate set. While ensuring the quality and feasibility of the solution, it significantly reduces the number of algorithm iterations and computation time, making this method applicable to rapid planning and dynamic adjustment of large-scale urban areas.

[0090] 3. Achieving a multi-objective balance between service coverage and construction economic costs. This invention constructs a dual-objective optimization model that can simultaneously quantify and evaluate the network's service capacity and investment costs. Objective one is to maximize the number of serviced air segments, and objective two is to minimize the total cost of facility construction. A multi-objective evolutionary algorithm (such as NSGA-II) is used to solve for the Pareto optimal solution set. This provides decision-makers with multiple optimal solutions under different preferences, avoiding the problems of excessive costs or insufficient service caused by single-objective optimization, and improving the scientific rigor and flexibility of the planning scheme. Attached Figure Description

[0091] Figure 1 A flowchart of the urban air traffic vertical take-off and landing airport site selection method provided in this embodiment of the invention;

[0092] Figure 2This is a schematic diagram of a hierarchical network design scheme considering different types of vertical take-off and landing airports, provided for an embodiment of the present invention. Detailed Implementation

[0093] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0094] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0095] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0096] Furthermore, for clarity and brevity, descriptions of well-known structures, functions, and configurations may have been omitted. Those skilled in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of this disclosure.

[0097] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0098] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0099] Example 1

[0100] like Figure 1 As shown in the figure, this embodiment provides a method for selecting the site of a vertical take-off and landing airport for urban air traffic, which specifically includes the following steps:

[0101] S1. Data Preprocessing and Spatial Discretization: Obtain historical ground traffic data and urban geographic information data of the target city. Based on the urban geographic information data, the spatial grid of the target city is divided into several discretized units, and the historical ground traffic data is mapped to the corresponding discretized units to obtain the demand point set.

[0102] Since urban air transport is not yet in large-scale operation and direct operational data is lacking, this embodiment uses historical ground traffic data and urban geographic information data as basic data to indirectly reflect potential air transport demand through data mapping. The specific processing flow is as follows:

[0103] S11. Data Cleaning: Obtain raw travel order data from historical ground transportation data, remove data with abnormal trips and trajectories (such as trajectory drift and extremely short trips), retain trip data that meets the potential application scenarios of urban air transportation (such as long-distance travel trips with a trip distance greater than a preset threshold), and form an effective trip set.

[0104] S12. Spatial gridding: In order to reduce the computational complexity of the subsequent optimization model and eliminate spatial directional bias, the target city space is gridded based on urban geographic information data. The spatial gridding algorithm is used to discretize the continuous study area of ​​the target city into several discretized units.

[0105] For example, a regular hexagonal grid system with side lengths of a specific scale can be established. Regular hexagonal grids are isotropic and can more uniformly cover urban space. In other embodiments, the discretization unit can also be a square grid, a triangular grid, or directly use administrative street boundaries or geofences as the discretization unit.

[0106] S13. Demand Mapping: Map the coordinates of the origin and destination of the cleaned valid trip set to the corresponding discretized units (e.g., to the centroid of a hexagonal grid or the center of an administrative unit). Spatial aggregation is performed on the travel data within each discretized unit to obtain the weight of each discretized unit. Each discretized unit is treated as a demand point, ultimately forming a weighted set of demand points. The weight can be the number of trips, the number of travelers, or the intensity of potential demand within that discretized unit.

[0107] S2. Cluster-based candidate location identification: Based on the clustering algorithm, the set of demand points is clustered to select several candidate take-off and landing airport points and generate a set of candidate take-off and landing airport points.

[0108] Directly optimizing site selection on all discretized units is computationally too intensive. Therefore, this step utilizes spatial clustering to filter candidate facility locations with high potential demand from a massive number of demand points, generating a candidate set of takeoff and landing airports as input to the subsequent optimization model, thereby reducing the spatial dimensionality of the solution. The specific steps are as follows:

[0109] S21. Cluster Analysis: A clustering algorithm is used to perform cluster analysis on the weighted set of demand points generated in step S1. For example, the K-means++ clustering algorithm can be used. This algorithm improves clustering stability by optimizing the selection probability of the initial centers, avoiding the traditional K-means algorithm from getting trapped in local optima. In some specific implementations, the clustering algorithm can also be the DBSCAN algorithm, hierarchical clustering algorithm, or Gaussian mixture model. The core is to identify the centers of high-density demand areas through clustering.

[0110] S22. Determining the Optimal Number: Iteratively calculate the clustering evaluation index under different numbers of cluster centers within a preset range. By plotting the change curve of the clustering evaluation index, identify the abrupt change points (inflection points or elbows) of the curve, and determine the number of cluster centers corresponding to the abrupt change point as the optimal number of candidate facilities. The clustering evaluation index is preferably the weighted sum of squared errors within clusters; when using other clustering algorithms, it can also be an index that can evaluate the clustering effect, such as the silhouette coefficient or the Davidson-Boudin index.

[0111] S23. Candidate Set Generation: The coordinates of the cluster center points or high-density points obtained from clustering are used as candidate vertical take-off and landing (VTOL) airports. The candidate airports corresponding to the optimal number of candidate facilities are selected as the basis for the solution space of the subsequent optimization model. Subsequent steps only perform type selection and network construction on these candidate points, rather than searching the entire continuous urban space, thereby improving computational efficiency.

[0112] Step S3: Construct a vertical take-off and landing airport site selection optimization model: Based on the technical service capabilities of the candidate take-off and landing airport sites, determine the types of candidate take-off and landing airport sites. Based on the set of candidate take-off and landing airport sites, the types of candidate take-off and landing airport sites, the set of demand points, and historical ground traffic data, define the basic data set required for the vertical take-off and landing airport site selection optimization model, as well as the model constraints, including the hierarchical dependency constraints between different types of candidate take-off and landing airport sites.

[0113] Since the site selection and network design of urban air transport vertical takeoff and landing (VTOL) airports are still in the strategic planning stage, the design of urban air transport networks must, on the one hand, serve more users of urban air transport, and on the other hand, achieve this with the lowest possible construction cost. Therefore, a bi-objective mixed-integer programming model is established for further optimization. The specific steps are as follows:

[0114] S31. Defining Different Types of Vertical Take-Off and Landing Airports: Considering the heterogeneity of facility functions in actual operation, vertical take-off and landing airports are divided into different functional types. Different types of facilities (in this article, facilities refer to vertical take-off and landing airports) have significant differences in technical service capabilities (such as construction costs, maximum daily processing capacity, energy supply capabilities, and maintenance support capabilities). In some preferred embodiments, vertical take-off and landing airports are divided into three categories: vertical take-off and landing hubs (high-level nodes) with the highest level of technical service capabilities, vertical take-off and landing bases (medium-level nodes) with medium-level technical service capabilities, and vertical take-off and landing stations (low-level nodes) with basic-level technical service capabilities. It should be noted that the classification names and number of levels in this embodiment are only examples. In other embodiments of the present invention, they can also be divided into two or more levels, or named as Class I facilities, Class II facilities, etc. The core is to distinguish the technical service capabilities of different facilities and their hierarchical roles in the network.

[0115] The specific technical definitions of various facilities are as follows:

[0116] ① Vertical Take-Off and Landing Hubs (High-Level Nodes): Vertical take-off and landing hubs are core nodes in the network, possessing the highest level of technical service capabilities. As the backbone of the urban air traffic network, they are responsible for passenger flow distribution over a large area and for the operation and maintenance support of lower-level nodes. They possess multiple take-off and landing stands and parking positions, with comprehensive charging, battery swapping, and full-service maintenance capabilities, including major overhauls, and the highest maximum daily processing capacity. They are typically located in easily accessible distribution centers, such as near international airports, large train stations, or transportation hubs on the city's outskirts, to connect with ground transportation networks and other aviation networks. They occupy the highest level in the hierarchical network, providing maintenance spare parts support and operational scheduling support for vertical take-off and landing bases.

[0117] ② Vertical Take-Off and Landing (VTOL) Bases (Intermediate-Level Nodes): VTOL bases are the main nodes in the network, possessing intermediate-level technical service capabilities. They are widely distributed in high-demand urban areas, primarily serving the urban air traffic needs of medium- to high-density areas. They have a moderate number of take-off and landing stands, primarily functioning to meet passenger waiting and security check needs, and possessing fast charging or battery swapping capabilities to support continuous aircraft operation, but typically lack major overhaul capabilities. They are commonly found on the rooftops of urban skyscrapers, in large shopping malls, or in business districts—areas with high passenger traffic. They occupy an intermediate level in the hierarchical network, relying on VTOL hubs for in-depth operational support, while also providing route connections and capacity organization support for VTOL stations.

[0118] ③ Vertical Take-Off and Landing Stations (Low-Level Nodes): These stations are end-point access nodes in the network, possessing basic technical service capabilities. As a supplement to large and medium-sized facilities, they offer flexible location options and are primarily used for rapid passenger boarding and alighting, achieving "last-mile" air traffic access. They typically have only a limited number of take-off and landing positions, mainly for quick stop-and-go use, not providing long-term parking or complex maintenance services, and have limited energy replenishment capabilities. They are commonly found in community plazas, the rooftops of standalone office buildings, or small public spaces. At the lowest level of the hierarchical network, they must rely on vertical take-off and landing bases or hubs, obtaining access to energy and maintenance support through flight routes, and cannot sustain long-term operation independently.

[0119] S32. Define decision variables:

[0120] To transform the problem of physical site selection and urban air transport network design in a target city into a computable technical solution, a set of decision variables is constructed to characterize facility status, network topology, and trip assignment. In a preferred embodiment of this application, the decision variables include:

[0121] 1. Facility Site Selection Variables: These are used to characterize the construction status and type configuration of candidate takeoff and landing airport sites. For example, variables... Indicates candidate take-off and landing airports Is it constructed as a type? The variable represents the vertical takeoff and landing (VTOL) airport. Essentially, this variable determines the spatial layout and functional level of the physical facilities. In other implementations, this variable can also be represented using multi-level encoding (e.g., 0 represents no construction, 1 represents a site, 2 represents a base, and 3 represents a hub) or a binary vector group, as long as it can distinguish the type and status of the facility.

[0122] 2. Route connectivity variables: These are used to characterize the network connectivity between vertical takeoff and landing (VTOL) airports. For example, variables... Indicates the first The candidate take-off and landing airports and the first Whether to open air routes between the candidate take-off and landing airports.

[0123] 3. Trip Decision Variables: These are used to characterize the traffic allocation methods under different travel modes. For example, they include air flight segment variables (such as...). ): indicates itinerary Should I select from the candidate takeoff and landing airports? Fly to candidate take-off and landing airports Variables of travel using purely ground transportation modes (such as...) ): indicates itinerary Whether to use purely ground transportation from the origin to the destination, and connecting variables (such as...) , ): Indicates itinerary Whether to use ground transportation From the place of departure Entering candidate take-off and landing airports , Indicates the use of a trip Whether to use ground transportation From candidate take-off and landing airports Departure to the destination. The above decision variables are used to simulate the "ground-air-ground" multimodal transport process to ensure that the model reflects the real travel chain.

[0124] S33, Define a set

[0125] Based on the data processed in steps S1 and S2, a basic dataset is constructed to support the model. In a specific implementation, the basic dataset includes:

[0126] 1. Candidate take-off and landing airport point set That is, the set of candidate vertical takeoff and landing airports generated in step S2;

[0127] 2. Itinerary assembly That is, the set of travel itineraries represented by the weighted demand point set generated in step S1;

[0128] 3. Set of candidate take-off and landing airport types This includes the three categories defined above: vertical takeoff and landing hubs, vertical takeoff and landing bases, and vertical takeoff and landing stations.

[0129] 4. Other auxiliary sets: including the administrative division set Z and the ground transportation mode set. Gathering at the starting point of the journey Meeting at the destination wait.

[0130] S34. Construct the objective function:

[0131] This application employs a multi-objective optimization strategy to balance the number of services and construction investment. The model includes the following two main objective functions:

[0132] ① First objective function: Maximize the number of air segments served. This objective aims to maximize the travel demand fulfilled using the urban air transport network, reflecting the network's service capacity.

[0133]

[0134] By optimizing facility layout and route connections, more trips that were previously inefficient due to ground congestion or long distances can be transformed into trips that can be served by in-flight internet services.

[0135] ② Second objective function: Minimize the infrastructure construction costs for the system operator. This objective aims to minimize the total investment in infrastructure, reflecting the economic feasibility of network construction.

[0136]

[0137] in, The type is The construction cost of candidate take-off and landing airport sites, Indicate whether to include candidate take-off and landing airports Type construction as Different types of facilities (hubs, bases, stations) have significantly different construction costs. This goal-oriented model prioritizes combinations of facility types with higher cost-effectiveness while meeting service demands, avoiding resource waste caused by over-construction of high-level facilities.

[0138] S35. Set model constraints

[0139] To ensure that the generated site selection and network design schemes meet the basic constraints, this application sets the following constraints in the optimization model. It should be noted that the mathematical expression of the following constraints is only a preferred embodiment; in other embodiments of the invention, other equivalent logical expressions can be used to achieve the same technical effect.

[0140] ① Uniqueness constraint: Ensure that each physical candidate point can only be assigned one functional type in space, avoiding facility type conflicts in the same geographical location.

[0141]

[0142] ② Administrative Division Constraints: To prevent the excessive concentration of high-level facilities within specific administrative regions, leading to uneven resource allocation or service coverage blind spots, and to meet the spatial balance requirements of urban planning, the types of facilities restricted within each administrative division of the target city are limited to the number of candidate high-level node take-off and landing airport sites:

[0143]

[0144] in, This represents the set of candidate take-off and landing airports within administrative region z. Indicates the first Whether the candidate take-off and landing airport sites are to be developed into high-level node types (vertical take-off and landing hubs), hub );

[0145] ③ Hierarchical Constraints: Due to significant differences in charging support capabilities, maintenance capabilities, and service capabilities among different types of vertical takeoff and landing (VTOL) airports, the operation of urban air traffic networks must follow a clear hierarchical dependency relationship. Lower-level facilities physically depend on higher-level facilities for operation and maintenance support; therefore, connections must be established on the constructed traffic network topology to ensure accessibility.

[0146] It should be noted that the term "connection" in this application has different physical meanings at different levels. For the operation and maintenance support level (such as base-dependent hubs), "connection" is defined as a physically direct route. This is because maintenance spare parts and energy replenishment have high timeliness requirements, and direct transportation can avoid delays or secondary losses caused by stopovers and transfers, ensuring that lower-level facilities can receive direct operation and maintenance support from higher-level facilities. For the passenger service level (such as station-served passengers), "connection" can theoretically be network reachable, that is, allowing passengers to reach higher-level nodes through transfers to meet a wider range of travel needs. In the preferred embodiment, to simplify model complexity and improve operation and maintenance reliability, the hierarchical dependency constraint is uniformly set to the existence of a direct route.

[0147] Base-dependent hubs: Each selected vertical takeoff and landing (VTOL) base must have a direct flight route to at least one selected VTOL hub.

[0148]

[0149] in, Indicates the first The candidate take-off and landing airports and the first Whether air routes can be opened between the candidate take-off and landing airports;

[0150] As an end-point access facility, a vertical take-off and landing station must be connected to a base or hub to ensure accessibility for energy and maintenance support.

[0151] Each low-level node, acting as a terminal access facility, must have a direct flight path with at least one high-level or mid-level node, defined as:

[0152]

[0153] in, Indicates the first Is the type of candidate take-off and landing airport site to be developed into a high-level node (vertical take-off and landing hub)? ， Indicates the first Whether each candidate take-off and landing airport site is to be developed into a medium-level node base ; Indicates the first Whether each candidate take-off and landing airport site is to be developed into a medium-level node (vertical take-off and landing base);

[0154] Through the above constraints, the model enforces that lower-level facilities do not become "islands" on the physical network and must depend on higher-level facilities to exist, thereby ensuring the operational feasibility of the entire air traffic network.

[0155] ④ Consistency and symmetry constraints of the route: Ensure that the establishment of the route is based on the actual construction of facilities at both ends and conforms to the physical characteristics of two-way traffic.

[0156] Facilities (candidate take-off and landing airports) must be built at both ends of the route:

[0157]

[0158]

[0159] The route direction is symmetrical (suitable for two-way operation scenarios):

[0160]

[0161] If applicable to a one-way route scenario, the symmetry constraint can be adjusted to a directed graph constraint.

[0162] ⑤ Trip Selection Constraints: Simulating the entire process of real-world urban air services (ground transfer - air flight - ground transfer) ensures that the allocation of travel demand conforms to intermodal transport logic and does not exceed the carrying capacity of the urban air transport network. In this embodiment, travel modes are divided into two mutually exclusive modes: "pure ground transportation" and "ground-air-ground intermodal transport".

[0163] a. Each traveler may only choose one mode of transport: pure ground transportation or intermodal transport.

[0164]

[0165] b. When an air segment is selected, the air segment... Candidate take-off and landing airports and routes between It is enabled:

[0166]

[0167] c. For trips using intermodal transport, candidate departure and arrival airports for both departure and destination ground connections must be established:

[0168]

[0169] d. The ground connection method should be consistent with the air travel method:

[0170]

[0171] e. Each traveler may choose a maximum of one combination of candidate departure and arrival airports:

[0172]

[0173]

[0174] in, Indicates itinerary Whether to use ground transportation From the point of origin to the candidate take-off and landing airport point , Indicates itinerary Whether to use ground transportation From candidate take-off and landing airports Depart for your destination.

[0175] ⑥ Daily maximum processing capacity constraints (physical capacity limitations) for each candidate takeoff and landing airport.

[0176]

[0177] in, Indicates the air flight segment y :journey For candidate take-off and landing airports Arrival at candidate take-off and landing airports ; Indicates the air flight segment y :journey For candidate take-off and landing airports Arrival at candidate take-off and landing airports ; Indicates the first The types are Candidate take-off and landing airports, Representation type The maximum daily processing capacity of the candidate take-off and landing airports. Specifically, the processing capacity may include indicators such as take-off and landing times and passenger throughput. The specific values ​​can be set according to the type of candidate take-off and landing airports.

[0178] S4. Model solution based on multi-objective evolutionary algorithm: The multi-objective evolutionary algorithm is used to solve the vertical take-off and landing airport site selection optimization model and output the optimal layout scheme.

[0179] Since the aforementioned vertical takeoff and landing airport site selection and urban air traffic network design model is an NP-hard problem with a huge solution space and complex hierarchical constraints, this invention utilizes a multi-objective evolutionary algorithm (in this preferred embodiment, an improved non-dominated sorting genetic algorithm, i.e., NSGA-II), to obtain a Pareto optimal solution set that balances service coverage and construction cost. The detailed program flow is as follows:

[0180] S41, Chromosome Coding

[0181] To transform physical location schemes into individual entities that the algorithm can process, an initial population is generated using an encoding method representing the facility type status. Multi-base integer encoding is used to generate the initial population. The chromosome length is equal to the number of candidate takeoff and landing airports. (Number of optimal candidate facilities), each gene locus in the chromosome The set of values ​​is {0,1,2,3}, which corresponds to the functional type of each candidate take-off and landing airport point, representing: no construction, construction of a site, construction of a base, and construction of a hub, respectively.

[0182] S42. Constraint Feasibility Repair

[0183] Since randomly generated individuals may violate constraints (such as hierarchical dependencies and administrative division restrictions), after each new individual is generated, it is checked whether the initial population satisfies the model constraints. A repair operator is immediately executed to force the constraints to be satisfied, thus performing constraint feasibility repair on the initial population to obtain a feasible population and ensure the physical feasibility of the solution. This is a key technical means in this application to improve the algorithm's convergence efficiency. It includes:

[0184] ① Hub Quantity Restoration: Traverse all administrative divisions. If the number of nodes with a gene value of 3 (hub) in a certain division exceeds a preset threshold (e.g., 1), randomly retain one and downgrade the rest to 2 (base) or 1 (site). This step ensures the spatial balance of high-level facilities and complies with urban planning constraints.

[0185] ② Hierarchical Support Repair: Traverse all nodes with a gene value of 2 (Base) and check if there is at least one node with a gene value of 3 (Hub) within its maximum reach range. If not, upgrade the node to a Hub or downgrade it to a Station according to a probabilistic strategy to ensure that the Base receives Hub support.

[0186] S43, Heuristic Decoding and Fitness Evaluation

[0187] To accelerate computation and accurately evaluate the merits of different solutions, a greedy heuristic strategy is employed to decode the feasible population, simulate the actual operational process, and calculate the fitness of each individual in the feasible population based on the objective function value. Specifically, this includes:

[0188] 1. Constructing the route network: Based on the layout of candidate take-off and landing airports corresponding to the current chromosome individual, and according to the maximum physical range limit of the electric vertical take-off and landing aircraft, generate a set of all feasible routes. ;

[0189] 2. Calculate generalized cost: for a set of trips For each trip in the process, calculate the generalized cost of "pure ground transportation" and the generalized cost of "ground-air-ground" intermodal transportation.

[0190] For travelers choosing connecting flights, the time consumed... With monetary cost It consists of three parts: ground connection ( and ), aerial flight ( and ), ground connection ( and ).

[0191]

[0192]

[0193] Each component is calculated in the following way:

[0194] Time cost of ground transfer from departure point to vertical takeoff and landing airport and monetary costs The specific value depends on the ground transportation method selected from the departure point to the takeoff point and the candidate takeoff and landing airports where the takeoff point is located:

[0195]

[0196]

[0197] in, Indicates the use of ground transportation. a From the place of departure m Entering candidate take-off and landing airports Time cost Indicates the use of ground transportation. a From the place of departure m Entering candidate take-off and landing airports The monetary cost;

[0198] Time and cost of air travel: Time cost The cost is determined by the flight path distance between vertical takeoff and landing airports, transfer time, and takeoff and landing physical maneuver time; monetary cost. It depends on the route distance and the unit flight operating cost parameters;

[0199]

[0200]

[0201] in, Indicates the transfer time. Indicates the candidate take-off and landing airport points k arrive d The time consumed Indicates the time of takeoff and landing. Indicates the candidate take-off and landing airport points k arrive d The monetary cost of flying;

[0202] The time and cost of reaching the destination from a vertical takeoff and landing airport, i.e., the time and cost of ground connection to the destination. and monetary costs The specific value depends on the vertical takeoff and landing airport point at the time of departure and the ground transportation connection method selected when leaving the candidate takeoff and landing airport point.

[0203]

[0204]

[0205] in, , These represent the use of ground transportation. From candidate take-off and landing airports Leaving and reaching the destination The time and money costs.

[0206] 3. Mode Selection and Allocation: Compare the generalized costs of the two modes. If air intermodal transport is cheaper, and there are feasible air segments (meeting the range requirements), and the remaining physical capacity of the departure and arrival airports meets the demand, then the trip is allocated to air intermodal transport, and the daily processing capacity of the corresponding candidate departure and arrival airports is deducted; otherwise, the trip remains a pure ground transportation trip.

[0207] 4. Statistical objective function value: sum the number of trips of the serviced air segments (to obtain objective one) and the cost of building each candidate take-off and landing airport point (to obtain objective two), and calculate the fitness of the individual accordingly.

[0208] S44, Genetic Evolution Operations

[0209] The system executes an evolutionary loop based on fitness, iteratively solving for the optimal addressing scheme step by step:

[0210] 1) Fast non-dominated sorting and crowding calculation: The feasible population is stratified, and solutions with Pareto fronts are retained first. Within the same stratum, solutions with large crowding distances are retained first to ensure the diversity of the solution set.

[0211] 2) Selection: The parent generation is selected using a binary tournament selection method.

[0212] 3) Crossover: The uniform crossover operator is used to allow the facility layout features of different parents to be recombined in the offspring.

[0213] 4) Mutation: Randomly reset mutation is used to reset the gene position to other values ​​in {0,1,2,3} with a certain probability, and trigger the repair operator in step S42 again.

[0214] S45, Output the optimal layout scheme

[0215] When the algorithm reaches the preset termination condition (e.g., reaching the maximum number of iterations, objective function convergence, or solution set stability), it outputs a Pareto optimal solution set containing multiple candidate solutions. Each candidate solution represents the optimal service coverage configuration under a specific cost constraint.

[0216] 1. Target Solution Determination: Based on preset preference parameters, the system determines a target layout scheme from the Pareto optimal solution set as the output optimal layout scheme. Preference parameters can be input by the user through an interactive interface or automatically set by the system according to urban planning standards. The core objective is to select the final scheme that meets the actual constraints from the multi-objective optimization results.

[0217] 2. Scheme Content Generation: Based on the target layout scheme, generate configuration data containing the following parameters:

[0218] 1) Facility spatial layout data: The specific geographic coordinates of all selected vertical take-off and landing airports;

[0219] 2) Facility type attribute data: Functional type identifier for each airport (e.g., in this embodiment, it is identified as a vertical take-off and landing hub, vertical take-off and landing base, or vertical take-off and landing station, corresponding to different construction standards and capacity parameters).

[0220] 3) Urban air traffic network topology data: the connection relationship between airport routes and route attributes (such as flight distance and estimated flight time).

[0221] 3. Application of Technological Achievements: The above configuration data will be output to external systems to guide the physical implementation or operational configuration of urban air traffic infrastructure. Specific application scenarios include:

[0222] Planning visualization: Visualize the site selection layout and route network structure corresponding to the solution on a geographic information system (GIS) or computer-aided design (CAD) map for planners to review. Figure 2 This diagram illustrates the urban air traffic network structure formed in a target city under a vertical takeoff and landing (VTOL) airport configuration with three functional types: VTOL stations, VTOL bases, and VTOL hubs. VTOL hubs are primarily located in high-demand-density areas of the city, VTOL bases are mainly located in medium-demand-density areas, and VTOL stations serve as a supplement to medium- and high-level facilities while also covering suburban needs. This demonstrates a hierarchical network layout with clear coordination and division of labor among the three types of facilities.

[0223] Construction guidance: Generates a list of facilities to be built and location drawings, which can be used directly to guide the physical construction and equipment deployment of vertical take-off and landing airports;

[0224] System configuration: Import route network structure data into the air traffic management system database for aircraft route planning, capacity scheduling, and navigation services.

[0225] Through the above steps, this application can provide a scientific planning scheme that balances coverage and cost while ensuring that the network structure meets hierarchical constraints and physical operation conditions, and realize a closed-loop design from demand data to physical facility layout.

[0226] Understandably, this application breaks through the limitations of traditional site selection models that treat facilities as homogeneous nodes by using a hierarchical network dependency constraint mechanism based on facility heterogeneity. It proposes a hierarchical network design method that considers facilities of different functional levels and establishes strict hierarchical dependency constraints, namely, facilities of lower functional levels must be connected to facilities of higher functional levels in the network topology. This is the key to ensuring the feasibility of air traffic networks at the physical operation level, such as energy and maintenance.

[0227] Meanwhile, this application establishes a two-stage planning framework that integrates spatial clustering pre-screening and multi-objective optimization. To address the problem of high computational complexity in site selection for large-scale urban areas, it proposes a two-stage solution strategy that combines candidate point screening with global optimization. In the first stage, spatial clustering is used to identify high-potential candidate locations from a massive number of discrete demand units, reducing the solution space. In the second stage, a multi-objective optimization model is constructed and solved only on the candidate set, thus solving the technical problem of low computational efficiency on a large scale in this planning scenario.

[0228] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Based on the technical essence of the present invention, any simple modifications, equivalent substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for site selection of vertical take-off and landing airports for urban air traffic, characterized in that, Specifically, the following steps are included: S1. Obtain historical ground transportation data and urban geographic information data of the target city. Based on the urban geographic information data, divide the spatial grid of the target city into several discretized units, and map the historical ground transportation data to the corresponding discretized units to obtain the demand point set. S2. Based on the clustering algorithm, perform cluster analysis on the demand point set, select several candidate take-off and landing airport points, and generate a candidate take-off and landing airport point set; S3. Construct a vertical take-off and landing airport site selection optimization model. Based on the technical service capabilities of the candidate take-off and landing airport sites, determine the types of candidate take-off and landing airport sites. Based on the set of candidate take-off and landing airport sites, the types of candidate take-off and landing airport sites, the demand point set, and historical ground traffic data, define the basic data set required for the vertical take-off and landing airport site selection optimization model, as well as the model constraints, including the hierarchical dependency constraints between different types of candidate take-off and landing airport sites. The basic dataset includes a set of candidate takeoff and landing airports. H Itinerary Collection P Candidate take-off and landing airport type set L The set of administrative divisions Z, and the model constraints include: Uniqueness constraint: Ensure that each candidate take-off and landing airport point can only be assigned one type in space; Indicate whether to include candidate take-off and landing airports i Type construction as l ; Administrative division constraints: Within each administrative division of the target city, the number of high-level node candidate take-off and landing airport sites is limited. in, Indicates administrative region z The set of candidate take-off and landing airports within the country. Indicates the first i Whether each candidate take-off and landing airport site is to be developed into a high-level node hub ; Hierarchical constraints: There are hierarchical dependencies between different types of candidate take-off and landing airports; The candidate takeoff and landing airport locations must include at least high-level nodes with the highest level of technical service capabilities, medium-level nodes with medium-level technical service capabilities, and low-level nodes with basic level technical service capabilities. The hierarchical dependency relationship is as follows: Each selected mid-level node must have a direct flight path to at least one selected high-level node, defined as: in, Indicates the first i The candidate take-off and landing airports and the first j Whether to open air routes between the candidate take-off and landing airports. Indicates the first j Whether the candidate take-off and landing airport sites are designated as high-level nodes. hub , Indicates the first i Whether each candidate take-off and landing airport site is to be developed into a medium-level node base ; Each low-level node, acting as a terminal access facility, must have a direct flight path with at least one high-level or mid-level node, defined as: in, Indicates the first i Whether each candidate take-off and landing airport site is constructed as a low-level node. ; Consistency and symmetry constraints of flight routes: Flight routes between candidate take-off and landing airports must meet the requirement of two-way traffic. Trip selection constraints: Simulate the entire process of real urban air transportation services to ensure that the system's allocation of travel demand conforms to intermodal transport logic and does not exceed the carrying capacity of the urban air transportation network; Maximum daily processing capacity constraints for each candidate takeoff and landing airport: in, Indicates the air flight segment y : journey p For candidate take-off and landing airports i Arrival at candidate take-off and landing airports d ; Indicates the air flight segment y : journey p For candidate take-off and landing airports k Arrival at candidate take-off and landing airports i ; Indicates the first i The types are l Candidate take-off and landing airports, Representation type The maximum daily processing capacity of candidate take-off and landing airports; S4. Use a multi-objective evolutionary algorithm to solve the vertical take-off and landing airport site selection optimization model and output the optimal layout scheme.

2. The method for selecting a site for a vertical take-off and landing airport for urban air traffic according to claim 1, characterized in that, The specific process of obtaining the demand point set is as follows: S11. Perform data cleaning on historical ground transportation travel data, remove data with abnormal travel routes and trajectories, and obtain a valid travel set; S12. Based on urban geographic information data, the space of the target city is gridded and divided into several discrete units. S13. Map the coordinates of the starting point and ending point of the valid trip set to the corresponding discretized units, perform spatial aggregation on the travel data in each discretized unit, obtain the weight of each discretized unit, treat each discretized unit as a demand point, and finally form a weighted demand point set.

3. The method for site selection of a vertical take-off and landing airport for urban air traffic according to claim 1, characterized in that, The specific process of obtaining the candidate take-off and landing airport point set is as follows: S21. Use a clustering algorithm to perform cluster analysis on the demand point set, and iteratively calculate the cluster evaluation index under different numbers of cluster centers within a preset range; S22. By plotting the curve of the change of cluster evaluation index, identify the abrupt change point of the curve, and determine the number of cluster centers corresponding to the abrupt change point as the optimal number of candidate facilities. S23. Select high-density demand points from the demand point set as candidate take-off and landing airport points, and select the demand points with the best number of candidate facilities from the high-density demand points and store them in the candidate take-off and landing airport point set.

4. The method for selecting a site for a vertical take-off and landing airport for urban air traffic according to claim 1, characterized in that, The construction of a vertical takeoff and landing airport site selection optimization model also includes: The objective function for constructing a vertical takeoff and landing (VTOL) airport site selection optimization model includes a first objective function for maximizing the number of air segments served and a second objective function for minimizing the facility construction costs for the system operator. The first objective function Z1 is: in, Indicates itinerary Should I select from the candidate takeoff and landing airports? To candidate take-off and landing airports The air flight segment; The second objective function Z2 is: in, The type is The construction cost of candidate take-off and landing airport sites, Indicate whether to include candidate take-off and landing airports i Type construction as l .

5. The method for selecting a site for a vertical takeoff and landing airport for urban air traffic according to claim 4, characterized in that, The basic dataset includes a set of ground transportation modes. The simulation of a real-world urban air service process includes: departure ground transfer, air flight, and destination ground transfer. Travel modes include purely ground transportation and ground-air-ground intermodal transport. Trip selection constraints include: 1) Each traveler may only choose between purely ground transportation and ground-air-ground intermodal transportation: in, Indicates itinerary Whether to use purely ground transportation; 2) Aerial flight segment Candidate take-off and landing airports k and d routes between It is enabled: 3) When selecting the ground-air-ground intermodal transport mode, both the departure and arrival airports must be operational: in, and These represent the candidate take-off and landing airports. k and d Whether it is built as a type l The takeoff and landing airports, and the candidate takeoff and landing airports. k and d They serve as departure and arrival airports respectively in ground-air-ground intermodal transport; 4) The ground connection method and the air flight method should be consistent: 5) Each traveler may only choose one combination of candidate departure and arrival airports: in, Indicates itinerary Whether to use ground transportation From the point of origin to the candidate take-off and landing airport point k , Indicates itinerary Whether to use ground transportation From candidate take-off and landing airports Depart for your destination.

6. The method for selecting a site for a vertical takeoff and landing airport for urban air traffic according to claim 5, characterized in that, The specific process of outputting the optimal layout scheme in step S4 is as follows: S41. Generate an initial population using multi-base integer encoding based on the number of candidate take-off and landing airports; S42. Check whether the initial population meets the model constraints, and perform constraint feasibility repair on the initial population. After each new individual is generated, execute the repair operator to force the model constraints to be met, including hub number repair and hierarchical constraint repair, to obtain a feasible population. S43. A greedy heuristic strategy is used to decode the feasible population, simulate the actual operation process, and calculate the fitness of each individual in the feasible population based on the objective function value. S44. Based on the fitness of each individual, execute an evolutionary loop to iteratively solve for the optimal addressing scheme; when the iteration reaches the preset termination condition, output a Pareto optimal solution set containing multiple candidate schemes; S45. Based on the preset preference parameters, select a candidate solution from the Pareto optimal solution set as the optimal layout solution and output it.

7. The method for selecting a site for a vertical takeoff and landing airport for urban air traffic according to claim 6, characterized in that, The specific process of step S43 is as follows: Each site selection scheme is treated as an individual in the feasible population. Based on the layout of candidate take-off and landing airports in the individual, and according to the maximum physical range limit of the electric vertical take-off and landing aircraft, a set of all feasible routes is generated. For each trip in the trip set, calculate the generalized cost of each trip under different travel modes, including pure ground transportation and air intermodal transportation. Compare the generalized costs of each trip under pure ground transportation and air intermodal transport modes. If the generalized cost of air intermodal transport is lower, and there are feasible air flight segments and the remaining physical capacity of the candidate take-off and landing airports meets the demand, then the trip is allocated to air intermodal transport, and the daily processing capacity of the corresponding origin and destination candidate take-off and landing airports is deducted. If the generalized cost of purely ground transportation is lower, then the trip will remain a purely ground transportation trip; The objective function value is obtained by summing the number of trips corresponding to the served air segments and the cost of building each candidate take-off and landing airport. The individual fitness is then calculated based on the objective function value.

8. The method for selecting a site for a vertical takeoff and landing airport for urban air traffic according to claim 7, characterized in that, The generalized cost of air intermodal transport includes the time cost. With monetary cost : Among them, the time cost of ground transfer from departure. and monetary costs It depends on the ground transportation method selected from the departure point to the takeoff point and the candidate takeoff and landing airports where the takeoff point is located; Time cost of air travel The monetary cost of air travel depends on the flight path distance between candidate airports, transfer time, and takeoff and landing physical maneuver time. It depends on the route distance and the unit flight operating cost parameters; Destination ground connection time cost and monetary costs It depends on the vertical takeoff and landing airport point at the time of departure and the ground transportation connection selected when leaving the candidate takeoff and landing airport point.

Images