Urban low-altitude vertical take-off and landing field site selection method fusing mobile phone signaling big data

Abstract

The invention belongs to the technical field of geographic information science space optimization, and particularly discloses a low-altitude vertical take-off and landing field site selection method fusing mobile phone signaling big data dynamic maximum coverage, which comprises the steps of fusing multi-source geographic space big data, and constructing an urban low-altitude vertical take-off and landing field candidate position set; defining a multi-time-sequence take-off and landing field maximum coverage site selection model, and determining an optimization objective function, constraint conditions and a decision-making method; a genetic algorithm is used as a basic algorithm model, and three limiting conditions of a multi-time-sequence influence factor, total quantity constraint and integer programming are added to form a multi-time-sequence integer constraint spatial evolution algorithm; based on MP-ICSEA, an MP-MCLP problem is solved, an optimal solution of service facility configuration is obtained, the effectiveness of vertical take-off and landing field site selection and vertical take-off and landing aircraft redeployment along with the change of the flow of people is verified, and therefore the reasonability of vertical take-off and landing field site selection and the efficiency of vertical take-off and landing aircraft redeployment according to the time sequence are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of spatial optimization technology in geographic information science, and more specifically, to a method for selecting urban low-altitude vertical take-off and landing sites by integrating mobile phone signaling big data. Background Technology

[0002] With accelerated urbanization and soaring population density, traditional ground transportation faces severe challenges such as heavy operational pressure, soaring carbon emissions, and insufficient carrying capacity. Urban air transportation is an effective solution to alleviate urban traffic problems, and the scientific layout of vertical take-off and landing (VTOL) sites directly determines network coverage, operating costs, and public acceptance. Under the constraints of complex urban geographical environments, the selection of urban low-altitude VTOL sites presents a challenge in balancing maximum population coverage with resource allocation. Furthermore, existing methods have the following shortcomings: First, models often focus on improving economic efficiency and internal system performance, or fail to adequately consider the fairness of service coverage within a static planning framework, making it difficult to adapt to urban population tidal effects; second, existing research rarely focuses on population coverage efficiency in VTOL site selection as an optimization objective; and third, facility layout schemes exhibit resource redundancy, failing to establish a quantitative balance mechanism between coverage efficiency and resource allocation efficiency.

[0003] Therefore, existing technologies have bottlenecks in optimizing the location of vertical take-off and landing sites, and there is an urgent need for solutions to address the synergistic optimization of maximizing time-varying population and balancing the allocation of low-altitude service resources under multiple constraints. Summary of the Invention

[0004] To address the challenge of balancing maximum coverage of time-varying populations with resource allocation in the site selection process for urban low-altitude vertical take-off and landing (LVT) fields, this invention discloses a method for LVT field site selection based on maximizing coverage through the integration of time-varying populations. The specific details are as follows:

[0005] S1. Integrate multi-source geospatial big data to construct a set of candidate locations for urban low-altitude vertical take-off and landing sites;

[0006] S2. Define the Multi-period Maximum CoveringLocation Problem for Vertiport Siting (MP-MCLP), and determine the optimization objective function, constraints, and decision-making methods;

[0007] S3. Using the genetic algorithm as the basic algorithm model, and adding three constraints—temporal constraints, total quantity constraints, and integer programming—a multi-period integer constrained spatial evolutionary algorithm (MP-ICSEA) is formed.

[0008] S4. Based on the MP-ICSEA geographical location method, solve the MP-MCLP problem of vertical take-off and landing fields to obtain the optimal solution for service facility configuration.

[0009] This invention uses GIS spatial analysis to screen potential vertical take-off and landing sites, and solves the MP-MCLP problem based on MP-ICSEA, which can obtain the optimal layout scheme of service facilities in a short time. By combining multi-source spatial data and integrating various spatial constraints, the site selection decision is optimized. By constructing the MP-MCLP model, the dynamic reconfiguration balance of population coverage and resources with time-series changes is improved, and service facility redundancy is reduced.

[0010] The beneficial effects of this invention compared to the prior art are as follows:

[0011] 1. Temporal Dimension Characteristics: The model output exhibits a "spatiotemporal imbalance" structure. Unlike traditional uniform or fixed-period resource allocation methods, this model, based on traffic flow peak patterns, concentrates resource allocation during morning and evening peak hours and strategically reduces resource allocation during off-peak hours. This setting maximizes resource utilization efficiency throughout the entire cycle, addressing sudden demands during peak hours while avoiding resource idleness during off-peak hours, thereby improving the overall system operating efficiency and economy.

[0012] 2. Spatial Layout Structure: The facilities present a hierarchical network with the six main urban districts as the core, gradually spreading outwards to secondary centers and the surrounding urban-rural transition zones. Unlike traditional models of uniform distribution or single-center layouts, this layout explicitly establishes high-density coverage around the six main urban districts, extending along secondary centers to support multi-center development, and adopting a sparse, nodal layout in the surrounding urban-rural transition zones. This structure ensures efficient operation in the core areas, optimizes the urban spatial structure through gradient diffusion, and achieves basic service coverage in the peripheral areas at a lower cost, balancing operational cost control with service space equity.

[0013] 3. Traditional integer programming methods struggle to address the challenges posed by spatiotemporal heterogeneity in solving the high-dimensional, dynamic optimization problem of "demand response - total resource constraint" in vertical take-off and landing (VTOL) field layout. Therefore, this paper improves the standard genetic algorithm by adding three constraints: temporal constraints, total resource constraints, and integer programming, forming a multi-temporal integer-constrained spatial evolutionary algorithm to enhance the optimization capability of layout schemes under complex real-world constraints. Attached Figure Description

[0014] Figure 1 A flowchart of an urban low-altitude vertical take-off and landing field location selection algorithm that integrates mobile phone signaling big data is provided for an embodiment of the present invention;

[0015] Figure 2 This invention provides a study area and data distribution map of Nanjing City ((a) represents the POI distribution in Nanjing City in 2021; (b) represents the building area distribution in Nanjing City in 2021; (c) represents the community center point distribution; (d) represents the population density distribution).

[0016] Figure 3 This invention provides a data preprocessing map ((a) represents a local magnification of candidate points; (b) represents a local magnification of candidate building surfaces).

[0017] Figure 4 This invention provides a flowchart of an integer-constrained space evolutionary algorithm.

[0018] Figure 5 This invention provides schematic diagrams illustrating the reduction of vertical take-off and landing (VTOL) facilities during different periods: (a1) represents the reduction of facilities during the morning peak compared to the daytime off-peak under experimental conditions of service radius R=5km and total facility activation count P=200; (b1) represents the reduction of facilities during the morning peak compared to the daytime off-peak under experimental conditions of R=5km and P=400; (c1) represents the reduction of facilities during the morning peak compared to the daytime off-peak under experimental conditions of R=5km and P=600; (a2) represents the increase of facilities during the evening peak compared to the daytime off-peak under experimental conditions of R=5km and P=200; (b2) represents the increase of facilities during the evening peak compared to the daytime off-peak under experimental conditions of R=5km and P=200; (c2) represents the increase in facilities during the evening peak compared to the daytime off-peak under the experimental conditions of R=5km and P=400; (a3) ​​represents the spatial distribution of tidal low-altitude vertical take-off and landing fields activated during the morning and evening peaks under the experimental conditions of R=5km and P=200; (b3) represents the spatial distribution of tidal low-altitude vertical take-off and landing fields activated during the morning and evening peaks under the experimental conditions of R=5km and P=400; (c3) represents the spatial distribution of tidal low-altitude vertical take-off and landing fields activated during the morning and evening peaks under the experimental conditions of R=5km and P=600. Detailed Implementation

[0019] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, and the terms "inner" and "outer" refer to directions toward or away from the geometric center of a specific component, respectively.

[0020] This embodiment of a method for selecting urban low-altitude vertical take-off and landing sites by integrating mobile phone signaling big data specifically includes the following steps.

[0021] S1. Integrate multi-source geospatial big data to construct a set of candidate locations for urban low-altitude vertical take-off and landing sites;

[0022] The set of candidate locations for constructing urban low-altitude vertical take-off and landing fields includes:

[0023] S1.1 Filtering of building facade geometry;

[0024] S1.2 Building surface function filtering;

[0025] S1.3 Extract candidate roofs for vertical take-off and landing sites;

[0026] S1.4 Constructing overlay network relationships.

[0027] The overall construction mainly includes three steps, as shown in the appendix. Figure 2 As shown in (a), S1.1, roof geometry screening. Roofs with an area greater than 720 square meters were selected according to the "China Civil Airports Association Group Standard" to meet the space requirements for simultaneous operation of 5 aircraft and ensure a reasonable layout of takeoff and landing buffer zones and safety isolation. The shortest side of the circumscribed rectangle of the roof was limited to no less than 12 meters and the longest side to no less than 60 meters to eliminate excessively long or irregularly shaped roofs.

[0028] S1.2, Roof Function Filtering. Based on POI data, the use of the rooftop is limited to categories suitable for constructing vertical take-off and landing sites, including large shopping malls, office buildings, hospitals, major transportation hubs such as train stations and airports, and public open spaces such as parks and squares.

[0029] S1.3 Extract candidate roofs for vertical take-off and landing sites. The center points of the eligible building roofs are clustered and generated into 1000 candidate points. They are then assigned unique numbers (ID 1 to 1000) according to their spatial distribution from west to east and from north to south. Each ID corresponds to the geographic coordinates of a candidate point and the attributes of the building roof (such as usage and area).

[0030] S1.4 Constructing Coverage Network Relationships. ① Calculate the distance between the community center point and the candidate building facades, and establish a coverage cost matrix. ② Construct a maximum coverage model for the low-altitude vertical take-off and landing field. The objective function is to maximize the coverage population, and the constraints include single-time demand allocation, facility activation priority, and a limit on the total number of facilities.

[0031] S2. Define the Multi-Period Maximum Covering Location Problem for Vertiport Siting (MP-MCLP), and determine the optimization objective function, constraints, and decision-making method; the definition of the vertical take-off and landing field maximum coverage location model includes:

[0032] S2.1 optimizes and maps the real vertical take-off and landing field location problem into a vertical take-off and landing field maximum coverage model;

[0033] S2.2 Establishes the set of service facility points and demand points for the vertical take-off and landing field, and sets an objective function to maximize coverage demand. The objective function no longer simply pursues maximizing the number of people covered, but rather maximizing the spatiotemporally weighted total service utility obtained by the covered population across all time periods. Inverse distance weights are introduced. This transforms the binary coverage relationship into a continuous service utility value. This allows spatial proximity to be quantified and incorporated into the model, enabling a characterization of coverage quality. The closer a demand point is to the facility, the higher its single-use service utility.

[0034] Represents the set of demand points ( ), Represents the set of facility points ( ), Represents a set of time periods ( ), For binary decision variables, in the time period At that time, facility point When selected, the value is 1; otherwise, it is 0. Assign a binary variable over a time period At that time, put the demand points Demand records are given to the facility. The value is 1 if it is positive and 0 otherwise. Indicates the time period Demand points The demand, For demand points to facility distance, This represents the maximum effective coverage distance of the vertical takeoff and landing (VTOL) facility point. It is to satisfy A collection of facilities This is the maximum number of times all facilities can be activated throughout the entire planning period. This is the distance attenuation coefficient; This coefficient follows the basic paradigm of the gravitational model, which is set in this paper. (That is, assuming that service attractiveness is inversely proportional to distance), the objective function of the model is defined as follows:

[0035] (1)

[0036] S2.3 Furthermore, define constraints to ensure consistency between the number of facilities and the coverage of demand points, in order to optimize the rationality of the geographic spatial layout. These constraints include:

[0037] 1) During the time period At that time, demand points The demand is recorded only once for each facility point within the specified range, i.e.:

[0038] (2)

[0039] 2) Unused facilities cannot be allocated demand, i.e.:

[0040] (3)

[0041] 3) In all time periods Within, the maximum number of times the facility can be selected, i.e.:

[0042] (4)

[0043] S3. Using the genetic algorithm as the basic algorithm model, and adding three constraints—temporal constraints, total quantity constraints, and integer programming—a multi-period integer constrained spatial evolutionary algorithm (MP-ICSEA) is formed.

[0044] In one embodiment, genetic algorithms, as a typical intelligent evolutionary algorithm, are widely used in service facility site selection problems due to their global search capability, strong adaptability to complex nonlinear problems, and natural advantages in handling multi-objective optimization. For the collaborative optimization problem of "demand response-total resource constraints" in vertical take-off and landing field layout, traditional integer programming methods struggle to cope with the high-dimensional solution space and dynamic spatiotemporal heterogeneity. Therefore, this paper selects the standard genetic algorithm as the basic algorithm model, adding three constraints: temporal constraints, total resource constraints, and integer programming, to form a multi-temporal integer-constrained spatial evolutionary algorithm that can approximate the global optimum in a finite time.

[0045] The methods include:

[0046] S3.1 Real-number coding design. Each chromosome represents a vertical takeoff and landing field layout scheme, by... It consists of 1 gene locus, and each gene locus corresponds to a candidate vertical take-off and landing field facility point;

[0047] S 3.2 Initialize Population Generation. The initial population generation employs a "random generation-integer balancing" strategy to ensure that individuals meet the total population constraint. First, the number of facilities is randomly assigned to each candidate point. ,Require ; Calculate the total deviation of facilities The system is adjusted using an integer balancing algorithm to meet the facility quantity constraints, and this process is repeated until the generated scale is [value missing]. The initial population;

[0048] S3.3 Individual Fitness Evaluation. The fitness function aims to maximize the population coverage. The fitness value directly reflects the coverage efficiency of the layout scheme; a higher value indicates a better individual. During the evaluation process, the total population coverage of each chromosome's corresponding scheme is quickly calculated using the coverage cost matrix to satisfy the objective of formula (1) and ensure algorithm efficiency.

[0049] S3.4 Selection and Crossover. The selection mechanism combines roulette wheel selection with an elite retention strategy, prioritizing the retention of individuals with high fitness. In the crossover operation, n crossover sites are randomly selected from both parents, and their corresponding gene loci are exchanged. After crossover, an integer balancing algorithm is used to correct for overall facility bias.

[0050] S3.5 Mutation Operation. To enhance population diversity, m mutation sites are randomly selected from each individual. There is a 10% probability of reducing one facility; if There is a 10% probability of adding one facility. After mutation, the integer balancing algorithm is called again to balance the total number of facilities, avoiding the failure of global constraints due to local adjustments;

[0051] S3.6, Integer Balancing Algorithm Module. The integer balancing algorithm corrects the total facility deviation through mandatory and optional parameters, adapting to different individual initialization needs. The input parameters of this algorithm are divided into mandatory and optional parameters, mainly to correspond to the two methods of individual initialization during the algorithm implementation process. The main body of the algorithm is divided into three parts: individual initialization, preliminary balancing, and fine balancing;

[0052] S3.7 Termination Conditions and Output. The algorithm terminates under one of the following two conditions: reaching the maximum number of iterations or the rate of change of the optimal fitness is less than 1%. The individual with the highest fitness is output as the optimal addressing scheme. This design balances computational efficiency and optimization accuracy, avoiding getting trapped in local optima or overcomputing.

[0053] S4. Based on the MP-ICSEA algorithm, solve the MP-MCLP to obtain the optimal solution for service facility configuration;

[0054] The process of obtaining the optimal solution for service facility configuration includes: solving MP-MCLP based on MP-ICSEA and outputting the optimal solution for service facility configuration.

[0055] The experimental data are as follows:

[0056] The experimental hardware and software environment is described as follows: (1) Intel Core i7 CPU, 32GB memory, NVIDIA GTX3060Ti GPU; (2) Windows 10 64-bit operating system; (3) ArcGIS 10.8 is used for geospatial data processing, and Python 3.9 is used to implement the integer constraint spatial evolution algorithm. This paper selects Nanjing City as the research target. Nanjing City is located in Jiangsu Province in eastern China. It is a mega-city in the Yangtze River Delta region with a resident population of over 9.4 million. Nanjing City has 11 districts under its jurisdiction, with a built-up area of ​​approximately 743 square kilometers. The population is densely distributed and the demand for transportation is strong. However, the central urban area is limited by historical blocks and high-rise buildings, and ground traffic is congested during peak hours all year round. Based on these characteristics, Nanjing City has incorporated the development of low-altitude economy and the construction of urban vertical take-off and landing field network into its urban planning in recent years. Choosing Nanjing City as the research area is of typical significance: on the one hand, Nanjing has a positive policy background and potential demand for promoting UAM; on the other hand, its high-density built-up environment brings representative challenges to the selection of vertical take-off and landing field sites. The population data is based on mobile operator signaling big data. Considering that the population distribution is more stable at night, mobile signaling data at 21:00 on April 12, 2021 was selected. Figure 2 (c) The time periods selected are the morning peak (08:00), the daytime off-peak (14:00) and the evening peak (17:00). The spatial distribution of population density shows a decreasing trend from the center to the periphery, and the temporal distribution shows a tidal fluctuation trend.

[0057] Figure 1 This is a flowchart of an algorithm for selecting urban low-altitude vertical take-off and landing sites that integrates mobile signaling big data, provided as an embodiment of the present invention.

[0058] Experiment 1: This study designed nine sets of computational examples covering three service radii (S=2, 5, 10 km) and three total activation counts of facilities (P=200, 400, 600). The MP-ICSEA algorithm was used for solving the problems. The evaluation focused on two dimensions: optimality of the solution (objective function value) and computational efficiency (CPU runtime). Specific experimental results are shown in Table 1. Regarding solution quality, the objective function value obtained by MP-ICSEA is extremely close to the optimal solution, indicating that the algorithm can obtain a near-optimal solution with minimal accuracy loss. In terms of computational efficiency, MP-ICSEA is slower than commercial solvers, but still within the same order of magnitude, demonstrating the effectiveness of MP-ICSEA.

[0059] Table 1. Performance comparison of various algorithms under different parameter configurations

[0060] Scale S Scale P MP-ICSEA target value MP-ICSEA time (s) 2 200 43913158 5.51 2 400 58926512 10.49 2 600 67732734 15.42 5 200 47863878 21.03 5 400 60878403 41.94 5 600 66097259 66.53 10 200 47913523 107.59 10 400 60324892 213.23 10 600 68887960 328.30

[0061] Experiment 2: Based on the results of the previous experiment, this study selects the greedy algorithm with the best overall performance and uses three sets of parameter combinations (S=5 km, P=200 / P=400 / P=600) to generate a vertical take-off and landing field location scheme. Figure 5 This was used to assess the layout characteristics under different resource configurations. For each set of parameters, a baseline network was constructed using facilities selected during daytime off-peak hours, and three key comparisons were further visualized: Figure 5 (a1 / b1 / c1) represents the reduction in facilities during the morning peak compared to the daytime off-peak; Figure 5 (a2 / b2 / c2) represent the increase in facilities during the evening peak compared to the daytime off-peak; and Figure 5 The spatial distribution of tidal vertical take-off and landing fields (a3 / b3 / c3) activated during morning and evening peak hours is analyzed. An intermediate-scale scenario (S=5 km, P=400) is selected as a typical case for in-depth analysis.

[0062] From a temporal perspective, the number of activated facilities and the population covered varied across different time periods. Specifically, during the morning peak hours, 139 facilities were activated, covering a population of 12.2559 million, with a coverage rate of 89.88%; during the daytime off-peak hours, the number of activated facilities decreased to 125, covering a population of 9.5923 million, with a coverage rate of 87.05%; and during the evening peak hours, 136 facilities were activated, covering a population of 11.7179 million, with a coverage rate of 88.71%. In-depth analysis of the dynamic adjustments to the facility network revealed that from the morning peak to the daytime off-peak hours, 15 facility locations were deactivated, and 1 new facility location was added. Figure 5(b1)). These deactivated facilities were not randomly distributed, but mainly concentrated on the outskirts of the main urban core and along some commuter corridors. This indicates that the model can identify areas where demand declines most significantly and manage resources by deactivating their service facilities. From the daytime off-peak to the evening peak, the facility network expanded again, activating 12 new facilities and reducing 1 facility location (b1). Figure 5 (b2) It is worth noting that this group of newly added facilities largely overlaps spatially with the facilities that are closed during the morning rush hour to off-peak hours, and functionally plays the role of tidal facilities. Figure 5 (b3): ​​These are activated during morning and evening rush hours to meet travel demand, while they are temporarily deactivated during off-peak hours to avoid resource redundancy. This time-series configuration pattern reflects the dynamic optimization characteristics of the model: that is, resources are concentrated during peak demand periods to maximize coverage, while resources are strategically reduced during periods of low demand to improve resource utilization efficiency, ultimately achieving efficient deployment of limited resources during critical periods.

[0063] From a spatial perspective, the overall pattern exhibits an agglomeration characteristic centered on the central urban area. It features the following characteristics: 1) Dense coverage of the main urban area: Facilities are highly concentrated in the main urban area (Xuanwu, Qinhuai, Jianye, Gulou, Qixia, and Yuhuatai districts), where the population density reaches 2560–17399 people / km². This area is the city's population and economic activity center, forming a deep coverage network over high-value areas to respond to their strong and continuous transportation demand. 2) Rational distribution of secondary agglomeration areas: The facility network breaks through the single-center agglomeration model, extending to secondary population agglomeration areas (Pukou and Jiangning districts, with population densities of 453–1269 people / km²). These areas, as emerging urban developments, have high population densities and strong economic vitality, effectively covering these secondary centers and reflecting a multi-center spatial development orientation. 3) Supplementation of peripheral area nodes: In the urban-rural transition zones with lower population densities (such as Lishui and Gaochun districts, with population densities of 453–556 people / km²), the facility layout exhibits a low-density, node-like sparse distribution. This layout effectively fills service gaps while controlling resources, balances resource input and coverage benefits, and helps to bridge the service gap between urban and rural areas.

[0064] In summary, this invention discloses a method for low-altitude vertical take-off and landing (VTOL) field site selection based on maximizing coverage by time-varying population. Addressing complex geographical constraints, this invention optimizes site selection decisions by combining multi-source spatial data and integrating various spatial constraints. For the issue of time-varying population coverage efficiency, it constructs a dynamic maximum coverage model for VTOL fields that incorporates distance attenuation effects, and solves the problem using the integer-constrained spatial evolution algorithm. This improves the balance between population coverage and resource allocation, reduces service facility redundancy, and its effectiveness and feasibility are experimentally verified. The VTOL field maximum coverage model can solve the problem of maximizing coverage demand points for facility locations. This invention is applicable to the study of solving large-scale VTOL field spatial site selection optimization problems.

[0065] The technical means disclosed in this invention are not limited to those disclosed in the above embodiments, but also include technical solutions composed of any combination of the above technical features.

Claims

1. A method for selecting urban low-altitude vertical take-off and landing sites by integrating mobile phone signaling big data, characterized in that: Specifically, the steps include the following: S1. Integrate multi-source geospatial big data to construct a set of candidate locations for urban low-altitude vertical take-off and landing sites; S2. Define the maximum coverage location model MP-MCLP for multi-time-series takeoff and landing fields, and determine the optimization objective function, constraints, and decision-making methods; S3. Using the genetic algorithm as the basic algorithm model, and adding three constraints—temporal constraints, total quantity constraints, and integer programming—a multi-temporal integer-constrained spatial evolutionary algorithm, MP-ICSEA, is formed. S4. Based on the MP-ICSEA geographical location method, solve the MP-MCLP problem of vertical take-off and landing fields to obtain the optimal solution for service facility configuration.

2. The method for selecting urban low-altitude vertical take-off and landing sites by integrating mobile phone signaling big data as described in claim 1, characterized in that: S1 specifically includes the following steps: S1.1 Roof geometry selection; S1.2 Building surface function filtering; S1.3 Extract candidate roofs for vertical take-off and landing sites; S1.4 Constructing overlay network relationships.

3. The method for selecting urban low-altitude vertical take-off and landing sites by integrating mobile phone signaling big data as described in claim 1, characterized in that: The maximum coverage location model for multi-time-series takeoff and landing fields (MP-MCLP) is defined, and its specific contents include the following: S2.1 optimizes and maps the real vertical take-off and landing site selection problem into an MP-MCLP problem; considering the significant tidal effect of urban population at different times, in order to improve the operational efficiency after the vertical take-off and landing site selection, it is necessary to dynamically redeploy the take-off and landing site selection and the vertical take-off and landing aircraft configured at the take-off and landing site, and design an MP-MCLP model. S2.2 Establish the set of service facility points and demand points for the vertical take-off and landing field, and set the objective function to maximize coverage demand; introduce inverse distance weights. The binary coverage relationship is transformed into a continuous service utility value; the objective function of the model is defined as follows: ；（1） S2.3 In addition, define constraints to ensure consistency between the number of facilities and the coverage of demand points, so as to optimize the rationality of the geographic spatial layout; the constraints include: 1) During the time period At that time, demand points The demand is recorded only once for each facility point within the specified range, i.e.: （2） 2) Unused facilities cannot be allocated demand, i.e.: （3） 3) In all time periods Within, the maximum number of times the facility can be selected, i.e.: （4） The symbols are defined as follows: Represents the set of demand points ( ), Represents the set of facility points ( ), Represents a set of time periods ( ), For binary decision variables, in the time period At that time, facility point When selected, the value is 1; otherwise, it is 0. Assign a binary variable over a time period At that time, put the demand points Demand records are given to the facility. The value is 1 if it is positive and 0 otherwise. Indicates the time period Demand points The demand, For demand points to facility distance, This represents the maximum effective coverage distance of the vertical takeoff and landing (VTOL) facility point. It is to satisfy A collection of facilities This is the maximum number of times all facilities can be activated throughout the entire planning period. This is the distance attenuation coefficient.

4. The method for selecting urban low-altitude vertical take-off and landing sites by integrating mobile phone signaling big data as described in claim 1, characterized in that: The Multi-Time Integer Constrained Space Evolutionary Algorithm (MP-ICSEA) specifically includes the following steps: S3.1 Real-number coding design; each chromosome represents a vertical takeoff and landing field layout scheme, by It consists of 1 gene locus, and each gene locus corresponds to a candidate vertical take-off and landing field facility point; S3.2 Initialize population generation; the initial population generation adopts a strategy of first generating randomly and then adjusting with integers to ensure that individuals meet the total population constraint; firstly, randomly allocate the number of facilities to each candidate point. ,Require ; Calculate the total deviation of facilities The system is adjusted using an integer balancing algorithm to meet the facility quantity constraints, and this process is repeated until the generated scale is [value missing]. The initial population; S3.3 Individual fitness assessment; The fitness function aims to maximize the coverage population; During the assessment process, the total population coverage of each chromosome corresponding to the scheme is quickly calculated through the coverage cost matrix to satisfy the objective of formula (1) and ensure the efficiency of the algorithm; S3.4 Selection and Crossover: The selection mechanism uses a combination of roulette wheel selection and elite retention strategy, prioritizing the retention of individuals with high fitness; in the crossover operation, n crossover sites of the parent individuals are randomly selected, and the corresponding gene locus information is exchanged; after crossover, the total facility deviation is corrected by an integer balancing algorithm; S3.5 Mutation Operation; To enhance population diversity, m mutation sites are randomly selected from each individual. There is a 10% probability of reducing one facility; if With a 10% probability, one facility is added; after mutation, the integer balancing algorithm is called again to balance the total number of facilities, so as to avoid the failure of global constraints due to local adjustments. S3.6 Integer Balancing Algorithm Module; The input parameters of the integer balancing algorithm module are divided into required parameters and optional parameters, corresponding to the two methods of individual initialization in the algorithm implementation process; The main body of the algorithm is divided into three parts: individual initialization, preliminary balancing, and fine balancing; S3.7 Termination Conditions and Output; The algorithm iteration terminates under one of the following two conditions: reaching the maximum number of iterations, or the rate of change of the optimal fitness is less than 1%; the final output is the individual with the highest fitness as the optimal addressing scheme.

5. The method for selecting urban low-altitude vertical take-off and landing sites by integrating mobile phone signaling big data as described in claim 1, characterized in that: The optimal solution for service facility configuration includes, in particular, [the following]. S4.1 performs parameter tuning on the MP-MCLP model, including setting different service radii for vertical take-off and landing fields and different numbers of vertical take-off and landing fields; S4.2 Spatial visualization of the data to evaluate the effectiveness of the site selection method.

Images