An integrated layout optimization method, device and medium for low-altitude traffic

Abstract

The application discloses an integrated layout optimization method for low-altitude traffic, equipment and medium, relates to the technical field of data processing method based on management purposes, and comprises the following steps: attribute labeling is performed on candidate positions in a target area, and demand distribution data of low-altitude traffic and technical parameters of various facilities are collected; a mixed integer optimization model is established based on the candidate positions, the demand distribution data and the technical parameters, and a global coverage constraint, a capacity constraint and a co-site optimization mechanism are set; a comprehensive optimization objective function is set, a cost discount weight is configured for the co-site deployment condition in cost calculation, a multi-objective weight configuration mechanism and a constraint violation penalty mechanism are established; the mixed integer optimization model is iteratively solved for multiple rounds to output an optimal facility layout scheme; the optimal facility layout scheme is comprehensively evaluated to generate a facility layout list and compared with a traditional planning scheme, and a visual layout diagram and a technical report are output.

Description

Technical Field

[0001] This application relates to the field of data processing methods for management purposes, and in particular to an integrated layout optimization method, device and medium for low-altitude transportation. Background Technology

[0002] With the rapid development of the urban low-altitude economy, emerging business models such as drone logistics and manned air transportation have created an urgent need for a safe and efficient ground support system. To ensure the large-scale operation of low-altitude airspace, it is necessary to rationally deploy take-off and landing stations on the ground, as well as four key support facilities: communication, navigation, surveillance, and meteorology, which together constitute the physical foundation of the low-altitude transportation network.

[0003] Currently, in the field of low-altitude transportation infrastructure planning, the common practice is to select sites for take-off and landing stations, communication base stations, navigation facilities, surveillance equipment, and meteorological stations separately based on their respective technical standards and coverage requirements. This fragmented planning approach results in a lack of spatial coordination among the facility networks, easily leading to coverage blind spots or unnecessary overlapping coverage, making it difficult to ensure that aircraft receive continuous and seamless service support along all pre-planned routes. Furthermore, independent planning makes it difficult to optimize opportunities for co-location of facilities, failing to fully utilize the site's resources such as space, power, and transmission, resulting in persistently high overall construction and operation costs. Summary of the Invention

[0004] This application provides an integrated layout optimization method, device, and medium for low-altitude transportation to solve the above-mentioned technical problems.

[0005] On the one hand, embodiments of this application provide an integrated layout optimization method for low-altitude transportation, including:

[0006] Acquire candidate locations within the target area for facility deployment, label each candidate location with attributes, and collect low-altitude traffic demand distribution data and technical parameters of various facilities;

[0007] Based on the candidate locations, the demand distribution data, and the technical parameters, a hybrid integer optimization model for joint site selection of multiple facilities is established, and a global coverage constraint and a capacity constraint, as well as a co-location optimization mechanism, are set in the hybrid integer optimization model.

[0008] A comprehensive optimization objective function is set for the mixed integer optimization model. In the cost calculation of the comprehensive optimization objective function, a cost preference weight is configured for the co-location deployment case, and a multi-objective weight configuration mechanism and a constraint violation penalty mechanism are established.

[0009] The mixed-integer optimization model is solved through multiple rounds of iterations to output the optimal facility layout scheme;

[0010] The optimal facility layout scheme is comprehensively evaluated to generate a facility layout list. The facility layout list is then compared and analyzed with traditional planning schemes, and a visual layout diagram and technical report are output.

[0011] In one implementation of this application, candidate locations for facility deployment within a target area are obtained, each candidate location is labeled with attributes, and low-altitude traffic demand distribution data and technical parameters of various facilities are collected, specifically including:

[0012] A systematic scan of the target area is performed to identify all candidate locations within the target area for the construction of low-altitude transportation infrastructure, resulting in a set of candidate locations;

[0013] Perform attribute labeling operations on each candidate location in the candidate location set; the attribute labeling operations include indicating the types of facilities that can be constructed at the candidate location, recording the site physical attributes of the candidate location, investigating the infrastructure conditions of the candidate location, and identifying regulatory and technical constraints;

[0014] Collect urban low-altitude traffic demand forecast data, analyze air traffic flow characteristics based on the demand forecast data, obtain distribution information of demand hotspot areas, and obtain a set of demand points.

[0015] A demand gridding model is established, which discretizes the continuous three-dimensional low-altitude airspace into a regular grid and assigns a corresponding demand weight to each three-dimensional grid cell in the regular grid to predict future demand growth trends. The demand gridding model is divided into square or hexagonal grids in the horizontal direction and into several height layers in the vertical direction according to the height stratification standard of low-altitude airspace.

[0016] Technical parameters of communication base stations, navigation stations, monitoring equipment, meteorological stations, and take-off and landing stations are collected, standardized, and cost-benefit analyses are performed to obtain the construction, operation, and maintenance costs of various facilities.

[0017] In one implementation of this application, a systematic scan of the target area is performed to identify all candidate locations within the target area for constructing low-altitude transportation infrastructure, resulting in a set of candidate locations, specifically including:

[0018] By integrating high-resolution satellite imagery, urban 3D models, geographic information system data, and urban planning drawings, a systematic scan of the target area is conducted to identify all potential candidate locations within the target area for the construction of low-altitude transportation infrastructure, thus obtaining a set of potential candidate locations.

[0019] The potential candidate locations in the potential candidate location set are verified on-site to determine whether the location information of the potential candidate locations is correct and to obtain the actual situation on site;

[0020] Based on preset regulatory requirements, technical constraints, and economic feasibility criteria, the set of potential candidate locations is screened to exclude infeasible locations, and the final set of candidate locations is obtained.

[0021] In one implementation of this application, a mixed-integer optimization model for multi-facility joint site selection is established based on the candidate locations, the demand distribution data, and the technical parameters. The mixed-integer optimization model includes setting global coverage constraints and capacity constraints, as well as a co-location optimization mechanism, specifically including:

[0022] Binary site selection decision variables are defined for each candidate location and each combination of facility types to indicate whether the candidate location should be used to construct the corresponding type of facility. Auxiliary binary variables are introduced to indicate whether the candidate location is selected, whether a specific facility co-location combination is adopted, and a hierarchical model architecture is constructed. The hierarchical model architecture includes a coverage layer, a capacity layer, and a co-location layer.

[0023] For each demand point and each facility type, a coverage relationship judgment logic is established. By calculating the distance between the candidate location and the demand point and comparing it with the service radius of the corresponding facility, a full-domain coverage constraint is constructed to ensure that each demand point is covered by at least one facility of the current type that meets the conditions. Redundant coverage constraints are also established for demand points in key areas.

[0024] For each candidate location and each facility type, a service demand calculation logic is established. By summarizing the demand weights of all demand points falling within the facility's service radius for the current type of facility, a capacity constraint is constructed to ensure that the service demand undertaken by the facility does not exceed the preset rated capacity. A load balancing constraint is also introduced to limit the load rate differences between facilities.

[0025] Construct a facility compatibility matrix, and based on the facility compatibility matrix, establish mutual exclusion constraints for each pair of non-co-located facility types at each candidate location, and establish combination selection variables and corresponding association constraints for facility combinations that are allowed to co-locate.

[0026] In one implementation of this application, a comprehensive optimization objective function is set for the mixed-integer optimization model. Cost preference weights are configured for co-location deployments in the cost calculation of the comprehensive optimization objective function. A multi-objective weight configuration mechanism and a constraint violation penalty mechanism are established, specifically including:

[0027] Based on the cost parameters in the technical parameters, a full life cycle cost calculation model is constructed for each facility at each candidate location, and a cost saving coefficient is introduced into the co-location combination of each facility at each candidate location in the cost calculation model;

[0028] The total cost of the facility co-location combination is calculated by multiplying the sum of the costs of each facility within the co-location combination by the cost-saving factor, thereby constructing a total cost minimization function; the total cost minimization function aims to minimize the sum of the costs of the facility co-location combination.

[0029] Construct a function to maximize coverage redundancy to calculate the weighted sum of the number of times each demand point is covered by similar facilities beyond the basic requirements, and construct a function to minimize load balancing to calculate the sum of the differences between the maximum and minimum load rates among similar facilities.

[0030] Assign corresponding weight coefficients to the total cost minimization function, the coverage redundancy maximization function, and the load balancing minimization function, and combine them into a single comprehensive objective function;

[0031] The degree of violation of the global coverage constraint and the capacity constraint is multiplied by a penalty coefficient to add a constraint violation penalty term to the comprehensive objective function.

[0032] In one implementation of this application, the mixed-integer optimization model is solved iteratively in multiple rounds to output the optimal facility layout scheme, specifically including:

[0033] In each iteration, the facility that can cover the most uncovered demand points is added to the facility layout scheme to obtain an initial feasible solution that satisfies the full coverage constraint.

[0034] The initial feasible solution is subjected to capacity verification in order to add similar facilities to overloaded nodes for diversion, supplement facilities to areas with insufficient coverage, and improve the initial feasible solution.

[0035] For the improved feasible solution, a chromosome encoding scheme is designed to represent the multi-facility combination decision for each candidate location. Population evolution is carried out by initializing the population, designing a fitness function, and performing genetic operations to output the optimal facility layout scheme.

[0036] In one implementation of this application, facilities that can cover the most uncovered demand points are added to the facility layout scheme in each iteration to obtain an initial feasible solution that satisfies the full coverage constraint, specifically including:

[0037] Before each iteration begins, based on the selected facility set, calculate the covered and uncovered states of all demand points and generate an uncovered demand point set.

[0038] For each candidate facility that was not selected in the facility layout scheme, the coverage benefit value of the candidate facility is calculated, and the coverage benefit values ​​of all candidate facilities are compared. The candidate facility with the highest coverage benefit value is added to the facility layout scheme. The coverage benefit value is obtained by the ratio of the number of uncovered demand points that the candidate facility can cover to the deployment cost.

[0039] The demand points covered by the newly added facility are removed from the set of uncovered demand points. The process of calculating the coverage benefit value, selecting the facility with the highest benefit, and updating the uncovered set is repeated until the set of uncovered demand points is empty. An initial feasible solution that satisfies the global coverage constraint is then output.

[0040] In one implementation of this application, for the improved feasible solution, a chromosome encoding scheme is designed to represent the multi-facility combination decision for each candidate location, and population evolution is performed by initializing the population, designing a fitness function, and executing genetic operations to output the optimal facility layout scheme, specifically including:

[0041] The multi-facility combination decision for each candidate location is encoded into a gene string to form a chromosome, and the population is initialized; the population contains random individuals and excellent individuals corresponding to the improved feasible solution;

[0042] In each generation of evolution, the comprehensive objective function value and constraint violation penalty value of the facility layout scheme corresponding to each individual chromosome in the population are calculated, and the fitness of each individual chromosome in the population is calculated based on the comprehensive objective function value and the constraint violation penalty value.

[0043] Based on the fitness of individual chromosomes, parent individuals are selected from the current population. A single-point crossover operation is performed on the selected parent individuals to generate offspring individuals. Random mutation operations are then performed on some gene loci in the offspring individuals to change the facility combination encoding.

[0044] Individuals in the population are merged with their offspring to form a new generation of population, and the evolutionary process is repeated until the optimal solution remains stable across multiple generations.

[0045] On the other hand, embodiments of this application also provide an integrated layout optimization device for low-altitude traffic, the device comprising:

[0046] At least one processor;

[0047] And, a memory communicatively connected to the at least one processor;

[0048] The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform an integrated layout optimization method for low-altitude traffic as described above.

[0049] On the other hand, this application embodiment also provides a non-volatile computer storage medium storing computer-executable instructions, which, when executed, implement the integrated layout optimization method for low-altitude traffic as described above.

[0050] This application provides an integrated layout optimization method, device, and medium for low-altitude traffic, which has at least the following beneficial effects:

[0051] By constructing a mixed-integer optimization model integrating site selection decisions for five types of facilities—take-off and landing stations, communication, navigation, surveillance, and meteorology—systematic and coordinated optimization of the spatial layout of multiple facilities was achieved. Multiple constraints such as coverage, capacity, and co-location, along with multi-dimensional objectives such as cost and service quality, were uniformly incorporated into the mathematical model for solution. This model can automatically generate layout schemes from a globally optimal perspective, solving problems such as coverage blind spots, resource overlap, or capacity imbalance caused by fragmented planning. By introducing facility co-location constraints and optimization mechanisms, and configuring cost-preferential weights for co-location deployments in the objective function, the model can proactively identify and utilize the compatibility between different facilities, reducing redundant investment in basic resources such as land occupation, power access, and transmission links, and reducing the total construction and operation costs. The phased intelligent optimization solution strategy combines the speed of ensemble coverage heuristic algorithms, the adjustment capability of local search, and the global optimization capability of genetic algorithms. This ensures that when dealing with large-scale, highly complex urban area planning problems, feasible solutions can be obtained within a reasonable time, and high-quality, near-globally optimal layout schemes can be obtained through deep search. The generation of facility layout lists and visualization results makes the optimization results more intuitive. Attached Figure Description

[0052] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0053] Figure 1 A flowchart illustrating an integrated layout optimization method for low-altitude transportation provided in an embodiment of this application;

[0054] Figure 2 This is a schematic diagram of the internal structure of an integrated layout optimization device for low-altitude traffic, provided as an embodiment of this application. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0056] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0057] Figure 1 This is a flowchart illustrating an integrated layout optimization method for low-altitude transportation provided in an embodiment of this application.

[0058] The analysis method involved in the embodiments of this application can be implemented by a terminal device or a server, and this application does not impose any special limitations on it. For ease of understanding and description, the following embodiments are all described in detail using a server as an example.

[0059] It should be noted that the server can be a single device or a system composed of multiple devices, i.e., a distributed server. This application does not make any specific limitations on this.

[0060] like Figure 1 As shown in the embodiment of this application, an integrated layout optimization method for low-altitude traffic includes:

[0061] Step 101: Obtain candidate locations for facility deployment within the target area, label each candidate location with attributes, and collect low-altitude traffic demand distribution data and technical parameters of various facilities.

[0062] Through systematic research of target cities or regions, candidate locations for constructing take-off and landing stations, as well as installing communication base stations, navigation stations, surveillance equipment, and meteorological stations, are identified and screened. It should be noted that candidate locations include rooftops of high-rise buildings, existing communication towers, vacant land at transportation hubs, and land reserved by relevant departments. Next, each candidate location is labeled with attributes. It should be noted that labeled attributes include key parameters such as the type of facility that can be constructed, site area, altitude, power supply conditions, and transportation accessibility. Then, low-altitude traffic demand distribution data is collected. It should be noted that demand distribution data includes expected air traffic flow density, major airway corridors, and demand hotspots. Furthermore, the technical parameters of various facilities are obtained. It should be noted that technical parameters include the coverage radius and capacity of communication base stations, the service range of navigation stations, the detection radius and target capacity of surveillance equipment, the monitoring range of meteorological stations, and the service capacity of take-off and landing stations.

[0063] In this embodiment, firstly, high-resolution satellite imagery or aerial photography data is used to quickly identify sites with large flat areas or specific structural features, such as large building rooftops or open spaces, from a macroscopic perspective. Secondly, urban 3D model data is accessed to further evaluate the vertical spatial structure, height, and relative positional relationship with the surrounding environment of potential candidate locations, in order to determine whether these locations are suitable for surveillance or communication equipment requiring good visibility. Thirdly, layer data from a Geographic Information System (GIS), such as land use planning maps, building ownership information, and topographic elevation maps, is combined to understand site attributes from a policy and physical geography perspective. Finally, urban planning maps and infrastructure special plans are referenced to identify areas reserved for future facilities or key future development areas. Then, information from different sources is overlaid, compared, and analyzed under the same geographic coordinate system to systematically scan the entire target area and preliminarily identify a set of potential candidate locations. It should be noted that potential candidate locations are used to indicate that these locations appear feasible in terms of spatial physical morphology, but their actual feasibility has not yet been verified and screened in the field.

[0064] It is understandable that any remote sensing data may contain errors, be outdated, or fail to reflect details. Therefore, on-site verification of potential candidate locations is necessary. The purpose of verification includes at least two aspects: confirmation and exploration. Confirmation refers to verifying the accuracy of information such as location, boundaries, and dimensions marked on satellite imagery or drawings. Exploration refers to obtaining actual on-site conditions that cannot be obtained from remote data, such as: the actual load-bearing structure of building roofs and equipment installation conditions, the current ownership of open spaces and the presence of temporary structures, the current status of roads leading to the site, and whether there are any major construction projects planned in the vicinity that may create new obstructions. This can filter out impossible locations caused by information distortion.

[0065] Finally, explicit rejection criteria are applied to filter potential candidate locations after on-site verification, forming the final candidate location set. Pre-set criteria typically fall into several categories: regulatory requirements, technical constraints, and economic feasibility. Regulatory requirements include, for example, that the site must be located within areas where the construction of such facilities is permitted by law, and must meet safety distance requirements from airports, military restricted areas, and ecological protection zones. Technical constraints include, for example, that the geological conditions of the site must be stable enough to meet the load-bearing requirements of the facility foundation, or, for facilities requiring reliable power supply, that the site must be within a range where a reliable mains power supply can be connected. Economic feasibility includes, for example, that sites whose estimated construction or leasing costs far exceed the project budget may be excluded even if technically feasible. Through a standardized screening process, a set of candidate locations with high feasibility in terms of regulations, technology, and economics can be obtained, improving the efficiency of subsequent optimization.

[0066] In this embodiment, through on-site surveys, satellite image analysis, and urban planning data review, all potential candidate locations within the target area for the construction of low-altitude transportation infrastructure are systematically identified. Emphasis is placed on locations with suitable construction conditions, such as rooftops of high-rise buildings, existing communication towers, radio and television towers, vacant land around transportation hubs, industrial parks, land reserved by relevant departments, and parks and green spaces. Then, the candidate locations are initially screened, excluding obviously infeasible locations, such as prohibited construction areas, areas with poor geological conditions, and areas with complex property rights. A candidate location database is established, assigning a unique identifier to each candidate point and recording basic information such as geographical coordinates and surrounding environment.

[0067] After obtaining a reliable set of candidate locations, attribute labeling is required to characterize the features of each location in a digital and structured manner. Specifically, based on the on-site verification results and general engineering principles, a professional judgment is made to indicate the type of facility that can be constructed at each location. For example, the rooftop of a reinforced concrete commercial building may be suitable for deploying lightweight communication antennas, small weather sensors, and visual surveillance cameras, but may not be suitable for deploying large, sophisticated mechanically scanned radars due to structural vibration or load-bearing limitations. Conversely, a large, reserved site by relevant authorities may be compatible with all types of facilities, including integrated take-off and landing station systems that require a large footprint and independent equipment rooms.

[0068] Record the site's physical attributes, including available area, load-bearing capacity, altitude, terrain slope, and surrounding obstacles. It's important to note that these physical attributes directly determine the type of equipment that can be installed, whether additional reinforcement work is needed, and whether the actual effective service range of the deployed equipment will be affected by obstructions.

[0069] The investigation of infrastructure conditions primarily assesses the external support conditions upon which construction and operation depend, including power network access, fiber optic cable access, road accessibility, and ease of maintenance. For example, it involves investigating the location, available capacity, and reliability of power connection points; assessing the accessibility and bandwidth of fiber optic or microwave transmission links; and examining road conditions to ensure accessibility for construction and subsequent maintenance vehicles. These conditions directly impact the project's unit construction costs and long-term operating costs.

[0070] Regulatory and technical constraints are identified, such as distance from airport airspace, separation requirements from residential areas, and electromagnetic compatibility limitations. For example, a location may have good physical conditions but be situated within an urban landscape protection zone, imposing specific restrictions on the appearance and height of facilities. Alternatively, a location may be too close to other existing wireless equipment, creating explicit electromagnetic compatibility constraints that limit the frequency bands and power of deployable communication equipment. These attributes are systematically entered into a database, forming a structured candidate location attribute dataset. Each field in this dataset can potentially serve as an input parameter for constraints or cost coefficients in subsequent optimization models. Furthermore, a candidate location suitability scoring mechanism is established to provide a reference for subsequent optimization.

[0071] Urban low-altitude transportation demand forecast data is collected through multiple channels. It should be noted that demand forecast data indicates where and how much service is needed, including takeoff and landing demand density in different areas, distribution of major air route corridors, freight logistics demand hotspots, and passenger travel demand distribution. Examples include macro-level forecasts of future urban air mobility (UAM) networks by urban planning departments, commercial analyses of drone delivery hotspots by logistics companies, and simulation model outputs from transportation research institutions based on population flow and activity patterns. Air traffic flow characteristics are analyzed based on demand forecast data. It should be noted that air traffic flow characteristics include peak-hour distribution, seasonal variation patterns, and the impact of special events. For example, the spatiotemporal patterns of commuter corridors during weekday evening peak hours, logistics hotspots in commercial areas at midday, and potential air bus routes connecting airports, high-speed rail stations, and city centers are identified.

[0072] To transform the continuous and complex demand distribution into a discrete form suitable for mathematical model calculation, a three-dimensional demand grid model needs to be established. Since low-altitude traffic is a typical three-dimensional operating scenario, with aircraft performing tasks at different altitudes, the airspace needs to be discretized in a three-dimensional manner.

[0073] Specifically, horizontally, the entire target geographical area is covered with a regular square or hexagonal grid. The grid side length is determined according to the facility's service accuracy requirements, typically ranging from 100 meters to 500 meters. Vertically, based on the altitude stratification standards stipulated in low-altitude airspace management regulations and the characteristics of aircraft operation, the airspace is divided into several altitude layers, such as 0-120 meters for the ultra-low altitude layer, 120-300 meters for the low altitude layer, and 300-1000 meters for the mid-low altitude layer. The intersection of each horizontal grid with each altitude layer forms a three-dimensional grid unit.

[0074] Based on demand forecasting and traffic flow characteristic analysis, each 3D grid cell is assigned a corresponding demand weight. The demand weight for the same horizontal location may differ at different altitude levels. For example, logistics drones mainly operate below 120 meters, so their demand weight is higher in the ultra-low altitude layer; while manned aircraft may mainly cruise at around 300 meters, resulting in a higher weight at the corresponding altitude level. Grids covering the central business district may be assigned a high weight, representing a strong demand for communication, navigation, and surveillance services in that area; while grids covering large bodies of water or parks may have a lower weight. This weight value comprehensively reflects expected flight frequency, safety level requirements, or business priorities.

[0075] This three-dimensional gridding process generates a structured set of demand points, where the geometric center of each grid cell is considered a demand point with a specific demand weight. The coordinates of these demand points include longitude, latitude, and altitude. This allows for the prediction of future demand growth trends, providing a basis for forward-looking planning of infrastructure layout, and taking into account the impact of factors such as new area construction and industrial restructuring on demand distribution in conjunction with urban development planning.

[0076] Then, collect detailed technical parameters such as coverage radius, service capacity, transmission power, and antenna gain of communication base stations; service range, positioning accuracy, and update frequency of navigation stations; detection radius, target capacity, resolution, and update rate of monitoring equipment; monitoring range, measurement accuracy, and data update frequency of meteorological stations; and service capacity, land area, and design standards of take-off and landing stations. Establish a unified parameter database to ensure the consistency and comparability of parameters for each facility. Also, consider technological development trends and select advanced and applicable equipment parameters as the basis for planning.

[0077] Because equipment may come from different manufacturers and follow different standards, the raw parameters collected directly may have inconsistent units and test conditions. Therefore, technical parameters are standardized. For example, all distance units are standardized to meters, all power units are standardized to watts, and performance parameters are normalized and corrected according to recognized standard test environments to ensure fair comparability between different options.

[0078] Furthermore, cost-benefit analyses of these facilities are necessary, including not only collecting equipment procurement costs but also systematically estimating the costs throughout the facility's lifecycle. This includes site preparation and civil engineering costs, equipment installation and commissioning costs, operating costs such as electricity and network leasing required for continuous operation, and regular maintenance and upkeep costs. Finally, these economic and technical parameters form the core input data for the objective function and some constraints in the subsequent optimization model. The objective function is primarily a cost function, and the constraints include the trade-off between capacity and cost.

[0079] Step 102: Based on candidate locations, demand distribution data, and technical parameters, establish a mixed integer optimization model for joint site selection of multiple facilities, and set full coverage constraints and capacity constraints in the mixed integer optimization model, as well as referencing the co-location optimization mechanism.

[0080] A mixed-integer optimization model is established to integrate the site selection decisions for takeoff and landing stations, communication base stations, navigation stations, surveillance equipment, and meteorological stations into a unified mathematical model. A full-coverage constraint is set to ensure that all predetermined flight airspace locations within the planning area are within the service range of communication, navigation, surveillance, and meteorology, achieving blind-spot-free coverage. Capacity constraints are established to ensure that the service load of each selected facility does not exceed its rated capacity, including the capacity of the airfield's sortie rate, the capacity of the number of connections of the communication base stations, and the capacity of the number of targets tracked by the surveillance equipment. A facility co-location constraint and optimization mechanism are introduced, allowing the co-location of multiple compatible facilities at the same candidate location, providing cost advantages for co-location, while prohibiting the deployment of incompatible facilities at the same location. This transforms the complex multi-facility layout problem into a computable mathematical optimization model.

[0081] In this embodiment, binary decision variables are defined for each candidate location and each facility type, and auxiliary variables are introduced. Indicates candidate position Whether to be selected to build any facilities, and establish a set of facility types. Candidate position set and the set of demand points It establishes a basic set of data and constructs a layered model architecture, organically combining the overlay layer, capacity layer, and co-location layer to form a complete mathematical model framework.

[0082] Specifically, when defining decision variables and model structure, a complete system of decision variables is established, including primary decision variables (facility location), auxiliary decision variables (location usage, co-location combination), and continuous variables (service allocation, load level), ensuring the integrity and solvability of the model. Model constraints are stratified according to their hardness and importance; hard constraints (such as coverage and capacity) must be strictly satisfied, while soft constraints (such as equilibrium and optimization) can be reflected in the objective function, improving the model's flexibility. Through reasonable variable design and constraint simplification, the model's size and complexity are controlled, ensuring high-quality solutions are obtained within an acceptable computational time.

[0083] For each demand point To ensure coverage by every necessary facility type, a full coverage constraint is established, expressed by the following formula:

[0084]

[0085] Among them, the candidate location set Set of facility types Set of demand points , Indicates the candidate position Should the first Such facilities Indicates the take-off and landing station. Indicates a communication base station. Indicates a navigation station. Indicates monitoring equipment. Indicates a weather station. Indicates candidate position To the point of demand distance, Indicates the first The service radius of such facilities.

[0086] It should be noted that the candidate positions Distance to demand point k The calculation needs to consider both horizontal and vertical distances. Let the candidate positions be... The coordinates are Demand points The coordinates are The formula for calculating three-dimensional distance is:

[0087]

[0088] The three-dimensional characteristics of the service range differ for different types of facilities. Communication base stations exhibit an inverted cone-shaped coverage pattern, requiring consideration of the antenna downtilt angle; surveillance radar's detection range is limited by elevation angle, with the effective vertical detection angle typically between -5° and +45°; navigation stations have relatively uniform signal coverage but have minimum elevation angle requirements. .

[0089] when When the elevation angle is too small (i.e. too close to the horizon), the signal is more easily blocked by terrain / buildings and affected by multipath and near-ground environment, making it difficult to guarantee the availability and accuracy of positioning / navigation. Therefore, an elevation angle threshold / elevation angle mask is set.

[0090] Assumption Station antenna height drone altitude Then height difference .Depend on The maximum permissible horizontal distance can be obtained. .

[0091] Therefore, even with a large service radius, as long as the drone is more than approximately 567m horizontally away from the station, its elevation angle will be lower than [the required angle]. In the model, this should be considered as not meeting the navigation coverage condition.

[0092] In practical calculations, the standard spherical coverage model can be adjusted by introducing a three-dimensional coverage correction coefficient related to facility type.

[0093] For critical areas, redundant coverage constraints are established, requiring that they be covered by at least two similar facilities, as expressed by the following formula:

[0094]

[0095] in, Represents the set of key requirements. This represents the set of facility types that require redundant coverage.

[0096] Specifically, when establishing full-area coverage constraints, the coverage relationship matrix between candidate locations and demand points is pre-calculated. For each facility type and each candidate location, the set of demand points that can be covered is determined, improving the efficiency of constraint establishment. Based on the importance of demand points and the safety requirements of low-altitude traffic, a hierarchical coverage standard is established, with higher coverage redundancy required for important areas and basic coverage sufficient for general areas. Furthermore, considering the mobility characteristics of aircraft, a dynamic coverage model is established to ensure that aircraft maintain effective connectivity with ground facilities throughout their flight.

[0097] For each selected facility, ensure that the demand for its services does not exceed the capacity limit, and establish a capacity constraint, expressed by the following formula:

[0098]

[0099] in, Indicate demand points For the The demand for such facilities Indicates the first The capacity limit for this type of facility (this value is designed based on peak demand; this design principle ensures that the facility can still provide stable and reliable service under extreme load conditions, avoiding a decline in service quality due to insufficient capacity). Represents a set of facility types. Represents the set of candidate positions. This represents the set of demand points.

[0100] Demand weight The specific calculation method is as follows:

[0101]

[0102] in, For demand points The basic demand is determined by low-altitude traffic flow forecast data, in units of flights / hour; For the first The service weight coefficient for different types of facilities is used to reflect the differences in service characteristics between different facility types. (The last part, "takeoff and landing station service weight coefficient," appears to be an unrelated fragment and is omitted from the translation.) =2.0, Service Weight Coefficient of Communication Base Station =1.2, service weight coefficient of navigation station =1.0, Service weight coefficient of monitoring equipment =1.5, service weight coefficient of weather stations =0.8 For demand points The importance coefficient is set at 1.5-2.0 for key areas such as business centers and transportation hubs, and 1.0 for general areas.

[0103] For example, for a communication demand point located in a commercial center area, its basic requirements are... Flights per hour, service weight of communication base stations Importance coefficient of this region Then the demand weight of this demand point for communication facilities The result is 100 × 1.2 × 1.8 = 216.

[0104] By introducing load balancing constraints, we can prevent some facilities from being overloaded while others remain idle. This can be expressed by the following formula:

[0105]

[0106] in, Indicates the load balancing coefficient. Indicates the first Average load rate of such facilities.

[0107] Load balancing coefficient The value ranges from 1.1 to 1.5, and is used to represent the maximum allowable load rate of a single facility that can reach the average load rate of that type of facility. times. The smaller the value, the more stringent the requirements for load balancing; The larger the value, the more lenient the allowable load fluctuation range.

[0108] For example, when This indicates that the load rate fluctuation of each facility does not exceed 30% of the average. If the average load rate of a certain type of facility is 60%, then the load rate of any single facility should not exceed 60% × 1.3 = 78%. In practical applications, for critical areas with high service reliability requirements, it is recommended to take [a certain value]. For general areas, the following can be adopted: .

[0109] Specifically, when establishing capacity constraints, a demand-to-facilities allocation model is created to determine which facility provides service for each demand point, achieving reasonable load distribution and effective capacity utilization. Based on the temporal variation characteristics of demand, facility capacity is designed according to peak demand to ensure sufficient service capacity even during peak periods. Interfaces for capacity expansion are reserved in the model; when demand growth exceeds current capacity, service capacity can be expanded by adding facilities or upgrading equipment.

[0110] Establish a facility compatibility matrix ,in Indicates the first Class and First Such facilities can be co-located. This indicates that the address cannot be shared.

[0111] The facility compatibility matrix is ​​determined based on the following technical rules:

[0112] (1) Electromagnetic compatibility rules: When communication base stations and navigation stations are deployed in the same location, electromagnetic interference may occur, affecting the accuracy of navigation signals. Therefore, it is necessary to maintain a certain physical isolation distance or adopt frequency band isolation measures. Facility compatibility matrices should be set up at candidate locations where isolation conditions are not available. ;

[0113] (2) Physical space requirements rules: Take-off and landing stations require a large independent operating space to ensure the safe take-off and landing of aircraft. Their clearance requirements may conflict with the antennas or sensors of other facilities. Therefore, take-off and landing stations are usually not co-located with other facilities, and a facility compatibility matrix is ​​set up. ;

[0114] (3) Operation and maintenance collaboration rules: Both the meteorological station and the monitoring equipment require a stable power supply and data transmission link, and are functionally complementary. They can share infrastructure to reduce costs and set up [specific rules]. Communication base stations can be co-located with weather stations and monitoring equipment, and a facility compatibility matrix can be set up. .

[0115] For example, the compatibility matrix A for five types of facilities—takeoff and landing stations (j=1), communication base stations (j=2), navigation stations (j=3), surveillance equipment (j=4), and meteorological stations (j=5)—is shown in Table 1 below:

[0116] Table 1 Compatibility Matrix

[0117]

[0118] It should be noted that diagonal elements are marked with "-", indicating that the problem of duplicate deployment of similar facilities in the same location does not fall under the scope of co-location compatibility. Instead, it is determined whether to add similar facilities in nearby locations based on capacity constraints and economic analysis.

[0119] The matrix is ​​symmetric about its diagonal, that is This reflects the two-way nature of co-location relationships.

[0120] For pairs of facilities that cannot be co-located, a mutual exclusion constraint is established, expressed by the following formula:

[0121]

[0122] in, , This indicates two different types of facilities. Indicates the candidate position Should the first Such facilities Indicates the candidate position Should the first Such facilities This represents the facility compatibility matrix, used to indicate whether facilities can be co-located based on specific values.

[0123] Introducing co-address indicator variables Indicates the location Whether to deploy a combination of facilities ,in A subset of compatible facilities is represented by a co-location constraint, expressed by the following formula:

[0124]

[0125]

[0126] in, Indicates a combination of facilities. This indicates a co-address indicator variable.

[0127] Specifically, when setting up facility co-location constraints and optimization mechanisms, a thorough analysis of the technical compatibility between different facility types is conducted, including electromagnetic interference, physical space requirements, and operational management requirements, to establish detailed compatibility rules. The cost savings and management conveniences brought by facility co-location are quantified, including infrastructure sharing, reduced maintenance costs, and improved management efficiency, accurately reflecting these benefits in the objective function. Through clever constraint design, the logical consistency of co-location decisions is ensured, avoiding infeasible combinations.

[0128] Step 103: Set a comprehensive optimization objective function for the mixed integer optimization model, configure cost preference weights for co-located deployment in the cost calculation of the comprehensive optimization objective function, and establish a multi-objective weight configuration mechanism and a constraint violation penalty mechanism.

[0129] With minimizing the total cost of infrastructure construction as the primary objective, and considering secondary objectives such as minimizing the number of facilities and maximizing service redundancy, a comprehensive optimization objective function is designed. Cost-preferential weights are assigned to co-located deployments in the cost function, encouraging integrated deployment by reducing the total construction cost of multiple facilities at the same site. A multi-objective weight allocation mechanism is established, allowing adjustment of the importance weights of different objectives based on actual planning needs. A constraint violation penalty mechanism is set to penalize violations of the objective function, ensuring that the optimized solution satisfies all hard constraints.

[0130] In this embodiment, a comprehensive cost function is established, including the construction cost, operating cost, and maintenance cost of various facilities. In position The total cost is calculated using the following formula:

[0131]

[0132] in, Indication facilities In position Total cost This indicates the construction cost of various facilities. This indicates the operating costs of various facilities. This indicates the maintenance costs of various facilities.

[0133] For co-located deployments, a cost-saving factor is introduced. ,Location Deployment facility portfolio The total cost is calculated using the following formula:

[0134]

[0135] in, Indicates position Deployment facility portfolio Total cost This indicates the cost savings resulting from co-location. The value is determined based on the size and type of the facility combination. Indication facilities In position The total cost.

[0136] Cost Savings Factor The specific calculation formula is as follows:

[0137]

[0138] in, S represents the number of facility types included in the co-located facility portfolio S. The total number of facility types in the system (in this embodiment) ), The basic cost-saving parameter reflects the marginal cost-saving effect brought about by co-location deployment, and its value ranges from 0.1 to 0.3 based on actual engineering experience.

[0139] The sources of cost savings mainly include: sharing of land leasing or purchase costs, sharing of power access and distribution facilities, reuse of communication transmission links, joint construction of computer rooms and supporting facilities, and unified management of operation and maintenance personnel.

[0140] For example, take :

[0141] When a single facility is deployed independently No cost savings;

[0142] When two facilities are deployed at the same site This can save 4% of the cost;

[0143] When the three facilities are deployed at the same site This can save 8% of the cost;

[0144] When four facilities are deployed at the same site This can save 12% of the cost.

[0145] The objective function for total cost is established, expressed by the following formula:

[0146]

[0147] in, This represents the total cost objective function. This is a co-address indicator variable, representing the location. Whether to deploy a combination of facilities .

[0148] Specifically, when constructing the cost function, facility costs are broken down into detailed items such as civil engineering costs, equipment procurement costs, installation and commissioning costs, operating personnel costs, power and communication costs, and maintenance costs, thereby improving the accuracy of cost estimation. A detailed co-location cost model is established to quantify the cost savings brought about by infrastructure sharing, collaborative management, and intensive maintenance, providing accurate economic incentives for co-location optimization. A full life-cycle cost analysis method is adopted to discount all costs during the construction, operation, and decommissioning phases to the planning base period, ensuring the fairness of cost comparisons.

[0149] A coverage redundancy objective function is established to encourage multiple coverage in important areas, expressed by the following formula:

[0150]

[0151] in, This represents the objective function for coverage redundancy. Indicate demand points For the The demand for such facilities, and the benefits of "redundant coverage," are weighted according to the importance of the demand points, with important areas (corresponding to...) When a large number of devices achieve multiple coverage, for The contribution is greater; general areas (corresponding to) Even with redundant coverage, the improvement to the target is relatively limited.

[0152] A load balancing objective function is established to promote load balance among various facilities, and it is expressed by the following formula:

[0153]

[0154] in, This represents the load balancing objective function. For position Upper Load rate of such facilities.

[0155] Specifically, when constructing the service quality objective function, a multi-dimensional coverage quality evaluation system is established, including indicators such as coverage range, coverage intensity, coverage continuity, and coverage reliability, to comprehensively reflect service quality. A response time optimization objective is incorporated into the objective function to encourage the deployment of facilities near high-demand areas, thereby reducing service response time. Furthermore, service fairness across different regions and user groups is considered to avoid situations where some areas receive excessive service while others receive insufficient service.

[0156] Multiple objectives are combined into a single comprehensive objective function using a weighted sum method:

[0157]

[0158] in, Represents a single comprehensive objective function. Represents the weighting coefficients, satisfying , The weighting coefficient can be adjusted according to the actual planning priority; it should be increased when cost is sensitive. Increase when service quality is prioritized .

[0159] The specific steps for determining weight coefficients using the Analytic Hierarchy Process (AHP) are as follows:

[0160] (1) Invite 3-5 experts with experience in low-altitude transportation planning to conduct pairwise importance comparisons of the three optimization objectives of minimizing cost, maximizing coverage redundancy, and minimizing load balancing, and construct a judgment matrix;

[0161] (2) Calculate the largest eigenvalue and the corresponding eigenvector of the judgment matrix, and normalize the eigenvector to obtain the weight coefficient of each target;

[0162] (3) Calculate the consistency ratio CR. When CR < 0.1, the judgment matrix is ​​considered to have satisfactory consistency and the weight result is acceptable; otherwise, the judgment matrix needs to be adjusted and recalculated.

[0163] Depending on the specific planning scenario, typical weight configuration schemes differ, for example:

[0164] In cost-sensitive scenarios (suitable for projects with limited budgets): , , Prioritize controlling construction costs;

[0165] In scenarios where service quality is prioritized (applicable to areas with high security requirements): , , Prioritize ensuring service redundancy;

[0166] In a balanced configuration scenario (applicable to general planning projects): , , At this point, it is necessary to balance cost and service quality;

[0167] In load balancing priority scenarios (suitable for regions with large demand fluctuations): , , Prioritize load balancing.

[0168] This application introduces a constraint violation penalty term to ensure that solutions that violate hard constraints are severely penalized, expressed by the following formula:

[0169]

[0170] in, This indicates the penalty for violating the constraint. This indicates a large penalty coefficient.

[0171] Specifically, in multi-objective weighted combination optimization, various methods such as the analytic hierarchy process (AHP), expert scoring, and historical case analysis are employed to determine the weights of each objective, ensuring the rationality and acceptability of the weight settings. Sensitivity analysis is performed on the optimization results under different weight combinations to identify key weight parameters, providing decision-makers with a reference for weight adjustment. The Pareto front of multi-objective optimization is calculated, providing decision-makers with multiple non-dominated solution options to support the selection of solutions under different preferences.

[0172] Step 104: Perform multiple rounds of iterative solutions on the mixed integer optimization model to output the optimal facility layout scheme.

[0173] Specifically, the first stage ignores capacity constraints and only considers coverage requirements, iteratively selecting facility addition schemes that can cover the most uncovered demand points to quickly obtain an initial feasible solution that satisfies full coverage. The second stage performs capacity verification and adjustment based on the initial feasible solution, adding similar facilities to overloaded nodes for load shedding and supplementing facilities in insufficiently covered areas, using a local search algorithm to optimize the quality of the solution. The third stage employs a genetic algorithm for global optimization, designing a specialized chromosome encoding scheme to represent the multi-facility combination decision for each candidate location, searching for a better layout scheme through population evolution, and setting a penalty function to handle constraint violations. This phased approach reduces problem complexity while combining intelligent algorithms to enhance global optimization capabilities.

[0174] In this embodiment, the set coverage algorithm based on a greedy strategy ignores capacity constraints and only considers coverage requirements. In each iteration, it selects the facility that can cover the most uncovered demand points to add to the facility layout scheme, and designs the coverage benefit function, which is expressed by the following formula:

[0175]

[0176] in, This represents the coverage benefit function, used in the first-stage set coverage (greedy strategy) to measure "in the first stage". Candidate positions No. Type of facilities That is, the "new coverage" portion (corresponding to the number of new coverage demand points) is the intersection between the current set of uncovered demand points and the set of demand points that the candidate facility can cover, and is further combined with cost-benefit indicators; This represents the set of demand points that are not currently covered. Indicates the location Deployment of the first The set of demand points that can be covered by such facilities.

[0177] Iterative execution selection The largest candidate facility is added to the facility layout scheme, and the set of uncovered demand points is updated until all demand points are covered. An initial feasible solution is output as the starting point for subsequent optimization. The initial feasible solution satisfies the coverage constraint but may not satisfy the capacity constraint.

[0178] The initial feasible solution undergoes capacity verification, calculating the actual load of each selected facility and identifying overloaded and underloaded facilities. For overloaded facilities, similar facilities are added to surrounding candidate locations for load diversion, using the nearest neighbor principle. For facilities with insufficient load, other similar facilities within their service range are considered for removal to reduce overall cost. Local search algorithms are employed to improve the solution, including neighborhood operations such as facility location adjustment, facility type replacement, and co-location optimization. Neighborhood structures are designed, including single-point movement (changing the location of a facility), type switching (changing the facility type combination at a location), and swapping operations (swapping the locations of two facilities). Simulated annealing or tabu search methods are used to avoid getting trapped in local optima and progressively improve the quality of the solution.

[0179] A specialized chromosome coding scheme is designed to encode the multi-facility combination decision for each candidate location into a gene string. Each candidate location corresponds to a gene bit, and the gene value represents the facility combination number deployed at that location. A facility combination coding table is established, assigning a unique number to all possible facility combinations (including empty combinations, single-facility combinations, and multi-facility co-location). The population is initialized, including the improved solutions from the second phase as superior individuals, with the remaining individuals generated randomly or heuristically. A fitness function is designed, expressed by the following formula:

[0180]

[0181] in, Represents the fitness function. and These represent the degree of violation of coverage constraints and capacity constraints, respectively. and This is the penalty coefficient.

[0182] The penalty coefficient serves to convert the degree of constraint violation into a fitness penalty value, causing solutions that violate the constraints to be eliminated during the evolutionary process. The penalty coefficient should be large enough to ensure that the fitness of any feasible solution is better than that of an infeasible solution.

[0183] It should be noted that the specific values ​​of the penalty coefficient are as follows:

[0184] (1) Penalty coefficient for covering constraints Used to penalize solutions that do not meet the coverage requirement; values ​​range from 10³ to 10. 5 Coverage constraints are hard constraints and must be strictly satisfied. Take the larger value. In this embodiment, take... =10 4 . The calculation method is the proportion of the number of uncovered demand points to the total number of demand points.

[0185] (2) Capacity constraint penalty coefficient Used to penalize solutions that exceed the capacity limit; values ​​range from 10² to 10. 4 This embodiment takes =10³. The calculation method is the sum of the overload ratios of all overloaded facilities.

[0186] (3) Large penalty coefficient Used in the constraint violation penalty term of the objective function, its value is 10-100 times the possible maximum value of the objective function; in this embodiment, it is... .

[0187] The magnitudes of the penalty coefficients should satisfy the following relationship: This is to reflect the priority of different constraints, that is, the priority of coverage constraints is higher than that of capacity constraints.

[0188] For example, if 5% of the requirements for a certain entity's layout scheme are not covered... Two facilities were overloaded, with an average overload rate of 20%. If the penalty value is 0.4, then its penalty value is... .

[0189] A convergence criterion is set so that evolution stops when the optimal solution shows no significant improvement over several consecutive generations, and the best individual is output as the final solution. The specific criteria for determining the convergence criterion are as follows:

[0190] Define convergence criterion parameter: stable algebra threshold and relative improvement threshold When continuous Within a generation, the relative improvement in the fitness value of the best individual in the population is less than the threshold. When the algorithm has converged, the evolution process is stopped.

[0191] The convergence criterion formula is as follows:

[0192]

[0193] in, Let represent the fitness value of the best individual in the g-th generation of the population, where g is the current generation number. Represents the stable algebraic threshold. This indicates the relative improvement threshold.

[0194] It should be noted that the stable algebra threshold The value range is 30-100. In this embodiment, This is used to represent 50 consecutive generations. (Relative Improved Threshold) The value range is 0.0001-0.01. In this embodiment, This is used to indicate an improvement of less than 0.1%.

[0195] Furthermore, when the maximum number of iterations is reached... When the above convergence conditions are met, the algorithm stops and outputs the current optimal solution.

[0196] For example, if the optimal fitness in generation 800 is 0.95 and the optimal fitness in generation 750 is 0.949, then the relative improvement is |0.95-0.949| / 0.949=0.00105>0.001, and the algorithm continues to evolve; if the optimal fitness in generation 850 is still 0.95 and the optimal fitness in generation 800 is 0.9495, then the relative improvement is 0.00053<0.001, and the algorithm determines that it has converged and stops.

[0197] Multiple strategies are employed to maintain population diversity, including diversity detection, similar individual penalties, and migration operations, to prevent the population from prematurely converging to local optima. Based on feedback information during the evolutionary process, the parameters of the genetic algorithm, such as crossover probability, mutation probability, and penalty coefficient, are adaptively adjusted to improve the algorithm's adaptability and effectiveness.

[0198] In this embodiment, before each iteration, the ensemble coverage algorithm needs to calculate the coverage status of all demand points (i.e., gridded demand units) based on the facility set of the currently selected schemes. This clarifies which demand points are covered and which are still uncovered, generating a dynamically updated set of uncovered demand points. For each candidate facility not yet selected for a layout scheme—that is, a combination of a specific candidate location and a specific facility type—its coverage benefit value needs to be calculated. The coverage benefit value is the ratio between the number of uncovered demand points that the candidate facility can cover and its deployment cost. For example, if a communication base station candidate scheme can newly cover 100 demand grids that are not currently covered by any selected base stations, and its construction and operation cost is estimated to be C, then its coverage benefit value is 100 / C. This considers both coverage capability and economic cost.

[0199] Then, the coverage benefit values ​​of all candidate facilities are compared, and the candidate facility with the highest coverage benefit value is added to the facility layout scheme. At each step, only the option that appears locally optimal at the current step is selected. After adding a new facility, all demand points that can be covered by that facility are removed from the set of uncovered demand points. Subsequently, the algorithm enters the next iteration, recalculates the coverage benefit values ​​of all remaining candidate facilities based on the updated uncovered set, and again selects the optimal one to add, repeating the above process until the set of uncovered demand points is empty. At this point, the algorithm outputs a scheme that ensures that every demand point is covered by at least one facility, thus obtaining an initial feasible solution that satisfies the global coverage constraint. It should be noted that since the initial feasible solution ignores the capacity constraint and only pursues a greedy decision based on coverage, it is usually not the economically or capacity-optimal solution, but it provides a crucial and feasible starting point for subsequent optimization.

[0200] In this embodiment, after obtaining the initial feasible solution, the second stage begins. The goal is to modify and optimize a solution that only satisfies coverage requirements into a higher-quality solution that satisfies all constraints (especially capacity constraints). First, capacity verification is performed. Based on the initial feasible solution and demand distribution data, the actual service load that each selected facility needs to bear is simulated and calculated. For example, for a selected communication base station, the demand weights of all demand grids within its signal coverage area are aggregated to obtain its total load. Then, this load is compared with the technical parameters (rated capacity) of the base station. Through this comparison, the algorithm can automatically identify two types of problem nodes: overloaded nodes, i.e., facilities whose actual load exceeds their rated capacity; and areas with insufficient coverage or extremely light load. This may be due to the locality of reference in the greedy algorithm, where some areas are covered but have very low redundancy, or some facilities have loads far below their capacity.

[0201] For identified overloaded nodes, a strategy of adding similar facilities to offload traffic is adopted. Specifically, within the geographical range surrounding the overloaded facility, other unused candidate locations are searched, and one or more suitable locations are selected to deploy similar facilities, thereby distributing some of the service demand from the original facility to the new facility. For areas with weak coverage or extremely uneven load, supplementary facilities or minor adjustments to facility locations are made accordingly.

[0202] After the initial corrections described above, a local search algorithm is used for further improvement. Starting with the current solution (i.e., the improved feasible solution), the local search algorithm searches for a better solution within its neighborhood. The neighborhood is defined by a series of predefined move operations, such as moving a facility from one candidate location to another nearby candidate location (position fine-tuning); replacing the facility type deployed at a certain location with another compatible type (type switching); swapping the locations of two facilities, etc. The algorithm attempts these move operations and evaluates the comprehensive objective function value of the new solution after each move. If a move brings improvement, such as reducing costs, increasing redundancy, or improving load balancing, the move is accepted, and the current solution is updated. Through repeated attempts and acceptance of improving moves, the solution is gradually refined locally, resulting in an improved feasible solution with significantly better quality than the initial solution and satisfying all constraints. This solution can be considered a high-quality local optimum.

[0203] In this embodiment, to escape local optima, a third stage is introduced: global optimization using a genetic algorithm. This algorithm simulates the principles of biological evolution, searching for the optimal solution by iteratively evolving a population representing multiple potential solutions. First, a chromosome encoding scheme is designed to represent the multi-facility combination decision for each candidate location.

[0204] For example, a direct encoding method is to number the candidate locations in the entire target area in a fixed order (from 1 to N). Then, a chromosome is a gene string of length N. The value of the i-th gene position in the gene string is not a simple 0 or 1, but a coded value interpreted according to a predefined facility combination coding table. For example, code 0 indicates that no facilities are deployed at that location; code 1 indicates that only a communication base station is deployed; code 2 indicates that only a weather station is deployed; code 3 indicates that a communication base station and a weather station are co-located, and so on. In this way, a chromosome uniquely determines a complete facility layout scheme.

[0205] Next, the population is initialized. It should be noted that the population contains two types of individuals: first, randomly generated individuals, where the facility combination encoding for each gene locus (candidate position) is randomly determined, which ensures the diversity of the population; second, the improved feasible solution obtained in the second stage is transformed into a chromosome according to the above encoding rules and added to the population as a superior individual. This is equivalent to injecting a high-quality initial seed into the evolutionary process, which can accelerate convergence.

[0206] Then, the iterative evolutionary process begins. In each generation, for each individual (chromosome) in the population, its corresponding facility layout scheme is decoded, and then the comprehensive objective function value and constraint violation penalty value are calculated. The fitness of an individual is usually designed to be negatively correlated with the comprehensive objective function value, that is, the better the scheme, the smaller the objective function value, the higher the fitness, and the more severe the penalty for constraint violation. This ensures that the evolutionary direction is towards a better and more feasible scheme.

[0207] Then, based on individual fitness levels, parent individuals are selected from the current population according to certain rules (such as roulette wheel selection or tournament selection). Individuals with higher fitness have a higher probability of being selected. For the selected parent individuals, a crossover operation (such as single-point crossover) is performed, where a random point is selected, and the gene sequences of two parent individuals after that point are exchanged, resulting in two new offspring individuals. This operation can combine the desirable traits of different parents. Subsequently, a mutation operation is performed on the offspring individuals with a certain probability, randomly changing the coding values ​​of one or more gene loci on their chromosomes, thereby introducing new variations that help explore new solution spaces and maintain population diversity. Finally, the resulting offspring individuals are merged with the parent individuals (or an elite preservation strategy is used to retain the best individuals of the current generation) to form a new generation of the population.

[0208] Step 105: Conduct a comprehensive evaluation of the optimal facility layout scheme, generate a facility layout list, compare and analyze the facility layout list with traditional planning schemes, and output a visual layout map and technical report.

[0209] The optimized layout scheme is comprehensively evaluated, and overall indicators such as the total number of stations, total cost, coverage, average redundancy, number of co-located nodes, and cost savings percentage are calculated. A detailed facility layout list is generated, indicating the combination of facility types deployed at each selected location. Furthermore, a comparative analysis is conducted with traditional separate planning schemes to quantify the improvement effect of integrated layout, thereby outputting a visual layout diagram and technical report, providing decision-makers with clear planning schemes and implementation guidance.

[0210] In this embodiment, the overall performance indicators of the layout scheme are calculated. It should be noted that the overall performance indicators include basic indicators: the total number of websites built. Total construction cost Coverage Average redundancy ,in For demand points The amount of redundant coverage. Calculate optimization metrics, such as the number of co-located nodes. Colocation rate Cost savings rate ,in The total cost is planned separately. Service quality indicators are calculated, such as average service response time, load balancing, and network reliability, and an evaluation system is established to comprehensively assess the technical performance, economic benefits, and social benefits of the solution.

[0211] Specifically, a multi-level performance indicator system should be established, including technical indicators (coverage rate, capacity utilization rate, response time), economic indicators (construction cost, operating cost, investment benefit), and social indicators (service equity, environmental impact, user satisfaction). A scientific benchmark comparison system should be established to compare the proposed solutions with similar projects at home and abroad, industry standards, and theoretical optimal values, so as to objectively evaluate the merits of each solution. Furthermore, the impact of parameter uncertainty on the evaluation results should be considered, and methods such as Monte Carlo simulation should be used to analyze the stability and reliability of the results.

[0212] This application generates a facility layout list based on the comprehensive evaluation results of the facility layout plan. It should be noted that the facility layout list includes detailed information for each selected location, such as location coordinates, address description, combination of deployed facility types, service coverage, load level, and construction cost. Subsequently, a corresponding visual layout map is drawn to display the spatial distribution of all facilities on the geographic information system platform, using different symbols and colors to distinguish different types of facilities and co-location situations. A service coverage map is also generated to display the service range and coverage overlap of various facilities, verifying the achievement of full-area coverage. A facility statistics table is created, statistically analyzing the number and cost distribution of facilities by facility type, region, and importance, and an implementation sequence diagram is compiled, arranging the construction timeline of facilities according to construction priority and resource constraints.

[0213] Specifically, advanced visualization technologies, including 3D modeling, dynamic demonstrations, and interactive queries, are employed to improve the intuitiveness and understandability of the layout scheme. Unified data standards and formats are established to ensure the compatibility and exchangeability of layout scheme data with other systems. Standardized document templates and formats are developed to ensure the completeness, consistency, and professionalism of layout scheme documents.

[0214] The generated optimal facility layout scheme is compared with the traditional separate planning scheme. The traditional scheme plans each type of facility independently, determining the optimal layout for each type separately. The improvement effect of the integrated layout is quantitatively analyzed, including specific values ​​for reduced facility quantity, cost savings, improved coverage, and enhanced redundancy. Furthermore, sensitivity analysis is conducted to examine the impact of changes in key parameters on the layout scheme, such as the stability of the scheme under scenarios of increased demand, cost changes, and technological upgrades. In addition, the implementation risks of the scheme are assessed, including technological risks, economic risks, and policy risks. Corresponding risk mitigation measures are proposed, and economic indicators such as the investment payback period, net present value, and internal rate of return are calculated to provide a basis for investment decisions.

[0215] Specifically, a comparative analysis is conducted from multiple dimensions, including technology, economy, society, and environment, to comprehensively evaluate the advantages and disadvantages of the integrated layout. Scientific quantitative methods are employed to accurately calculate the various benefits brought about by the integrated layout, including both direct and indirect benefits. A comprehensive risk assessment system is established to identify and evaluate various potential risks and formulate corresponding risk response strategies.

[0216] Finally, a detailed technical report is prepared, and decision support materials are produced. The layout plan is presented in intuitive formats such as charts, maps, and animations to facilitate understanding by decision-makers without a technical background. Multiple alternative plans are provided, based on different weight settings and constraints, offering decision-makers a variety of choices. Implementation guidelines are developed to guide project implementation, and a plan update mechanism is established to regularly update the layout plan as needs change and technology advances, maintaining the plan's timeliness and adaptability.

[0217] It should be noted that the technical report includes complete content such as project background, technical solution, optimization results, implementation suggestions, and risk analysis, while the implementation guidelines include construction standards, technical specifications, management systems, and acceptance standards.

[0218] Specifically, design a reasonable report structure to ensure the logical consistency, completeness, and readability of the content, meeting the needs of different readers. Develop specialized decision support tools, including alternative comparison tools, sensitivity analysis tools, and scenario analysis tools, to assist decision-makers in making informed decisions. Establish a continuous improvement mechanism for the plan, constantly optimizing and refining the layout plan based on implementation feedback and environmental changes.

[0219] The above are embodiments of the method proposed in this application. Based on the same inventive concept, embodiments of this application also provide an integrated layout optimization device for low-altitude traffic, the structure of which is as follows: Figure 2 As shown.

[0220] Figure 2 This is a schematic diagram of the internal structure of an integrated layout optimization device for low-altitude traffic, provided as an embodiment of this application. Figure 2 As shown, the device includes:

[0221] At least one processor;

[0222] And, a memory that is communicatively connected to at least one processor;

[0223] The memory stores instructions that can be executed by at least one processor, and the instructions, when executed by at least one processor, enable at least one processor to:

[0224] Acquire candidate locations within the target area for facility deployment, label each candidate location with attributes, and collect low-altitude traffic demand distribution data and technical parameters of various facilities;

[0225] Based on candidate locations, demand distribution data, and technical parameters, a mixed-integer optimization model for joint site selection of multiple facilities is established. In the mixed-integer optimization model, full coverage constraints and capacity constraints are set, as well as a co-location optimization mechanism is referenced.

[0226] A comprehensive optimization objective function is set for the mixed-integer optimization model. In the cost calculation of the comprehensive optimization objective function, a cost preference weight is configured for the co-location deployment case, and a multi-objective weight configuration mechanism and a constraint violation penalty mechanism are established.

[0227] The mixed-integer optimization model is solved through multiple rounds of iterations to output the optimal facility layout scheme;

[0228] The optimal facility layout scheme is comprehensively evaluated, a facility layout list is generated, and the facility layout list is compared and analyzed with traditional planning schemes to output a visual layout map and technical report.

[0229] This application also provides a non-volatile computer storage medium storing computer-executable instructions, which, when executed, can:

[0230] Acquire candidate locations within the target area for facility deployment, label each candidate location with attributes, and collect low-altitude traffic demand distribution data and technical parameters of various facilities;

[0231] Based on candidate locations, demand distribution data, and technical parameters, a mixed-integer optimization model for joint site selection of multiple facilities is established. In the mixed-integer optimization model, full coverage constraints and capacity constraints are set, as well as a co-location optimization mechanism is referenced.

[0232] A comprehensive optimization objective function is set for the mixed-integer optimization model. In the cost calculation of the comprehensive optimization objective function, a cost preference weight is configured for the co-location deployment case, and a multi-objective weight configuration mechanism and a constraint violation penalty mechanism are established.

[0233] The mixed-integer optimization model is solved through multiple rounds of iterations to output the optimal facility layout scheme;

[0234] The optimal facility layout scheme is comprehensively evaluated, a facility layout list is generated, and the facility layout list is compared and analyzed with traditional planning schemes to output a visual layout map and technical report.

[0235] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0236] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

Claims

1. An integrated layout optimization method for low-altitude transportation, characterized in that, The method includes: Acquire candidate locations within the target area for facility deployment, label each candidate location with attributes, and collect low-altitude traffic demand distribution data and technical parameters of various facilities; Based on the candidate locations, the demand distribution data, and the technical parameters, a hybrid integer optimization model for joint site selection of multiple facilities is established, and a global coverage constraint and a capacity constraint, as well as a co-location optimization mechanism, are set in the hybrid integer optimization model. A comprehensive optimization objective function is set for the mixed integer optimization model. In the cost calculation of the comprehensive optimization objective function, a cost preference weight is configured for the co-location deployment case, and a multi-objective weight configuration mechanism and a constraint violation penalty mechanism are established. The mixed-integer optimization model is solved through multiple rounds of iterations to output the optimal facility layout scheme; The optimal facility layout scheme is comprehensively evaluated to generate a facility layout list. The facility layout list is then compared and analyzed with traditional planning schemes, and a visual layout diagram and technical report are output. Acquire candidate locations within the target area for facility deployment, label each candidate location with attributes, and collect low-altitude traffic demand distribution data and technical parameters of various facilities, specifically including: A systematic scan of the target area is performed to identify all candidate locations within the target area for the construction of low-altitude transportation infrastructure, resulting in a set of candidate locations; Perform attribute labeling operations on each candidate location in the candidate location set; the attribute labeling operations include indicating the types of facilities that can be constructed at the candidate location, recording the site physical attributes of the candidate location, investigating the infrastructure conditions of the candidate location, and identifying regulatory and technical constraints; Collect urban low-altitude traffic demand forecast data, analyze air traffic flow characteristics based on the demand forecast data, obtain distribution information of demand hotspot areas, and obtain a set of demand points. A demand gridding model is established, which discretizes the continuous three-dimensional low-altitude airspace into a regular grid and assigns a corresponding demand weight to each three-dimensional grid cell in the regular grid to predict future demand growth trends. The demand gridding model is divided into square or hexagonal grids in the horizontal direction and into several height layers in the vertical direction according to the height stratification standard of low-altitude airspace. Technical parameters of communication base stations, navigation stations, monitoring equipment, meteorological stations and take-off and landing stations are collected, the technical parameters are standardized, and cost-benefit analysis is performed to obtain the construction cost, operating cost and maintenance cost of various facilities; Based on the candidate locations, the demand distribution data, and the technical parameters, a mixed-integer optimization model for joint site selection of multiple facilities is established. This model includes setting global coverage constraints and capacity constraints, as well as a co-location optimization mechanism, specifically including: Binary site selection decision variables are defined for each candidate location and each combination of facility types to indicate whether the candidate location should be used to construct the corresponding type of facility. Auxiliary binary variables are introduced to indicate whether the candidate location is selected, whether a specific facility co-location combination is adopted, and a hierarchical model architecture is constructed. The hierarchical model architecture includes a coverage layer, a capacity layer, and a co-location layer. For each demand point and each facility type, a coverage relationship judgment logic is established. By calculating the distance between the candidate location and the demand point and comparing it with the service radius of the corresponding facility, a full-domain coverage constraint is constructed to ensure that each demand point is covered by at least one facility of the current type that meets the conditions. Redundant coverage constraints are also established for demand points in key areas. For each candidate location and each facility type, a service demand calculation logic is established. By summarizing the demand weights of all demand points falling within the facility's service radius for the current type of facility, a capacity constraint is constructed to ensure that the service demand undertaken by the facility does not exceed the preset rated capacity. A load balancing constraint is also introduced to limit the load rate differences between facilities. Construct a facility compatibility matrix, and based on the facility compatibility matrix, establish mutual exclusion constraints for each pair of non-co-located facility types at each candidate location, and establish combination selection variables and corresponding association constraints for facility combinations that are allowed to co-locate; A comprehensive optimization objective function is set for the mixed-integer optimization model. In the cost calculation of the comprehensive optimization objective function, a cost-preferential weight is configured for co-location deployment. A multi-objective weight configuration mechanism and a constraint violation penalty mechanism are established, specifically including: Based on the cost parameters in the technical parameters, a full life cycle cost calculation model is constructed for each facility at each candidate location, and a cost saving coefficient is introduced into the co-location combination of each facility at each candidate location in the cost calculation model; The total cost of the facility co-location combination is calculated by multiplying the sum of the costs of each facility within the co-location combination by the cost-saving factor, thereby constructing a total cost minimization function; the total cost minimization function aims to minimize the sum of the costs of the facility co-location combination. Construct a function to maximize coverage redundancy to calculate the weighted sum of the number of times each demand point is covered by similar facilities beyond the basic requirements, and construct a function to minimize load balancing to calculate the sum of the differences between the maximum and minimum load rates among similar facilities. Assign corresponding weight coefficients to the total cost minimization function, the coverage redundancy maximization function, and the load balancing minimization function, and combine them into a single comprehensive objective function; The degree of violation of the global coverage constraint and the capacity constraint is multiplied by a penalty coefficient to add a constraint violation penalty term to the comprehensive objective function.

2. The integrated layout optimization method for low-altitude transportation according to claim 1, characterized in that, A systematic scan of the target area is performed to identify all candidate locations within the target area for the construction of low-altitude transportation infrastructure, resulting in a set of candidate locations, specifically including: By integrating high-resolution satellite imagery, urban 3D models, geographic information system data, and urban planning drawings, a systematic scan of the target area is conducted to identify all potential candidate locations within the target area for the construction of low-altitude transportation infrastructure, thus obtaining a set of potential candidate locations. The potential candidate locations in the potential candidate location set are verified on-site to determine whether the location information of the potential candidate locations is correct and to obtain the actual situation on site; Based on preset regulatory requirements, technical constraints, and economic feasibility criteria, the set of potential candidate locations is screened to exclude infeasible locations, and the final set of candidate locations is obtained.

3. The integrated layout optimization method for low-altitude transportation according to claim 1, characterized in that, The mixed-integer optimization model is solved iteratively in multiple rounds to output the optimal facility layout scheme, specifically including: In each iteration, the facility that can cover the most uncovered demand points is added to the facility layout scheme to obtain an initial feasible solution that satisfies the full coverage constraint. The initial feasible solution is subjected to capacity verification in order to add similar facilities to overloaded nodes for diversion, supplement facilities to areas with insufficient coverage, and improve the initial feasible solution. For the improved feasible solution, a chromosome encoding scheme is designed to represent the multi-facility combination decision for each candidate location. Population evolution is carried out by initializing the population, designing a fitness function, and performing genetic operations to output the optimal facility layout scheme.

4. The integrated layout optimization method for low-altitude transportation according to claim 3, characterized in that, In each iteration, facilities that can cover the most uncovered demand points are added to the facility layout scheme to obtain an initial feasible solution that satisfies the full coverage constraint, specifically including: Before each iteration begins, based on the selected facility set, calculate the covered and uncovered states of all demand points and generate an uncovered demand point set. For each candidate facility that was not selected in the facility layout scheme, the coverage benefit value of the candidate facility is calculated, and the coverage benefit values ​​of all candidate facilities are compared. The candidate facility with the highest coverage benefit value is added to the facility layout scheme. The coverage benefit value is obtained by the ratio of the number of uncovered demand points that the candidate facility can cover to the deployment cost. The demand points covered by the newly added facility are removed from the set of uncovered demand points. The process of calculating the coverage benefit value, selecting the facility with the highest benefit, and updating the uncovered set is repeated until the set of uncovered demand points is empty. An initial feasible solution that satisfies the global coverage constraint is then output.

5. The integrated layout optimization method for low-altitude transportation according to claim 3, characterized in that, For the improved feasible solution, a chromosome encoding scheme is designed to represent the multi-facility combination decision for each candidate location. Population evolution is then performed through population initialization, fitness function design, and genetic operations to output the optimal facility layout scheme. Specifically, this includes: The multi-facility combination decision for each candidate location is encoded into a gene string to form a chromosome, and the population is initialized; the population contains random individuals and excellent individuals corresponding to the improved feasible solution; In each generation of evolution, the comprehensive objective function value and constraint violation penalty value of the facility layout scheme corresponding to each individual chromosome in the population are calculated, and the fitness of each individual chromosome in the population is calculated based on the comprehensive objective function value and the constraint violation penalty value. Based on the fitness of individual chromosomes, parent individuals are selected from the current population. A single-point crossover operation is performed on the selected parent individuals to generate offspring individuals. Random mutation operations are then performed on some gene loci in the offspring individuals to change the facility combination encoding. Individuals in the population are merged with their offspring to form a new generation of population, and the evolutionary process is repeated until the optimal solution remains stable across multiple generations.

6. An integrated layout optimization device for low-altitude transportation, characterized in that, The device includes: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform an integrated layout optimization method for low-altitude traffic as described in any one of claims 1-5.

7. A non-volatile computer storage medium storing computer-executable instructions, characterized in that, When the computer-executable instructions are executed, an integrated layout optimization method for low-altitude transportation as described in any one of claims 1-5 is implemented.

Images