Site site selection method for low-altitude manned aircraft

Abstract

The invention relates to a site selection method for a low-altitude manned aircraft. The method comprises the following steps: obtaining a primarily selected low-altitude manned aircraft site based on potential demand points and control factors; obtaining candidate low-altitude manned aircraft stations through a clustering algorithm; a hierarchical service range model is introduced, the user demand quantity of each candidate low-altitude manned aircraft station is obtained in combination with the effective load of the low-altitude manned aircraft, and a low-altitude manned aircraft station site selection model is constructed; and solving the site selection model by adopting a Newton-Raphson optimization algorithm, wherein the target function of the site selection model is that the user demand quantity is maximum and the cost is minimum. Compared with the prior art, the site selection method has the advantages of realizing scientific, reasonable and accurate site selection of the low-altitude manned aircraft and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of urban low-altitude transportation, and particularly to a method for site selection of low-altitude manned aircraft. Background Art

[0002] As an important field for deepening the demonstration application of general aviation equipment, urban air transportation is a key infrastructure to support the "national 123 travel traffic circle" (1-hour commuting within the metropolitan area, 2-hour access within the urban agglomeration, and 3-hour coverage of major cities across the country). Among them, the large-scale application of low-altitude manned aircraft (mainly electric vertical take-off and landing aircraft eVTOL) urgently requires the construction of an efficient urban air transportation network, and the rationality of site selection directly determines the service quality and operation efficiency of the network.

[0003] However, the research on the site selection of low-altitude manned aircraft in the prior art still faces multiple challenges. First, the demand for urban air transportation travel is unknown. Urban air transportation is an unprecedented mode of transportation, so the site selection of low-altitude manned aircraft must be forward-looking to meet possible future user needs. Second, the low-altitude airspace control policy is not clear. Existing regulations such as the "Interim Regulations on the Flight Management of Unmanned Aerial Vehicles" do not clearly define the restrictions on the flight airspace of low-altitude manned aircraft. Third, it is difficult to balance costs and benefits. The construction cost of low-altitude manned aircraft is unknown and needs to be estimated based on the land price of the site where the station is located in order to make the most cost-effective choice within a limited budget. Finally, the application of optimization algorithms lags behind. Optimization algorithms are updated and iterated rapidly, and new and efficient algorithms have not been fully applied to the problem of low-altitude manned aircraft site selection.

[0004] After retrieval, Chinese Patent Application Publication No. CN118365026A discloses a method, device and medium for site selection and route selection of a manned electric vertical take-off and landing flying car, belonging to the field of urban low-altitude transportation. The method includes: obtaining information of all demand points and using a clustering algorithm to determine a candidate set of electric vertical take-off and landing platform sites; generating user travel routes driven by traffic big data based on the traffic information of the demand points; constructing a mathematical model according to the candidate set and the user travel routes, and optimizing the model with the total cost of operators and users as the optimization goal, and finally obtaining the decision on the optimal site selection of the flying car vertical take-off and landing platform and the operation of the system line network route. This existing patent application has the problem of unscientific site selection because the demand of candidate sites is not considered.

[0005] How to achieve more scientific and reasonable site selection of low-altitude manned aircraft has become a technical problem to be solved. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for site selection of low-altitude manned aircraft to overcome the defects existing in the above-mentioned prior art.

[0007] The object of the present invention can be achieved by the following technical solutions:

[0008] According to one aspect of the present invention, a method for site selection of a low-altitude manned aircraft is provided, and the method includes the following steps:

[0009] S1: Obtain a preliminary low-altitude manned aircraft site based on potential demand points and control factors;

[0010] S2: Obtain candidate low-altitude manned aircraft sites through a clustering algorithm;

[0011] S3: Introduce a hierarchical service range model, and in combination with the effective payload of the low-altitude manned aircraft, obtain the user demand of each candidate low-altitude manned aircraft site, and construct a site selection model for the low-altitude manned aircraft site;

[0012] S4: Use the Newton-Raphson optimization algorithm to solve the site selection model, where the objective function of the site selection model is: the maximum user demand and the minimum cost.

[0013] Preferably, the process of S1 includes:

[0014] Construct the potential demand points based on the urban POI data;

[0015] Construct a Hadamard product control matrix to determine the preliminary low-altitude manned aircraft site.

[0016] More preferably, the process of determining the preliminary low-altitude manned aircraft site includes

[0017] Construct a Hadamard product control matrix, specifically: Denote the east longitude and latitude coordinate point of City A as (x E , y E ), the west longitude and latitude coordinate point as (x W , y W ), the north longitude and latitude coordinate point as (x N , y N ), the south longitude and latitude coordinate point as (x S , y S ), and the formula is as follows:

[0018] m = x E - x W

[0019] n = y N - y S

[0020] x1 = x W

[0021] x m = x E

[0022] y1 = y N

[0023] y n = y S

[0024] AX = (x1, …, x m )

[0025] AY = (y1, …, y n )

[0026] i = x - (x w - 1)x ∈ AX

[0027] j = y - (y n - 1)y ∈ AY

[0028]

[0029] Wherein, m is the difference between the longitude of the easternmost coordinate point and the westernmost coordinate point of City A, n is the difference between the latitude of the northernmost coordinate point and the southernmost coordinate point of City A, AX = (x W , …, x E ), AY = (y N , …, y S ) are the latitude and longitude vectors of City A respectively, x and y are the longitude and latitude of the coordinates respectively, i and j are the matrix subscripts after the x and y coordinate conversions respectively, A mn is the coordinate schematic matrix of City A, represents the coordinate schematic matrix of k different control factors of City A;

[0030] Assume that the coordinates of a certain point are x and y, and its corresponding matrix subscripts i and j are obtained through conversion. When (x, y) is the coordinate point of City A, the element η mn of the coordinate schematic matrix A ij of City A is assigned 1, otherwise, it is 0;

[0031] Use the same method to determine whether the coordinate point (x, y) is the coordinate point of the kth control factor;

[0032] Convert all the coordinates of City A to obtain the coordinate schematic matrix A mn of City A and the coordinate schematic matrices of k different control factors of City A Then perform the Hadamard product of all the control factor coordinate schematic matrices and the coordinate schematic matrix of City A to obtain the Hadamard product control matrix H mn :

[0033]

[0034] Among them, the points where the data is 0 are no-fly points, and vice versa, they are the preliminary selected low-altitude manned aircraft sites, and finally v preliminary selected low-altitude manned aircraft sites are screened out.

[0035] Preferably, the process of obtaining the user demand for each candidate low-altitude manned aircraft site includes: taking all the preliminary selected low-altitude manned aircraft sites as the centers and the maximum candidate low-altitude site service range r3 as the radius, dividing v demand circles, and using the taxi data of each demand circle as the user demand for the candidate low-altitude manned aircraft site, and calculating the user demand for all the preliminary selected low-altitude manned aircraft sites. The formula is as follows:

[0036] n k ={o t}

[0037] o t =SCar+ECar×MCar 1≤t≤v,k∈[1,K]

[0038] In the formula, n k is the set of user demands for the preliminary selected low-altitude manned aircraft sites in different clusters, o t is the user demand for the t-th preliminary selected low-altitude manned aircraft site in a cluster, SCar is the number of orders starting from a certain location within the demand circle, ECar and MCar are respectively the number of orders and the amount with a certain location within the demand circle as the end point, and v is the number of preliminary selected low-altitude manned aircraft sites.

[0039] More preferably, the process of obtaining the user demand for each candidate low-altitude manned aircraft site further includes: combining the effective payload of the low-altitude manned aircraft and recalculating the user demand for the candidate low-altitude manned aircraft site. The formula is as follows:

[0040]

[0041] In the formula, S k represents the set of distances between the preliminary selected low-altitude manned aircraft sites in each cluster and the candidate low-altitude manned aircraft sites, CX k , CY k represent the latitude and longitude of the k-th candidate low-altitude manned aircraft site respectively, represent the latitude and longitude of the preliminary selected low-altitude manned aircraft sites in each cluster respectively; s t is the Euclidean distance between the t-th preliminary selected low-altitude manned aircraft site and the k-th candidate low-altitude manned aircraft site; R=(r1,r2,r3) is the candidate low-altitude site service range set based on three levels according to the different effective payloads of the low-altitude manned aircraft, N kis the set of user demands for candidate low-altitude manned aircraft sites, that is, the set of user demands for low-altitude manned sites based on the hierarchical service scope; both β1 and β2 are demand reduction coefficients.

[0042] More preferably, the process of obtaining the user demand for each candidate low-altitude manned aircraft site further includes: obtaining the corresponding land price according to the coordinate address of the candidate low-altitude manned aircraft site, and constructing a site selection model with high demand and low cost, specifically:

[0043] D=(d1,…,d K )

[0044]

[0045] where D is the candidate low-altitude manned aircraft site land price vector, and d k is the land price of the k-th candidate low-altitude manned aircraft site, is the initialized low-altitude manned aircraft site, where is a binary variable, that is, if then select this point as the low-altitude manned aircraft site, otherwise, abandon this point as the low-altitude manned aircraft site.

[0046] More preferably, the method further includes evaluating the demand satisfaction rate of the site location after calculating the site location, specifically:

[0047]

[0048] τ k =N k / B k

[0049] where B k is the total user demand of the k-th candidate low-altitude manned aircraft site, τ represents the user demand satisfaction rate of the k-th candidate low-altitude manned aircraft site, and N k is the set of user demands for candidate low-altitude manned aircraft sites.

[0050] Preferably, the process of solving the site selection model using the Newton-Raphson optimization algorithm includes:

[0051] Population initialization: Construct an initialized low-altitude manned aircraft site matrix Its mathematical expression is as follows:

[0052]

[0053] where ub and lb are the upper and lower limits of the population search position respectively; N pis the number of different combinations of low-altitude manned aircraft stations; K is the dimension, that is, the number of candidate low-altitude manned stations;

[0054] Improve the Newton-Raphson search, find the optimal solution during the search monitoring, and finally converge to the global solution; the specific improvement is as follows:

[0055] Define the parameter ρ to guide the population in the correct direction, specifically:

[0056]

[0057] where a and b are random numbers between (0,1), is the best solution obtained so far, is the current solution, and the subscripts k1 and k2 are different integers randomly selected from the population;

[0058] Design an adaptive coefficient δ to enhance the algorithm, specifically:

[0059]

[0060] where t represents the number of the current iteration, and Max_t represents the maximum number of iterations;

[0061] In order to maintain a balance between the exploration and development stages, the parameter δ will adjust itself during the iteration process, and the value of δ varies between 1 and -1;

[0062]

[0063]

[0064] where, and is the new vector position obtained by updating and NRSR represents the Newton-Raphson search rule, y w and y b are the positions of two vectors generated using Z k+1 and and is the worst solution so far, is the objective function, is the difference between the best solution and the current solution, and λ represents a random number between (0,1);

[0065] Finally, a new vector for the next iteration is obtained

[0066] More preferably, the process of using the Newton-Raphson optimization algorithm to solve the site selection model further includes: introducing a trap avoidance operator to obtain the optimal solution guided by the objective function.

[0067] Preferably, the process of obtaining candidate low-altitude manned aircraft sites through the clustering algorithm includes:

[0068] S21, data preparation: randomly select K candidate low-altitude manned aircraft sites;

[0069] S22, traverse the positions of all initially selected low-altitude manned aircraft sites, and calculate the distance between each initially selected low-altitude manned aircraft site and the candidate low-altitude manned aircraft sites according to the Euclidean distance;

[0070] S23, form new clusters: assign each initially selected low-altitude manned aircraft site to the candidate low-altitude manned aircraft site with the closest distance. At this time, a candidate low-altitude manned aircraft site is assigned some initially selected low-altitude manned aircraft sites, that is, a new cluster is formed;

[0071] S24: update the centroid: calculate the new centroid of all initially selected low-altitude manned aircraft sites in the new cluster;

[0072] Return to S21 until the centroid no longer changes or exceeds the specified number of iterations, and use all the new centroids as candidate low-altitude manned aircraft sites again.

[0073] Compared with the prior art, the present invention has the following beneficial effects:

[0074] 1) The present invention proposes a three-stage method for selecting low-altitude manned aircraft sites, which deterministically screens out initially selected low-altitude manned aircraft sites according to control factors; considering the demand of all initially selected low-altitude manned aircraft sites, clustering candidate low-altitude manned aircraft sites based on the initially selected low-altitude manned aircraft sites; using the Newton-Raphson optimization algorithm to solve the site selection model to determine the low-altitude manned aircraft site address, where the model considers the hierarchical service range, payload, airspace control, user demand, and cost of low-altitude manned aircraft, and finally comprehensively considers multiple factors to achieve scientific, reasonable and fast-converging site planning.

[0075] 2) The control matrix based on the Hadamard product in the present invention accurately excludes no-fly zones and improves the compliance of site selection.

[0076] 3) The present invention considers taxi users as potential low-altitude manned aircraft users, takes the taxi data near each initially selected low-altitude manned aircraft site as its user demand, and determines the user demand of different low-altitude manned aircraft sites based on the service range of three levels of initially selected low-altitude manned aircraft, which is beneficial to improving the utilization rate of low-altitude resources and also provides strong data support for demand prediction and subsequent modeling.

[0077] 4) The present invention improves the Newton-Raphson optimization algorithm, achieving greater improvements in convergence speed and obtaining the optimal solution compared to the particle swarm and PSO algorithms, resulting in a faster solution speed and more accurate site selection for the stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 is a schematic flowchart of the method for site selection of stations in the present invention;

[0079] Figure 2 is a schematic flowchart of using the Newton-Raphson optimization algorithm to solve the site selection model of the stations in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0080] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0081] Embodiment 1

[0082] This embodiment relates to a method for site selection of low-altitude manned aircraft stations, as Figure 1 , and the method includes the following steps:

[0083] S1: Construct potential demand points and obtain the preliminary selected low-altitude manned aircraft stations. Determine the addresses of nine categories of urban POI data such as company enterprises, shopping services, and transportation facility services as potential demand points, extract four types of control factors including classified areas, electromagnetic areas, natural and cultural areas, and flammable areas, and based on the Hadamard product control matrix, screen out the preliminary selected low-altitude manned aircraft stations;

[0084] S2: Cluster to obtain candidate low-altitude manned aircraft stations. Through the clustering algorithm, cluster the preliminary selected low-altitude manned aircraft stations with similar locations into one category, and use the centroid point of this category of preliminary selected low-altitude manned aircraft stations as the candidate low-altitude manned aircraft stations;

[0085] S3: Obtain the user demand volume of each candidate low-altitude manned aircraft station and construct a site selection model for low-altitude manned aircraft stations. With the candidate low-altitude manned aircraft station as the center, construct a hierarchical service range model based on the payload of the low-altitude manned aircraft, and convert the taxi data within the range into the user demand volume of each candidate low-altitude manned aircraft station through model calculation, that is, the user demand volume of the location where the candidate low-altitude manned aircraft station is located, which is the user demand volume of those who travel by low-altitude manned aircraft. Finally, construct a site selection model for the low-altitude manned aircraft based on the user demand volume of each candidate low-altitude manned aircraft station and its construction cost.

[0086] S4: Solve the site selection model using the Newton - Raphson optimization algorithm and evaluate the demand satisfaction rate of the site location. Solve the low - altitude manned aircraft site selection model using the Newton - Raphson optimization algorithm, calculate the site location, and evaluate the demand satisfaction rate of the site location.

[0087] The process of constructing potential demand points and obtaining the preliminary selected low - altitude manned aircraft sites in S1 includes:

[0088] S11: Obtain potential demand points: Collect locations with high population density and certain consumption potential as potential demand points, that is, use 9 categories of urban POI features, namely company enterprises, shopping services, transportation facility services, financial insurance services, science, education and culture services, commercial residences, sports and leisure services, medical and health services, and accommodation services, and set the potential demand points as u.

[0089] S12: Construct a Hadamard product control matrix to determine the preliminary selected low - altitude manned aircraft sites.

[0090] Denote the most east longitude and latitude coordinate point of City A as (x E , y E ), the most west longitude and latitude coordinate point as (x W , y W ), the most north longitude and latitude coordinate point as (x N , y N ), and the most south longitude and latitude coordinate point as (x S , y S ). The formulas are as follows:

[0091] m = x E - x W

[0092] n = y N - y S

[0093] x1 = x W

[0094] x m = x E

[0095] y1 = y N

[0096] y n = y S

[0097] AX = (x1,…, x m )

[0098] AY = (y1,…, y n )

[0099] i = x - (xw -1) x ∈ AX

[0100] j = y - (y n -1) y ∈ AY

[0101]

[0102] where m is the difference between the longitude of the easternmost coordinate point and the longitude of the westernmost coordinate point in City A, n is the difference between the latitude of the northernmost coordinate point and the latitude of the southernmost coordinate point in City A, AX = (x W , …, x E ), AY = (y N , …, y S ) are the latitude and longitude vectors of City A respectively, x and y are the longitude and latitude of the coordinates respectively, i and j are the matrix subscripts after the x and y coordinate conversions, A mn is the coordinate schematic matrix of City A, represents the coordinate schematic matrix of k different control factors in City A, and H mn represents the Hadamard product control matrix.

[0103] Assume that the longitude and latitude coordinates are all integers. By finding the four outermost coordinate points in City A, the longitude and latitude vectors of City A are constructed. Since the address of City A remains unchanged, the sizes of all schematic matrices (coordinate schematic matrix, control factor coordinate schematic matrix) are the same.

[0104] Assume that the coordinates of a point are x and y at this time. Through conversion, the matrix subscripts i and j corresponding to this point are obtained. When (x, y) is a coordinate point in City A, the element η mn of the coordinate schematic matrix A ij of City A is assigned a value of 1, otherwise, 0. Similarly, it is judged whether (x, y) is a coordinate point of the kth control factor. After converting all the coordinates of City A, the coordinate schematic matrix A mn of City A and the coordinate schematic matrices of k different control factors in City A are finally obtained. The Hadamard product of all the control factor coordinate schematic matrices and the coordinate schematic matrix of City A is calculated to obtain the Hadamard product control matrix. Among them, the points with data of 0 are no-fly points, and vice versa, they are the preliminary selected low-altitude manned aircraft sites. Finally, v preliminary selected low-altitude manned aircraft sites are screened out.

[0105] Determine the candidate low-altitude manned aircraft sites in S2: Obtain the candidate low-altitude manned aircraft sites through clustering.

[0106] Through the KMeans clustering algorithm, the preliminary selected low-altitude manned aircraft sites with similar geographical locations are clustered into different clusters, and the centroid points of these clusters are obtained and used as the candidate low-altitude manned aircraft sites. The specific steps and mathematical formulas are as follows:

[0107] S21, Data Preparation: Randomly select K candidate low-altitude manned aircraft stations.

[0108] S22, Calculate Distances: Traverse the positions of all initially selected low-altitude manned aircraft stations, and calculate the distance between each initially selected low-altitude manned aircraft station and the candidate low-altitude manned aircraft stations according to the Euclidean distance. The calculation formula is as follows:

[0109]

[0110] In the formula, lk is the distance between the initially selected low-altitude manned aircraft station with coordinates (x i , y j ) and the k-th candidate low-altitude manned aircraft station. CX k and CY k respectively represent the latitude and longitude of the k-th candidate low-altitude manned aircraft station. K is the number of candidate low-altitude manned aircraft stations, that is, calculate the Euclidean distance between each candidate low-altitude manned aircraft station and each initially selected low-altitude manned aircraft station.

[0111] S23, Form New Clusters: Assign each initially selected low-altitude manned aircraft station to the nearest one among the K candidate low-altitude manned aircraft stations. At this time, a candidate low-altitude manned aircraft station is assigned some initially selected low-altitude manned aircraft stations, that is, a new cluster is formed. Denote the K new clusters as δ k , as follows:

[0112] δ k = {(x i , y j ) t} 1 ≤ t ≤ v, k ∈ [1, K]

[0113] In the formula, t represents the number of initially selected low-altitude manned aircraft stations in the k-th cluster, and v is the number of low-altitude manned aircraft stations. In addition, denote the latitude of the initially selected low-altitude manned aircraft stations in each new cluster as and the longitude as

[0114] S24: Update the Centroid: Calculate the new centroid of all initially selected low-altitude manned aircraft stations in a new cluster. The calculation formula is as follows:

[0115]

[0116] Calculate the new centroids of all K new clusters in the same way, and use these K new centroids as the candidate low-altitude manned aircraft stations again;

[0117] Return to S21 until the centroid no longer changes or exceeds the specified number of iterations.

[0118] In S3, the sub-steps for obtaining the user demand volume of each candidate low-altitude manned aircraft site are as follows:

[0119] S31: Introduce a hierarchical service scope model and construct the user demand volume of the candidate low-altitude manned aircraft site in combination with the payload of the low-altitude manned aircraft.

[0120] First, with all the primary selected low-altitude manned aircraft sites as the centers and r3 as the radius, divide v demand circles. Use the taxi data of each demand circle as the user demand volume of the candidate low-altitude manned aircraft site. The types of vehicles include taxis, online car-hailing, etc. Calculate the user demand volume of all the primary selected low-altitude manned aircraft sites. The formula is as follows:

[0121] n k ={o t}

[0122] o t =SCar + ECar × MCar 1 ≤ t ≤ v, k ∈ [1, K]

[0123] In the formula, n k is the set of user demand volumes of the primary selected low-altitude manned aircraft sites in different clusters, o t is the user demand volume of the t-th primary selected low-altitude manned aircraft site in a cluster. SCar is the number of orders starting from a certain location within the demand circle, and ECar and MCar are the number of orders and the amount with a certain location within the demand circle as the end point respectively.

[0124] In addition, since the service distance of the candidate low-altitude manned aircraft site is limited and may not be able to meet the demands of all sites in the cluster, the user demand volume of the candidate low-altitude manned aircraft site is recalculated in combination with the payload of the low-altitude manned aircraft. The formula is as follows:

[0125] S k ={s t}

[0126]

[0127] R=(r1, r2, r3)

[0128]

[0129] In the formula, S k represents the set of distances between the primary selected low-altitude manned aircraft sites and the candidate low-altitude manned aircraft sites in each cluster. CX k , CY k represent the latitude and longitude of the k-th candidate low-altitude manned aircraft site respectively. Represent the latitude and longitude of the initially selected low-altitude manned aircraft sites in each new cluster respectively. R = (r1, r2, r3) is the service range of candidate low-altitude manned sites based on three levels (r1, r2, r3) set according to the different payloads of low-altitude manned aircraft, N k is the set of user demand quantities of candidate low-altitude manned aircraft sites, that is, the set of user demand quantities of low-altitude manned sites based on the hierarchical service range.

[0130] For the k-th candidate low-altitude manned aircraft site, its user demand quantity is the sum of the user demand quantities of all initially selected low-altitude manned aircraft sites within the service range. Within the r1 range, all aircraft models are flyable and can meet the demand quantities of all initially selected low-altitude manned aircraft sites. t . Within the range of r1 and r2, medium and heavy low-altitude manned aircraft are flyable, that is, the demand quantities of some initially selected low-altitude manned aircraft sites cannot be met. Set the demand reduction coefficient β1 to obtain the demand quantity β1 of the finally satisfied potential demand points. t . Within the range of r2 and r3, only heavy low-altitude manned aircraft are flyable, and the demand quantities of some potential demand points cannot be met. Set the demand reduction coefficient β2 to obtain the demand quantity β2 of the finally satisfied potential demand points. t . Outside the r3 range, this low-altitude manned aircraft site cannot meet the demand quantity of potential demand points, so its value is 0. Thus, the user demand quantities of K candidate low-altitude manned aircraft sites are obtained.

[0131] S32: According to the coordinate address of the candidate low-altitude manned aircraft site, obtain the land price of this place and construct a site selection model with large demand and small cost. Its mathematical expression is as follows:

[0132] D = (d1, …, d K )

[0133]

[0134] In the formula, D is the land price vector of candidate low-altitude manned aircraft sites, and d k is the land price of the k-th candidate low-altitude manned aircraft site. is the initialized low-altitude manned aircraft site, where is a binary variable, that is, if then select this point as the low-altitude manned aircraft site, otherwise, abandon this point as the low-altitude manned aircraft site.

[0135] In S4, the process of determining the low-altitude manned aircraft site based on the site selection model of the low-altitude manned aircraft using the Newton-Raphson optimization algorithm is as follows:

[0136] The Newton-Raphson based optimizer (NRBO) is used to solve this model. Through two operators, the Newton-Raphson search rule (NRSR) and the Trap Avoidance Operator (TAO), the search domain is explored, and the Newton-Raphson Method (NRM) is applied to discover the search area, thereby defining the search path. Among them, the Newton-Raphson optimization algorithm is a new type of meta-heuristic algorithm (intelligent optimization algorithm), and its performance and convergence speed far exceed those of the traditional particle swarm algorithm, with higher solution accuracy, far exceeding the PSO algorithm in smooth and low-dimensional optimization problems.

[0137] As Figure 2 , the specific steps are as follows:

[0138] S41: Population initialization, constructing an initialization matrix Its mathematical expression is as follows:

[0139]

[0140] Among them, ub and lb are the upper and lower limits of the population search position, N p is the initialized population size, and k is the dimension, that is, the candidate low-altitude manned sites.

[0141] S42: Newton-Raphson search, NRSR is based on NRM. NRM is proposed to promote the exploration trend and accelerate convergence. Starting from a supposed initial solution, it advances along a definite direction to the next position. This method solves nonlinear equations and is an approximate method for linearizing nonlinear equations. Expand the function into a Taylor series in the neighborhood of the point : Take the first two terms and let

[0142]

[0143] to get: We have:

[0144]

[0145] Transforming the minimization of the objective function into finding the solution where the derivative of the objective function is zero, that is, finding the solution of , the formula is:

[0146]

[0147] Furthermore,

[0148]

[0149]

[0150] Among them, is the best solution obtained so far, is the current solution. Compared with after substituting and into the objective function, the better one is set as The value of the worse one is set as Next:

[0151]

[0152] Among them, r1 represents a random number between (0, 1), the Mean method is to find the mean value of the function, randn represents a normally distributed random number with a mean of 0 and a variance of 1, y w and y b are the positions of two vectors generated using Z k+1 and to calculate NRSR. Next, the exploitation of the proposed NRBO is improved by including another parameter called ρ, which guides the population in the correct direction. The expression of ρ is as follows:

[0153]

[0154] Among them, a and b are random numbers between (0, 1), is the best solution obtained so far, is the current solution, and the subscripts k1 and k2 are different integers randomly selected from the population.

[0155] According to experience, the proposed algorithm must be able to achieve a balance between diversity and aggregation in order to find the optimal solution in the search space and finally converge to the global solution. The algorithm can be enhanced by applying an adaptive coefficient called δ, and the expression of δ is as follows:

[0156]

[0157] Among them, t represents the number of the current iteration, and Max_t represents the maximum number of iterations. In order to maintain a balance between the exploration and exploitation phases, the parameter δ adjusts itself during the iteration process, and the value of δ varies between 1 and -1.

[0158]

[0159]

[0160] Finally, a new vector for the next iteration is obtained

[0161] S43: TAO is a technique that enhances the performance of optimization algorithms by introducing randomness and diversity. It helps the algorithm escape from local optima and improve global search ability by combining the current solution and the best solution and generating new solutions using random parameters. The TAO operator significantly changes the position by combining the best vector the current vector to produce a solution with enhanced quality If the value of the random number is less than DF, it is calculated as follows

[0162]

[0163] μ1 = 3β * rand+(1 - β)

[0164] μ2 = β * rand+(1 - β)

[0165]

[0166] where Δ is a control parameter used to adjust the generation method of the random parameters μ1 and μ2, and can be defined as the distance metric between the current solution and the current optimal solution . is the Euclidean distance between the current solution and the optimal solution. is the norm of the optimal solution (for normalization). rand represents a uniform random number between (0, 1), θ1 and θ2 are uniform random numbers between (-1, 1) and (-0.5, 0.5) respectively, DF represents the determinant that controls the performance of NRBO, μ1 and μ2 are random parameters, and β represents a binary number, i.e., 1 or 0. Due to the randomness in the selection of the parameters μ1 and μ2, the population becomes more diverse and escapes from local optima, which helps to improve its diversity.

[0167] The degree of optimization is controlled by setting the maximum number of iterations. The optimization is performed according to the Newton - Raphson search rule, and the trap avoidance operator is used to avoid local optima to obtain the optimal solution guided by the objective function.

[0168] The selected sites are evaluated, and the user satisfaction rate is calculated. Its mathematical expression is as follows:

[0169]

[0170] τ k = N k / B k

[0171] where Bk is the total user demand for the k-th candidate low-altitude manned aircraft site, τ represents the user demand satisfaction rate of the k-th candidate low-altitude manned aircraft site, and N k is the set of user demands for candidate low-altitude manned aircraft sites.

[0172] Embodiment 2

[0173] This embodiment also relates to a method for site selection of low-altitude manned aircraft, and a specific scheme of this method is illustrated by an example.

[0174] S1. Obtain potential demand points, including the following steps:

[0175] S11. Collect locations with high population density and certain consumption potential as potential demand points, that is, adopt 9 major categories of urban POI features, namely corporate enterprises, shopping services, transportation facility services, financial insurance services, science and education cultural services, commercial residences, sports and leisure services, medical and health care services, and accommodation services, and set the potential demand points as u. Among them, the types selected for each major category of urban POI features are shown in Table 1:

[0176] Table 1

[0177]

[0178]

[0179]

[0180] S12. Construct a Hadamard product control matrix to determine the primary selected low-altitude manned aircraft sites.

[0181] Low-altitude manned aircraft are restricted in some airspaces. In the present invention, the airspace types where low-altitude manned aircraft are prohibited from passing are called control factors, including but not limited to those shown in Table 2.

[0182] Table 2

[0183]

[0184] S2. Determine the candidate low-altitude manned aircraft sites.

[0185] S3. Obtain the user demand for each candidate low-altitude manned aircraft site, including:

[0186] S31: Introduce a hierarchical service range model, and combine with the effective payload of the low-altitude manned aircraft to construct the user demand for the candidate low-altitude manned aircraft site.

[0187] Since the service distance of the candidate low-altitude manned aircraft sites is limited and may not be able to meet the needs of all sites in the cluster, the effective payload of the low-altitude manned aircraft is combined. It is known that according to the design payload dimension, eVTOL can be divided into three categories: light, medium, and heavy, as shown in Table 3 specifically.

[0188] Table 3

[0189] Payload of the eVTOL Design Load Latitude Range Light: 100 - 200 kg 35 km Medium: 300 - 500 kg 60 km Heavy ≥ 1000 kg 250 km

[0190] The flight ranges of eVTOLs with different payloads are different, so the service scopes of their sites must vary. The site service scope of an eVTOL with a longer flight range is larger. Apply this classification method to all low-altitude manned aircraft, and set the service scopes of low-altitude manned sites based on three levels, and recalculate the user demand of the candidate low-altitude manned aircraft sites.

[0191] S32: Obtain the land price of this place according to the coordinate address of the candidate low-altitude manned aircraft site, and construct a model with high demand and low cost.

[0192] S4. Determine the low-altitude manned aircraft site based on the low-altitude manned aircraft site selection model using the Newton-Raphson optimization algorithm.

[0193] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for site selection of a low-altitude manned aircraft, characterized in that The method includes the following steps: S1: Obtain the primary low-altitude manned aircraft sites based on potential demand points and regulatory factors; S2: Obtain the candidate low-altitude manned aircraft sites through a clustering algorithm; S3: Introduce a hierarchical service range model, and combine with the effective payload of the low-altitude manned aircraft to obtain the user demand of each candidate low-altitude manned aircraft site, and construct a site selection model for low-altitude manned aircraft sites; S4: Use the Newton-Raphson optimization algorithm to solve the site selection model, where the objective function of the site selection model is: the maximum user demand and the minimum cost.

2. The site selection method for a low-altitude manned aircraft according to claim 1, wherein The process of S1 includes: Construct the potential demand points based on urban POI data; Construct a Hadamard product control matrix to determine the primary low-altitude manned aircraft sites.

3. The site selection method of a low-altitude manned aircraft according to claim 2, wherein, The process of determining the primary low-altitude manned aircraft sites includes Construct a Hadamard product control matrix, specifically: Denote the easternmost longitude and latitude coordinate point of City A as (x E , y E ), the westernmost longitude and latitude coordinate point as (x W , y W ), the northernmost longitude and latitude coordinate point as (x N , y N ), and the southernmost longitude and latitude coordinate point as (x S , y S ). The formula is as follows: m = x E -x W n = y N -y S x1 = x W x m = x E y1 = y N y n = y S AX = (x1, …, x m ) AY = (y1, …, y n ) i = x - (x w - 1) x ∈ AX j = y - (y n - 1) y ∈ AY A mn = {η ij} where m is the difference in longitude between the easternmost and westernmost coordinate points of City A, n is the difference in latitude between the northernmost and southernmost coordinate points of City A, AX = (x W , …, x E ), AY = (y N , …, y S ) are the latitude and longitude vectors of City A respectively, x and y are the longitude and latitude of the coordinates respectively, i and j are the matrix subscripts after the x and y coordinate conversions respectively, A mn is the coordinate schematic matrix of City A, represents the coordinate schematic matrix of k different control factors in City A; Assume that the coordinates of a certain point are x and y, and through conversion, the subscripts i and j of its corresponding matrix are obtained. When (x, y) is the coordinate point of City A, the element η mn of the schematic matrix A of the coordinates of City A ij is assigned 1, and vice versa, 0; Use the same method to determine whether the coordinate point (x, y) is the coordinate point of the kth regulatory factor; Convert all the coordinates of City A to obtain the coordinate schematic matrix A of City A mn and the coordinate schematic matrix of k different control factors in City A Then perform the Hadamard product on all the coordinate schematic matrices of the control factors and the coordinate schematic matrix of City A to obtain the Hadamard product control matrix H mn : Among them, the point with data 0 is a no-fly zone, otherwise, it is a primary low-altitude manned aircraft site, and finally v primary low-altitude manned aircraft sites are selected.

4. The site selection method for a low-altitude manned aircraft according to claim 1, wherein The process of obtaining the user demand of each candidate low-altitude manned aircraft site includes: taking all the primary low-altitude manned aircraft sites as the centers and the maximum service range r3 of the candidate low-altitude manned sites as the radius, dividing v demand circles, and using the taxi data of each demand circle as the user demand of the candidate low-altitude manned aircraft site, and calculating the user demand of all the primary low-altitude manned aircraft sites. The formula is as follows: n k = {o t} o t = SCar + ECar × MCar for 1 ≤ t ≤ v, k ∈ [1, K] Where n k is the set of user demand quantities for the initially selected low-altitude manned aircraft sites in different clusters, and o t is the user demand quantity for the t-th initially selected low-altitude manned aircraft site in a cluster. SCar is the number of orders starting from a certain location within the demand circle, and ECar and MCar are the number of orders and the amount with a certain location within the demand circle as the end point, respectively. v is the number of initially selected low-altitude manned aircraft sites.

5. A site selection method for a low-altitude manned aircraft according to claim 4, characterized in that The process of obtaining the user demand of each candidate low-altitude manned aircraft site also includes: combining with the effective payload of the low-altitude manned aircraft, and recalculating the user demand of the candidate low-altitude manned aircraft site. The formula is as follows: S k = {s t} R＝(r1,r2,r3) In the formula, S k represents the set of distances between the primary low-altitude manned aircraft sites and the candidate low-altitude manned aircraft sites in each cluster. CX k , CY k represent the latitude and longitude of the k-th candidate low-altitude manned aircraft site respectively. represent the latitude and longitude of the primary low-altitude manned aircraft sites in each cluster respectively; s t is the Euclidean distance between the t-th primary low-altitude manned aircraft site and the k-th candidate low-altitude manned aircraft site; R = (r1, r2, r3) is the service range of candidate low-altitude manned sites set based on three levels according to the different payloads of low-altitude manned aircraft. N k is the set of user demand quantities of candidate low-altitude manned aircraft sites, that is, the set of user demand quantities of low-altitude manned sites based on the hierarchical service range; both β1 and β2 are demand reduction coefficients.

6. The site selection method of a low-altitude manned aircraft according to claim 5, characterized in that, The process of obtaining the user demand of each candidate low-altitude manned aircraft site also includes: obtaining the corresponding land price according to the coordinate address of the candidate low-altitude manned aircraft site, and constructing a site selection model with large demand and small cost. Specifically: D = (d1, …, d K ) where D is the land price vector of candidate low-altitude manned aircraft sites, and d k is the land price of the k-th candidate low-altitude manned aircraft site, is the initialized low-altitude manned aircraft site, where is a binary variable, that is, if then select this point as the low-altitude manned aircraft site, otherwise, abandon this point as the low-altitude manned aircraft site.

7. A site selection method for a low-altitude manned aircraft according to claim 5, characterized in that The method also includes evaluating the demand satisfaction rate of the site location after calculating the site location. Specifically: τ k = N k / B k Among them, B k is the total user demand of the k-th candidate low-altitude manned aircraft site, τ represents the user demand satisfaction rate of the k-th candidate low-altitude manned aircraft site, and N k is the set of user demands of the candidate low-altitude manned aircraft sites.

8. A site selection method for a low-altitude manned aircraft according to claim 1, characterized in that The process of using the Newton-Raphson optimization algorithm to solve the site selection model includes: Population initialization: Construct an initial matrix of low-altitude manned aircraft stations The mathematical expression is as follows: wherein, ub and lb are respectively the upper and lower limits of the population search position; N p is the number of different combinations of low-altitude manned aircraft sites; K is the dimension, that is, the number of candidate low-altitude manned sites; Improve the Newton-Raphson search, find the optimal solution in the search monitoring, and finally converge to the global solution; the improvement is specifically: Define the parameter ρ to guide the population in the correct direction. Specifically: where a and b are random numbers between (0, 1), is the best solution obtained so far, is the current solution, and the subscripts k1 and k2 are different integers randomly selected from the population; Design an adaptive coefficient δ to enhance the algorithm. Specifically: Where t represents the current number of iterations, and Max_t represents the maximum number of iterations; In order to maintain a balance between the exploration and development stages, the parameter δ will adjust itself during the iteration process, and the value of δ changes between 1 and -1; Among them, and are the new vector positions obtained by updating , NRSR represents the Newton-Raphson search rule, y w and y b are the positions of two vectors generated using Z k+1 and ; is the current worst solution, is the objective function, is the difference between the best solution and the current solution, and λ represents a random number between (0, 1); Finally, a new vector for the next iteration is obtained 9. The site selection method of a low-altitude manned aircraft according to claim 8, characterized in that The process of using the Newton-Raphson optimization algorithm to solve the site selection model also includes: introducing a trap avoidance operator to obtain the optimal solution guided by the objective function.

10. The site selection method of a low-altitude manned aircraft according to claim 1, characterized in that, The process of obtaining the candidate low-altitude manned aircraft sites through a clustering algorithm includes: S21, Data preparation: Randomly select K candidate low-altitude manned aircraft sites; S22. Traverse the positions of all initially selected low-altitude manned aircraft sites, and calculate the distance between each initially selected low-altitude manned aircraft site and the candidate low-altitude manned aircraft site according to the Euclidean distance; S23. Form new clusters: Assign each initially selected low-altitude manned aircraft site to the candidate low-altitude manned aircraft site with the closest distance. At this time, a candidate low-altitude manned aircraft site is assigned some initially selected low-altitude manned aircraft sites, thus forming a new cluster; S24: Update the centroid: Calculate the new centroid of all initially selected low-altitude manned aircraft sites in the new cluster; Return to S21 until the centroid no longer changes or exceeds the specified number of iterations, and use all the new centroids as candidate low-altitude manned aircraft sites again.

Images