Method, apparatus and device for selecting slow traffic station location based on multi-model coupling

Abstract

The present invention relates to a method, device and equipment for site selection of a slow-moving traffic station based on multi-model coupling, and belongs to the technical field of site selection of traffic stations. The method, device and equipment loosely couple a two-step mobile search model, an immune optimization algorithm model, an investment allocation model and a single judgment model by means of multi-model coupling, thereby constructing a site selection model; by using comprehensive analysis technology by means of model coupling, the two-step mobile search model is improved, and spatial element site selection indicators are added by means of the model, and then the immune optimization algorithm site selection result is optimized with the help of the investment allocation model and the single judgment model; finally, by means of multi-model coupling, the problems of one-sided site selection and excessive scaling workload caused by too few or too many influencing factors are optimized, and a relatively reasonable site selection result can be obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic station location, and particularly relates to a slow traffic station location method, device and equipment based on multi-model coupling. Background Art

[0002] There are more and more studies on the field of low-carbon slow traffic at home and abroad. In terms of the location of slow traffic stations, domestic and foreign scholars have conducted research on location methods to provide a basis for determining the relationship between the demand and supply of slow traffic stations.

[0003] In traditional technologies, several GIS location methods such as the analytic hierarchy process, gravity method, network coverage model, and simulation method are usually used for location. However, traditional location methods use a single model and determine the final location by increasing different location indicators. Too many location indicators will lead to time-consuming and laborious work, while too few location indicators will lead to unreasonable location results. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a slow traffic station location method, device and equipment based on multi-model coupling to overcome the problem that the current method uses a single model and determines the final location by increasing different location indicators. Too many location indicators will lead to time-consuming and laborious work, while too few location indicators will lead to unreasonable location results.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] On the one hand, a slow traffic station location method based on multi-model coupling includes:

[0007] Determine the facility reach range data of the target area; the facility reach range data includes the bus stop data within the maximum range corresponding to the straight-line distance of each travel mode;

[0008] Input the facility reach range data into the two-step floating catchment area model to obtain the rail transit accessibility and the supply-demand ratio; wherein, the supply-demand ratio is the ratio of the service supply of each bus stop to the number of potential demand people;

[0009] Determine the usage demand of shared bicycles in the target area, and determine the estimated demand points according to the usage demand and the rail transit accessibility;

[0010] Determine the statistical data related indicators of the target area, and calculate the demand according to the statistical data related indicators;

[0011] Determine the existing slow traffic network data and determine the current demand points;

[0012] Input the estimated demand points, demand quantities, and current demand points into the immune optimization algorithm model to obtain new demand quantities;

[0013] Input the new demand quantities into the combined model to obtain the optimal traffic station allocation quantities; wherein, the combined model is a combined model of an investment allocation model and a single judgment model;

[0014] Calculate the most optimal site location based on the urban road network data and urban building contour data of the target area and the optimal traffic station allocation quantities.

[0015] Optionally, the determining the facility reach range data of the target area includes:

[0016] Obtain rail transit station data and perform buffer processing for a preset distance;

[0017] Determine slow traffic station data, where the slow traffic station data includes: starting points and destination points; wherein, the starting points are rail transit station data, and the target points are mobile phone signaling data, parking lots, and bus station data;

[0018] Calculate the straight-line distance from rail transit stations to slow traffic stations within the maximum transfer capacity range; wherein, the maximum transfer capacity range includes: a walking distance of 1 km, a cycling distance of 2 km, and a bus travel distance of 5 km;

[0019] Determine the public transportation station data within the maximum transfer capacity range corresponding to the straight-line distance of each travel mode, and use the public transportation station data within the maximum transfer capacity range corresponding to the straight-line distance of each travel mode as the facility reach range data.

[0020] Optionally, the inputting the facility reach range data into the two-step floating catchment area method model to obtain rail transit-bus accessibility and supply-demand ratio includes:

[0021] Determine the ratio of the service supply of each existing bus station to the number of potential demand people as the supply-demand ratio;

[0022] Determine the bus stations that can provide bus services for rail stations, and calculate the rail transit-bus accessibility based on the supply-demand ratio and the bus stations.

[0023] Optionally, the determining the bike-sharing usage demand in the target area and determining the estimated demand points according to the usage demand and the rail transit-bus accessibility includes:

[0024] Obtain the bike-sharing usage demand data in the target area, perform density clustering on the usage demand data to obtain clustering points; merge each clustering point with the existing stations as preselected demand points;

[0025] Compare and screen the rail transit accessibility with the preselected demand points, and ignore the preselected demand points with rail transit accessibility lower than the preset threshold to obtain the estimated demand points.

[0026] Optionally, determining the statistical data related indicators of the target area and calculating the demand quantity according to the statistical data related indicators includes:

[0027] Determine the statistical data related indicators of the target area, and the statistical data related indicators include: age, income, travel scale, education level, and shared travel experience;

[0028] Calculate the weight of each statistical data related indicator;

[0029] Calculate the product of the coefficient of each statistical data related indicator and the corresponding weight, and sum all the products. Multiply the sum of all the products by the population quantity of the target area to obtain the demand group data;

[0030] Calculate the demand quantity according to the demand group data.

[0031] Optionally, determining the existing slow traffic network data and determining the current demand points includes:

[0032] Obtain the existing slow traffic network data based on an open platform;

[0033] Delete the unnecessary stations in the existing slow traffic network data according to the actual road network situation and the government's urban planning policies to obtain the current demand points.

[0034] Optionally, the combined model is a combined model constructed by using 0.62 and 0.38 as the weights of the investment allocation model and the single judgment model respectively as the combined weights.

[0035] On the other hand, a slow traffic station location device based on multi-model coupling includes:

[0036] A preprocessing module for determining the facility reach range data of the target area; the facility reach range data includes bus stop data within the maximum range corresponding to the straight-line distance of each travel mode;

[0037] A two-step moving search module for inputting the facility reach range data into a two-step moving search model to obtain rail transit accessibility and supply-demand ratio; wherein, the supply-demand ratio is the ratio of the service supply of each bus stop to the number of potential demand people;

[0038] A first determination module for determining the vehicle usage demand of shared bicycles in the target area and determining the estimated demand points according to the vehicle usage demand and the rail transit accessibility;

[0039] A second determination module is used to determine the statistical data related indicators of the target area, and calculate the demand according to the statistical data related indicators;

[0040] The third determination module is used to determine the existing network point data of slow traffic and determine the current demand points;

[0041] An immune optimization module is used to input the estimated demand point, demand amount and current demand point into the immune optimization algorithm model to obtain a new demand amount;

[0042] An allocation module, used for inputting the new demand into a combination model to obtain an optimal traffic station allocation; wherein the combination model is a combination model of an investment allocation model and a single judgment model;

[0043] The calculation module is used to calculate the most preferred site location based on the urban road network data and urban building outline data of the target area and the optimal transportation station allocation.

[0044] On the other hand, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute any of the above-mentioned methods for selecting a slow-moving traffic station based on multi-model coupling.

[0045] In another aspect, a device for selecting a site for a slow-moving traffic station based on multi-model coupling includes a processor and a memory, wherein the processor is connected to the memory:

[0046] Wherein, the processor is used to call and execute the program stored in the memory;

[0047] The memory is used to store the program, and the program is used to execute at least any one of the above-mentioned slow-moving traffic station site selection methods based on multi-model coupling.

[0048] The technical solution provided by the present invention includes at least the following beneficial effects:

[0049] The two-step mobile search model, immune optimization algorithm model, investment allocation model and single judgment model are loosely coupled through multi-model coupling to construct a site selection model; the two-step mobile search model is improved by using comprehensive analysis technology through model coupling, and the spatial element site selection index is added through this model, and then the site selection results of the immune optimization algorithm are optimized with the help of the investment allocation model and the single judgment model; finally, the one-sidedness of site selection and excessive scaling workload caused by too few or too many considerations of influencing factors are optimized through multi-model coupling, and a more reasonable site selection result can be obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0051] Figure 1 It is a schematic flowchart of a method for selecting a slow traffic station site with multi-model coupling provided by an embodiment of the present invention;

[0052] Figure 2 It is a schematic diagram of the convergence curve of an immune optimization algorithm provided by an embodiment of the present invention;

[0053] Figure 3 It is a schematic diagram of the network points in the area before optimization provided by the verification embodiment of the present invention, where the squares are the existing network points;

[0054] Figure 4 Provided by an embodiment of the present invention Figure 3 Schematic diagram of the network points in the optimized area, where the dots are the optimized network points;

[0055] Figure 5 It is a schematic structural diagram of a device for selecting a slow traffic station site with multi-model coupling provided by an embodiment of the present invention;

[0056] Figure 6 It is a schematic structural diagram of a device for selecting a slow traffic station site based on multi-model coupling provided by an embodiment of the present invention. Detailed implementation manners

[0057] To make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other implementation manners obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by the present invention.

[0058] In the aspect of mature slow traffic data mining and analysis, the mining of massive traffic operation data is insufficient, resulting in a certain degree of waste of data resources. The fusion depth of spatio-temporal data, traffic situation data and statistical data is not enough, and the application analysis of data mainly focuses on methods such as the analytic hierarchy process, the gravity method, the network coverage model and the simulation method for site selection.

[0059] In the prior art, most studies on the location selection of slow traffic networks make location decisions by establishing an evaluation system for slow traffic stations. Some studies conduct network layout by establishing a target model or a bi-level programming model, but less spatial elements are considered in the process of model construction. That is, in traditional technologies, location selection is usually carried out by several GIS location selection methods such as the analytic hierarchy process, the gravity method, the network coverage model, and the simulation method. However, traditional location selection methods use a single model and determine the final location by adding different location selection indicators. Too many location selection indicators will lead to time-consuming and laborious work, while too few location selection indicators will lead to unreasonable location selection results.

[0060] Based on this, the present invention couples a two-step mobile search model, an immune optimization algorithm model, an investment allocation model, and a single judgment model to perform the location selection of slow traffic stations.

[0061] Figure 1 For the flowchart of a multi-model coupled slow traffic station location selection method provided by an embodiment of the present invention, refer to Figure 1 , the method provided by the embodiment of the present invention may include the following steps:

[0062] Step S1: Determine the data of the facility reach range in the target area; the data of the facility reach range includes the data of public transport stations within the maximum range corresponding to the straight-line distance of each travel mode.

[0063] Specifically, preprocessing before the two-step mobile search model can be performed, which may specifically include the following steps:

[0064] Step (1): Obtain the rail transit station data of the Amap open platform through python technology and perform buffer processing within 200m (that is, irradiate traffic facilities within a radius of 200m with the rail transit station as the center).

[0065] Step (2): Select the slow traffic station data. Here, the slow traffic stations are divided into the starting point and the destination point. The starting point is the rail transit station data, and the destination point is the mobile phone signaling data (for example, it can be provided by a data technology company), the parking lot, and the bus stop data (the parking lot and bus stop data can come from the Amap open platform).

[0066] Step (3): Obtain the straight-line distance from the rail transit station to the slow traffic station within the maximum transfer capacity range (walking distance of 1km, cycling distance of 2km, and bus driving distance of 5km) by constructing an OD (origin-destination) matrix.

[0067] Among them, the OD matrix is a matrix sorted by all traffic zones in rows (origin zones) and columns (destination zones), with the travel volume (OD volume) between any two zones as elements. In this application, the establishment of the OD matrix will not be elaborated. Please refer to the prior art.

[0068] Step (4): Use SQL language to filter out the bus stop data within the corresponding range of the straight-line distance of each travel mode respectively, to obtain the facility reachable range data for the calculation of the two-step floating catchment area method.

[0069] Step S2: Input the facility reachable range data into the two-step floating catchment area model to obtain the street supply reachability, service supply, and the ratio of supply to demand.

[0070] Specifically, the two-step floating catchment area algorithm can be performed based on the obtained data. Taking bus stops and rail transit stations as examples, the operation of the two-step floating catchment area model will be described. Optionally, it may include the following specific steps:

[0071] Step (1): Determine the ratio of the service supply of each existing bus stop to the number of potential demand people as the supply-demand ratio.

[0072] The first search is used to determine the busyness of the existing bus stops. That is, the supply-demand ratio within the service area of each bus stop. Search for the rail transit stations within the search distance threshold (d0, service radius) of the bus stops, and obtain the population distribution grid data from the worldpop website (global population data website) to determine the population P of the stations k , and then apply the Gaussian equation G(d k,j , d0)P k for weighting. The weighted sum of the population is the potential demand population within the area. Calculate the ratio R of the service supply of the bus stop to the potential demand population, j as shown in formulas (1)-(3).

[0073]

[0074] In formula (1), P k is the population of the rail transit station; d k,j is the straight-line distance between positions k, j; G(d k,j , d0)P k is the Gaussian equation considering the spatial friction problem. The setting method of the present invention overcomes to a certain extent the problem that the reachability within the search radius is equal and does not conform to the actual law of decreasing with distance. The calculation method of the Gaussian equation is as shown in formula (2).

[0075]

[0076] Step (2): Determine the bus stops that can provide bus services for rail stations, and calculate the rail transit accessibility based on the supply-demand ratio and the bus stops.

[0077] The second search is used to calculate the accessibility of each rail station. Search for the location of the stops that can provide bus services for the rail station, and use the supply-demand ratio R obtained from the first search j Use the Gaussian equation to weight and sum. For each rail transit station, search for all bus stops within the distance threshold, add up the weighted supply-demand ratios to obtain the rail transit accessibility. The calculation method is as shown in formula (3):

[0078]

[0079] where A j is the accessibility of the rail transit. When the value of A j is larger, it means that the accessibility of the rail transit is better. In the application of demand points, the original demand points in places with low accessibility can be ignored.

[0080] Step S3: Determine the usage demand of shared bicycles in the target area, and determine the estimated demand points based on the usage demand and the rail transit accessibility.

[0081] Specifically, the specific implementation process of step S3 may include the following steps:

[0082] Obtain the usage demand data of shared bicycles in the target area, perform density clustering on the usage demand data to obtain clustering points; merge each clustering point with the existing stops as the preselected demand points;

[0083] Compare and screen the rail transit accessibility with the preselected demand points, and ignore the preselected demand points where the rail transit accessibility is lower than the preset threshold to obtain the estimated demand points.

[0084] For example, the usage demand data of shared bicycles can be loaded (for example, the data can come from the data of the big data track topic of the 2021 Digital China Innovation Contest), and then the data is subjected to density clustering to obtain clustering points. Then, through the model builder, noise points are removed. Finally, the clustering center points are merged with the existing stops as the preselected demand points. Then, the preselected demand points are compared and screened with the street supply accessibility calculated by the two-step moving search method, and the places with lower accessibility are ignored, and finally the estimated demand points are obtained as parameter three.

[0085] Step S4: Determine the relevant indicators of the statistical data in the target area, and calculate the demand based on the relevant indicators of the statistical data.

[0086] In some specific embodiments, step S4 can be implemented through the following steps:

[0087] Determine the statistical data-related indicators of the target area. The statistical data-related indicators include: age, income, travel scale, education level, and shared travel experience;

[0088] Calculate the weight of each statistical data-related indicator;

[0089] Calculate the product of the coefficient of each statistical data-related indicator and the corresponding weight, and sum all the products. Multiply the sum of all the products by the population of the target area to obtain the demand group data;

[0090] Calculate the demand based on the demand group data.

[0091] For example, consult relevant papers to design and construct a demand prediction index system, and obtain the statistical data-related indicators of the target area as shown in Table 1:

[0092] Table 1 Statistical Data-Related Indicators of the Target Area

[0093] Index Meaning Age Indicates the age range of potential users who meet the age requirements for a national motor vehicle driver's license and have a strong willingness to use shared mobility Income Indicates the income range sufficient to support the rental cost of shared bicycles for a long time Travel scale Indicates the number of people who often travel among potential users of shared bicycles Educational level Indicates the educational attainment range of shared bicycle users Shared mobility experience Indicates the group of people who have used shared mobility methods

[0094] Use the SPSS software to calculate the AHP judgment matrix for the obtained index data to obtain the index weights. The specific calculation process is not elaborated in this application. Refer to the existing calculation methods. Then construct a function model to obtain the demand group data, as shown in formula (4), where h i represents the number of shared bicycle demand groups in the i-th area, w j represents the weight of the j-th statistical data-related indicator, f j represents the coefficient of the j-th statistical data-related indicator, ∑ i w j f j represents the shared bicycle usage potential index, and Z j represents the social population quantity of the i-th area. Among them, the statistical data-related indicator weights and statistical data-related indicator coefficients can be set by users according to their needs.

[0095] h i = z i ∑ j w j f j (4)

[0096] Predict the demand based on the demand group data according to each weight index to obtain the demand, which is used as parameter two.

[0097] L = S’ - C (5)

[0098] T = (s + L) (6)

[0099] g i = h i T (7)

[0100] wherein, g i represents the demand for new energy vehicles in the i-th area, s’ represents the proportion of the population who are unwilling to use shared source vehicles under the existing conditions, s represents the proportion of the population who are willing to use them, c represents the number of people who still will not use them after the change of the new energy vehicle conditions, L represents the proportion of the population who may use them when the conditions change, and T represents the proportion of the actual demand population.

[0101] Different demand quantities can be used according to requirements.

[0102] Step S5: Determine the existing network data of slow traffic and determine the current demand points.

[0103] In some embodiments, step S5 may include the following specific implementation processes:

[0104] Obtain the existing network data of slow traffic based on an open platform;

[0105] Delete the unnecessary stations in the existing network data of slow traffic according to the actual road network conditions and the government's urban planning policies to obtain the current demand points.

[0106] For example, delete the unnecessary stations from the existing network data of slow traffic obtained from the Amap open platform according to the actual road network conditions and the government's urban planning policies to obtain the current demand points (two), which are used as parameter one.

[0107] Step S6: Input the estimated demand points, demand quantities, and current demand points into the immune optimization algorithm model to obtain new demand quantities.

[0108] After obtaining the estimated demand points, demand quantities, and current demand points, input the three parameters into the MATLAB software, optimize the output results of the immune optimization algorithm by adjusting parameters such as population size, memory bank capacity, iteration times, crossover probability, mutation probability, diversity evaluation parameter, and the number of alternative network points, and obtain the new site demand quantity data through iterative processing and in combination with alternative network points, which is denoted as the new demand quantity.

[0109] In this application, the immune optimization algorithm model is described as follows:

[0110] Taking the application of the shared new energy rental network points in the immune optimization algorithm model as an example, use the site optimization problem to replace the antigen; use the solution set of the feasible solutions of the site selection location to replace the antibody (B cell); use the quality of the feasible solution to replace the affinity, determine the optimal solution by evaluating the expected reproduction probability P of the individual, and bring it into the immune optimization algorithm process to solve.

[0111] In this application, comprehensive analysis is carried out on the existing rental network data of shared new energy vehicles in a certain city obtained from the Evcard platform. The data of POI points in a certain city obtained from the Gaode Open Platform, the street data and road network data of a certain city obtained from the OSM platform, the sixth population census data of a certain city obtained from the National Bureau of Statistics, and the popular travel data of a certain city obtained from the biendata competition, etc. are used as the research basis.

[0112] Note: Since antibodies are produced by B cells, in the immune algorithm, antibodies and B cells are not distinguished and both correspond to the feasible solutions of the optimization problem.

[0113] Table 2 Parameter Settings of Immune Optimization Algorithm

[0114]

[0115] Referring to Table 2, in this application, the population size sizepop = 50, the memory bank capacity overbest = 20, the number of iterations MAXGEN = 170, the crossover probability pcross = 0.95, the mutation probability pmutation = 0.55, the diversity evaluation parameter ps = 0.95, and the number of alternative network points length = 106 can be set as the benchmark. With the help of MATLAB software, it is brought into the immune optimization algorithm model for solution to obtain the optimal network point distribution interval.

[0116] The specific implementation steps of the immune optimization algorithm are as follows in steps ① - ⑦:

[0117] ① Analyze the problem. Analyze the problem and the characteristics of its solutions, and design a suitable expression form for the solutions;

[0118] ② Generate the initial antibody population. Randomly generate N individuals and extract m individuals from the memory bank to form the initial population, where m is the number of individuals in the memory bank;

[0119] ③ Evaluate each antibody in the above population. In the algorithm, the evaluation of an individual is based on the expected reproduction rate P of the individual;

[0120] ④ Form the parental population. Arrange the initial population in descending order according to the expected reproduction rate P, and select the first N individuals to form the parental population; at the same time, take out m individuals and store them in the memory bank;

[0121] ⑤ Judge whether the end condition is met. If so, end; otherwise, continue with the next operation;

[0122] ⑥ Generation of a new population. Perform selection, crossover, and mutation operations on the antibody population to obtain a new population; at the same time, take out the memorized individuals from the memory bank to jointly form a new generation population;

[0123] Go back to execute step ③.

[0124] The evaluation of the diversity of solutions of the immune optimization algorithm is as follows:

[0125] The formula for calculating the affinity between an antibody and an antigen is as recorded in Formula (8). Among them, F v represents the objective function; the second term in the denominator represents a penalty for solutions that violate the distance constraint; C takes a relatively large positive number.

[0126]

[0127] The formula for calculating the affinity between antibodies is as in Formula (9). Among them, K v,s represents the number of identical bits in the antibody and antibody s; L represents the length of the antibody.

[0128]

[0129] The formula for calculating the antibody concentration is as in Formula (10). Among them, N represents the total number of antibodies;

[0130] T represents a preset threshold.

[0131]

[0132] The formula for the expected reproduction probability is as in Formula (11). Among them, α represents a constant; A v represents the affinity between the antibody and the antigen; C v represents the antibody concentration.

[0133]

[0134] Figure 2 This is a schematic diagram of the convergence curve of an immune optimization algorithm provided by an embodiment of the present invention. Refer to Figure 2 , the solid line represents the optimal fitness value, and the dashed line represents the average fitness value. It is calculated that the optimal fitness value is lower than the average fitness value. The deviation between the two is relatively large at the 20th iteration, and the curve amplitude is also relatively large; the variation deviation is smaller at 20 - 60 iterations, and the curve amplitude is more obvious; the deviation and the curve amplitude are both relatively small at 60 - 160 iterations, and the two tend to balance. It can be seen from the convergence curve that the immune optimization algorithm converges to the optimal solution after 170 iterations.

[0135] Step S7: Input the new demand volume into the combined model to obtain the optimal traffic station allocation volume; among them, the combined model is a combined model of an investment allocation model and a single judgment model.

[0136] In some embodiments, the combined model is a combined model constructed by using 0.62 and 0.38 as the weights of the investment allocation model and the single judgment model respectively as the combined weights.

[0137] In this application, the investment allocation model is described as follows:

[0138]

[0139]

[0140]

[0141]

[0142]

[0143] In Formulas 12 - 16, V represents the optimized set of outlets, i.e., V = {1, 2 ……, v}; u represents the u-th outlet; Ru is the population served by the u-th outlet, and Q f u represents the facility capacity of each outlet, i.e., the number of parking spaces; B u represents the service population influence coefficient of point u; C u represents the cost of a single parking space at point u; k represents the total capacity of all outlets.

[0144] In this application, the single judgment model is described as follows:

[0145] Based on the population served by each outlet, an automobile allocation model is constructed. First, according to the effective using customers of the u-th outlet, the weight v of the effective using customer index of the u-th outlet is obtained u ;

[0146]

[0147] According to the usage index weight v u , a single judgment model is constructed.

[0148]

[0149] Q u d is the allocated quantity of shared new energy vehicles at the u-th outlet under the single judgment model.

[0150] The combined investment allocation model and the single judgment model are as shown in Formula (19):

[0151] The two sets of outlet allocation results of the allocation model and the single judgment model are comprehensively calculated. According to the query of thesis materials, the default value uses the golden section point value 0.618 as the combined weight, which can be adjusted according to the actual demand comparison value. A linear weighted function is constructed. The following is the comprehensive allocation algorithm, and Q U is the allocated quantity data after the model combination, and Q u d is the allocated quantity (two), and Qu f For the allocation quantity (1).

[0152]

[0153] Step S8: Calculate the most optimal site location based on the urban road network data, urban building contour data, and the best traffic station allocation quantity of the target area.

[0154] Taking the slow and low-carbon traffic station location as an example, this invention inputs subway station data, bus station data, shared bicycle parking point data, urban traffic road network data, 100-meter precision population distribution lattice data, etc. into the two-step mobile search model. At the same time, it improves the two-step mobile search algorithm by adding a Gaussian decay function considering the spatial friction problem to the two-step mobile search algorithm, overcoming the problem that the accessibility is equal within a certain search radius and does not conform to the actual law of decreasing with distance. In the first step, the ratio of service supply to potential population demand is calculated. In the second step, based on the calculation of the first step and adding the weighted values of the Gaussian equation, the accessibility distribution of each street in the city (i.e., rail transit accessibility) is obtained, and finally, the comparison data of the slow traffic accessibility in the city is obtained.

[0155] Compare and screen the accessibility results with the site vehicle usage demand data. According to the rule of ignoring the demand points in places with low accessibility, the processed site demand points are obtained as one of the input parameters (parameter three) of the immune optimization algorithm model; perform index weight processing on the urban area statistical data to obtain the site demand quantity data as one of the input parameters (parameter two) of the immune optimization algorithm model; perform deletion and deactivation processing on the existing slow traffic network points in the city to obtain the demand point data as one of the input parameters (parameter one) of the immune optimization algorithm model; then input the three parameters into the MATLAB software, and optimize the output result of the immune optimization algorithm by adjusting parameters such as population size, memory bank capacity, iteration times, crossover probability, mutation probability, diversity evaluation parameter, and the number of alternative network points. Through iterative processing and combining with alternative network points, new site demand quantity data is obtained.

[0156] Furthermore, the site demand data is optimized by means of an investment allocation model and a single judgment model. Among them, the investment allocation model processes to obtain the site allocation quantity (I), which is the allocation quantity obtained by each network point with the construction investment cost and vehicle demand as the allocation targets; each transportation site combines the demand data with the construction cost data and performs single judgment model processing to finally obtain the site allocation quantity (II), that is, the vehicle allocation quantity of each network point based on the served population. The site allocation quantity (I) and the site allocation quantity (II) are subjected to model combination processing. The combination processing uses an approximate golden section point value of 0.62 as the combination weight to construct a linear weighted function to finally obtain the optimal transportation site allocation quantity, and then with the help of urban road network data and urban building contour data, visualization processing is performed through ArcMap software to obtain the optimal site selection location.

[0157] In order to verify the effect of the slow traffic site selection method with multi-model coupling provided by the embodiments of the present invention, the present application also provides a verification embodiment.

[0158] Taking the Qixia District, Gulou District, Qinhuai District of a certain city and the surrounding areas of the three districts as the research scope for analysis and research, taking the parking points of shared new energy vehicles as an example, the original technical solution and the technical solution after coupling the models are respectively analyzed and processed.

[0159] Figure 3 This is a schematic diagram of the network points in the area before optimization provided for the verification embodiment of the present invention. Among them, the squares are the existing network points. Figure 4 Provided for the embodiments of the present invention Figure 3 Schematic diagram of the network points in the optimized area, where the dots are the optimized network points.

[0160] Refer to Figure 3 , which is the scope of Fuzimiao Street in a certain city, with an area of 2.94 square kilometers, ranking ninth in Qinhuai District in terms of economic strength, being a non-central economic region, and having relatively few existing site demands. However, the original number of network points in this street is as high as 6, which does not match its site demands. Refer to Figure 4 , after the immune optimization algorithm, the overlap rate of the network points in the research area is 42.85%. Among them, there are only 2 duplicate network points in Fuzimiao Street, and the number of optimized network points is only 2. It can be concluded that the site selection model after multi-model coupling is more scientific and reasonable than that after single model processing.

[0161] Based on a general inventive concept, the embodiments of the present invention also provide a slow traffic site selection device with multi-model coupling.

[0162] Figure 5 This is a schematic structural diagram of a slow traffic site selection device with multi-model coupling provided for the embodiments of the present invention. Refer to Figure 5 , the device provided for the embodiments of the present invention may include the following structures:

[0163] A preprocessing module 51, configured to determine facility reach range data of a target area; the facility reach range data includes bus stop data within the maximum range corresponding to the straight-line distance of each travel mode.

[0164] A two-step movement search module 52, configured to input the facility reach range data into a two-step movement search model to obtain rail transit accessibility and supply-demand ratio; wherein, the supply-demand ratio is the ratio of the service supply of each bus stop to the number of potential demanders.

[0165] A first determination module 53, configured to determine the vehicle usage demand of shared bicycles in the target area, and determine an estimated demand point according to the vehicle usage demand and the rail transit accessibility.

[0166] A second determination module 54, configured to determine indicators related to statistical data of the target area, and calculate the demand quantity according to the indicators related to the statistical data.

[0167] A third determination module 55, configured to determine existing network point data of slow traffic and determine current demand points.

[0168] An immune optimization module 56, configured to input the estimated demand point, the demand quantity, and the current demand point into an immune optimization algorithm model to obtain a new demand quantity.

[0169] An allocation module 57, configured to input the new demand quantity into a combined model to obtain the optimal traffic stop allocation quantity; wherein, the combined model is a combined model of an investment allocation model and a single judgment model.

[0170] A calculation module 58, configured to calculate the optimal site selection location according to the urban road network data, the urban building contour data, and the optimal traffic stop allocation quantity of the target area.

[0171] Optionally, the preprocessing module is specifically configured to obtain rail transit station data and perform buffer processing with a preset distance.

[0172] Determine slow traffic station data, where the slow traffic station data includes: a starting point and a destination point; wherein, the starting point is the rail transit station data, and the target point is mobile phone signaling data, a parking lot, and bus stop data.

[0173] Calculate the straight-line distance from a rail transit station to a slow traffic station within the maximum connection capacity range; wherein, the maximum connection capacity range includes: a walking distance of 1 km, a cycling distance of 2 km, and a bus driving distance of 5 km.

[0174] Determine public transportation station data within the maximum connection capacity range corresponding to the straight-line distance of each travel mode, and use the public transportation station data within the maximum connection capacity range corresponding to the straight-line distance of each travel mode as the facility reach range data.

[0175] Optionally, the two-step movement search module is specifically configured to determine the ratio of the service supply of each existing bus stop to the potential demand population as the supply-demand ratio.

[0176] Determine the bus stops that can provide bus services for rail stations, and calculate the rail transit accessibility based on the supply-demand ratio and the bus stops.

[0177] Optionally, the first determination module is specifically configured to obtain the vehicle usage demand data of shared bicycles in the target area, perform density clustering on the vehicle usage demand data to obtain clustering points; merge each clustering point with the existing stations as preselected demand points.

[0178] Compare and screen the rail transit accessibility with the preselected demand points, and ignore the preselected demand points with rail transit accessibility lower than the preset threshold to obtain the estimated demand points.

[0179] Optionally, the second determination module is specifically configured to determine the statistical data related indicators of the target area. The statistical data related indicators include: age, income, travel scale, education level, and shared travel experience.

[0180] Calculate the weight of each statistical data related indicator.

[0181] Calculate the product of the coefficient of each statistical data related indicator and the corresponding weight, and sum all the products. Multiply the sum of all the products by the population of the target area to obtain the demand group data.

[0182] Calculate the demand based on the demand group data.

[0183] Optionally, the third determination module is specifically configured to obtain the existing network data of slow traffic based on the open platform.

[0184] Delete the unnecessary stations in the existing network data of slow traffic according to the actual road network situation and the government urban planning policy to obtain the current demand.

[0185] Regarding the device in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0186] Based on a general inventive concept, an embodiment of the present invention further provides a slow traffic site selection device based on multi-model coupling.

[0187] Figure 6 For the structural schematic diagram of a slow traffic site selection device based on multi-model coupling provided by an embodiment of the present invention, refer to Figure 6, the device provided in this embodiment may include: a processor 61 and a memory 62, and the processor 61 is connected to the memory 62. Among them, the processor 61 is used to call and execute the program stored in the memory 62; the memory 62 is used to store the program, and the program is at least used to execute the slow traffic station location selection method based on multi-model coupling in the above embodiments.

[0188] The specific implementation of the slow traffic station location selection device based on multi-model coupling provided in the embodiments of the present application may refer to the implementation manner of the slow traffic station location selection method based on multi-model coupling in any of the above embodiments, and will not be elaborated here.

[0189] Based on one general inventive concept, the embodiments of the present invention also provide a non-transitory computer-readable storage medium storing computer instructions, and the computer instructions are used to cause a computer to execute the slow traffic station location selection method based on multi-model coupling in any one of the above.

[0190] It can be understood that the same or similar parts in the above embodiments can be referred to each other, and the content not detailed in some embodiments can be seen in the same or similar content in other embodiments.

[0191] It should be noted that in the description of the present invention, terms such as "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, the meaning of "a plurality" means at least two.

[0192] Any process or method description in the flowchart or described in other ways herein can be understood as representing a module, segment, or part of code including one or more executable instructions for implementing a specific logical function or process, and the scope of the preferred embodiments of the present invention includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in an order opposite to that shown or discussed, according to the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0193] It should be understood that each part of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in the memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0194] Those of ordinary skill in the art can understand that all or part of the steps carried by the method of the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0195] In addition, in each of the embodiments of the present invention, each functional unit may be integrated in a processing module, may exist physically separately for each unit, or two or more units may be integrated in one module. The above integrated module may be implemented in the form of hardware or in the form of a software functional module. When the above integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0196] The above-mentioned storage medium may be a read-only memory, a magnetic disk, an optical disc, or the like.

[0197] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0198] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A slow traffic station location method based on multi-model coupling, characterized in that Including: Determine the facility reach range data of the target area; The facility reach range data includes bus stop data within the maximum range corresponding to the straight-line distance of each travel mode; Input the facility reach range data into a two-step mobile search model to obtain rail transit bus accessibility and supply-demand ratio; wherein, the supply-demand ratio is the ratio of the service supply of each bus stop to the potential number of demanders; Determine the usage demand of shared bicycles in the target area, and determine the estimated demand points according to the usage demand and the rail transit bus accessibility; Determine the statistical data related indicators of the target area, and calculate the demand quantity according to the statistical data related indicators; Determine the existing network data of slow traffic and determine the current demand points; Input the estimated demand points, demand quantity and current demand points into an immune optimization algorithm model to obtain a new demand quantity; Input the new demand quantity into a combined model to obtain the optimal traffic station allocation quantity; wherein, the combined model is a combined model of an investment allocation model and a single judgment model; Calculate the optimal site location according to the urban road network data and urban building contour data of the target area and the optimal traffic station allocation quantity.

2. The method according to claim 1, wherein The determination of the facility reach range data of the target area includes: Obtain rail transit station data and perform buffer processing with a preset distance; Determine slow traffic station data, and the slow traffic station data includes: starting point and destination point; wherein, the starting point is rail transit station data, and the destination point is mobile phone signaling data, parking lot, bus stop data; Calculate the straight-line distance from the rail transit station to the slow traffic station within the maximum connection capacity range; wherein, the maximum connection capacity range includes: walking distance of 1 km, cycling distance of 2 km, and bus driving distance of 5 km; Determine the public transportation station data within the maximum connection capacity range corresponding to the straight-line distance of each travel mode, and use the public transportation station data within the maximum connection capacity range corresponding to the straight-line distance of each travel mode as the facility reach range data.

3. The method according to claim 1, characterized in that, The input of the facility reach range data into a two-step mobile search model to obtain rail transit bus accessibility and supply-demand ratio includes: Determine the ratio of the service supply of each currently existing bus stop to the potential number of demanders as the supply-demand ratio; Determine the bus stops that can provide bus services for rail stations, and calculate the rail transit bus accessibility based on the supply-demand ratio and the bus stops.

4. The method according to claim 1, characterized in that, The determination of the usage demand of shared bicycles in the target area and the determination of the estimated demand points according to the usage demand and the rail transit bus accessibility include: Obtain the usage demand data of shared bicycles in the target area, perform density clustering on the usage demand data to obtain clustering points; merge each clustering point with the existing stations as preselected demand points; Compare and screen the rail transit bus accessibility with the preselected demand points, and ignore the preselected demand points with rail transit bus accessibility lower than the preset threshold to obtain the estimated demand points.

5. The method according to claim 1, wherein The determination of the statistical data related indicators of the target area and the calculation of the demand quantity according to the statistical data related indicators include: Determine the statistical data-related indicators of the target area, where the statistical data-related indicators include: age, income, travel scale, education level, and shared travel experience; Calculate the weight of each statistical data-related indicator; Calculate the product of the coefficient of each statistical data-related indicator and the corresponding weight, and sum all the products. Multiply the sum of all the products by the population of the target area to obtain the demand group data; Calculate the demand quantity based on the demand group data.

6. The method according to claim 1, wherein For the determination of the existing slow traffic network data and the determination of the current demand points, it includes: Obtain the existing slow traffic network data based on an open platform; Delete the unnecessary stations in the existing slow traffic network data according to the actual road network situation and the government's urban planning policies to obtain the current demand points.

7. The method according to claim 1, wherein The combined model is a combined model constructed with 0.62 and 0.38 as the weights of the investment allocation model and the single judgment model respectively as the combined weights.

8. A slow traffic station location device based on multi-model coupling, characterized in that, It includes: A preprocessing module for determining the facility reach range data of the target area; The facility reach range data includes the bus stop data within the maximum range corresponding to the straight-line distance of each travel mode; A two-step mobile search module for inputting the facility reach range data into a two-step mobile search model to obtain the rail transit accessibility and the supply-demand ratio; where the supply-demand ratio is the ratio of the service supply of each bus stop to the number of potential demanders; A first determination module for determining the vehicle usage demand of shared bicycles in the target area and determining the estimated demand points based on the vehicle usage demand and the rail transit accessibility; A second determination module for determining the statistical data-related indicators of the target area and calculating the demand quantity based on the statistical data-related indicators; A third determination module for determining the existing slow traffic network data and determining the current demand points; An immune optimization module for inputting the estimated demand points, the demand quantity, and the current demand points into an immune optimization algorithm model to obtain a new demand quantity; An allocation module for inputting the new demand quantity into the combined model to obtain the optimal traffic station allocation quantity; where the combined model is a combined model of an investment allocation model and a single judgment model; A calculation module for calculating the optimal site selection location based on the urban road network data and the urban building contour data of the target area and the optimal traffic station allocation quantity.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to execute the slow traffic station location selection method based on multi-model coupling according to any one of claims 1-8.

10. A slow traffic station location device based on multi-model coupling, characterized in that, It includes a processor and a memory, and the processor is connected to the memory: Wherein, the processor is used to call and execute the program stored in the memory; The memory is used to store the program, and the program is at least used to execute the slow traffic station location selection method based on multi-model coupling according to any one of claims 1-7.

Images