A shared bicycle parking point site selection method and system based on multi-point joint distribution

Abstract

The application relates to a shared bicycle parking point location method and system based on multi-point joint distribution, which comprises the following steps: acquiring information of multiple demand points and multiple alternative parking points, and determining optimization targets and / or constraint conditions of a non-dominated sorting genetic model of multi-point joint distribution according to the information; generating an initial population of the non-dominated sorting genetic model based on a binary and integer combined chromosome coding mode; calculating non-dominated levels and crowded distances of each chromosome; screening the initial population by using a tournament selection method, and performing crossover and variation on the screened population based on a position vector crossover operator and a mutation operator to obtain a child population; and iteratively processing the parent population based on an elite strategy and the non-dominated levels and the crowded distances until an optimal solution meeting one or more constraint conditions is found. The application establishes an NSGA-II model based on multi-point joint distribution, and optimizes genetic operators in the model, so that the practicability of the model is improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of shared bicycle operation or traffic planning, and particularly relates to a shared bicycle parking point site selection method and system based on multi-point joint distribution. BACKGROUND

[0002] In order to facilitate the standardized management of shared bicycles in a certain region by the local government, the number of shared bicycle parking points in the region, the total number of bicycles in the region and the alternative parking positions are initially planned. A shared bicycle operation enterprise needs to make decisions on the parking points and the number of bicycles to be put under the planning of the government, so as to meet the user demand and reduce the cost of the enterprise as much as possible. Generally, the average demand for the number of bicycles in a certain region does not change much in a period of time, and the operation party will initialize the bicycles at each point before the operation starts every day. For the demand in a whole day, if the number of bicycles at each parking point can meet the demand in the peak period, the demand in other periods in the region can be basically met. Therefore, the peak time demand of each demand point in a day is designed as the target, and the demand for the number of bicycles at each demand point can be obtained according to the average data in the past.

[0003] In addition, the usual parking site selection model considers that the user only stops and takes the bicycle at the nearest parking point to the demand point. However, in the actual situation, on the one hand, when there is no bicycle at the nearest parking point of a demand point, the user of the demand point will consider to find other parking points within the maximum walking distance, but the satisfaction degree of the user will decrease with the increase of the walking distance, and the parking points beyond the maximum walking distance are not considered. On the other hand, the user will select the most suitable parking point within the maximum walking distance according to the travel direction. Therefore, it is necessary to study the parking site selection problem under the condition of multiple parking points, assuming that the demand for each parking point and the distance between each parking point and the demand point are related within the walking allowed range of each demand point. In summary, in order to ensure the satisfaction degree of the user, reduce the cost of the enterprise and ensure that the demand of each demand point is met as much as possible, while considering the constraints such as the number of bicycles at each point, the distance between each parking point, the number of parking points, the total number of bicycles in the region and the like, a reasonable mathematical model needs to be established and a corresponding algorithm needs to be used to solve the model.

[0004] Pareto solution is also called non-dominated solution or non-dominating solution: when there are multiple objectives, due to the existence of conflicts between objectives and the phenomenon of incomparable, a solution is the best in a certain objective, and can be the worst in other objectives. These solutions, while improving at least one objective function, will inevitably weaken at least one other objective function, which are called non-dominated solutions or Pareto solutions. In 1995, Srinivas and Deb proposed a non-dominated sorting genetic algorithm (NSGA). It is a genetic algorithm based on the concept of Pareto optimality. SUMMARY

[0005] To solve the problem of shared bicycle parking point location considering multiple objective constraints such as multi-point joint distribution, a shared bicycle parking point location method based on multi-point joint distribution is provided in the first aspect of the application, comprising: obtaining information of multiple demand points and multiple alternative parking points, and determining multiple optimization objectives and / or multiple constraint conditions of a non-dominated sorting genetic model of multi-point joint distribution according to the information; the constraint conditions include the maximum total amount of bicycles in the alternative parking points, the minimum distance from other alternative parking points and the minimum satisfaction; based on the binary and integer combined chromosome coding mode and the constraint conditions, an initial population of the non-dominated sorting genetic model is generated; the multiple objective function values of each chromosome in the initial population are calculated, and the non-dominated level and the crowding distance of each chromosome are obtained according to the multiple objective function values; the initial population is screened by using the tournament selection method, and the screened population is crossed and mutated by using the position vector-based crossover operator and the mutation operator to obtain one or more offspring populations; based on the elite strategy and the non-dominated level and the crowding distance, the optimal N chromosomes are selected from the initial population and all offspring populations, and the optimal N chromosomes are used as the parent population for the next iteration, and the iteration is performed until the iteration number reaches a threshold value or the optimal solution meeting one or more constraint conditions is found, wherein N is a positive integer.

[0006] In some embodiments of the application, the initial population of the non-dominated sorting genetic model is generated based on the binary and integer combined chromosome coding mode and the constraint conditions, comprising: encoding whether the alternative parking point is selected based on binary; the number of parking spaces of the alternative parking point is encoded by integer.

[0007] In some embodiments of the present application, the crossover operator and the mutation operator based on the position vector are used to cross and mutate the screened population to obtain one or more offspring populations, which includes: for two parent chromosomes X1 and X2 to be crossed, the crossover operator is represented by two first position vectors, each element of the position vector includes 1 or 0, and one of the first position vectors represents a randomly selected crossover position; based on the two parent chromosomes X1 and X2 to be crossed, the crossover operator is used to calculate to obtain the crossed offspring Y1 and Y2; for the parent chromosome X3 to be mutated, the mutation operator is represented by a second position vector, and the elements of the second position vector include 1 or 0, wherein the position with element 1 represents the mutation position; based on the parent chromosome X3 to be mutated, the crossover operator is used to calculate to obtain the mutated offspring Y3.

[0008] Further, the calculation of the crossover operator based on the two parent chromosomes X1 and X2 to be crossed to obtain the crossed offspring Y1 and Y2 includes:

[0009]

[0010] wherein ma and mb represent the first position vectors with a length of n, and the elements of the position vectors include 1 or 0; mb represents the randomly selected crossover position, and ma+mb=1; 1 represents the unit vector with a length of n, n represents the number of alternative parking points, and · represents dot product.

[0011] Further, the shared bicycle parking point location method based on multi-point joint distribution, the calculation of the crossover operator based on the parent chromosome X3 to be mutated to obtain the mutated offspring Y3 is represented by the following method:

[0012] Y3={(1-mc); (1-mc)}·X3+{mc; mc}·X m ,

[0013] wherein the i-th column element in X m is:

[0014]

[0015] wherein mc represents the second position vector; δ and δ' are both random numbers between 0 and 1, δ is used to determine whether the chromosome of the parking point position or the chromosome of the parking point number is mutated, and δ' is used to mutate the parking number; is the upper limit of the number of bicycles allowed to be parked at the mc i th parking point.

[0016] In the above embodiment, the obtaining the information of the plurality of demand points and the plurality of candidate parking points, and determining the plurality of optimization objectives and / or the plurality of constraint conditions of the non-dominated sorting genetic model based on the multi-point joint allocation includes: obtaining the information of the plurality of demand points and the plurality of candidate parking points, and calculating the satisfaction degree of each candidate parking point according to the information; determining the plurality of optimization objectives of the non-dominated sorting genetic model according to the satisfaction degree of each candidate parking point; determining the plurality of parameters of the non-dominated sorting genetic model, and determining the plurality of constraint conditions of the non-dominated sorting genetic model according to the plurality of parameters.

[0017] In a second aspect, the application provides a shared bicycle parking point location system based on multi-point joint allocation, comprising: an obtaining module configured to obtain information of a plurality of demand points and a plurality of candidate parking points, and determine a plurality of optimization objectives and / or a plurality of constraint conditions of a non-dominated sorting genetic model based on the multi-point joint allocation; the constraint conditions include a maximum total number of bicycles to be put in each candidate parking point, a minimum distance between the candidate parking point and other candidate parking points, and a minimum satisfaction degree; a calculation module configured to generate an initial population of the non-dominated sorting genetic model based on a binary and integer combined chromosome coding mode and the constraint conditions; calculate a plurality of objective function values of each chromosome in the initial population, and obtain a non-dominated level and a crowding distance of each chromosome according to the plurality of objective function values; an evolution module configured to filter the initial population by using a tournament selection method, and perform crossover and mutation on the filtered population based on a position vector crossover operator and a mutation operator to obtain one or more offspring populations; and an iteration module configured to select N optimal chromosomes from the initial population and all the offspring populations based on an elitist strategy and the non-dominated level and the crowding distance, and use the N optimal chromosomes as parent chromosomes for the next iteration until a threshold number of iterations is reached or an optimal solution satisfying one or more constraint conditions is found, wherein N is a positive integer.

[0018] In a third aspect, the application provides an electronic device, comprising: one or more processors; and a storage device storing one or more programs configured to be executed by the one or more processors to cause the one or more processors to implement the shared bicycle parking point location method based on the multi-point joint allocation provided in the first aspect.

[0019] In a fourth aspect, the application provides a computer-readable medium having stored thereon a computer program, wherein the computer program is configured to be executed by a processor to implement the shared bicycle parking point location method based on the multi-point joint allocation provided in the first aspect.

[0020] The application has the following beneficial effects:

[0021] The application is directed to the problem of shared bicycle parking point location, based on the actual situation that users will comprehensively select parking points to take bicycles in the maximum walking distance range considering various factors such as travel direction, and considering that the deployment of shared bicycles involves the influence of multiple factors such as users, government and bicycle operation enterprises, the relevant research on the parking location problem in the shared bicycle system is carried out. Therefore, the method has the following advantages: (1) considering that users will determine the bicycle taking situation within the maximum walking distance according to factors such as travel direction, a multi-point joint distribution method for satisfying user demand of bicycle quantity distribution is proposed, and multiple factors such as user demand, government requirements and enterprise operation cost are considered, and a parking location multi-objective optimization model is established; (2) an improved NSGA-II algorithm, namely INSGA-II algorithm, is proposed, since there is a certain relationship between the parking point selection and the corresponding bicycle quantity in the parking location problem, a chromosome coding method is proposed, and the corresponding genetic operator is improved. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 It is a basic flowchart of the multi-point joint distribution based shared bicycle parking point location method in some embodiments of the application.

[0023] Figure 2 It is a chromosome diagram in some embodiments of the application.

[0024] Figure 3 It is a multi-parking point joint demand satisfaction diagram in some embodiments of the application.

[0025] Figure 4 It is a crossover operator diagram in some embodiments of the application.

[0026] Figure 5 It is a mutation operator diagram in some embodiments of the application.

[0027] Figure 6 It is a shared bicycle parking point location division effect diagram of a certain area in a certain city.

[0028] Figure 7 It is a target function convergence curve in some embodiments of the application.

[0029] Figure 8 It is a Pareto solution set in some embodiments of the application.

[0030] Figure 9 It is an INSGA-II algorithm and NSGA-II algorithm result comparison diagram.

[0031] Figure 10 It is a specific flowchart of the multi-point joint distribution based shared bicycle parking point location method in some embodiments of the application.

[0032] Figure 11 Structure diagram of a shared bicycle parking point locating system based on multi-point joint distribution in some embodiments of the present application;

[0033] Figure 12 Structure diagram of an electronic device in some embodiments of the present application. DETAILED DESCRIPTION

[0034] The principles and features of the present application are described below in conjunction with the accompanying drawings, and the examples are only used to explain the present application and not to limit the scope of the present application.

[0035] Embodiment 1

[0036] Reference Figure 1 In a first aspect of the present application, a shared bicycle parking point locating method based on multi-point joint distribution is provided, comprising: S100. obtaining information of a plurality of demand points and a plurality of alternative parking points, and determining a plurality of optimization objectives and / or a plurality of constraint conditions of a non-dominated sorting genetic model of multi-point joint distribution according to the information; the constraint conditions include a maximum total amount of bicycles to be put in the alternative parking points, a minimum distance from other alternative parking points, and a minimum satisfaction degree; S200. generating an initial population of the non-dominated sorting genetic model based on a binary and integer combined chromosome coding mode and the constraint conditions; calculating a plurality of objective function values of each chromosome in the initial population, and obtaining a non-dominated level and a crowding distance of each chromosome according to the plurality of objective function values; S300. screening the initial population by using a tournament selection method, and performing crossover and mutation on the screened population based on a position vector crossover operator and a mutation operator to obtain one or more offspring populations; S400. selecting the optimal N chromosomes from the initial population and all offspring populations based on an elitist strategy and the non-dominated level and the crowding distance, and taking the optimal N chromosomes as a parent population for the next iteration, and iterating until the number of iterations reaches a threshold value or an optimal solution satisfying one or more constraint conditions is found, wherein N is a positive integer.

[0037] In the above embodiment S100, the obtaining of the information of the plurality of demand points and the plurality of alternative parking points, and the determination of the plurality of optimization objectives and / or the plurality of constraint conditions of the non-dominated sorting genetic model of multi-point joint distribution according to the information includes:

[0038] S101. obtaining information of a plurality of demand points and a plurality of alternative parking points, and calculating a satisfaction degree of each alternative parking point according to the information; specifically, the satisfaction degree between different parking points for a demand point is related to the distance between the demand point and the parking point, and can be obtained by the following formula:

[0039]

[0040] In the formula, R represents the maximum walking distance of the demand point user; R A represents the maximum walking distance of the demand point user fully satisfied, and d represents the straight-line distance between the demand point and the parking point.

[0041] (2) Multi-point joint satisfaction of demand Figure 2 As shown in the figure, a demand point has three parking points within the maximum walking distance for selection, and the satisfaction of the demand point to different parking points is different according to the distance between the parking points and the demand point, such as Figure 2 As shown in the figure, the satisfaction of the demand point to the three parking points A, B and C is 0.8, 0.6 and 0.1 respectively, in combination with the travel direction of different parking points and the demand point, it is assumed that the travel demand of the demand point allocated to a parking point within the maximum walking distance is related to the satisfaction of the point, such as Figure 2 As shown in the figure, there are three parking points A, B and C within the maximum walking distance of the demand point, and the total amount of bicycle demand of the demand point is allocated to A, B and C according to the proportion of the total satisfaction of each point, and the bicycle demand allocated to each point is 16, 12 and 2 respectively. The initial satisfaction matrix is calculated and determined according to this method.

[0042] S102. Determine multiple optimization objectives of the non-dominated sorting genetic model according to the satisfaction of each alternative parking point; S103. Determine multiple parameters of the non-dominated sorting genetic model, and determine multiple constraint conditions of the non-dominated sorting genetic model according to the same.

[0043] Specifically, according to the set of all demand points and parking alternative points, the initial satisfaction matrix, the distance between each alternative parking point, the initial parameters of the model are determined, including the upper limit of the number of parking spaces of each alternative parking point, the upper limit of the number of bicycles in the target area, the operation cost of the bicycle, etc., and the model is established. The site selection model is:

[0044] maxF1=∑ I ∑ M f(d im )X m (2),

[0045] minF2=∑ M αX m +β∑ M N m +γ∑ M (|∑ I Z im -N m |) (3),

[0046] s.t.

[0047]

[0048] ∑ M C m ≤C a (5),

[0049] ∑ M X m ≤P (6),

[0050]

[0051]

[0052]

[0053]

[0054] X m ∈[0,1] (11),

[0055] C m Z im ≥0 and all are integers

[0056] Equation (2) represents the maximum sum of satisfaction of all demand points in the target area; Equation (3) represents the minimum cost required for bicycle operators to build parking points and deploy bicycles in the area. The first part of Equation (3) is the sum of the fixed construction and operation costs of the selected parking points, the second part of Equation (3) is the operation cost of each bicycle, and the third part of Equation (3) is the penalty cost for controlling the number of vehicles in the parking points. If there are too few vehicles in the alternative points, it is the penalty cost for not meeting the demand. If there are too many vehicles in the alternative points, exceeding the required upper limit, it is the penalty cost for wasting resources.

[0057] Equations (4) to (12) represent constraints, where equation (4) indicates that the number of vehicles parked at each selected parking spot in the area shall not exceed the maximum number of vehicles allowed at that parking spot; equation (5) indicates that the sum of the number of vehicles parked at all alternative parking spots shall not exceed the maximum number of vehicles allowed by the government in the area; equation (6) indicates that the final number of parking spots determined in the area shall not exceed the maximum number of parking spots allowed by the government; equation (7) indicates that only parking spots within the maximum walking distance of the demand point are considered by the corresponding demand point; equation (8) indicates that the distance between all determined parking spots shall not be less than the minimum allowed distance; equation (9) indicates that vehicles are allowed to be deployed only when selected as parking spots; equation (10) indicates that the satisfaction of each demand point shall not be less than the given minimum satisfaction; equations (11) and (12) indicate the range of values ​​for the variables.

[0058] It can be understood that the multi-objective optimization is performed for the parking site selection problem, and the purpose is to determine the suitable parking site and the corresponding parking quantity from the candidate parking sites, and the single vehicle quantity at the candidate sites which are not selected as the parking sites can only be zero. The genetic operations such as crossover and mutation in the evolution process must also meet the above conditions, and therefore, whether to be selected as the parking site and the corresponding single vehicle quantity are jointly considered to form the chromosome in the NSGA-II algorithm.

[0059] Therefore, with reference to Figure 3 In step S200 of some embodiments of the present application, the chromosome coding mode based on the combination of binary and integer, and the constraint condition generate the initial population of the non-dominated sorting genetic model, including: S201. Coding whether the candidate parking site is selected based on binary; S202. Coding the parking quantity of the candidate parking site by integer.

[0060] Specifically, binary coding is used to represent whether the corresponding parking site is selected, and if 0 represents that the site is not selected, and if 1 represents that the site is selected as the parking site. Real number coding is used to represent the corresponding parking quantity, and the single vehicle quantity of each site does not exceed the upper limit of the specified quantity. It should be noted that when a candidate site is not selected as the final parking site, the single vehicle quantity of the site must be zero. A chromosome is shown in Figure 3 The chromosome represents that the candidate parking sites 2, 4, 5 and 7 are parking sites, and the single vehicle quantities are 20, 15, 30 and 23 respectively.

[0061] It should be understood that the simulated binary crossover (SBX) is a crossover operator used in the original NSGA-II algorithm. When the SBX is used to process the integer optimization problem, the child variable may be non-integer, which further causes the child solution to be difficult to meet the constraint condition or a specific condition. The original SBX crossover operator formula is as follows:

[0062] c1 / 2, j = (x1, j + x2, j) / 2 ± β × (x1, j - x2, j) / 2.

[0063] The chromosome coding method proposed for the parking site selection problem needs to perform the crossover operation on the parking site and the single vehicle quantity.

[0064] Therefore, with reference to Figure 4 and Figure 5In step S300 of some embodiments of the present application, the crossover operator and the mutation operator based on the position vector are used to perform crossover and mutation on the screened population to obtain one or more offspring populations, including: S301. For two parent chromosomes X1 and X2 to be crossed, the crossover operator is represented by two first position vectors, each element of the position vector including 1 or 0, and one of the first position vectors represents a randomly selected crossover position; based on the two parent chromosomes X1 and X2 to be crossed, the crossover operator is used to calculate to obtain the crossed offspring Y1 and Y2; S301. For the parent chromosome X3 to be mutated, the mutation operator is represented by a second position vector, and the elements of the second position vector include 1 or 0, wherein the positions with elements of 1 represent the mutation positions; based on the parent chromosome X3 to be mutated, the crossover operator is used to calculate to obtain the mutated offspring Y3.

[0065] Specifically, in step S301, the crossover operator is calculated based on the two parent chromosomes X1 and X2 to be crossed, and the two parent chromosomes X1 and X2 to be crossed are represented by two position vectors ma and mb, and each element of the position vector includes 1 or 0. And The calculation of the crossover operator includes:

[0066]

[0067] Where ma and mb represent the first position vectors of length n, and each element includes 1 or 0; mb represents a randomly selected crossover position, and ma+mb=1; 1 represents a unit vector of length n, n represents the number of candidate parking points, and · represents dot product. The specific operation is as shown in Figure 4 As shown in the figure, the 2nd parking point of the parent 1 is not selected, and the 2nd parking point of the parent 2 is selected and the number of single vehicles is 20. In the crossover operation, the 2nd parking point is randomly selected as the crossover position, so that the 2nd parking point of the parent 1 is changed to be selected and the number of single vehicles is 20 to become the offspring 1, and the 2nd parking point of the offspring 2 is changed to be not selected to become the offspring 2. At this time, the corresponding vectors ma and mb are ma={1, 0, 1, 1, 1, 1, 1, 1, 1} and mb={0, 1, 0, 0, 0, 0, 0, 0, 0}. It is worth noting that in order to improve the diversity of the population, the crossover probability is set to 1, and in the crossover process, there may be a case that both of the two parents of the crossover position are 0, so it is necessary to compare the offspring after the crossover with the parents. If it is the same as the parent, the corresponding offspring is discarded, and if it is different from the parent, it becomes the offspring for subsequent calculation.

[0068] It should be understood that, similar to the crossover operator, the mutation operator also needs to be improved for the multi-objective optimization problem of parking site selection. The original NSGA-II algorithm uses a polynomial method for single-point mutation operation. The form of mutation is c i =x i +β PM . Where βPM is obtained from the following formula:

[0069]

[0070] where rand is a random number uniformly distributed in the interval (0, 1), η PM The custom non-negative real number. The above method may also cause the value of some positions in the chromosome to be not an integer in the offspring individual, on the one hand, and the parking point and the corresponding number of bicycles have a certain correspondence, so it is necessary to improve the mutation operator according to the chromosome coding method in the foregoing.

[0071] Therefore, in step S302, the shared bicycle parking point location method based on multi-point joint distribution, the parent chromosome X3 to be mutated, and the offspring Y3 obtained by using the crossover operator for calculation are represented by the following method:

[0072] Y3 = {(1-mc); (1-mc)} · X3 + {mc; mc} · X m ,

[0073] where the i-th column element in X m is:

[0074]

[0075] In the formula, mc represents the second position vector; δ and δ' are random numbers between 0 and 1, δ is used to determine whether the chromosome of the parking point position or the chromosome of the number of parked bicycles is mutated, and δ' is used to mutate the number of parked bicycles; is the upper limit of the number of bicycles allowed to be parked in the mc i th parking point. The improved mutation operator operation is shown in Figure 4 , the 3rd alternative parking point of the parent 3 is selected and the number of bicycles is 20, in the mutation operation, the 3rd position is randomly selected as the mutation position, and the position 3 contains two variables of whether the parking point is selected and the number of bicycles corresponding to the selected parking point, at this time, a random number r ∈ [0, 1] is generated, and according to the size of r, different variables are operated: if r ≤ 0.5, the mutation of whether to be selected as a parking point is performed, for the parent 3 shown in Figure 4 , the alternative parking point 3 is unselected, and the corresponding number of bicycles is cleared to zero, that is, the offspring 4 shown in Figure 4 ; if r > 0.5, the 3rd parking position of the parent 3 is kept selected, and the number of bicycles corresponding to the alternative point 3 is mutated, that is, Figure 4The offspring 5 is shown. Similarly, similar to the improved crossover operator, the mutated offspring needs to be compared with the parent, if the same as the parent, the corresponding offspring is discarded, if not the same as the parent, it becomes the offspring for subsequent calculation.

[0076] It should be noted that the chromosomes operated by the improved crossover and mutation operators need to be checked for constraint conditions, and the chromosomes that do not meet the constraint conditions are removed, and the chromosomes that meet all the constraint conditions are retained for subsequent population recombination and iterative calculation. After obtaining the offspring population through evolution operation, the elite strategy is applied to combine the offspring population and the parent population to obtain a new population, and then the chromosomes in the population are sorted, crowdedness is calculated, and a number of chromosomes with low non-dominated level and large crowded distance are selected as the parent population for the next iteration calculation.

[0077] According to the steps S200-S400, the maximum number of iterations is reached. Through the above algorithm steps, a plurality of solutions meeting the constraint conditions can be obtained.

[0078] In one embodiment of the present application, in combination with the demand point type, the division results of each branch in the region to the target region are as shown in Figure 6 The distance between the target region candidate parking point and the demand point is shown in Table 1. According to the different types of demand points, the bicycle demand of each point is shown in Table 2.

[0079] Table 1: Distance between candidate points and demand points

[0080]

[0081] Table 2: Bicycle demand of each demand point

[0082]

[0083] After investigation, other parameters of the model are shown in Table 3: the fixed construction and operation cost of each parking point is 2000 yuan; the fixed operation cost of each bicycle per day is 5 yuan; the penalty cost of exceeding the number of parking points and not meeting the demand of bicycles is 10 yuan per vehicle; the completely satisfied distance and the maximum walking distance of the demand point are 100m and 200m respectively; the maximum number of parking points is initially set to 28; the upper limit of the number of vehicles of a single candidate parking point is 50; the maximum number of bicycles allowed in the target region is 600; the minimum satisfaction degree of the demand point is 0.2; and the minimum distance between two parking points is 100m.

[0084] Table 3: Selection of model parameters

[0085]

[0086] The improved NSGA-II algorithm is used to solve the parking site selection model. The initial population size is set to 50, the maximum iteration step is 1000, and the mutation probability is 0.05. The convergence curves of the two objective functions are shown in Figure 7 , and the obtained Pareto solution set is shown in Figure 8 . The related program is written by MATLAB and the average calculation time is 546 s. In addition, it is worth noting that the objective of the model is to make the value of objective function 1 as large as possible while ensuring that the value of objective function 2 is as small as possible. However, the optimization problem generally solves the minimum value of the objective function, so the reciprocal of objective function 1 is taken as the optimization objective.

[0087] It can be seen from Figure 7 that the convergence of objective function 1 and objective function 2 starts around the 142th and 435th iteration steps, respectively. The maximum value of objective function 1 in the final Pareto solution set is 54.88, the minimum value is 25.86, and the average value is 45.26. The minimum value of objective function 2 is 30280, the maximum value is 90140, and the average value is 63183. From the Pareto solution set, 9 solutions with the largest crowding distance are selected as shown in Table 4. Solution 1 selects the least number of parking points, the value of objective function 2 is the smallest, the construction cost is the smallest, and the demand satisfaction degree is the lowest. Conversely, solution 2 selects the most number of parking points, the value of objective function 1 is the largest, i.e. more parking points can make the demand point satisfaction degree the highest, but the corresponding cost will increase. For example, the number of single cars at alternative points 1, 23 and 27 is less than 20, which is much smaller than the given upper limit of the number of single cars, which is not conducive to the sufficient use of parking spaces at parking points.

[0088] Table 4 Partial solutions of Pareto solution set

[0089]

[0090]

[0091] To compare the effectiveness of the improved INSGA-II algorithm, the original NSGA-II algorithm is used to solve the same model. The SBX operator is used for crossover and mutation, and the parameters of the NSGA-II algorithm are the same as those of the INSGA-II algorithm. The convergence of the two objective functions of the model is shown in Figure 9 . The maximum value of objective function 1 obtained by the original NSGA-II algorithm is 53.23, which is lower than 54.88 obtained by the INSGA-II algorithm. The minimum value of objective function 2 obtained by the original NSGA-II algorithm is 31960, which is higher than the minimum value 30280 obtained by the INSGA-II algorithm. That is, the values of the two objective functions obtained by the original NSGA-II algorithm are worse than those obtained by the INSGA-II algorithm. From Figure 9It is also known that the objective function 1 obtained by the two methods starts to converge at the 145th and 667th iteration steps, respectively, and the objective function 2 starts to converge at the 435th and 524th iteration steps, respectively, indicating that the efficiency of the INSGA-II algorithm is superior to that of the NSGA-II algorithm. In summary, the INSGA-II algorithm proposed in this paper is superior to the original NSGA-II algorithm in solving the parking location problem.

[0092] Reference Figure 10 In another embodiment of the present application, the shared bicycle parking point location method based on multi-point joint distribution includes the following steps:

[0093] Step one, according to all demand points and parking candidate point set, determine the initial satisfaction matrix, the distance between each candidate parking point, determine the initial parameters of the model, including the upper limit of the number of parking in each candidate parking point, the upper limit of the number of bicycles in the target area, the bicycle operation cost, etc., and establish the model; Step two, according to the improved chromosome coding mode, combined with the constraint condition, generate an initial population with a size of N; Step three, calculate the values of multiple objective functions of all initial chromosomes, get the non-dominated level and crowding distance of all chromosomes; Step four, use the tournament strategy to select the initial population, and apply the improved crossover and mutation operators to get the offspring population; Step five, use the elite strategy, calculate the non-dominated level and crowding distance of all chromosomes in the new population, select the better N chromosomes as the parent population for the next iteration. Step six, repeat steps three to five until the maximum number of iterations is reached.

[0094] Example 2

[0095] Reference Figure 11In a second aspect, the application provides a shared bicycle parking point location system based on multi-point joint allocation, comprising: an acquisition module 11 configured to acquire information of a plurality of demand points and a plurality of alternative parking points, and determine a plurality of optimization objectives and / or a plurality of constraint conditions of a non-dominated sorting genetic model of multi-point joint allocation according to the information; the constraint conditions include a maximum total amount of bicycles to be put in the alternative parking points, a minimum distance from other alternative parking points, and a minimum satisfaction degree; a calculation module 12 configured to generate an initial population of the non-dominated sorting genetic model based on a binary and integer combined chromosome coding mode and the constraint conditions; calculate a plurality of objective function values of each chromosome in the initial population, and obtain a non-dominated level and a crowding distance of each chromosome according to the plurality of objective function values; an evolution module 13 configured to screen the initial population by using a tournament selection method, and perform crossover and mutation on the screened population based on a position vector crossover operator and a mutation operator to obtain one or more offspring populations; and an iteration module 14 configured to select N optimal chromosomes from the initial population and all offspring populations based on an elitist strategy and the non-dominated level and the crowding distance, and use the N optimal chromosomes as parent populations for the next iteration until the number of iterations reaches a threshold or an optimal solution satisfying one or more constraint conditions is found, wherein N is a positive integer.

[0096] Further, the acquisition module 11 comprises: a calculation unit configured to acquire information of a plurality of demand points and a plurality of alternative parking points, and calculate a satisfaction degree of each alternative parking point according to the information; determine a plurality of optimization objectives of the non-dominated sorting genetic model according to the satisfaction degree of each alternative parking point; and a determination unit configured to determine a plurality of parameters of the non-dominated sorting genetic model, and determine a plurality of constraint conditions of the non-dominated sorting genetic model according to the plurality of parameters.

[0097] Embodiment 3

[0098] Reference Figure 12 In a third aspect, the application provides an electronic device, comprising: one or more processors; and a storage device storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of the first aspect of the application.

[0099] The electronic device 500 can include a processing device (such as a central processor, a graphics processor, etc.) 501, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 502 or programs loaded from a storage device 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0100] Generally, the following devices can be connected to the I / O interface 505: input devices 506, including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 507, including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 508, including, for example, a hard disk, etc.; and communication devices 509. The communication devices 509 can allow the electronic device 500 to communicate wirelessly or via a wire with other devices to exchange data. Although Figure 12 The electronic device 500 is shown with various devices, but it is understood that all of the shown devices are not required to be implemented or present. More or fewer devices can alternatively be implemented or present. Figure 12 Each block shown in the middle can represent one device or, as desired, multiple devices.

[0101] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for executing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication device 509, or installed from the storage device 508, or installed from the ROM 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the methods of embodiments of the present disclosure are executed. It should be noted that the computer readable medium described in embodiments of the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In embodiments of the present disclosure, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus or device. In embodiments of the present disclosure, the computer readable signal medium can include a data signal carried in a baseband or as part of a carrier wave, in which the computer readable program code is carried. Such a propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate or transport a program for use by or in connection with an instruction execution system, apparatus or device. The program code contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to a wire, cable, RF (radio frequency), or the like, or any suitable combination of the above.

[0102] The computer readable medium described above can be included in the electronic device described above; or can exist separately from the electronic device and be not assembled in the electronic device. The computer readable medium described above carries one or more computer programs, which, when executed by the electronic device, cause the electronic device to:

[0103] Computer program code for carrying out operations of embodiments of the present disclosure can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++, Python, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0104] The computer program instructions can also be loaded onto a computer or other programmable information processing apparatus to cause a series of operations to be performed on the computer or other programmable information processing apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable information processing apparatus implement the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0105] The foregoing is merely illustrative of the principles of the application, and the application should not be limited to such detail. Rather, the spirit and scope of the application are limited solely by the claims.

Claims

1. A method for selecting parking spots for shared bicycles based on multi-point joint allocation, characterized in that, include: Information on multiple demand points and multiple alternative parking points is obtained, and multiple optimization objectives and / or multiple constraints are determined based on this information for a non-dominated sorting genetic model for multi-point joint allocation. The constraints include the maximum total number of vehicles deployed per alternative parking point, the minimum distance between alternative parking points, and the minimum satisfaction level. Based on the chromosome encoding method combining binary and integers, and the aforementioned constraints, an initial population for the non-dominated sorting genetic model is generated; multiple objective function values ​​for each chromosome in the initial population are calculated, and the non-dominated level and crowding distance of each chromosome are obtained accordingly. The initial population is selected using a tournament selection method. Crossover and mutation operators based on position vectors are then applied to the selected population to obtain one or more offspring populations. For two parent chromosomes to be crossovered... and The crossover operator is represented by two first position vectors, each containing either 1 or 0 elements, with one of the first position vectors representing a randomly selected crossover position; based on two parent chromosomes to be crossed. and The crossover offspring are obtained by using the crossover operator. and ; For the paternal chromosome to be mutated The mutation operator is represented by a second position vector, and the elements of the second position vector include 1 or 0, where the position with an element of 1 represents the mutation position; based on the parent chromosome to be mutated. The mutated offspring are obtained by using the crossover operator. ; wherein, the basis is two parent chromosomes to be crossed and The crossover offspring are obtained by using the crossover operator. and include: ， in, ma and mb Indicates length is n The first position vector, whose elements all include 1 or 0; mb This represents a randomly selected intersection position, and ma + mb =1; 1 indicates a length of 1. n unit vector, n This indicates the number of alternative parking spots. Dot product; Based on the elitist strategy and non-dominated hierarchy and crowding distance, the optimal N chromosomes are selected from the initial population and all offspring populations, and used as the parent population for the next iteration until the number of iterations reaches a threshold or an optimal solution that satisfies one or more constraints is found, where N is a positive integer.

2. The shared bicycle parking spot selection method based on multi-point joint allocation according to claim 1, characterized in that, The chromosome encoding method based on a combination of binary and integers, and the constraints, generate the initial population for the non-dominated sorting genetic model, including: The selection of a parking spot is encoded using binary code. The number of parking spaces at each of the candidate parking spots is encoded using integers.

3. The shared bicycle parking spot selection method based on multi-point joint allocation according to claim 1, characterized in that, The parent chromosome to be mutated The mutated offspring are obtained by using the crossover operator. This can be represented as follows: ， in, X m The first in i The column elements are: ， In the formula, mc Indicates the second position vector; and All are random numbers between 0 and 1. The chromosome used to determine whether the mutation occurred in the location of the parking spot or the number of parking spots. Used to vary the number of parking spaces; For the first mc i The maximum number of bicycles that can be parked at each parking spot.

4. The shared bicycle parking spot selection method based on multi-point joint allocation according to any one of claims 1 to 3, characterized in that, The process of acquiring information on multiple demand points and multiple alternative parking points, and determining multiple optimization objectives and / or multiple constraints for the non-dominated sorting genetic model of multi-point joint allocation based on this information, includes: Information on multiple demand points and multiple alternative parking spots is obtained, and the satisfaction level of each alternative parking spot is calculated based on this information. Based on the satisfaction level of each alternative parking spot, multiple optimization objectives of the non-dominated ranking genetic model are determined. Determine multiple parameters of the non-dominated sorting genetic model, and determine multiple constraints of the non-dominated sorting genetic model based on them.

5. A shared bicycle parking spot selection system based on multi-point joint allocation, characterized in that, include: The acquisition module is used to acquire information on multiple demand points and multiple alternative parking points, and determine multiple optimization objectives and / or multiple constraints of the non-dominated sorting genetic model for multi-point joint allocation based on it; the constraints include the maximum total number of vehicles deployed per alternative parking point, the minimum distance between alternative parking points and other alternative parking points, and the minimum satisfaction level; The computation module is used to generate an initial population of the non-dominated sorting genetic model based on a chromosome encoding method combining binary and integers and the constraints; calculate multiple objective function values ​​for each chromosome in the initial population, and obtain the non-dominated level and crowding distance of each chromosome based on them. The evolution module is used to select the initial population using tournament selection. Based on position vector-based crossover and mutation operators, it performs crossover and mutation on the selected population to obtain one or more offspring populations. For two parent chromosomes to be crossovered... and The crossover operator is represented by two first position vectors, each containing either 1 or 0 elements, with one of the first position vectors representing a randomly selected crossover position; based on two parent chromosomes to be crossed. and The crossover offspring are obtained by using the crossover operator. and ; For the paternal chromosome to be mutated The mutation operator is represented by a second position vector, and the elements of the second position vector include 1 or 0, where the position with an element of 1 represents the mutation position; based on the parent chromosome to be mutated. The mutated offspring are obtained by using the crossover operator. ; wherein, the basis is two parent chromosomes to be crossed and The crossover offspring are obtained by using the crossover operator. and include: ， in, ma and mb Indicates length is n The first position vector, whose elements all include 1 or 0; mb This represents a randomly selected intersection position, and ma + mb =1; 1 indicates a length of 1. n unit vector, n This indicates the number of alternative parking spots. Dot product; The iteration module is used to select the best N chromosomes from the initial population and all offspring populations based on the elitist strategy, non-dominated hierarchy and crowding distance, and use them as the parent population for the next iteration until the number of iterations reaches a threshold or an optimal solution that satisfies one or more constraints is found, where N is a positive integer.

6. The shared bicycle parking spot selection system based on multi-point joint allocation as described in claim 5, characterized in that, The acquisition module includes: The calculation unit is used to acquire information on multiple demand points and multiple alternative parking points, and calculate the satisfaction level of each alternative parking point; based on the satisfaction level of each alternative parking point, it determines multiple optimization objectives of the non-dominated ranking genetic model. A determining unit is used to determine multiple parameters of a non-dominated sorting genetic model and, based thereon, to determine multiple constraints of the non-dominated sorting genetic model.

7. An electronic device, comprising: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the shared bicycle parking spot selection method based on multi-point joint allocation as described in any one of claims 1 to 4.

8. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by the processor, it implements the shared bicycle parking spot selection method based on multi-point joint allocation as described in any one of claims 1 to 4.

Images