A method for optimizing the layout of hydrogen refueling stations considering users' travel characteristics and traffic flow

Abstract

The present invention discloses a method for optimizing the layout of hydrogen refueling stations in consideration of user travel characteristics and traffic flow. The method comprises: starting from actual traffic flow, predicting the future hydrogen fuel vehicle ownership based on a bass diffusion model, obtaining trajectory data of hydrogen fuel vehicles through a public platform, preprocessing the trajectory data, analyzing user travel characteristics, constructing a hydrogen fuel vehicle energy replenishment demand model, and forming a basic network for hydrogen refueling station site selection; combining the demand replenishment model and the urban traffic network model to form an objective function guided by the hydrogen refueling demand of hydrogen fuel vehicle users and minimizing the sum of the investment and operation cost of the hydrogen refueling station operator and the user's hydrogen refueling cost, constructing an optimal site selection model for charging stations, and using an improved inertia-free reverse particle swarm algorithm to solve the model so that the result can jump out of the local optimum and find the global optimal solution.

Description

Technical Field

[0001] The present invention belongs to the technical field of new energy, and relates to an optimization layout method for hydrogen refueling stations considering user travel characteristics and traffic flow. Background Art

[0002] With the shortage of traditional energy (oil) resources and higher requirements for environmental quality, new energy vehicles, especially hydrogen fuel cell vehicles with more mature technologies, have gradually entered our lives. With the increase in the number of hydrogen fuel cell vehicles, there is a demand for the construction of hydrogen refueling stations. How to arrange hydrogen refueling stations in an economical and efficient way and improve the hydrogen refueling efficiency of new energy vehicles has become an urgent problem to be solved.

[0003] From the perspective of new energy vehicles, it mainly involves minimizing the hydrogen consumption during the journey of the vehicle to the hydrogen refueling station. Therefore, when designing and optimizing the location of urban hydrogen refueling stations, it is mainly considered from two aspects: the residential location of residents and the location of hydrogen refueling stations.

[0004] In addition, considering the investment cost of operators is equally important. Scientifically and reasonably determining the spatial layout of hydrogen refueling stations can not only directly reduce the fixed cost investment of hydrogen refueling stations and the operating cost after production, but also has important practical significance for attracting vehicle owners to refuel and relieving urban traffic pressure. Summary of the Invention

[0005] Objective: To overcome the deficiencies in the prior art, the present invention provides an optimization layout method for hydrogen refueling stations considering user travel characteristics and traffic flow.

[0006] Technical Solution: To solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0007] In the first aspect, an optimization layout method for hydrogen refueling stations considering user travel characteristics and traffic flow is provided, including:

[0008] According to historical urban traffic flow data, a traffic network hydrogen fuel vehicle trajectory dataset is obtained, where the traffic network hydrogen fuel vehicle trajectory dataset contains the state S of the fuel cell at the travel time of the hydrogen fuel vehicle i,o ;

[0009] According to historical urban traffic flow data, based on the potential consumer group conversion rate C set based on policies and environmental protection factors r , based on the Bass diffusion model, the future hydrogen fuel vehicle ownership is predicted;

[0010] According to the state S of the fuel cell at the travel time of the hydrogen fuel vehicle in the traffic network hydrogen fuel vehicle trajectory dataset i,o and the pre-constructed hydrogen fuel vehicle energy consumption model f e, the hydrogen consumption S of the hydrogen fuel vehicle at this moment is obtained i,t , it is judged whether the hydrogen fuel vehicle has a hydrogen refueling demand K, and an energy supply demand model for the hydrogen fuel vehicle is formed;

[0011] According to the traffic network hydrogen fuel vehicle trajectory data set, the future hydrogen fuel vehicle ownership, and the open source map of the target city, a topological map of the urban traffic network is constructed to form the basic network for hydrogen refueling station site selection;

[0012] According to the basic network for hydrogen refueling station site selection and the energy supply demand model of the hydrogen fuel vehicle, a target model is constructed. The target model includes the objective function and constraint conditions with the minimization of the sum of the hydrogen refueling station investment and operation cost and the user hydrogen refueling cost;

[0013] An improved inertia-free reverse particle swarm optimization algorithm is used to solve the target model, and the optimal layout of the hydrogen refueling station is obtained, including the number of hydrogen refueling stations and the location coordinate information.

[0014] In some embodiments, obtaining the traffic network hydrogen fuel vehicle trajectory data set includes:

[0015] The vehicle trajectory data packets provided by the Gaia platform for D days are grouped by "day", and then the orders on the same day are integrated:

[0016] The GPS coordinates of all vehicle trajectories on the x-th day are compared with the range of the research area, and the trajectory data that does not belong to the research scope is removed, where x ∈ D; the trajectories with a time interval exceeding 100 s between adjacent trajectory coordinate points of the same vehicle in the GPS are removed; according to the vehicle driving speed V provided by the data packet i , the average speed of each section of the trajectory is calculated The trajectories with an average speed greater than 120 km / h are removed; the trajectories with a starting straight-line distance of less than 500 meters for the entire trajectory are removed;

[0017] The remaining qualified trajectory data is used for traffic origin-destination (OD) data extraction. The leuvenmapmatching tool in python is used to match the GPS data with the road network in the research area. At the same time, according to the 0-1 change of the passenger-carrying parameter P, the empty and passenger-carrying states are distinguished to form the trajectory network on the x-th day.

[0018] In some embodiments, the traffic network hydrogen fuel vehicle trajectory data set includes the number N of each vehicle i , the vehicle operation time T i , the vehicle GPS positioning (X i , Y i ), the vehicle speed V i and whether it is passenger-carrying P (1 for passenger-carrying, 0 for not passenger-carrying), and the state S of the fuel cell of the hydrogen fuel vehicle at the travel time i,oInformation

[0019] In some embodiments, predicting the future ownership of hydrogen fuel vehicles includes:

[0020] Obtaining the number of potential consumers through statistical research data;

[0021] Determining the conversion rate C of the potential consumer group according to policies and environmental protection factors r ;

[0022] Based on the number of potential consumers and the conversion rate C of the potential consumer group r , the future ownership of hydrogen fuel vehicles is obtained based on the Bass diffusion model.

[0023] Since the Bass diffusion model mainly uses three parameters, namely the maximum market potential, the innovation coefficient, and the imitation coefficient, when making predictions; that is, the calculation of the future ownership of hydrogen fuel vehicles is transformed into the conversion rate of potential consumers of hydrogen fuel vehicles into consumers in the next few years, and relevant policy promotion and potential consumer selection factors should be fully considered;

[0024] Using the parameter settings of the agent change state provided by the Anylogic platform, setting its potential users and conversion rate Cr, so as to obtain the future ownership of hydrogen fuel vehicles.

[0025] In some embodiments, the hydrogen fuel vehicle energy consumption model f e , includes:

[0026]

[0027] Among them, f e is the hydrogen consumption per 100 kilometers, E wh is the energy used to resist the driving resistance of the vehicle, E rgb is the energy charged into the battery by regenerative braking, η tr is the average comprehensive transmission efficiency, η bat_discθrg is the average discharge efficiency of the battery, b e_avg is the average hydrogen consumption rate of the fuel cell, and C is the hydrogen-electric conversion coefficient.

[0028] In some embodiments, determining whether a hydrogen fuel vehicle generates a hydrogen refueling demand K to form a hydrogen fuel vehicle energy replenishment demand model includes:

[0029] According to the mileage traveled by the vehicle at this moment and the pre-constructed hydrogen fuel vehicle energy consumption model f e , obtaining the hydrogen consumption S of the fuel vehicle at this moment i,t , using the state S of the fuel cell at the departure time of the hydrogen fuel vehicle i,oSubtract the hydrogen consumption S of the fuel cell vehicle at this moment i,t to obtain the remaining energy S at this moment t Compare the remaining energy S t with the set warning value L alert If it is lower than the warning value, hydrogen addition is required, and the OD segment of the traffic start and end data from this moment until it returns above the warning value is extracted to obtain the straight-line distance of the traffic start and end data OD

[0030] In some embodiments, a basic network for hydrogen refueling station site selection is formed, including:

[0031] According to the open-source map of the target city, using the python and Kepler platforms, places such as airports, railway stations, bus stops, supermarkets, parks, art galleries, stadiums, and independent parking lots with transfer situations in the area are marked as candidate points, and based on the traffic network hydrogen fuel vehicle trajectory dataset, high-frequency points where hydrogen fuel vehicles have hydrogen addition requirements are merged to form Cp candidate points, forming a basic network for hydrogen refueling station site selection, and subsequent planning is carried out on the basic candidate points

[0032] In some embodiments, the target model includes:

[0033] Objective function:

[0034] minC = α(C1 + C2 + C3K year - C4)+β(T1 + T2)K year

[0035] C1 = d·e -bx ·s

[0036] C2 = b + k·q

[0037] C3 = μC2

[0038] C4 = 0.2C2

[0039]

[0040]

[0041] Among them, C is the total cost; C1 is the land cost of the hydrogen refueling station; d is the land price at the center of the research area; b is the change rate of the distance to the center of the research area and the price; s is the occupied area of the hydrogen refueling station; C2 is the construction cost of the hydrogen refueling station; b is the cost of the fixed equipment of the hydrogen refueling station; q is the number of hydrogen dispensers; k is the proportionality coefficient; C3 is the annual operating cost of the hydrogen refueling station, which includes equipment maintenance and staff salaries, and is proportional to the construction scale μ; C4 is the government subsidy cost, which is subsidized at 20% of the actual equipment investment amount, with a maximum of no more than 4 million; T1 is the time cost required for users from generating hydrogen refueling demand to arriving at the hydrogen refueling station; among them, δ is the road straight line coefficient, θ is the road congestion coefficient, is the straight-line distance from the demand generation point to the hydrogen refueling station, is the average speed of the fuel cell vehicle for this section of the journey, t u is the unit time cost of the user; T2 is the user queuing time cost; among them, r b is the service capacity coefficient of this hydrogen refueling station, w b is the queuing time of this hydrogen refueling station; K year is the preset operating life of the hydrogen refueling station; α, β are the weight coefficients of the investment cost and the user cost;

[0042] Constraints:

[0043] C4 ≤ 400

[0044] cp ≤ X b ·Cp

[0045] a pmin ≤ a p ≤ a pmax

[0046] Among them, the government subsidy C4 cannot exceed 4 million; cp is the number of hydrogen refueling stations, Cp is the candidate site, X b judges whether to build a station at this candidate point; a p is the number of hydrogen dispensers of the hydrogen refueling station.

[0047] In some embodiments, an improved inertia-free reverse particle swarm optimization algorithm is used to solve the target model, including:

[0048] The velocity and position update formulas of the general particle swarm optimization algorithm are:

[0049] v i,j (t + 1) = ωv i,j (t) + c1rand1(pbest i,j - x i,j (t))

[0050] + c2rand2(gbest i,j - x i,j (t))

[0051] x i,j (t + 1) = x i,j (t) + v i,j (t + 1)

[0052] where v i,j (t), x i,j (t) is the velocity and position at time t; ω ∈ [0, 1] is the inertia weight; c1, c2 ∈ [0, 2] are the individual learning factor and the social learning factor respectively; rand1, rand2 ∈ [0, 1] are two random numbers obeying the uniform distribution; pbest i,j is the individual optimum; gbest i,j is the global optimum;

[0053] To prevent falling into local optimum, it is necessary to absorb a large number of population environmental factors. Therefore, the reverse learning strategy is used to expand the search range to achieve global optimum. From the velocity iteration formula of the general particle swarm optimization (PSO), it can be seen that the next iteration is affected by both the individual and the global parts. Among them, the inertial movement belongs to the individual influencing factor and needs to be reduced. Therefore, it is proposed to remove the inertial part and use a new velocity iteration formula to provide direction for the next generation of particles;

[0054] Among them, the formula of the reverse learning strategy is as follows:

[0055] Set as x i = (x i,1 , x i,2 , … x i,D )'s reverse solution, then 's definition is:

[0056]

[0057] Among them, D is the space dimension, j = 1, 2, ···, D;

[0058] k ∈ (0, 1) is a random number obeying the uniform distribution, da j , db j are the upper and lower dynamic boundaries of the j - th dimensional search space for particle ;

[0059] The improved velocity iteration formula is:

[0060]

[0061] Among them, u(t - 1) is the mean value of the positions of all particles in the population at time (t - 1), s ∈ (0, 1) is the difference coefficient; ω ∈ [0, 1] is the inertia weight; c1, c2 ∈ [0, 2] are the individual learning factor and the social learning factor respectively; rand1, rand2 ∈ [0, 1] are two random numbers following a uniform distribution; pbest i,j is the individual optimum; gbest i,j is the global optimum; N is the total number of the population;

[0062] By introducing the mean value of the positions of all particles at time (t - 1), the position variance between two adjacent times is calculated to provide a guiding direction for the movement of the next generation of particles;

[0063] In addition, by adjusting the α and β coefficients in the objective function, the weights between the investors and users are changed, thereby changing the specific demand plan and providing more plan options for the construction of hydrogen refueling stations.

[0064] In a second aspect, the present invention provides a hydrogen refueling station optimization layout device considering user travel characteristics and traffic flow, including a processor and a storage medium;

[0065] The storage medium is used to store instructions;

[0066] The processor is used to operate according to the instructions to execute the steps of the method according to the first aspect.

[0067] In a third aspect, the present invention provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method according to the first aspect are realized.

[0068] Beneficial effects: The hydrogen refueling station optimization layout method considering user travel characteristics and traffic flow provided by the present invention has the following advantages: Using the actual traffic flow as the basis for predicting the number of hydrogen fuel cell vehicles, regarding the places with transfer situations such as airports, railway stations, bus stops, supermarkets, parks, art galleries, stadiums, and independent parking lots in the target city as candidate site selection addresses for hydrogen refueling stations, and combining the road topology of the target city to predict the hydrogen refueling demand characteristics of hydrogen fuel cell vehicles; Taking the minimum sum of the hydrogen refueling time cost of hydrogen fuel cell vehicle users and the investment and operation cost of hydrogen refueling stations as the goal, a most optimal site selection model for hydrogen refueling stations is established, and an improved inertia - less reverse particle swarm optimization algorithm is used to solve the model, so that finally it can jump out of the local optimal solution and obtain the global optimum, scientifically select the sites of hydrogen refueling stations in the region, thereby providing a reliable reference for the subsequent optimization and planning of the service site selection of hydrogen fuel cell vehicles, and further promoting the active development of the hydrogen fuel cell vehicle industry. Brief Description of the Drawings

[0069] Figure 1 It is a technical route diagram of a hydrogen refueling station site selection method according to an embodiment of the present invention;

[0070] Figure 2 Flow chart of improved inertia - free inverse particle swarm optimization according to an embodiment of the present invention;

[0071] Figure 3 Vehicle trajectory processing result diagram according to an embodiment of the present invention. Detailed implementation manners

[0072] The present invention will be further described below with reference to the accompanying drawings and embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and cannot be used to limit the protection scope of the present invention.

[0073] In the description of the present invention, the meaning of several is more than one, the meaning of multiple is more than two, greater than, less than, exceeding, etc. are understood as not including the number itself, and above, below, within, etc. are understood as including the number itself. If the first and the second are described only for the purpose of distinguishing technical features, they cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features or implicitly indicating the sequence relationship of the indicated technical features.

[0074] In the description of the present invention, the description with reference to terms such as "an embodiment", "some embodiments", "schematic embodiments", "examples", "specific examples", 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 descriptions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0075] Embodiment 1

[0076] According to an embodiment of the present application, referring to Figure 1 , a method for optimizing the layout of hydrogen refueling stations considering user travel characteristics and traffic flow includes:

[0077] S1. Obtain a traffic network hydrogen fuel vehicle trajectory data set according to historical urban traffic flow data, where the traffic network hydrogen fuel vehicle trajectory data set includes the number N of each vehicle i , the vehicle operation time T i , the vehicle GPS positioning (X i , Y i ), the vehicle speed V i , and whether it is carrying passengers P (1 means carrying passengers, 0 means not carrying passengers), and the state S of the fuel cell at the time of travel of the hydrogen fuel vehicle i,o information;

[0078] S2. Based on the historical urban traffic flow data, through the potential consumer group conversion rate C set based on policies and environmental protection factors, predict the future hydrogen fuel vehicle ownership based on the Bass diffusion model; r

[0079] S3. According to the state S of the fuel cell at the travel time of the hydrogen fuel vehicle in the hydrogen fuel vehicle trajectory data set of the transportation network i,o and the pre - constructed hydrogen fuel vehicle energy consumption model f e obtain the hydrogen consumption S of the hydrogen fuel vehicle at this moment i,t judge whether the hydrogen fuel vehicle has a hydrogen refueling demand K, and form a hydrogen fuel vehicle energy replenishment demand model;

[0080] S4. According to the hydrogen fuel vehicle trajectory data set of the transportation network, the future hydrogen fuel vehicle ownership, and the open - source map of the target city, construct an urban traffic network topology map to form the basic network for hydrogen refueling station location;

[0081] S5. According to the basic network for hydrogen refueling station location and the hydrogen fuel vehicle energy replenishment demand model, construct a target model, where the target model includes minimizing the sum of the hydrogen refueling station investment and operation cost and the user hydrogen refueling cost as the objective function and constraint conditions;

[0082] S6. Use the improved inertia - less reverse particle swarm optimization algorithm to solve the target model to obtain the optimal layout of hydrogen refueling stations, including the number and location coordinate information of hydrogen refueling stations.

[0083] The above steps are described in detail as follows:

[0084] In some embodiments, in step S1, to obtain the hydrogen fuel vehicle trajectory data set of the transportation network, the specific operations are as follows:

[0085] Group the vehicle trajectory data packets provided by the Gaia platform in units of "days", and then integrate the orders on the same day.

[0086] Compare the GPS coordinates of all vehicle trajectories on the x - th day (x ∈ D) with the range of the research area, and remove the trajectory data that does not belong to the research scope; remove the trajectories where the time interval between adjacent trajectory coordinate points of the same vehicle by GPS exceeds 100s. Generally, the data collection frequency is controlled within 2 - 20 seconds. Exceeding 100s indicates a GPS positioning failure; according to the vehicle driving speed V i provided by the data packet, calculate the average speed of each section of the trajectory Remove the trajectories with an average speed greater than 120 km / h; remove the trajectories with a starting straight - line distance of less than 500 meters for the entire trajectory.

[0087] ​Perform OD extraction on the remaining eligible trajectory data. Use the leuvenmapmatching tool in Python to match the GPS data with the road network in the research area. At the same time, distinguish between the empty and occupied states according to the change of the passenger-carrying parameter P from 0 to 1, and form the trajectory network for the x-th day, as Figure 3 shown.

[0088] In some embodiments, in step S2, the method for predicting the future hydrogen fuel vehicle ownership is as follows:

[0089] Obtain the potential consumer volume through research data statistics;

[0090] Determine the conversion rate C of the potential consumer group according to policies and environmental protection factors r ;

[0091] Based on the potential consumer volume and the conversion rate C of the potential consumer group r , obtain the future hydrogen fuel vehicle ownership based on the Bass diffusion model.

[0092] Since the Bass diffusion model mainly uses three parameters, namely the maximum market potential, the innovation coefficient, and the imitation coefficient, when making predictions. That is, the calculation of the future ownership of hydrogen fuel vehicles is transformed into the conversion rate of potential consumers of hydrogen fuel vehicles into consumers in the next few years, and relevant policy promotion and potential consumer selection factors should be fully considered.

[0093] Use the parameter settings of the agent change state provided by the Anylogic platform to set its potential users, users, and conversion rate C r , so as to obtain the future hydrogen fuel vehicle ownership.

[0094] In some embodiments, the state S of the fuel cell at the travel time of the hydrogen fuel vehicle in step S3 i,o and the hydrogen fuel vehicle energy consumption model f e :

[0095] Considering the commercial nature of taxis, the operator will complete energy replenishment before departure, so the initial fuel cell state S of the commercial vehicle i,o is set to 100%; and the hydrogen fuel vehicle energy consumption model is as follows:

[0096]

[0097] where f e is the hydrogen consumption per 100 kilometers, E wh is the energy used to resist the driving resistance of the vehicle, E rgb is the energy charged into the battery by regenerative braking, and η tris the average comprehensive transmission efficiency, η bat_dischrg is the average discharge efficiency of the battery, b e_avg is the average hydrogen consumption rate of the fuel cell, and C is the hydrogen-electric conversion coefficient.

[0098] In some embodiments, in step S4, the method for judging whether a fuel vehicle has a hydrogen refueling demand K is as follows:

[0099] According to the mileage (trajectory movement distance) of the vehicle at this moment and the hydrogen consumption per 100 kilometers f of the above hydrogen fuel vehicle e , obtain the percentage S of the hydrogen consumption in the total energy i,t , use the initial S i,o subtract the S at this moment i,t , obtain the remaining energy S t , compare this energy with the warning value L alert , if it is lower than the warning value, hydrogen refueling is required, and extraction is performed from this moment until the OD section above the warning value is reached, and the straight-line distance of this OD is obtained

[0100] In some embodiments, in step S4, the method for constructing the basic network for hydrogen refueling station location is as follows:

[0101] According to the open-source map of the target city, using python and the Kepler platform, mark the places with transfer situations such as airports, railway stations, bus stops, supermarkets, parks, art galleries, stadiums, and independent parking lots in the area as candidate points, and merge the high-frequency points where hydrogen fuel vehicles have hydrogen refueling demands according to the vehicle trajectory data to form Cp candidate points, forming the basic network for hydrogen refueling station location, and subsequent planning is carried out on the basic candidate points.

[0102] In some embodiments, in step S5, the target model includes:

[0103] Objective function:

[0104] minC = α(C1 + C2 + C3K year - C4) + β(T1 + T2)K year

[0105] C1 = d·e -bx ·s

[0106] C2 = b + k·q

[0107] C3 = μC2

[0108] C4 = 0.2C2

[0109]

[0110]

[0111] Among them, C is the total cost; C1 is the land cost of the hydrogen refueling station; d is the land price at the center of the research area; b is the change rate of the distance to the center of the research area and the price; s is the occupied area of the hydrogen refueling station;; C2 is the construction cost of the hydrogen refueling station; b is the cost of the fixed equipment of the hydrogen refueling station (such as hydrogen storage, hydrogen production, etc.); q is the number of hydrogen dispensers; k is the proportionality coefficient; C3 is the annual operating cost of the hydrogen refueling station, which includes equipment maintenance and staff salaries, and is proportional to the construction scale μ; C4 is the government subsidy cost, which is subsidized at 20% of the actual equipment investment amount, with a maximum of no more than 4 million; T1 is the time cost required for users to generate hydrogen refueling demand and reach the hydrogen refueling station; among them, δ is the road straight line coefficient, θ is the road congestion coefficient, is the straight-line distance from the demand generation point to the hydrogen refueling station, is the average speed of the fuel cell vehicle on this section of the journey, t u is the user's unit time cost; T2 is the user's queuing time cost; among them, r b is the service capacity coefficient of this hydrogen refueling station, w b is the queuing time of this hydrogen refueling station; K year is the preset operating life of the hydrogen refueling station; α, β are the weight coefficients of the investment cost and the user cost;

[0112] Constraint conditions:

[0113] C4 ≤ 400

[0114] cp ≤ X b ·Cp

[0115] a pmin ≤ a p ≤ a pmax

[0116] Among them, the government subsidy C4 cannot exceed 4 million; cp is the number of hydrogen refueling stations, Cp is the candidate site, X b judges whether to build a station at this candidate point; a p is the number of hydrogen dispensers of the hydrogen refueling station.

[0117] In some embodiments, in step S6, the improved inertia-free reverse particle swarm optimization algorithm is as follows:

[0118] The velocity and position update formulas of the general particle swarm optimization algorithm are:

[0119] v i，j (t + 1) = ωv i，j (t) + c1rand1(pbest i，j -x i，j (t)) + c2rand2(gbest i，j -x i，j (t))xi，j (t + 1) = x i，j (t) + v i，j (t + 1)

[0120] where ω ∈ [0, 1] is the inertia weight; c1, c2 ∈ [0, 2] are the individual learning factor and the social learning factor respectively; rand1, rand2 ∈ [0, 1] are two random numbers following a uniform distribution; pbest i，j is the individual optimum; gbest i，j is the global optimum.

[0121] To prevent falling into local optima, a large number of population environmental factors need to be absorbed. Therefore, the reverse learning strategy is used to expand the search range to achieve global optimality. From the velocity iteration formula of the general PSO, it can be seen that the next iteration is affected by both the individual and the global parts. Among them, the inertial movement belongs to the individual influencing factor and needs to be reduced. Therefore, it is proposed to remove the inertial part and use a new velocity iteration formula to provide directions for the next generation of particles. The reverse learning strategy formula is as follows:

[0122] Set to be the reverse solution of x i =(x i,1 , x i,2 ,…x i,D ), then the definition of is:

[0123]

[0124] where D is the space dimension, j = 1, 2, ···, D;

[0125] k ∈ (0, 1) is a random number following a uniform distribution, da j , db j are the upper and lower dynamic boundaries of the j-th dimensional search space for particle ;

[0126] The improved velocity iteration formula is:

[0127]

[0128] where u(t - 1) is the mean value of the positions of all particles in the population at time (t - 1), s ∈ (0, 1) is the difference coefficient, and the other parameters are still defined as before.

[0129] By introducing the mean value of the positions of all particles at time (t - 1), the position variance between two adjacent times is calculated to provide a guiding direction for the movement of the next generation of particles.

[0130] In addition, by adjusting the α and β coefficients in the objective function, the weights between investors and users can be changed, thereby changing the specific demand plan and providing more options for the construction of hydrogen refueling stations.

[0131] To further elaborate on the specific process of the above vehicle trajectory data processing, this application gives a specific example. The following is the specific operation process, and the results are shown as Figure 3 :

[0132] Taking days as the time unit, the overall data provided by the online car-hailing platform is divided into single-day data. By integrating the single-day data, the disorder of the data is reduced, providing data support for the subsequent spatio-temporal distribution characteristics of hydrogen demand.

[0133] The first step is to remove the trajectory data that does not belong to the research area. Using the positioning system built into the online car-hailing service, trajectory points are formed within the research area, and the trajectories beyond the range are deleted.

[0134] The second step is to remove the duplicate data within 0.5 kilometers and within 1 minute of the same order. Considering that there may be personal reasons for drivers or passengers to cancel orders of this type, or actual traffic obstacles to cause repeated detours, resulting in duplicate vehicle trajectories, such complex situations are not considered for the time being.

[0135] The third step is to remove the trajectories where the time interval between adjacent trajectory coordinate points of the same vehicle by GPS exceeds 100 seconds. Generally, the data acquisition frequency is controlled within 2 - 20 seconds. If it exceeds 100 seconds, it indicates that there is a GPS positioning failure.

[0136] The fourth step is to remove the trajectory data with a speed exceeding 120 kilometers per hour. According to the traffic regulations in the research area, the maximum driving speed within the urban area is 70 kilometers per hour. If the trajectory data with a speed exceeding 120 kilometers per hour appears, it indicates that the vehicle may operate across urban areas with too long a distance, which does not conform to the relevant settings of this solution.

[0137] The present invention takes the actual traffic flow as the basis for predicting the number of hydrogen fuel cell vehicles, selects the places with transfer situations such as airports, railway stations, bus stops, supermarkets, parks, art galleries, stadiums, independent parking lots, etc. in the target city as candidate site selection addresses for hydrogen refueling stations, combines the road topology of the target city to predict the hydrogen refueling demand characteristics of hydrogen fuel cell vehicles; aims to minimize the sum of the hydrogen refueling time cost of hydrogen fuel cell vehicle users and the investment and operation cost of hydrogen refueling stations, establishes an optimal site selection model for hydrogen refueling stations, and uses an improved inertia-free reverse particle swarm optimization algorithm to solve the model, so that finally it can jump out of the local optimal solution and obtain the global optimum, scientifically select the sites of hydrogen refueling stations within the region, thereby providing a reliable reference for the subsequent optimization and planning of the service site selection of hydrogen fuel cell vehicles, and further promoting the active development of the hydrogen fuel cell vehicle industry.

[0138] Example 2

[0139] In a second aspect, this embodiment provides a hydrogen refueling station optimal layout device that takes into account user travel characteristics and traffic flow, including a processor and a storage medium;

[0140] The storage medium is used to store instructions;

[0141] The processor is configured to operate according to the instructions to execute the steps of the method described in Embodiment 1.

[0142] Embodiment 3

[0143] In a third aspect, this embodiment provides a storage medium with a computer program stored thereon, and when the computer program is executed by a processor, it implements the steps of the method described in Embodiment 1.

[0144] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0145] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0146] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device that implements the functions specified in Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0147] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to produce a computer-implemented process, thereby the instructions executed on the computer or other programmable apparatus provide steps for realizing the functions specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 and / or steps for realizing the functions specified in one block or a plurality of blocks.

[0148] The above are only the preferred embodiments of the present invention, and it should be pointed out that: for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A method for optimizing the layout of hydrogen refueling stations considering user travel characteristics and traffic flow, characterized in that Including: According to historical urban traffic flow data, obtain a traffic network hydrogen fuel vehicle trajectory dataset, where the traffic network hydrogen fuel vehicle trajectory dataset contains the state S of the fuel cell at the time of hydrogen fuel vehicle travel i,o ; According to the historical urban traffic flow data, the conversion rate of potential consumer groups C is set based on policy and environmental factors. r , based on the Bass diffusion model, predict the future number of hydrogen fuel cell vehicles; According to the state S of the fuel cell at the travel time of the hydrogen fuel vehicle in the traffic network hydrogen fuel vehicle trajectory dataset i,o and the pre-built hydrogen fuel vehicle energy consumption model f e , the hydrogen consumption S of the hydrogen fuel vehicle at this moment is obtained i,t , determine whether the hydrogen fuel vehicle has a hydrogen refueling demand K, and form a hydrogen fuel vehicle energy replenishment demand model; Construct a city traffic network topology map based on the traffic network hydrogen fuel vehicle trajectory dataset, the future hydrogen fuel vehicle ownership, and the open-source map of the target city to form the basic network for hydrogen refueling station site selection; Construct a target model based on the basic network for hydrogen refueling station site selection and the hydrogen fuel vehicle energy replenishment demand model. The target model includes minimizing the sum of the hydrogen refueling station investment and operation cost and the user hydrogen refueling cost as the objective function and constraint conditions; Use an improved inertia-free reverse particle swarm optimization algorithm to solve the target model to obtain the optimal layout of hydrogen refueling stations, including the number and location coordinate information of hydrogen refueling stations.

2. The hydrogen refueling station optimization layout method considering user travel characteristics and traffic flow according to claim 1, wherein Obtain the traffic network hydrogen fuel vehicle trajectory dataset, including: Group the vehicle trajectory data packets provided by the Gaia platform in units of "days", and then integrate the orders on the same day: Compare the GPS coordinates of all vehicle trajectories on the x-th day with the scope of the study area, and remove the trajectory data that does not belong to the study scope, where x ∈ D; remove the trajectories with a time interval exceeding 100 s between adjacent trajectory coordinate points of the same vehicle in GPS; calculate the average speed of each segment of the trajectory according to the vehicle driving speed V provided by the data packet i , and calculate the average speed of each segment of the trajectory Remove the trajectories with an average speed greater than 120 km / h; remove the trajectories with a starting straight-line distance of less than 500 meters for the entire trajectory; Extract the origin-destination (OD) data of the remaining eligible trajectory data, use the leuvenmapmatching tool in Python to match the GPS data with the road network in the research area, and distinguish between the empty and occupied states according to the 0 / 1 change of the passenger-carrying parameter P to form the trajectory network for the x-th day; Among them, the traffic network hydrogen fuel vehicle trajectory dataset contains the number N of each vehicle i , the vehicle operation time T i , the vehicle GPS positioning, the vehicle speed V i , and whether it is carrying passengers P, the state S of the fuel cell at the time of the hydrogen fuel vehicle's trip i,o information.

3. The optimized layout method of a hydrogen refueling station considering the travel characteristics of users and traffic flow according to claim 1, characterized in that Predict the future hydrogen fuel vehicle ownership, including: Obtain the number of potential consumers through research data statistics; Determine the conversion rate C of potential consumer groups according to policies and environmental protection factors r ; Based on the potential consumer volume and the conversion rate C of the potential consumer group r , the future ownership of hydrogen fuel vehicles is obtained based on the Bass diffusion model.

4. The hydrogen refueling station optimization layout method considering user travel characteristics and traffic flow according to claim 1, characterized in that Hydrogen fuel cell vehicle energy consumption model f e , including: Among them, f e is the hydrogen consumption per 100 kilometers, E wh is the energy used to resist the driving resistance of the vehicle, E rgb is the energy regeneratively braked and charged into the battery, η tr is the average comprehensive transmission efficiency, η bat_dischrg is the average discharge efficiency of the battery, b e_avg is the average hydrogen consumption rate of the fuel cell, and C is the hydrogen-electric conversion coefficient.

5. The hydrogen refueling station optimization layout method considering user travel characteristics and traffic flow according to claim 1, characterized in that Judge whether the hydrogen fuel vehicle generates a hydrogen refueling demand K to form a hydrogen fuel vehicle energy replenishment demand model, including: According to the mileage traveled by the vehicle at that moment and the pre-constructed hydrogen fuel vehicle energy consumption model f e , obtain the hydrogen consumption S of the fuel vehicle at that moment i,t , subtract the hydrogen consumption S of the fuel vehicle at that moment from the state S of the fuel cell at the travel time of the hydrogen fuel vehicle i,o to obtain the remaining energy S at that moment i,t , compare the remaining energy S t with the set warning value L t . If it is lower than the warning value, hydrogen addition is required, and the OD segment of the traffic start and end data from that moment until it returns above the warning value is extracted to obtain the straight-line distance of the traffic start and end data OD alert ​ 6. The hydrogen refueling station optimization layout method considering user travel characteristics and traffic flow according to claim 1, characterized in that Form the basic network for hydrogen refueling station site selection, including: Based on the open-source map of the target city, use Python and the Kepler platform to mark the places with transfer situations such as airports, railway stations, bus stops, supermarkets, parks, art galleries, stadiums, and independent parking lots in the area as candidate points, and combine the high-frequency points where hydrogen fuel vehicles generate hydrogen refueling demands according to the traffic network hydrogen fuel vehicle trajectory dataset to form Cp candidate points to form the basic network for hydrogen refueling station site selection, and subsequent planning is carried out on the basic candidate points.

7. The optimized layout method of hydrogen refueling stations considering user travel characteristics and traffic flow according to claim 1, characterized in that The target model includes: Objective function: minC = α(C1 + C2 + C3K year - C4) + β(T1 + T2)K year C1 = d·e -bx ·s C2 = b + k·q C3 = μC2 C4 = 0.2C2 Among them, C is the total cost; C1 is the land cost of the hydrogen refueling station; d is the land price at the center of the research area; b in the formula of C1 is the change rate of the distance to the center of the research area and the price; s is the occupied area of the hydrogen refueling station; C2 is the construction cost of the hydrogen refueling station; b in the formula of C2 is the cost of the fixed equipment of the hydrogen refueling station; q is the number of hydrogen dispensers; k is the proportionality coefficient; C3 is the annual operating cost of the hydrogen refueling station, which includes equipment maintenance and staff salaries, and is proportional to the construction scale μ; C4 is the government subsidy cost, which is subsidized at 20% of the actual equipment investment amount, with a maximum of no more than 4 million; T1 is the time cost required for users to generate hydrogen refueling demand and reach the hydrogen refueling station; among them, δ is the road straight-line coefficient, θ is the road congestion coefficient, is the straight-line distance from the demand generation point to the hydrogen refueling station, is the average speed of the fuel cell vehicle on this section of the journey, t u is the user's unit time cost; T2 is the user queuing time cost; among them, r b is the service capacity coefficient of this hydrogen refueling station, w b is the queuing time of this hydrogen refueling station; K year is the preset operating life of the hydrogen refueling station; α is the weight coefficient of the investment cost, β is the weight coefficient of the user cost; Constraint conditions: C4≤400 cp ≤ X b · Cp a pmin ≤ a p ≤ a pmax Among them, the government subsidy cost C4 cannot exceed 4 million; cp is the number of hydrogen refueling stations, Cp is the number of candidate sites, X b Judge whether to build a station at this candidate site; a p is the number of hydrogen dispensers at the hydrogen refueling station.

8. The optimized layout method of a hydrogen refueling station considering user travel characteristics and traffic flow according to claim 7, characterized in that, Use an improved inertia-free reverse particle swarm optimization algorithm to solve the target model, including: The velocity and position update formulas of the general particle swarm optimization algorithm are: v i,j (t + 1) = ωv i,j (t) + c1rand1(pbest i,j -x i,j (t)) + c2rand2(gbest i,j -x i,j (t)) x i,j (t + 1)=x i,j (t)+v i,j (t + 1) Among them, v i,j (t), x i,j (t) are the velocity and position at time t; ω ∈ [0, 1] is the inertia weight; c1, c2 ∈ [0, 2] are the individual learning factor and the social learning factor respectively; rand1, rand2 ∈ [0, 1] are two random numbers obeying the uniform distribution; pbest i,j is the individual optimum; gbest i,j is the global optimum; To prevent falling into local optima, a large number of population environmental factors need to be absorbed. Therefore, the reverse learning strategy is used to expand the search range to achieve global optimality; and from the velocity iteration formula of the general particle swarm optimization algorithm PSO, the next iteration is affected by both the individual and the global parts. Among them, the inertial motion belongs to the individual influence factor and needs to be reduced. Therefore, it is proposed to remove the inertial part and use a new velocity iteration formula to provide directions for the next generation of particles; Among them, the reverse learning strategy formula is as follows: Set as x i =(x i,1 , x i,2 , … x i,D ) is the reverse solution, then is defined as: where D is the spatial dimension, j = 1, 2, ···, D; k ∈ (0, 1) is a random number following a uniform distribution, da j , db j are the upper and lower dynamic boundaries of the particle in the j-th dimensional search space; The improved velocity iteration formula is: Among them, u(t - 1) is the mean value of the positions of all particles in the population at time (t - 1), s ∈ (0, 1) is the difference coefficient; ω ∈ [0, 1] is the inertia weight; c1, c2 ∈ [0, 2] are the individual learning factor and the social learning factor respectively; rand1, rand2 ∈ [0, 1] are two random numbers following a uniform distribution; pbest i,j is the individual optimum; gbest i,j is the global optimum; N is the total number of the population; By introducing the position mean of all particles at the (t - 1) moment, calculate the position variance between two adjacent moments to provide a guiding direction for the movement of the next generation of particles; In addition, by adjusting the α and β coefficients in the objective function, change the weights between investors and users, thereby changing the specific demand plan and providing more plan options for the construction of hydrogen refueling stations.

9. An optimization layout device for a hydrogen refueling station considering user travel characteristics and traffic flow, characterized in that Comprising a processor and a storage medium; The storage medium is used for storing instructions; The processor is used for operating according to the instructions to execute the steps of the method according to any one of claims 1 to 8.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images