A Method for Selecting the Location of Hydrogen Refueling Stations Based on the Data of Hydrogen Energy Vehicles and Hydrogen Production Plants

Abstract

The present invention provides a method for selecting the location of a hydrogen refueling station based on the data of hydrogen energy vehicles and hydrogen production plants. An objective optimization problem is established from two aspects of hydrogen refueling demand and cost. Combining big data of hydrogen energy vehicles with a suitable algorithm provides an effective way to solve the above optimization problem. When selecting the location of the hydrogen refueling station, the present invention comprehensively considers multiple factors such as the hydrogen refueling behavior of hydrogen energy vehicles, the hydrogen refueling demand of alternative locations of the hydrogen refueling station, and the production, storage, transportation costs of hydrogen energy and the vehicle hydrogen refueling cost, etc., thereby overcoming the technical problem that the location selection of the hydrogen refueling station and the hydrogen production plant do not cooperate with each other, affecting the hydrogen refueling efficiency, and having many beneficial effects not possessed by the prior art.

Description

Technical Field

[0001] The present invention belongs to the technical field of hydrogen energy industry facility planning, and particularly relates to a method for selecting a hydrogen refueling station location based on big data related to hydrogen energy applications such as hydrogen energy vehicles and hydrogen production plants. Background Art

[0002] As a truly clean energy, hydrogen energy is one of the important development directions in the field of new energy technologies. In recent years, it has also been strongly promoted by the state. The number of hydrogen energy vehicles in some demonstration operation cities has reached a scale of thousands. However, in the current field of new energy vehicles, chemical batteries such as lithium batteries are still the mainstream energy source, and the overall industry surrounding chemical batteries is gradually improving. The power generation, power transmission, and power supply network with the power grid, charging facilities, and vehicles as nodes has taken initial shape. Although hydrogen energy has many advantages over other new energy sources, due to the relatively recent start of the construction of the current hydrogen energy industry chain, the basic supporting facilities such as hydrogen refueling facilities and hydrogen production plants are relatively lagging behind. The fuel supply of hydrogen energy vehicles has become the main problem faced at present, which has hindered the use and popularization efficiency of hydrogen energy vehicles to a certain extent.

[0003] There are the following two main problems in the selection of hydrogen refueling station locations at the present stage:

[0004] 1. The location selection of hydrogen refueling stations does not match the distribution area of hydrogen energy vehicles, and it is impossible to meet the vehicle's need for nearby hydrogen refueling. Hydrogen energy vehicles often need to travel from a long distance to refuel, and then return to the vehicle's destination after refueling, reducing the vehicle's use efficiency.

[0005] 2. The radiation ranges of hydrogen refueling stations and hydrogen production plants do not match. At present, the service coverage distance of hydrogen production plants is 200 - 250 kilometers, but the main basis for the location selection of hydrogen refueling stations and hydrogen production plants still stays in urban construction planning, without considering the mutual coverage between the two. For hydrogen refueling stations at a long distance, sufficient hydrogen supply cannot be provided, resulting in frequent shortage of hydrogen sources at hydrogen refueling stations and affecting the operation of hydrogen refueling stations.

[0006] Therefore, how to solve the above problems of the location selection of hydrogen refueling stations and hydrogen production plants, and provide a suitable method for selecting the location of hydrogen refueling stations for the radiation area of hydrogen production plants to meet the supply - demand balance of hydrogen energy is a technical problem urgently to be solved in this field. Summary of the Invention

[0007] In view of the above - mentioned technical problems existing in this field, the present invention provides a method for selecting the location of a hydrogen refueling station based on data of hydrogen energy vehicles and hydrogen production plants, specifically including the following steps:

[0008] Step 1: Collect the operation data of hydrogen energy vehicles in a certain area through the hydrogen energy vehicle big data platform, and obtain the hydrogen refueling demand-related information of each vehicle, including the daily average mileage, daily average hydrogen consumption, vehicle operation area distribution, hydrogen remaining amount at different location nodes, vehicle hydrogen refueling time distribution, etc.;

[0009] Step 2: For each alternative address of the planned hydrogen refueling station, establish a hydrogen refueling demand function faced by the corresponding hydrogen refueling station. Taking the alternative address of the hydrogen refueling station as the decision variable and aiming at the maximum total hydrogen refueling demand faced by the hydrogen refueling stations in the area, use the particle swarm optimization algorithm to solve and obtain the optimal set of alternative addresses of the hydrogen refueling station considering the hydrogen refueling demand in different time periods and regions;

[0010] Step 3: Through the hydrogen energy vehicle big data platform, collect the operation information of hydrogen production plants in the area, including the number of tube trailers, hydrogen production capacity, the number of hydrogen refueling stations served, the alternative addresses of hydrogen refueling stations within a specific radiation range centered on each hydrogen production plant, scheduling strategies, etc., and obtain the hydrogen supply capacity-related information, including the available hydrogen volume, the number of mobilizable tube trailers, and the maximum service distance range under the full-load service of each hydrogen production plant;

[0011] Step 4: Establish a cost function considering the common costs of the hydrogen refueling of hydrogen energy vehicles, the hydrogen transportation by tube trailers, the hydrogen storage in hydrogen refueling stations, and the hydrogen production in hydrogen production plants. Taking the alternative address of the hydrogen refueling station as the decision variable and aiming at the minimum of this cost function, use the particle swarm optimization algorithm to solve and obtain the optimal set of alternative addresses of the hydrogen refueling station considering the cost;

[0012] Step 5: Integrate the optimal set of alternative addresses of the hydrogen refueling station considering the hydrogen refueling demand and the optimal set of alternative addresses of the hydrogen refueling station considering the cost obtained by solving, and determine the final set of hydrogen refueling station site selections.

[0013] Further, the objective function of the maximum total hydrogen refueling demand faced by the hydrogen refueling stations in the area in Step 2 specifically adopts the following form:

[0014] max∑W OD ·y·p

[0015] In the formula, W OD is the traffic flow from the hydrogen refueling departure node to the alternative hydrogen refueling station address in the vehicle operation area, y is the real-time hydrogen refueling demand of the vehicle, p is the probability set of the vehicle selecting each alternative address, and p can be expressed in the form of the following matrix:

[0016]

[0017]

[0018] Among them, M and N respectively represent the number of hydrogen energy vehicles and the number of planned alternative addresses of hydrogen refueling stations, i represents the i-th alternative address, j represents the j-th vehicle, and pij denotes the probability that the \(j\)-th vehicle goes to the hydrogen refueling station at the \(i\)-th alternative address;

[0019] Extract three types of data from the hydrogen refueling demand-related information, including: (1) vehicle characteristics, (2) characteristics of alternative hydrogen refueling station addresses, and (3) joint characteristics of vehicles and alternative hydrogen refueling station addresses, and encode them as the feature vector \(X\). ij Build a vehicle hydrogen refueling behavior prediction model based on logistic regression:

[0020]

[0021] where \(W\) T denotes the weights of each feature, \(X\) ij is the feature vector value, \(b\) is the model bias value, and softmax is the normalization function; solve the model parameters \(W\) T and \(b\) using the historical data of the three types of features collected, and obtain the vehicle hydrogen refueling decision probability under different conditions, that is, fit the hydrogen refueling behavior of hydrogen energy vehicles;

[0022] For the above objective function with the largest total hydrogen refueling demand, use the particle swarm optimization algorithm to solve and obtain the optimal alternative hydrogen refueling station addresses considering the hydrogen refueling demand.

[0023] Furthermore, the specific implementation process of the above particle swarm optimization algorithm is as follows: Let the longitude and latitude coordinates \(x\) i \(=(L\) 1i , \(L\) 2i ) \(i = 1, 2, 3,\cdots, 50\) be the decision variables, where the subscripts 1 and 2 are used to distinguish longitude and latitude, and the particle swarm size is 50; First, randomly initialize the decision variables and calculate the objective function value. The individual maximum value is the optimal solution \(pbest\) of each particle, and find the global optimal value \(gbest\) from the optimal solutions of the 50 individuals. Compare with the global optimal value and update. The update formula is as follows:

[0024] \(v\) i \(=\omega\times v\) i \(+c1\times rand\times(pbest\) i \(-x\) i ) \(+c2\times rand\times(gbest\) i \(-x\) i )

[0025] \(x\) i \(=x\) i \(+v\) i

[0026] where rand represents a random number in the interval [0, 1], and w, c1, and c2 represent the inertia factor, the individual learning factor, and the social learning factor, respectively, which are set to 0.8, 2, and 2. When the update condition is met, the iteration is terminated, and the optimized 50 decision variables are output as the optimal set of alternative hydrogen refueling station addresses.

[0027] Further, the above vehicle characteristics specifically include: vehicle type T, vehicle operation area distribution U, hydrogen surplus S at different location nodes, vehicle hydrogen refueling time distribution t; the characteristics of alternative hydrogen refueling station addresses specifically include: coordinates P of each alternative address, traffic density D near each alternative address; the joint characteristics of vehicle - alternative hydrogen refueling station addresses specifically include: whether the vehicle and each alternative address are on the same line B, the distance d from the vehicle to each alternative address, and the traffic condition s of the optional routes for the vehicle to each alternative address.

[0028] Further, the objective function of minimizing the total cost of the following links: hydrogen refueling of hydrogen - powered vehicles, hydrogen transportation by tube trailers, hydrogen storage in hydrogen refueling stations, and hydrogen production in hydrogen production plants in step four specifically adopts the following form:

[0029]

[0030]

[0031] where a x is the cost coefficient of transportation mode x, representing the transportation cost per unit distance when using the x - th transportation mode; d ij is the driving distance from the i - th alternative address to the j - th hydrogen production plant; b z is the cost coefficient of hydrogen storage mode z in the hydrogen refueling station, that is, the total cost of the hydrogen storage system, usage system, and energy consumption per unit hydrogen when using the z - th storage mode; w ij is the hydrogen energy demand of the i - th hydrogen refueling station for the j - th hydrogen production plant; u j is the hydrogen production cost of the j - th hydrogen production plant; a i，i+1 is the driving distance between any two alternative hydrogen refueling station addresses; D1 and D2 are the maximum distance constraints between two alternative hydrogen refueling station addresses and the maximum service distance from the hydrogen production plant to the hydrogen refueling station, respectively; L j is the production capacity limit of the j - th hydrogen production plant; K is the area where hydrogen refueling stations cannot be built; N is the number of mobilizable tube trailers;

[0032] For the above objective function of minimizing the cost, following the solution process of step two, changing the objective function and conditional constraints, the particle swarm algorithm is used to solve for the optimal alternative hydrogen refueling station addresses considering the cost.

[0033] Further, Step Five specifically includes: For the obtained optimal hydrogen refueling station alternative address set Y1 considering hydrogen refueling demand and the optimal hydrogen refueling station alternative address set Y2 considering cost, traverse the distances between any alternative address coordinates (x1, y1) in Y1 and any alternative address coordinates (x2, y2) in Y2, and taking less than a predetermined value as an index, determine the comprehensive optimal hydrogen refueling station alternative address set, and adjust according to the actual situation to determine the final hydrogen refueling station site selection set.

[0034] The method provided by the present invention starts from two aspects of hydrogen refueling demand and cost to establish an objective optimization problem, and combines hydrogen energy vehicle big data with a suitable algorithm to provide an effective way to solve the above optimization problem. When selecting the location of the hydrogen refueling station, the present invention comprehensively considers various factors such as the hydrogen refueling behavior of hydrogen energy vehicles, the hydrogen refueling demand of hydrogen refueling station alternative addresses, and the production, storage, transportation costs of hydrogen energy and vehicle hydrogen refueling costs, thereby overcoming the technical problem that the location selection of hydrogen refueling stations and hydrogen production plants does not cooperate with each other, affecting the hydrogen refueling efficiency, and having many beneficial effects not possessed by the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a general flow schematic diagram of the method provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] Next, the technical solutions of the present invention will be clearly and completely described in conjunction with the drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0037] A method for hydrogen refueling station site selection based on hydrogen energy vehicle and hydrogen production plant data provided by the present invention specifically includes the following steps:

[0038] Step One: Collect the operation data of hydrogen energy vehicles in a certain area through the hydrogen energy vehicle big data platform, and obtain the hydrogen refueling demand-related information of each vehicle, including daily average mileage, daily average hydrogen consumption, vehicle operation area distribution, hydrogen remaining amount at different position nodes, vehicle hydrogen refueling time distribution, etc.;

[0039] Step Two: For each hydrogen refueling station alternative address set in the setting plan, establish a hydrogen refueling demand function faced by the corresponding hydrogen refueling station, use the hydrogen refueling station alternative address as a decision variable, and with the goal of maximizing the total hydrogen refueling demand faced by the hydrogen refueling stations in the area, use the particle swarm algorithm to solve and obtain the optimal hydrogen refueling station alternative address set considering hydrogen refueling demand in different time periods and regions;

[0040] Step 3: Collect the operation information of hydrogen production plants in the region through the hydrogen energy vehicle big data platform, including the number of tube trailers, hydrogen production capacity, the number of hydrogen refueling stations served, the alternative addresses of hydrogen refueling stations within a specific radiation range centered on each hydrogen production plant, dispatching strategies, etc., and obtain the hydrogen supply capacity-related information such as the allocable hydrogen volume, the number of mobilizable tube trailers, and the maximum service distance range under the full-load service condition of each hydrogen production plant;

[0041] Step 4: Establish a cost function considering the common costs of several links including hydrogen refueling of hydrogen energy vehicles, hydrogen transportation by tube trailers, hydrogen storage in hydrogen refueling stations, and hydrogen production in hydrogen production plants. Taking the alternative addresses of hydrogen refueling stations as decision variables and aiming at minimizing this cost function, use the particle swarm algorithm to solve and obtain the set of optimal alternative addresses of hydrogen refueling stations considering costs;

[0042] Step 5: Integrate the set of optimal alternative addresses of hydrogen refueling stations considering hydrogen refueling demand and the set of optimal alternative addresses of hydrogen refueling stations considering costs obtained by solving, and determine the final set of hydrogen refueling station locations.

[0043] In a preferred embodiment of the present invention, the objective function of maximizing the total hydrogen refueling demand faced by hydrogen refueling stations in the region aims to ensure that the construction of hydrogen refueling stations conforms to the hydrogen refueling behaviors of each vehicle as much as possible, avoid the low probability and opportunity of some hydrogen refueling stations being selected after completion, resulting in idleness, and at the same time prevent serious congestion and queuing at some other hydrogen refueling stations. Its specific form is as follows:

[0044] max∑W OD ·y·p

[0045] In the formula, W OD is the traffic flow from the hydrogen refueling starting node to the alternative hydrogen refueling station address in the vehicle operation area, y is the real-time hydrogen refueling demand of the vehicle, p is the probability set of the vehicle choosing each alternative address, and p can be expressed in the form of the following matrix:

[0046]

[0047]

[0048] Among them, M and N respectively represent the number of hydrogen energy vehicles and the number of planned alternative hydrogen refueling station addresses, i represents the i-th alternative address, j represents the j-th vehicle, and p ij represents the probability that the j-th vehicle goes to the hydrogen refueling station at the i-th alternative address for hydrogen refueling;

[0049] In a specific preferred embodiment, three types of data are extracted from the hydrogenation demand-related information, including: (1) vehicle characteristics: vehicle type T (discrete variable, category-encoded), regional vehicle hydrogen demand U (continuous variable, numerically encoded), vehicle activity range distribution S (including the longitude and latitude of the activity range center position, the maximum activity radius, and the number of nodes), different position information R of different on-vehicle hydrogen storage amounts (extracting the activity distribution with hydrogen remaining amounts of 100%, 95%,..., 5%, 0%, encoded as a continuous variable), vehicle hydrogenation time distribution t (extracting the hydrogenation clock count from historical hydrogenation times, discrete variable, numerically encoded); (2) characteristics of alternative hydrogenation station addresses: the location P of the to-be-built hydrogenation station (longitude and latitude, numerically encoded), the traffic density D near the to-be-built hydrogenation station (continuous variable, numerically encoded); (3) vehicle-alternative hydrogenation station address joint characteristics: whether the vehicle and the hydrogenation station are on the same route B (binary encoding, 0 indicates not on the same route, 1 indicates on the same route), the distance d of the vehicle from the to-be-built hydrogenation station (continuous variable, the shorter the distance, the smaller the value), and the regional traffic condition s (the average value of the regional traffic index). After encoding them as a feature vector, a vehicle hydrogenation behavior prediction model based on logistic regression is established:

[0050]

[0051] X ij =(T, U, S, R, t, P, D, B, d, s)

[0052] where W T represents the feature weight, X ij is the eigenvalue, b is the model bias value, and softmax is the normalization function. The model parameters W T and b are solved using the three types of historical data collected, and the vehicle hydrogenation decision probabilities under different conditions are obtained, that is, the hydrogenation behavior of hydrogen energy vehicles is fitted.

[0053] Those skilled in the art should be aware that the specific selection and encoding of the three types of feature data are not limited to the above specific methods. The above methods are only optional and do not limit the protection scope of the claims of the present invention. Under the teaching of the present invention, selecting diverse feature data and encoding methods can solve the technical problems to be solved by the present invention.

[0054] For the above objective function with the largest total hydrogenation demand, the particle swarm algorithm is used to solve for the optimal alternative hydrogenation station address considering the hydrogenation demand. The specific implementation process is as follows: Let the longitude and latitude coordinates x i =(L 1i , L 2i) i = 1, 2, 3..., 50 are decision variables, and the particle swarm size is 50. First, randomly initialize the decision variables and calculate the objective function value. The individual maximum value is the optimal solution pbest for each particle, and the global optimal value gbest is found from the optimal solutions of the 50 individuals. Compare with the global optimum and perform updates. The update formulas are as follows:

[0055] v i = ω × v i + c1 × rand × (pbest i - x i ) + c2 × rand × (gbest i - x i )

[0056] x i = x i + v i

[0057] In the formula, rand represents a random number in the interval [0, 1], and w, c1, and c2 respectively represent the inertia factor, individual learning factor, and social learning factor, which are usually set to 0.8, 2, and 2. When the update condition is met, the iteration is terminated, and the optimized 50 decision variables are output as the set of optimal hydrogen refueling station alternative addresses.

[0058] Those skilled in the art should also know, according to the teachings of the present invention, how to adopt other various different objective function forms with the goal of maximizing the total hydrogen refueling demand faced by the hydrogen refueling station, and which specific algorithm to select for solving. The specific objective function and solution process adopted above do not constitute a limitation on the protection scope of the claims of the present invention.

[0059] In a preferred embodiment of the present invention, the above vehicle characteristics specifically include: vehicle type, vehicle operation area distribution, hydrogen remaining amount at different location nodes, vehicle hydrogen refueling time distribution; the hydrogen refueling station alternative address characteristics specifically include: coordinates of each alternative address, traffic density near each alternative address; the vehicle-hydrogen refueling station alternative address joint characteristics specifically include: whether the vehicle and each alternative address are on the same line, the distance between the vehicle and each alternative address, and the traffic conditions of the optional routes for the vehicle to each alternative address. Of course, those skilled in the art can also flexibly select the data specifically corresponding to the three types of data of vehicle characteristics, hydrogen refueling station alternative address characteristics, and vehicle-hydrogen refueling station alternative address joint characteristics according to actual needs. The data of each characteristic type listed above does not constitute a limitation on the protection scope of the claims of the present invention.

[0060] In a preferred embodiment of the present invention, the objective function of minimizing the total cost of the following several links: hydrogen refueling of hydrogen energy vehicles, hydrogen transportation by tube trailers, hydrogen storage in hydrogen refueling stations, and hydrogen production in hydrogen production plants, specifically adopts the following form:

[0061]

[0062]

[0063] Among them, a x is the cost coefficient of transportation mode x, representing the transportation cost per unit distance when using the x-th transportation mode; d ij is the driving distance from the i-th alternative address to the j-th hydrogen production plant; b z is the cost coefficient of the hydrogen storage mode z of the hydrogen refueling station, that is, the total cost of the storage system, usage system, and energy consumption per unit hydrogen when using the z-th storage mode; w ij is the hydrogen energy demand of the i-th hydrogen refueling station for the j-th hydrogen production plant; u j is the hydrogen production cost of the j-th hydrogen production plant; d i，i+1 is the driving distance between any two alternative addresses of the hydrogen refueling stations; D1 and D2 are the maximum distance constraints between the two alternative addresses of the hydrogen refueling stations and the maximum service distance from the hydrogen production plant to the hydrogen refueling station, respectively; L j is the production capacity limit of the j-th hydrogen production plant; K is the area where hydrogen refueling stations cannot be built; N is the number of mobilizable tube trailers;

[0064] For the above objective function with the minimum cost, following the optimization and solution process in Step 2, the particle swarm optimization algorithm is used to solve and obtain the optimal alternative addresses of the hydrogen refueling stations considering the cost.

[0065] Those skilled in the art should also know according to the teachings of the present invention how to adopt other various different objective function forms with the minimum cost, and which specific algorithm to select for solution. The specific objective function and solution process adopted above do not constitute a limitation on the protection scope of the claims of the present invention.

[0066] In a preferred embodiment of the present invention, Step 5 specifically includes: for the obtained set Y1 of optimal alternative addresses of the hydrogen refueling stations considering the hydrogen refueling demand and the set Y2 of optimal alternative addresses of the hydrogen refueling stations considering the cost, traverse the distances between any alternative address coordinates (x1, y1) in Y1 and any alternative address coordinates (x2, y2) in Y2, and taking being less than a predetermined value as an index, determine the set of comprehensively optimal alternative addresses of the hydrogen refueling stations, and adjust and determine the final set of hydrogen refueling station locations in combination with the actual situation.

[0067] It should be understood that the magnitudes of the sequence numbers of the steps in the embodiments of the present invention do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0068] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for hydrogen refueling station site selection based on data of hydrogen energy vehicles and hydrogen production plants, characterized in that: Specifically, it includes the following steps: Step 1: Collect the operation data of hydrogen energy vehicles in a certain area through the hydrogen energy vehicle big data platform, and obtain the hydrogen refueling demand-related information of each vehicle, including the daily average mileage, daily average hydrogen consumption, vehicle operation area distribution, hydrogen remaining amount at different position nodes, and hydrogen refueling time distribution of the vehicle; Step 2: For each alternative address of the hydrogen refueling station set in the plan, establish the hydrogen refueling demand function faced by the corresponding hydrogen refueling station. Taking the alternative address of the hydrogen refueling station as the decision variable and aiming at the maximum total hydrogen refueling demand faced by the hydrogen refueling stations in the area, use the particle swarm optimization algorithm to solve and obtain the set of optimal alternative addresses of the hydrogen refueling station considering the hydrogen refueling demand at different times and in different areas. The objective function of the maximum total hydrogen refueling demand faced by the hydrogen refueling stations in the area specifically adopts the following form: Where W OD is the traffic flow of the vehicle operation area from the hydrogen refueling departure node to the address of the alternative hydrogen refueling station, y is the real-time hydrogen refueling demand of the vehicle, p is the probability set of the vehicle choosing each alternative address, and p can be expressed in the form of the following matrix: Among them, M and N respectively represent the number of hydrogen energy vehicles and the number of candidate sites for planned hydrogen refueling stations. i represents the i-th candidate site, j represents the j-th vehicle, and p ij represents the probability that the j-th vehicle goes to the hydrogen refueling station at the i-th candidate site; Extract three types of data from the hydrogen refueling demand-related information, including: (1) vehicle characteristics, (2) characteristics of alternative hydrogen refueling station addresses, and (3) joint characteristics of vehicles and alternative hydrogen refueling station addresses. After encoding them as feature vectors, establish a vehicle hydrogen refueling behavior prediction model p based on logistic regression ij = F(X ij ), where F(·) represents the specific functional equation of the prediction model; use the three types of historical data collected to fit the hydrogen refueling behavior of hydrogen energy vehicles reflected by the probability For the above objective function of the maximum total hydrogen refueling demand, use the particle swarm optimization algorithm to solve and obtain the optimal alternative address of the hydrogen refueling station considering the hydrogen refueling demand; Step 3: Collect the operation information of the hydrogen production plants in the area through the hydrogen energy vehicle big data platform, including the number of tube trailers, hydrogen production capacity, the number of hydrogen refueling stations served, the alternative addresses of the hydrogen refueling stations within a specific radiation range centered on each hydrogen production plant, and the dispatching strategy, and obtain the information related to the hydrogen supply capacity, such as the allocable hydrogen amount under the full-load service of each hydrogen production plant, the number of mobilizable tube trailers, and the farthest service distance range; Step 4: Establish a cost function considering the common costs of several links, including hydrogen refueling of hydrogen energy vehicles, hydrogen transportation by tube trailers, hydrogen storage in hydrogen refueling stations, and hydrogen production in hydrogen production plants. Taking the alternative address of the hydrogen refueling station as the decision variable and aiming at the minimum of this cost function, use the particle swarm optimization algorithm to solve and obtain the set of optimal alternative addresses of the hydrogen refueling station considering the cost; Step 5: Integrate the set of optimal alternative addresses of the hydrogen refueling station considering the hydrogen refueling demand and the set of optimal alternative addresses of the hydrogen refueling station considering the cost obtained by solving, and determine the final set of hydrogen refueling station site selections.

2. The method according to claim 1, wherein: The above vehicle characteristics specifically include: vehicle type, vehicle operation area distribution, hydrogen remaining amount at different position nodes, and hydrogen refueling time distribution of the vehicle; the characteristics of the alternative addresses of the hydrogen refueling station specifically include: the coordinates of each alternative address, the traffic density near each alternative address; the joint characteristics of the vehicle - alternative address of the hydrogen refueling station specifically include: whether the vehicle and each alternative address are on the same route, the distance between the vehicle and each alternative address, and the traffic conditions of the optional routes for the vehicle to each alternative address.

3. The method according to claim 1, wherein: The objective function of the minimum common cost of several links, including hydrogen refueling of hydrogen energy vehicles, hydrogen transportation by tube trailers, hydrogen storage in hydrogen refueling stations, and hydrogen production in hydrogen production plants, in Step 4 specifically adopts the following form: Among them, a x is the cost coefficient of transportation mode x, representing the transportation cost per unit distance when using the x-th transportation mode; d ij is the driving distance from the i-th alternative address to the j-th hydrogen production plant; b z is the cost coefficient of the hydrogen storage mode z of the hydrogen refueling station, that is, the total cost of the storage system, use system, and energy consumption per unit of hydrogen when using the z-th storage mode; w ij is the demand for hydrogen energy from the i-th hydrogen refueling station to the j-th hydrogen production plant; u j is the hydrogen production cost of the j-th hydrogen production plant; d i,i+1 is the driving distance between any two alternative addresses of hydrogen refueling stations; D1 and D2 are the maximum distance constraints between two alternative addresses of hydrogen refueling stations and the farthest service distance from the hydrogen production plant to the hydrogen refueling station; L j is the production capacity limit of the j-th hydrogen production plant; K is the area where hydrogen refueling stations cannot be built; N is the number of mobilizable tube trailers; For the above objective function of the minimum cost, use the particle swarm optimization algorithm to solve and obtain the optimal alternative address of the hydrogen refueling station considering the cost.

4. The method according to claim 1, characterized in that: Step 5 specifically includes: For the obtained optimal hydrogen refueling station alternative address set Y1 considering hydrogen addition demand and the optimal hydrogen refueling station alternative address set Y2 considering cost, traverse the distances between any alternative address coordinates (x1, y1) in Y1 and any alternative address coordinates (x2, y2) in Y2, and taking less than a predetermined value as an indicator, determine the comprehensive optimal hydrogen refueling station alternative address set, and adjust according to the actual situation to determine the final hydrogen refueling station site selection set.

Images