Optimal scheduling strategy for electric-hydrogen hybrid charging stations based on charging decision prediction

Abstract

The purpose of this invention is to provide an optimized scheduling strategy for electric-hydrogen hybrid charging stations based on charging decision prediction. First, a corresponding charging strategy is developed for new energy vehicles. Two charging plans are designed for electric vehicle users to choose from. Electricity price incentives are used to guide and adjust consumer behavior. A psychology-based Weber-Fechner (W-F) incentive-response decision model is established to rationally describe drivers' charging plan selection behavior. Subsequently, the NSGA-II algorithm is used to solve this multi-objective problem, taking into account the interests of both users and suppliers, with the operating revenue of the electric-hydrogen hybrid charging station and the user's waiting time as the objective function.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric-hydrogen composite charging stations, and in particular relates to an optimization scheduling strategy for electric-hydrogen composite charging stations based on charging decision prediction. Background Art

[0002] As new energy vehicles, electric vehicles and hydrogen fuel cell vehicles offer significant advantages and potential for reducing carbon dioxide emissions and alleviating the energy crisis. As the energy supply link for electric and hydrogen fuel cell vehicles, new energy vehicle charging stations are gradually evolving into integrated, comprehensive charging stations. The addition of hydrogen production, storage, and charging equipment to new energy vehicle charging stations can meet the hydrogen charging needs of hydrogen fuel cell vehicles, enabling low-carbon travel. The charging station's electric-hydrogen energy system can provide an ideal location for absorbing intermittent renewable energy sources such as wind and solar power. Furthermore, hydrogen fuel cell power generation, as a flexible and dispatchable power generation method, facilitates the safe and stable operation of electric-hydrogen hybrid charging station systems. Therefore, hydrogen / electric vehicle charging stations that simultaneously address the charging needs of electric and hydrogen fuel cell vehicles, renewable energy generation, and energy storage have broad application prospects.

[0003] Scholars have proposed novel energy management systems for optimizing the power supply operation of new energy vehicle charging stations. Others have proposed optimization management algorithms for charging stations that integrate photovoltaic and energy storage units, leveraging these units to optimize the station's operating costs. Furthermore, some have proposed the concept of electric vehicle aggregators, whereby large numbers of electric vehicles would like to operate in the electricity market under the direction of an aggregator. While these studies examine the design and operation of electric-hydrogen hybrid charging stations, they overlook the impact of charging decisions made by electric and hydrogen vehicles on the overall operational efficiency of electric-hydrogen hybrid charging stations. Summary of the Invention

[0004] To address the shortcomings and deficiencies of existing technologies and enhance scheduling flexibility, this paper proposes an optimized scheduling strategy for electric-hydrogen hybrid charging stations based on charging decision prediction. This strategy comprehensively considers the charging behavior of electric and hydrogen fuel cell vehicles, enabling rapid and effective predictions and rational scheduling of energy devices and vehicles within the station, thereby improving the overall operational efficiency of the hybrid station.

[0005] First, a charging strategy was developed for new energy vehicles. Two charging plans were designed for electric vehicle users. Electricity price incentives were used to guide user behavior adjustment. A Weber-Fechner (WF) incentive-response decision model, based on psychology, was established to rationally describe drivers' charging plan selection behavior. Subsequently, the NSGA-II algorithm was used to solve this multi-objective problem, taking into account the interests of both users and suppliers, with the operating revenue of the electric-hydrogen hybrid charging station and the user's waiting time as the objective functions.

[0006] The technical solution specifically adopted by the present invention to solve the technical problem is:

[0007] An optimization scheduling strategy for electric-hydrogen hybrid charging stations based on charging decision prediction is proposed. The strategy is characterized by: adjusting consumption behavior based on the charging strategy of new energy vehicles through electricity price incentives, and establishing a Weber-Fechner incentive response decision model based on psychology to describe drivers' charging plan selection behavior; comprehensively considering the interests of both the user side and the energy supply side, taking the operating income of the electric-hydrogen hybrid charging station and the user's queuing time as the objective function, and using the NSGA-Ⅱ algorithm to solve the problem.

[0008] Furthermore, the specific process of solving the problem using the NSGA-Ⅱ algorithm is as follows:

[0009] Step 1: Based on the local weather forecast data, we obtain the wind speed and sunlight intensity for the next day. Then, we combine the wind power and photovoltaic models to generate the output power at different times of the day.

[0010] Step 2: Calculate the total charging power of electric vehicles and the total hydrogen demand of hydrogen fuel cell vehicles. Use the Monte Carlo method to simulate the daily travel and charging process of new energy vehicle users, and obtain data including the arrival time, charging demand, and starting SOC of electric and hydrogen fuel cell vehicles. Based on this data, a WF-based energy storage decision model is established to estimate the type of charging solution selected by charging users.

[0011] Step 3: Randomly generate the electrolyzer hydrogen production power for each time period according to the constraints; obtain the hydrogen production power in one day based on the hydrogen production power in each time period;

[0012] Step 4: Based on the power balance relationship, calculate the power supply of the AC distribution network at time t and the discharge power of the fuel cell group at time t;

[0013] Step 5: Calculate the comprehensive operating cost of one of the objective functions; obtain the electricity sales cost, hydrogen sales cost, and electricity purchase cost through steps 1 to 4 and sum them to obtain the comprehensive operating income;

[0014] Step 6: Calculate the total queuing time of electric vehicle users, which is the second objective function; obtain the total queuing time of electric vehicle users in one day according to the dispatch status of electric vehicles in each time period;

[0015] Step 7: Based on steps 1 to 6, generate an initial population P of size G t , the offspring population Q is generated under the process of selection, crossover and mutation t , combined with the generated population R of size 2G t ;

[0016] Step 8: For population Rt Perform fast non-dominated sorting to form a non-dominated set Z i And calculate the crowding degree of individuals in each non-dominated layer, select suitable individuals according to the sorting results and crowding degree, until the number of individuals is G, the new generation parent population P t +1 formation;

[0017] Step 9: New generation parent population P t +1 generates a subpopulation Q under the process of selection, crossover, and mutation t +1, combined with the generated population R of size 2G t +1; Repeat steps 7 to 9 until the genetic generation is reached.

[0018] Furthermore, the electric-hydrogen hybrid charging station consists of a photovoltaic solar power generation system, a wind turbine system, an electrolyzer, a hydrogen storage tank and a fuel cell, and is connected to the public power grid so that electricity can be purchased when energy demand is not met; the electricity supplied to the power station from photovoltaic / wind power generation and the public power grid all passes through a common coupling point; the hydrogen in the station is produced by the electrolyzer and stored in the hydrogen tank; the stored hydrogen can be injected into the hydrogen fuel cell or converted into electrical energy through the fuel cell; electric vehicles are allowed to be charged directly using the electricity from the coupling point and the electricity generated by the fuel cell.

[0019] Furthermore, two charging solutions are designed for electric vehicle users to choose from:

[0020] (1) Normal charging mode: Charge the electric vehicle at the maximum constant power until the battery reaches the expected state of charge (SOC).

[0021] (2) Agreement charging mode: The operator of the electric-hydrogen hybrid charging station signs an agreement with the electric vehicle driver, and the integrated station charges to the expected SOC value within a certain period of time; the electric vehicle driver authorizes the operator of the electric-hydrogen hybrid charging station, allowing the operator of the electric-hydrogen hybrid charging station to freely dispatch authorized electric vehicles for charging and discharging during this period.

[0022] Furthermore, in both charging modes, the electric-hydrogen hybrid charging station can only dispatch authorized electric vehicles that select the protocol charging mode for charging, discharging, and inactivity. During the parking period of electric vehicles, the dispatch system sets the charging amount and dispatch time of each electric vehicle according to the corresponding dispatch schedule. It is assumed that the charging and discharging power is rated and remains unchanged during the charging period of each electric vehicle, and the power loss is entirely from the AC / DC module and DC / DC module, and no power loss is generated in the other processes. The electric vehicles participating in the dispatch should be subject to the following restrictions:

[0023]

[0024] in Indicates the charging power of the electric vehicle scheduled by the protocol numbered m at time t; and Represent the charging and discharging power of electric vehicles respectively; and They represent the arrival time and departure time of the electric car numbered m respectively.

[0025] Furthermore, in the Weber-Fechner incentive-response decision model based on psychology, the probability of an electric vehicle driver accepting dispatch is expressed as:

[0026]

[0027] ρ i =0.7ρ dsc,i +0.3ρ cha,i

[0028] where k represents the Weber fraction, s0 represents the stimulus constant; ρ l Indicates the minimum analog value; when the analog value is less than ρ l When ρ, all electric vehicle drivers will not accept the dispatch; h Indicates the maximum sensory value; when the analog value is greater than ρ h When the probability p n Corresponding to p h ; The simulation includes two influencing factors, among which ρ dsc,i represents the discount simulation amount, ρ cha,i Indicates the analog value of charging demand.

[0029] Furthermore, the integrated station operator earns revenue by providing electricity and hydrogen to users. When electricity cannot meet operational needs, the operator purchases electricity from the AC distribution network at peak and valley prices. The optimization goal is to maximize the daily revenue of the integrated station. The objective function is expressed as:

[0030]

[0031] Among them, P t ev_u and P t ev_p denote the charging demands of electric vehicle drivers who do not participate in the agreement and those who participate in the agreement at time t; λ denotes the incentive discount, EP t ev represents the electricity price at time t, HP represents the price of hydrogen, EP t pg Indicates the price of purchasing electricity from the grid.

[0032] Furthermore, considering the user's charging experience, the second goal is to minimize the total queuing time of the user. Under the condition of satisfying the constraints, the scheduling time of each vehicle is optimized to reduce the occupancy of charging piles and shorten the average queuing time of users. The queuing time of electric vehicle n at time t is The remaining charging time of the vehicle being charged at time t and the number of vehicles in the queue N q Related to; expressed as:

[0033]

[0034]

[0035]

[0036]

[0037] Among them, N ev represents the number of electric vehicles charged in the composite station; It is expected that electric vehicles u Replenishment power threshold; x n Represents a state variable. When its value is 1, it indicates that the vehicle volume accepts the protocol scheduling. When it is 0, it indicates that the vehicle volume does not accept the protocol scheduling. ev Indicates the fast charging power of the charging pile; represents the ascending set of real-time remaining charging time of electric vehicles being charged at time t, N k represents the number of vehicles being charged in the composite station; h represents N q / M remainder, M represents the number of charging piles in the composite station; T t left {h+1} represents the real-time remaining charging time of the {h+1}th vehicle at time t after ascending order; N b Indicates the number of electric vehicles that have priority over other electric vehicles in the queue. Represents the time required for the queuing vehicle to charge to the required SOC threshold.

[0038] Furthermore, the constraints specifically include: the sum of the power supplied to the electric vehicle and the electrolyzer is limited by the output power of the photovoltaic power generation; the power supplied to the electric vehicle should be constrained by the demand of the electric vehicle; the amount of hydrogen in the hydrogen storage tank at the beginning of the day corresponds to the same time of the next day; the scheduling time Δt of the electric vehicles participating in the protocol n Subject to the following constraints:

[0039] Δt min ≤Δt n ≤Δt max

[0040] Δtmin and Δt max Represent the minimum and maximum scheduling time respectively.

[0041] Compared with the prior art, the present invention and its preferred embodiments have at least the following beneficial effects:

[0042] (1) Two charging modes are designed for electric vehicle users to choose from, and the demand response potential of electric vehicles is fully utilized through incentive agreements.

[0043] (2) Based on the WF law, an incentive-response charging decision prediction method is proposed to reasonably describe the relationship between external factors and charging decision response. An EV charging decision model that integrates charging economic cost and time cost is established to effectively predict the charging choice behavior of electric vehicles and improve the accuracy of the prediction.

[0044] (3) By rationally dispatching the charging and discharging of energy storage devices and electric vehicles in the station, the impact of EV fast charging load and the randomness and volatility of renewable energy output can be reduced, thereby improving the comprehensive benefits of the electric-hydrogen hybrid charging station. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0046] Figure 1 This is a flowchart of the solution using the NSGA-II algorithm in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] To make the features and advantages of this patent more clearly understood, the following embodiments are specifically described in detail as follows:

[0048] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in this specification have the same meaning as commonly understood by those skilled in the art to which this application belongs.

[0049] Renewable energy generation model

[0050] The electric-hydrogen hybrid charging station consists of a photovoltaic solar power generation system, a wind turbine system, an electrolyzer, a hydrogen storage tank, and a fuel cell. It is connected to the utility grid to purchase electricity when energy demand is insufficient. Electricity supplied to the station from both the photovoltaic / wind power generation system and the utility grid passes through a common coupling point. Hydrogen is produced within the station by the electrolyzer and stored in hydrogen tanks. This stored hydrogen can be injected into hydrogen fuel cells or converted into electricity by the fuel cell. Electric vehicles can be charged directly using electricity from the coupling point and electricity generated by the fuel cell.

[0051] The probability density function is used to model the uncertain solar radiation and predict the photovoltaic power generation. The normal distribution function can be used to generate solar radiation scenarios, where the available power generated by the photovoltaic power station in each period can be expressed as:

[0052] P t RE =P t PV +P t WT (1)

[0053]

[0054] Where: P t RE is the total output power of photovoltaic and wind power. t PV and P t WT The output power of photovoltaic and wind turbine are shown respectively. pv Represents the converter efficiency between photovoltaic and electrolyzer. P pvr is the rated capacity of the photovoltaic module, is the solar radiation at time t, G s is the solar radiation under reference conditions.

[0055] The power generated by a wind turbine depends on the wind speed. The relationship between them can be expressed as:

[0056]

[0057] Where: P wtr Indicates the rated power of the fan, v t represents the wind speed at time t. In addition, v r 、v cin and v cout They respectively represent the rated wind speed, cut-in wind speed and cut-out wind speed of the selected wind turbine.

[0058] Fuel cell model

[0059] Fuel cells generate electricity by consuming hydrogen stored in a tank. Because fuel cells are energy conversion devices, energy loss during the conversion process is inevitable. Therefore, their primary function is to work with hydrogen storage tanks to provide power when power is insufficient.

[0060] The electrical energy generated by the fuel cell is expressed as:

[0061]

[0062] P FC,min ≤P t FC_EV ≤PFC,max (5)

[0063] Where: V t FC Indicates the amount of hydrogen consumed by the fuel cell. t FC_EV Indicates the power provided by the fuel cell to the electric vehicle. FC Indicates the efficiency of the fuel cell. FC,min and P FC,max represent the minimum and maximum power of the fuel cell respectively.

[0064] Hydrogen production and storage model

[0065] Electric hydrogen production technology is a hydrogen production method based on water electrolysis, which decomposes water into hydrogen and oxygen through water electrolysis. It has been applied in many fields, such as hydrogen fuel cell vehicles, fuel cell power generation, etc. Among them, electric hydrogen production technology can be used to produce hydrogen, making hydrogen fuel cell vehicles a clean alternative energy source and reducing dependence on traditional petroleum energy. In addition, electric hydrogen production technology can be used to produce hydrogen to drive fuel cells, thereby generating electricity. The advantages of this power generation method are that it can provide clean electricity, does not produce harmful gases, and improves the flexibility of the energy system. Assuming that the power of the electrolyzer per unit time is constant, the amount of hydrogen produced by the electrolyzer can be expressed as:

[0066]

[0067] P ELE,min ≤P t RE-ELE ≤P ELE,max (7)

[0068] Where: V t ST Indicates the amount of hydrogen in the hydrogen storage tank at time t. η ELE Represents the efficiency of the electrolyzer. t RE_ELE represents the power provided by the photovoltaic system to the electrolyzer. Δt represents the time constant.

[0069] The hydrogen produced by the electrolysis of water in the electrolyzer is first stored in a high-pressure hydrogen storage tank to provide hydrogen for hydrogen fuel cell vehicles and fuel cells. During the hydrogen supply process for hydrogen fuel cell vehicles, the hydrogen is pressurized and cooled by a hydrogenation machine to form liquid hydrogen, which is then injected into the vehicle. The energy and time consumed in the cooling and hydrogenation processes are negligible. The dynamic equation for the volume of hydrogen in the hydrogen tank can be expressed as:

[0070]

[0071] V ST,min ≤V t ST ≤VST,max (9)

[0072] Where: V t HV It represents the amount of hydrogen provided by the hydrogen storage tank to the hydrogen fuel vehicle at time t. ST,min and V ST,max Respectively represent the minimum and maximum storage capacity of the storage tank.

[0073] Charging Strategy

[0074] Within a time period (24 hours), a corresponding hydrogen production strategy is formulated based on the total daily demand for hydrogen fuel cell vehicles. When a hydrogen fuel cell vehicle is ready to charge at time t, the hydrogen storage capacity of the charging station can be obtained through the interactive system. If the demand of hydrogen fuel cell vehicles cannot be met at this time, the driver will not choose this charging station. The constraints can be expressed as:

[0075] V t ST ≥V t HVN (10)

[0076] The electricity required for hydrogen production in the electrolyzer comes from wind power, photovoltaic power generation, and power provided to the DC bus by the AC distribution network. By scheduling the electrolyzer's hydrogen production power at different times, the optimal objective function can be met.

[0077] The present invention designs two charging modes:

[0078] (1) Normal charging mode: The normal charging mode is to charge the electric vehicle at the maximum constant power until the battery reaches the expected state of charge (SOC) value.

[0079] (2) Agreement charging mode: In the agreement charging mode, the electric-hydrogen hybrid charging station operator signs an agreement with the electric vehicle driver, and the integrated station charges to the expected SOC value within a certain period of time. The electric vehicle driver authorizes the electric-hydrogen hybrid charging station operator to freely dispatch the authorized electric vehicles for charging and discharging during this period.

[0080] In both charging modes, the electric-hydrogen hybrid charging station can only dispatch authorized electric vehicles that choose the protocol charging mode for charging, discharging, and non-operation. During the parking period of electric vehicles, the dispatch system sets the charging amount and dispatch time of each electric vehicle according to the corresponding dispatch schedule. It is assumed that the charging and discharging power is rated and remains unchanged during the charging period of each electric vehicle. In addition, the power loss comes entirely from the AC / DC module and the DC / DC module, and no power loss is generated in the other processes. Electric vehicles participating in the dispatch should be subject to the following restrictions:

[0081]

[0082] in Indicates the charging power of the electric vehicle scheduled by the protocol numbered m at time t. and Represent the charging and discharging power of electric vehicles respectively. and They represent the arrival time and departure time of the electric car numbered m respectively.

[0083]

[0084]

[0085] SOC min ≤SOC t,m ≤SOC max (14)

[0086]

[0087] Where: SOC t,m represents the SOC of electric vehicle m at time t, Δt m represents the scheduling time, k represents the charge and discharge coefficient, and C represents the battery capacity of the electric vehicle. In addition, SOC min and SOC max Represent the minimum and maximum values ​​of the electric vehicle SOC respectively. It is expressed as the expected SOC value of the electric vehicle numbered m when it leaves.

[0088] WF charging decision

[0089] To reveal drivers' charging behavior and measure the relationship between external factors and drivers' charging decisions, this embodiment proposes a stimulus-responsive charging decision estimation method based on the WF law. The World Economic Forum law has been applied in many fields, including psychology and marketing. This law has been widely used in the conversion of sensory stimulus values ​​to physical quantities. The WF law states that when the stimulus intensity increases exponentially, the sensory intensity increases geometrically. Therefore, the probability of an electric vehicle driver accepting a dispatch can be expressed as:

[0090]

[0091] ρ i =0.7ρ dsc,i +0.3ρ cha,i (17)

[0092] Where k represents the Weber fraction and s0 represents the stimulus constant. l Indicates the minimum analog value. This means that when the analog value is less than ρ lWhen , all electric vehicle drivers will not accept the dispatch. In addition, ρ h Indicates the maximum sensory value. When the analog value is greater than ρ h When the probability p n Corresponding to p h The simulation includes two influencing factors, among which ρ dsc,i represents the discount simulation amount, ρ cha,i Indicates the analog value of charging demand.

[0093] Objective function and constraints

[0094] The integrated station operator earns revenue by providing electricity and hydrogen to users. When electricity cannot meet operational needs, the operator purchases electricity from the AC distribution network at peak and valley prices. The optimization goal is to maximize the daily revenue of the integrated station. The objective function can be expressed as:

[0095]

[0096] Among them, P t ev_u and P t ev_p They represent the charging demands of electric vehicle drivers who do not participate in the agreement and those who participate in the agreement at time t. λ represents the incentive discount, EP t ev represents the electricity price at time t, HP represents the price of hydrogen, EP t pg Indicates the price of purchasing electricity from the grid.

[0097] Considering the user's charging experience, with the minimum total queuing time as the second goal, the scheduling time of each vehicle is optimized under the condition of meeting the constraints, reducing the occupancy of charging piles, shortening the average queuing time of users, and improving user experience. The queuing time of electric vehicle n at time t The remaining charging time of the vehicle being charged at time t and the number of vehicles in the queue N q It can be expressed as:

[0098]

[0099]

[0100]

[0101]

[0102] Among them, N ev Represents the number of electric vehicles charging in the compound station. It is expected that electric vehicles uThe battery threshold for replenishment. n Represents a state variable. When its value is 1, it indicates that the vehicle volume accepts the protocol scheduling. When it is 0, it indicates that the vehicle volume does not accept the protocol scheduling. ev Indicates the fast charging power of the charging pile. represents the ascending set of real-time remaining charging time of electric vehicles being charged at time t, N k represents the number of vehicles being charged in the composite station. h represents N q / M remainder, M represents the number of charging piles in the composite station. t left {h+1} represents the real-time remaining charging time of the {h+1}th vehicle at time t after ascending. In addition, N b Indicates the number of electric vehicles that have priority over other electric vehicles in the queue. Represents the time required for the queuing vehicle to charge to the required SOC threshold.

[0103] The sum of the power supplied to the electric vehicle and the electrolyzer is limited by the output power of the photovoltaic power generation, as shown in Equation (25). Equation (26) shows that the power supplied to the electric vehicle should be constrained by the demand of the electric vehicle. The amount of hydrogen in the hydrogen storage tank at the beginning of the day corresponds to the same time of the next day, which is constrained in Equation (27). The scheduling time of the electric vehicles participating in the protocol is constrained by Equation (28), as follows:

[0104]

[0105]

[0106] 0≤P t RE_ELE +P t RE_EV ≤P t RE (25)

[0107] 0≤P t RE_EV +P t FC_EV ≤P t ev_u +P t ev_p (26)

[0108]

[0109] Δt min ≤Δt n ≤Δt max (28)

[0110] n u and n prepresent the number of electric vehicle drivers who do not participate in the agreement and those who participate in the agreement, respectively. and Respectively represent the hydrogen weight of the gas tank at the initial time and the final time of a certain day. Δt min and Δt max Represent the minimum and maximum scheduling time respectively.

[0111] Solution process:

[0112] The present invention adopts the NSGA-Ⅱ algorithm to solve the above problems. NSGA-Ⅱ is a multi-objective optimization algorithm with high efficiency, diversity and adaptability. It uses the non-dominated sorting algorithm and the concept of congestion to maintain the diversity among individuals, which can prevent the population from converging to a single solution and promote better exploration of the search space. NSGA2 also uses innovative crossover and mutation operators to adjust the probability of crossover and mutation according to the current distribution and diversity of the population to increase the adaptability and performance of the algorithm. Among them, the difficulty in solving the optimization model lies in calculating the objective function, that is, optimizing the operating revenue and the driver's waiting time during the scheduling period. The calculation result of the objective function is the basis for fast non-dominated sorting of individuals in the NSGA-Ⅱ algorithm population. As Figure 1 As shown, the specific process of solving the problem using the NSGA-II algorithm in this embodiment is as follows:

[0113] Step 1: Based on the region’s weather forecast data, the wind speed and sunlight intensity for the next day are obtained. Combined with wind power and photovoltaic models, the output power at different times of the day is generated.

[0114] Step 2: Calculate the total charging power of electric vehicles and the total hydrogen demand for hydrogen fuel cell vehicles. Using the Monte Carlo method, simulate the daily travel and charging process of new energy vehicle users to obtain relevant data such as arrival time, charging demand, and starting SOC for electric and hydrogen fuel cell vehicles. Based on this data, a WF-based energy storage decision model is established to estimate the type of charging solution selected by charging users.

[0115] Step 3: Randomly generate the electrolyzer hydrogen production power for each time period based on the constraints. Obtain the hydrogen production power for one day based on the hydrogen production power for each time period.

[0116] Step 4: Based on the power balance relationship, calculate the power supply of the AC distribution network at time t and the discharge power of the fuel cell group at time t.

[0117] Step 5: One of the objective functions is to calculate the comprehensive operating cost. The electricity sales cost, hydrogen sales cost, and electricity purchase cost are obtained through steps 1 to 4 and summed to obtain the comprehensive operating income.

[0118] Step 6: Calculate the second objective function as the total queuing time of electric vehicle users. Obtain the total queuing time of electric vehicle users in one day based on the dispatch status of electric vehicles in each period.

[0119] Step 7: Based on the theoretical basis of steps 1 to 6, generate an initial population P of size G t , the offspring population Q is generated under the process of selection, crossover and mutation t , combined with the generated population R of size 2G t .

[0120] Step 8: For population R t Perform fast non-dominated sorting to form a non-dominated set Z i And calculate the crowding degree of individuals in each non-dominated layer, select suitable individuals according to the sorting results and crowding degree, until the number of individuals is G, the new generation parent population P t +1 formed.

[0121] Step 9: New generation parent population P t +1 generates a subpopulation Q under the process of selection, crossover, and mutation t +1, combined with the generated population R of size 2G t +1. Repeat steps 7 to 9 until the genetic generation is reached.

[0122] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0123] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0124] These computer program instructions may 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, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0126] This patent is not limited to the above-mentioned best implementation method. Anyone can derive various other forms of electric-hydrogen hybrid charging station optimization scheduling strategies based on charging decision prediction under the inspiration of this patent. All equal changes and modifications made within the scope of the patent application of this invention should be covered by this patent.

Claims

1. An optimized scheduling strategy for electric-hydrogen hybrid charging stations based on charging decision prediction, characterized by: Based on the charging strategy of new energy vehicles, the paper uses electricity price incentives to guide users to adjust their consumption behavior and establishes a Weber-Fechner incentive-response decision model based on psychology to describe drivers' choice of charging options. Taking into account the interests of both the user side and the energy supply side, the operating income of the electric-hydrogen hybrid charging station and the user's queuing time are used as the objective function, and the NSGA-Ⅱ algorithm is used to solve it. The specific process of solving the problem using the NSGA-Ⅱ algorithm is as follows: Step 1: Based on the local weather forecast data, we obtain the wind speed and sunlight intensity for the next day. Then, we combine the wind power and photovoltaic models to generate the output power at different times of the day. Step 2: Calculate the total charging power of electric vehicles and the total hydrogen demand of hydrogen fuel cell vehicles. Use the Monte Carlo method to simulate the daily travel and charging process of new energy vehicle users, and obtain data including the arrival time, charging demand, and starting SOC of electric and hydrogen fuel cell vehicles. Based on this data, a WF-based energy storage decision model is established to estimate the type of charging solution selected by charging users. Step 3: Randomly generate the electrolyzer hydrogen production power for each time period according to the constraints; obtain the hydrogen production power in one day based on the hydrogen production power in each time period; Step 4: Based on the power balance relationship, calculate the power supply of the AC distribution network at time t and the discharge power of the fuel cell group at time t; Step 5: Calculate the comprehensive operating cost of one of the objective functions; obtain the electricity sales cost, hydrogen sales cost, and electricity purchase cost through steps 1 to 4 and sum them to obtain the comprehensive operating income; Step 6: Calculate the total queuing time of electric vehicle users, which is the second objective function; obtain the total queuing time of electric vehicle users in one day according to the dispatch status of electric vehicles in each time period; Step 7: Based on steps 1 to 6, generate an initial population P of size G t , the offspring population Q is generated under the process of selection, crossover and mutation t , combined with the generated population R of size 2G t ; Step 8: For population R t Perform fast non-dominated sorting to form a non-dominated set Z i And calculate the crowding degree of individuals in each non-dominated layer, select suitable individuals according to the sorting results and crowding degree, until the number of individuals is G, the new generation parent population P t +1 formation; Step 9: New generation parent population P t +1 generates a subpopulation Q under the process of selection, crossover, and mutation t +1, combined with the generated population R of size 2G t +1; Repeat steps 7 to 9 until the genetic generation is reached; The electric-hydrogen hybrid charging station consists of a photovoltaic solar power generation system, a wind turbine system, an electrolyzer, a hydrogen storage tank and a fuel cell, and is connected to the public power grid so that electricity can be purchased when energy demand is not met; all electricity supplied to the power station from photovoltaic / wind power generation and the public power grid passes through a public coupling point; hydrogen in the station is produced by the electrolyzer and stored in a hydrogen tank; the stored hydrogen can be injected into the hydrogen fuel cell or converted into electricity through the fuel cell; electric vehicles are allowed to be charged directly using the electricity from the coupling point and the electricity generated by the fuel cell.

2. The optimized scheduling strategy for the electric-hydrogen hybrid charging station based on charging decision prediction according to claim 1 is characterized by: Two charging solutions are designed for electric vehicle users to choose from: (1) Normal charging mode: charging the electric vehicle at the maximum constant power until the battery state of charge (SOC) reaches the expected value; (2) Agreement charging mode: The operator of the electric-hydrogen hybrid charging station signs an agreement with the electric vehicle driver, and the integrated station charges to the expected SOC value within a certain period of time; the electric vehicle driver authorizes the operator of the electric-hydrogen hybrid charging station, allowing the operator of the electric-hydrogen hybrid charging station to freely dispatch authorized electric vehicles for charging and discharging during this period.

3. The optimized scheduling strategy for the electric-hydrogen hybrid charging station based on charging decision prediction according to claim 2 is characterized by: In both charging modes, the electric-hydrogen hybrid charging station can only dispatch authorized electric vehicles that choose the protocol charging mode to charge, discharge, and not operate; during the parking period of electric vehicles, the dispatch system sets the charging amount and dispatch time of each electric vehicle according to the corresponding dispatch schedule; Assuming that the charging and discharging power is rated and remains constant during the charging period of each electric vehicle, the power loss is entirely from the AC / DC module and the DC / DC module, and no power loss occurs in the rest of the process; the electric vehicles participating in the dispatch should be subject to the following restrictions: in Indicates the charging power of the electric vehicle scheduled by the protocol numbered m at time t; and Represent the charging and discharging power of electric vehicles respectively; and They represent the arrival time and departure time of the electric car numbered m respectively.

4. The optimized scheduling strategy for the electric-hydrogen hybrid charging station based on charging decision prediction according to claim 1 is characterized by: In the Weber-Fechner incentive-response decision model based on psychology, the probability of an electric vehicle driver accepting a dispatch is expressed as: r i =0.7ρ dsc,i +0.3p cha,i where k represents the Weber fraction, s0 represents the stimulus constant; ρ l Indicates the minimum analog value; when the analog value is less than ρ l When ρ, all electric vehicle drivers will not accept the dispatch; h Indicates the maximum sensory value; when the analog value is greater than ρ h When the probability p n Corresponding to p h ; The simulation includes two influencing factors, among which ρ dsc,i represents the discount simulation amount, ρ cha,i Indicates the analog value of charging demand.

5. The optimized scheduling strategy for the electric-hydrogen hybrid charging station based on charging decision prediction according to claim 1 is characterized in that: The integrated station operator earns revenue by providing electricity and hydrogen to users. When electricity cannot meet operational needs, the operator purchases electricity from the AC distribution network at peak and valley prices. The optimization goal is to maximize the daily revenue of the integrated station. The objective function is expressed as: in and They represent the charging demands of electric vehicle drivers who do not participate in the agreement and those who participate in the agreement at time t; λ represents the incentive discount, represents the electricity price at time t, HP represents the price of hydrogen, Indicates the price of purchasing electricity from the grid.

6. The optimized scheduling strategy for the electric-hydrogen hybrid charging station based on charging decision prediction according to claim 1 is characterized in that: Considering the user's charging experience, with the minimum total queuing time of the user as the second goal, the scheduling time of each vehicle is optimized under the condition of meeting the constraints, the occupancy of the charging pile is reduced, and the average queuing time of the user is shortened; the queuing time of electric vehicle n at time t The remaining charging time of the vehicle being charged at time t and the number of vehicles in the queue N q Related to; expressed as: Among them, N ev represents the number of electric vehicles charged in the composite station; It is expected that electric vehicles u Replenishment power threshold; x n Represents a state variable. When its value is 1, it indicates that the vehicle volume accepts the protocol scheduling. When it is 0, it indicates that the vehicle volume does not accept the protocol scheduling. ev Indicates the fast charging power of the charging pile; represents the ascending set of real-time remaining charging time of electric vehicles being charged at time t, N k represents the number of vehicles being charged in the composite station; h represents N q / the remainder of M, where M represents the number of charging piles in the composite station; Indicates the real-time remaining charging time of the {h+1}th vehicle at time t after ascending order; N b Indicates the number of electric vehicles that have priority over other electric vehicles in the queue. Represents the time required for the queuing vehicle to charge to the required SOC threshold.

7. The optimized scheduling strategy for electric-hydrogen hybrid charging stations based on charging decision prediction according to claim 1 is characterized in that: The constraints specifically include: the sum of the power supplied to the electric vehicle and the electrolyzer is limited by the output power of the photovoltaic power generation; the power supplied to the electric vehicle should be constrained by the demand of the electric vehicle; the amount of hydrogen in the hydrogen storage tank at the beginning of the day corresponds to the same time of the next day; the scheduling time Δt of the electric vehicles participating in the agreement n Subject to the following constraints: Δt min ≤Δt n ≤Δt max Δt min and Δt max Represent the minimum and maximum scheduling time respectively.

Images