Master-slave game scheduling method considering interaction between charge-discharge-storage integrated station and electric vehicle

Abstract

The invention provides a master-slave game scheduling method considering interaction between a charging, discharging and storing integrated station and an electric vehicle, and the method comprises the steps: firstly, building a charging, discharging and storing integrated station model, and carrying out the segmentation setting of multiple scenes of the charging, discharging and storing integrated station; secondly, establishing a dynamic road network model and predicting EV charging load space-time distribution under the constraint of an urban area road network in combination with EV travel characteristics; then, a multi-target master-slave game optimization scheduling model of the EV and the charging, discharging and storing integrated station is established according to a prediction result, and multi-target coordination is carried out on EV users and revenue of the charging, discharging and storing integrated station; and finally, converting the EV and all-in-one station multi-target master-slave game optimization scheduling model into a mixed integer linear programming model by adopting a KKT condition and a duality theory, so as to obtain an optimal solution. According to the method, the charging, discharging and storing integrated station is taken as a main body, and the two parties are guided by using the electricity price, so that the optimal strategy in each scene state is obtained, and the total cost of system operation is effectively reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of charging scheduling for electric vehicles, and particularly to a master-slave game scheduling method considering the interaction between a charging-discharging-storage integrated station and electric vehicles. Background Art

[0002] The power of a single EV is low, different types of EVs will bring different problems, and there are obvious differences in the travel behaviors of EV users, resulting in the difficulty of determining the EV load. Therefore, adopting an effective scheduling strategy to reasonably guide the charging behavior of EVs is beneficial to increasing the overall economic and environmental benefits of the microgrid.

[0003] Due to the uncertainty of EV charging load, many scholars have conducted extensive research on EV charging guidance strategies. The literature [Zhan Hua, Jiang Changxu, Su Qinglie. An EV Charging Guidance Method Based on Hierarchical Reinforcement Learning [J]. Electric Power Automation Equipment, 2022, 42(10): 264-272.] proposed an EV charging guidance method based on hierarchical enhanced deep network reinforcement learning, which can decide the optimal EV charging destination and driving route under random conditions; the literature [Shao Yinchi, Mu Yunfei, Lin Jiaying, et al. A Fast Charging Guidance Strategy for Electric Vehicles under the Multiple Demands of "Vehicle-Station-Network" [J]. Automation of Electric Power Systems, 2019, 43(18): 60-66+101.] proposed a fast charging guidance strategy for electric vehicles facing the multiple demands of electric vehicles (vehicles), fast charging stations (stations), and distribution networks (networks), which not only saves the user's charging cost but also improves the operation efficiency of the charging station; the literature [Zhang Cong, Peng Ke, Xiao Chuanliang, et al. An EV Charging Guidance Strategy Based on "Vehicle-Road-Network" Collaboration [J]. Electric Power Automation Equipment, 2022, 42(10): 125-133.] proposed a real-time user charging guidance strategy considering the operating status of the road network, charging stations, and distribution networks, reasonably alleviating traffic pressure and ensuring that the distribution network voltage is within the normal operating range; the literature [Zhou Chenrui, Sheng Guangzong, Li Sheng. Multi-objective Optimal Scheduling of Microgrids Considering the Access of Electric Vehicles [J]. Journal of Electrical Engineering, 2023, 18(01): 211-218.] considered that the use of electric vehicles has strong randomness and flexibility on the basis of the traditional microgrid model, and proposed a reasonable orderly charging and discharging scheduling strategy for electric vehicles under different electricity price mechanisms, significantly reducing the user cost and the operation cost of the microgrid; the literature [Ge Xiaolin, Cao Shipeng, Fu Yang, et al. Spatiotemporal Dual-scale Optimization Scheduling of Electric Vehicles Based on Regional Decoupling [J]. Proceedings of the CSEE, 2022, 1-13.] considered the electrical characteristics and travel characteristics of EVs in different regions, and proposed a spatiotemporal dual-scale optimization scheduling method for electric vehicles based on regional decoupling. Different scheduling models are adopted according to different regional characteristics, which is more conducive to the implementation of actual scheduling strategies. The above literature has all pointed out the impact of EV travel characteristics and charging behavior on EV load, and proposed effective scheduling strategies considering the uncertainty of EV load. If the charging and discharging of EV charging and storage integrated stations with energy storage devices are considered on the basis of EV charging guidance, on the one hand, it can effectively alleviate the impact of the uncertainty of EV charging load on the grid side, and on the other hand, it can also improve the economic benefits of charging stations and EV users.

[0004] According to relevant literature research, compared with traditional charging stations, integrated stations can not only control the flow of electric energy through the dispatching center to achieve peak load shifting of the power grid, but also centrally compensate for harmonics within the station to optimize the power quality. On the other hand, when an emergency occurs in the power grid, the EV charging, discharging and storage integrated station can be disconnected from the power grid for island operation to reduce economic losses. Reference [Chu Haoxiang, Jie Da. Charging and discharging control strategy of integrated charging and discharging station for electric vehicles considering the operation status of the power grid [J]. Electric Power Automation Equipment, 2018, 38(04): 96-101.] The battery packs in the battery swap station and the tiered station are divided into several parts, and a new charging and discharging control strategy is proposed in combination with the real-time load level of the power grid. While ensuring that the power storage equipment in the station is fully charged, it can also provide value-added services to the power grid; Reference [Yuan Hongtao, Wei Gang, Zhang He, et al. Robust optimization scheduling of active distribution network considering integrated charging and storage station [J]. Proceedings of the CSEE, 2020, 40(08): 2453-2468.] A two-stage robust optimization scheduling model is proposed that comprehensively considers the integrated charging and storage station for electric vehicles and the active distribution network. The integrated station is regarded as a new type of controllable energy source and reasonably involved in the optimization scheduling strategy, which effectively reduces the total operating cost of the system. Reference [Yuan Hongtao, Wei Gang, Zhang He, et al. Optimal operation of distribution network with integrated charging and storage stations based on model predictive control [J]. Automation of Electric Power Systems, 2020, 44(05): 187-197.] In order to address the risks brought to the distribution network by the large-scale access of EV charging loads, an optimal scheduling model that comprehensively considers the integrated charging and storage stations for electric vehicles and the active distribution network is proposed. The addition of the integrated station not only meets the needs of intraday optimal scheduling, but also reduces the operation and maintenance costs of the active distribution network. Summary of the invention

[0005] In view of the deficiencies in the above-mentioned background technology, the present invention proposes a master-slave game scheduling method that takes into account the interaction between the integrated charging, discharging and storage station and electric vehicles. The method takes the integrated charging, discharging and storage station as the main body and uses electricity prices to guide both parties, thereby obtaining the optimal strategy under various scenario conditions; as a new type of controllable energy, the integrated station reasonably participates in the optimization scheduling strategy, effectively reducing the total system operation cost.

[0006] The technical solution of the present invention is achieved in this way:

[0007] A master-slave game scheduling method considering the interaction between a charging, discharging and storage integrated station and an electric vehicle, the steps of which are as follows:

[0008] Step 1: Establish a charging, discharging and storage integrated station model and set up segmented settings for multiple scenarios of the charging, discharging and storage integrated station;

[0009] Step 2: Establish a dynamic road network model and combine it with EV travel characteristics to predict the spatiotemporal distribution of EV charging load under the constraints of the urban area road network;

[0010] Step 3: Based on the prediction results, establish a multi-objective master-slave game optimization scheduling model for EVs and charging-discharging-storage integrated stations to coordinate the benefits of EV users and charging-discharging-storage integrated stations;

[0011] Step 4: Use the KKT conditions and duality theory to transform the multi-objective master-slave game optimization scheduling model for EVs and integrated stations into a mixed-integer linear programming model, so as to obtain the optimal solution.

[0012] Preferably, the multi-scenarios of the charging-discharging-storage integrated station include three behaviors, namely the electricity storage behavior of purchasing electricity in the day-ahead market, the discharging behavior of selling electricity to the grid side, and the charging behavior of EV charging in the integrated station;

[0013] The multi-scenarios of the charging-discharging-storage integrated station are set in two stages:

[0014] 1) When the distribution network system operates normally or at low load, the integrated station gives priority to the charging behavior; when the EV charging demand is met, consider its electricity storage behavior. At this time, the integrated station shows load characteristics;

[0015] 2) When the distribution network system reaches the peak load operation state, the integrated station gives priority to the discharging behavior; the integrated station sells electricity to the grid side. At this time, the integrated station shows the characteristics of distributed power sources.

[0016] Preferably, the method for establishing the charging-discharging-storage integrated station model is as follows:

[0017] The charging and discharging adjustment rate ψ(i0) of the integrated station is:

[0018]

[0019] In the formula, when i0 = 1, it represents the charging and discharging status of the charging system of the EV charging-discharging-storage integrated station; when i0 = -1, it represents the charging and discharging status of the energy storage system of the EV charging-discharging-storage integrated station; P I (i0) is the rated charging power of the charging system and the energy storage system; P Ie (i0) is the actual discharging power of the charging system and the energy storage system; when ψ(i0)>0, it means that the EV charging-discharging-storage integrated station is in the charging state, and when ψ(i0)<0, it means that the EV charging-discharging-storage integrated station is in the discharging state;

[0020] The energy storage system provides electrical energy support for the charging system through a Boost boost circuit, and its operating power is P B (t), P B (t) satisfies the relationship with the remaining power of the charging system and the remaining power of the energy storage system:

[0021]

[0022] In the formula, QC (t) is the remaining power of the charging system; Q F (t) is the remaining power of the energy storage system; is the charging power of the nth EV in the t time period; is the actual charging power of the EV; P C (t) is the power of the charging system at time t; P F (t) is the power of the energy storage system at time t;

[0023] Due to the limitations of the energy storage device and the converter device in the EV charging, discharging and energy storage integrated station, the energy storage system and the charging system will be restricted by boundary conditions, and their power constraints are as follows:

[0024] P C,Fmin ≤P C,F ≤P C,Fmax (3)

[0025] In the formula: P C,F is the charging and discharging power of the energy storage system of the charging, discharging and energy storage integrated station; P C,Fmin is the minimum charging and discharging power of the energy storage system of the charging, discharging and energy storage integrated station, P C,Fmax is the maximum charging and discharging power of the energy storage system of the charging, discharging and energy storage integrated station.

[0026] Preferably, the dynamic road network model includes a dynamic traffic road network model, an urban road resistance model, a distribution network model and an EV charging price responsiveness model;

[0027] The dynamic traffic road network model is expressed as:

[0028]

[0029] In the formula: G is the set of traffic road networks; V is the set of all nodes in the traffic road network; E is the set of all road sections in the traffic road network; H is the set of divided time, that is, the whole day is divided into T time periods; W is the set of road section weights, indicating the travel cost of vehicles passing through this road section, which can be quantified by travel time, passing speed and cost; the connection relationship between nodes in the traffic road network is described by the matrix D; the element d of the matrix D ij (t) The expression of is:

[0030]

[0031] In the formula: v ij is the connecting road section between the i-th and the j-th nodes; w ij (t) is the weight of the road section v in the t time period ij ;

[0032] The urban road resistance model is expressed as:

[0033]

[0034] Where: L ij is the length of road ij; R i is the waiting time for traffic lights at intersection i; v ij (t) represents the driving speed of the EV at time t; v ij,max is the zero-flow speed of road ij; C ij is the traffic capacity of road ij; Q ij (t) is the traffic flow of the section of road ij at time t; ω is the road traffic level; a, b, and γ are adaptive coefficients for different road levels;

[0035] The distribution network model is expressed as:

[0036]

[0037] Where: is the nth distribution network node; G Y is the position of the source node; C Y is the capacity position of the source node;

[0038] The total sum P of the total EV charging power connected to the nth distribution network node at time t n is:

[0039]

[0040] Where: is the charging power of the nth EV in the t time period; N is the total number of EV chargings in the t time period;

[0041] The EV charging price response model is expressed as:

[0042]

[0043] Where: λ c (t) is the EV demand response degree in the t time period; △c s,t (t) is the electricity price difference between the kth moment and the tth moment in the t time period; l s,t is the starting threshold, h s,t is the saturation threshold; λ c,max is the maximum EV demand response degree.

[0044] Preferably, the method for predicting the spatio-temporal distribution of EV charging load under the constraints of the urban area road network is:

[0045] Obtain the unit mileage △Cap according to the type of each EV, and combine the EV battery capacity with the initial SOC to obtain the initial electricity quantity Cap0; among them, the initial SOC is the battery state when the EV has just finished charging;

[0046] Use Monte Carlo sampling to allocate the starting node and the initial travel time t of each EVs , combined with the initial departure time t s , corresponding to the OD probability matrix, randomly sample to generate the destination node;

[0047] By repeatedly calling the OD probability matrix of each time period of the EV, depict the travel trajectory of each EV;

[0048] When the remaining power of the EV is greater than the charging threshold, complete the trip according to the specified route. Otherwise, charge quickly nearby to determine the spatio-temporal information of the fast charging load; repeat the above steps to traverse all EVs and count the charging load of each node in the distribution network.

[0049] Preferably, calculate the remaining power Cap of the EV at time t according to the unit mileage △Cap t :

[0050] Cap t = η(Cap t-1 -△l·△Cap) (12)

[0051] In the formula: Cap t-1 is the remaining power of the EV at time t-1; η is the energy consumption coefficient, indicating the power loss caused by starting and braking during vehicle driving; △l is the driving distance of the vehicle from time t-1 to t.

[0052] Preferably, the OD probability matrix is expressed as:

[0053]

[0054] In the formula: represents the number of EVs with the starting point i and the ending point j in the time period from T to T+1; represents the probability that the EV stays in place in the time period from T to T+1; represents the sum of the number of EVs from node i to any node in the time period from T to T+1.

[0055] Preferably, the multi-objective master-slave game optimization scheduling model of the EV and the integrated charging, discharging and storage station includes the objective function of the integrated station and the constraints of the EV and the integrated station, the objective function of the EV and the constraints of the EV;

[0056] The expression of the objective function of the integrated station is:

[0057]

[0058] In the formula: N is the number of EVs charging at the integrated station in a day; △t is the time spent by the vehicle due to road impedance; δ - (t) is the electricity purchase price of the integrated station from the power grid at time t, δ + (t) is the electricity selling price of the integrated station to the power grid at time t; E -(t) is the electricity purchase quantity of the integrated station from the real-time market during period t, E + (t) is the electricity sales quantity of the integrated station from the real-time market during period t; δ d (t) is the day-ahead contract electricity price during period t; E(t) is the contract electricity quantity in the day-ahead market during period t; is the charging electricity price of the nth EV during period t;

[0059] The constraints of the EV integrated station include:

[0060] Charging electricity price constraint:

[0061]

[0062] In the formula: is the highest electricity price during period t, is the lowest electricity price during period t; is the daily average electricity price;

[0063] Day-ahead electricity purchase quantity constraint and real-time electricity purchase and sales quantity constraint of the integrated station:

[0064]

[0065] In the formula: z(t) is a Boolean variable indicating the electricity trading status during period t; M is a positive number; is the energy discharge quantity of the integrated station's energy storage during period t;

[0066] Energy balance constraint:

[0067]

[0068] In the formula: is the charging quantity of the integrated station's energy storage equipment during period t;

[0069] Charging and discharging constraints of the integrated station's energy storage equipment:

[0070]

[0071] In the formula: u(t) is a Boolean variable indicating the status of the integrated station's energy storage facility during period t; is the maximum charging power of the integrated station's energy storage equipment, is the maximum discharging power of the integrated station's energy storage equipment;

[0072] Energy storage electricity quantity constraint of the integrated station:

[0073]

[0074] In the formula: S(t) is the energy storage electricity quantity of the integrated station's energy storage facility during period t; η + is the charging efficiency of the integrated station's energy storage equipment, η - is the discharging efficiency of the integrated station's energy storage equipment; Smax is the maximum capacity of the integrated station energy storage device; S(0) is the initial power of the integrated station energy storage device;

[0075] The EV objective function is expressed as:

[0076]

[0077] The EV constraint conditions include:

[0078] EV battery state of charge constraint:

[0079]

[0080] In the formula: ξ is the battery state of charge expected by the user; is the battery capacity of the nth EV; is the initial power of the nth EV; μ represents the EV charging efficiency; T a is the EV charging duration;

[0081] Charging power constraint:

[0082]

[0083] In the formula: is the maximum charging power of the nth vehicle.

[0084] Preferably, the method of transforming the EV and integrated station multi-objective master-slave game optimization scheduling model into a mixed-integer linear programming model by using the KKT conditions and duality theory is:

[0085] Equivalent nonlinear programming transformation of the master-slave game model:

[0086] For both sides of the game, the price is determined when the EV makes a decision. Replace the linear programming equations (20)-(22) with the corresponding KKT conditions to eliminate this optimization problem; Denote the dual variables as α1, ω it , and the KKT conditions corresponding to the linear programming equations (20)-(22) are:

[0087]

[0088]

[0089]

[0090]

[0091]

[0092] The constraints (25) and (26) are complementary slackness conditions, and \(x\perp y\) means that at most one of the scalars \(x\) and \(y\) can be strictly greater than 0; equations (23)-(27) transform the lower-level optimization into constraints.

[0093] Linearization of complementary slackness conditions:

[0094] Introduce Boolean variables and Transform the constraints (25) and (26) into linear inequalities:

[0095]

[0096]

[0097]

[0098]

[0099] Linearization of the objective function of the integrated station:

[0100] The source of the non-linearity of the objective function is the product of the electricity price and the charging power According to the duality theorem of linear programming, transform the linear programming equations (20)-(22) into:

[0101]

[0102] On the premise of satisfying the KKT conditions, the objective function equation (14) is equivalent to:

[0103]

[0104] Equivalent mixed-integer linear programming model of the master-slave game model:

[0105] The electricity price pricing game of the integrated station can be equivalently transformed into the following mixed-integer linear programming:

[0106]

[0107] s.t. equations (15)-(29), equations (23)-(24), equations (27)-(31)

[0108] In the optimal solution of the mixed-integer linear programming equation (34) It constitutes the Stackelberg equilibrium of the games (14)-(27), (20)-(22).

[0109] Preferably, combined with the EV charging spatio-temporal distribution prediction model and the multi-objective master-slave game optimization scheduling model of electric vehicles and integrated stations, an EV charging and discharging optimization scheduling model is proposed. The specific scheduling steps are as follows:

[0110] Obtain the spatio-temporal distribution of the charging demand of each node according to the prediction results in Step 2, and obtain the optimized electricity price through the master-slave game;

[0111] Formulate the charging electricity price based on the optimized electricity price according to the load conditions of each node;

[0112] Formulate the optimal charging plan according to the optimal electricity price.

[0113] Compared with the prior art, the beneficial effects produced by the present invention are as follows:

[0114] 1) A multi-objective master-slave game optimization scheduling model for EVs and charging, discharging, and energy storage integrated stations is established for the common interests of both EV users and charging stations. Taking the charging, discharging, and energy storage integrated station as the main body, with the maximization of the interests of both parties as the goal, and considering the interaction between EVs and the charging, discharging, and energy storage integrated station, a master-slave game optimization scheduling strategy considering the interaction between the charging, discharging, and energy storage integrated station and EVs is proposed. The response of the EV charging strategy to the electricity price is comprehensively considered, making the scheduling strategy closer to reality.

[0115] 2) This strategy effectively reduces the costs of EV users by segmentally optimizing multiple scenarios of the charging, discharging, and energy storage integrated station. The introduction of the charging, discharging, and energy storage integrated station greatly increases the income of the charging station, achieving a win-win situation for the charging, discharging, and energy storage integrated station and EV users. Compared with traditional charging stations, the charging, discharging, and energy storage integrated station is a new type of controllable energy source, which can not only reasonably participate in the optimization scheduling strategy but also effectively reduce the total system operation cost. By setting a reasonable planned energy storage device capacity and the lowest electricity price floor for the charging, discharging, and energy storage integrated station, the income of the charging, discharging, and energy storage integrated station can be increased, the total system operation cost can be reduced, and the construction cost of the charging, discharging, and energy storage integrated station can also be reduced. The charging, discharging, and energy storage integrated station can also sell electricity to the grid side, effectively increasing the income of the charging, discharging, and energy storage integrated station and playing a role in peak shaving and valley filling for the grid side. BRIEF DESCRIPTION OF THE DRAWINGS

[0116] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0117] Figure 1 It is the traffic network topology structure diagram of the present invention.

[0118] Figure 2 It is the flow chart of the spatio-temporal distribution prediction of the EV charging load of the present invention.

[0119] Figure 3 It is the flow chart of the EV charging and discharging optimization scheduling of the present invention.

[0120] Figure 4 This is the road network map adopted in the embodiments of the present invention.

[0121] Figure 5 This is the spatio-temporal distribution diagram of the charging load of each node measured by the method of the present invention.

[0122] Figure 6 This is the comparison result of the profits of the integrated station with different pricing lower limits by the method of the present invention.

[0123] Figure 7 This is the comparison result of the charging costs of EV users with different pricing lower limits by the method of the present invention.

[0124] Figure 8 This is the comparison result of the profits of the charging station under different Smax by the method of the present invention. Specific embodiments

[0125] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all 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.

[0126] The embodiments of the present invention provide a master-slave game scheduling method considering the interaction between the charging, discharging and storage integrated station and electric vehicles. First, by establishing a charging, discharging and storage integrated station model and segmentally setting for multiple scenarios of the charging, discharging and storage integrated station; second, establishing a dynamic road network model and combining with the EV travel characteristics to predict the spatio-temporal distribution of the EV charging load under the constraints of the urban regional road network; then, according to the prediction results, establishing a multi-objective master-slave game optimization scheduling model for EVs and the charging, discharging and storage integrated station to coordinate the multi-objectives of the EV users' and the charging, discharging and storage integrated station's revenues; finally, using the KKT conditions and duality theory to transform the multi-objective master-slave game optimization scheduling model for EVs and the integrated station into a mixed-integer linear programming model, so as to obtain the optimal solution. The proposed master-slave game model and scheduling strategy can enable both the EV users and the charging, discharging and storage integrated station to obtain the maximum revenue.

[0127] The multi-scenario setting is a solution method that transforms uncertain factors that are difficult to accurately describe in a stochastic process into deterministic factors through multiple scenarios. The present invention divides the charging, discharging and storage integrated station into three behaviors according to its characteristics, namely its electricity storage behavior of purchasing electricity in the day-ahead market, its discharging behavior of selling electricity to the grid side, and its charging behavior of EV charging in the integrated station. According to the above three behavior characteristics, a multi-scenario model of the integrated station is established in combination with the operation state of the distribution network system, as shown in Table 1.

[0128] Table 1 System multi-scenario description

[0129]

[0130] The multi-scenario setting of the integrated station is divided into two stages, and each stage is set separately.

[0131] 1) When the distribution network system is operating normally or at low load, the integrated station gives priority to the charging behavior; when the EV charging demand is met, its energy storage behavior is considered. At this time, the integrated station exhibits load characteristics.

[0132] 2) When the distribution network system reaches the peak load operation state, the integrated station gives priority to the discharging behavior; the integrated station sells electricity to the grid side. At this time, the integrated station acts as a distributed power source to achieve the effect of peak shaving. Compared with the first stage, the integrated station exhibits distributed power source characteristics at this time.

[0133] To better realize the two-way interaction ability between the EV charging station and electric energy and increase the scale of EV access to the grid, an EV charging, discharging, and energy storage integrated station is established. The integrated station can not only provide charging services for EV users but also discharge to the grid side through the energy storage system. The charge-discharge adjustment rate ψ(i0) of the integrated station is as follows:

[0134]

[0135] In the formula, when i0 = 1, it represents the charge-discharge status of the charging system of the EV charging, discharging, and energy storage integrated station; when i0 = -1, it represents the charge-discharge status of the energy storage system of the EV charging, discharging, and energy storage integrated station; P I (i0) is the rated charging power of the charging system and the energy storage system; P Ie (i0) is the actual discharging power of the charging system and the energy storage system; when ψ(i0) > 0, it means that the EV charging, discharging, and energy storage integrated station is in the charging state, and when ψ(i0) < 0, it means that the EV charging, discharging, and energy storage integrated station is in the discharging state.

[0136] The energy storage system provides electrical energy support for the charging system through a Boost boost circuit, and its operating power is P B (t), P B (t) satisfies the relationship with the remaining power of the charging system and the remaining power of the energy storage system as follows:

[0137]

[0138] In the formula, Q C (t) is the remaining power of the charging system; Q F (t) is the remaining power of the energy storage system; is the charging power of the nth EV in the t period; P C is the actual charging power of the EV; P C (t) is the power of the charging system at the t moment; P F (t) is the power of the energy storage system at the t moment.

[0139] Due to the limitations of the energy storage device and converter device of the EV charging and discharging integrated station, the energy storage system and charging system will be subject to boundary conditions. The power constraints are as follows:

[0140] P C,Fmin ≤P C,F ≤P C,Fmax (3)

[0141] Where: P C,F P is the charging and discharging power of the energy storage system of the charging and discharging integrated station; C,Fmin is the minimum charging and discharging power of the energy storage system in the charging and discharging integrated station, P C,Fmax It is the maximum charging and discharging power of the energy storage system of the integrated charging and discharging station.

[0142] The dynamic road network model includes a dynamic traffic network model, an urban road resistance model, a distribution network model and an EV charging electricity price responsiveness model.

[0143] The graph theory analysis method is used to model the traffic network. The traffic network topology structure diagram is as follows Figure 1 shown.

[0144] In order to reflect the interactive characteristics of "vehicle-road-network", the present invention adopts a dynamic road network model to update the traffic flow of each road section at fixed time periods. The dynamic road network model is described as follows:

[0145]

[0146] In the formula: G is the set of traffic network; V is the set of all nodes in the traffic network; E is the set of all road sections in the traffic network; H is the set of divided time, that is, the whole day is divided into T time periods; W is the set of road section weights, which represents the travel cost of a vehicle passing through the road section, which can be quantified by time consumption, passing speed and cost. The connection relationship between nodes in the traffic network is described by matrix D; the element d of matrix D is ij The expression of (t) is:

[0147]

[0148] Where: v ij is the connecting section between the i-th and j-th nodes; w ij (t) is the road section v during period t ij The weight of .

[0149] Since the urban road network has multiple intersections and changes in real time, the EV driving process will not only be affected by the road section impedance, but also the traffic lights will affect the driving time. In order to meet the EV charging needs on urban roads, the urban road network resistance model is introduced. This paper uses the driving time as the road resistance for modeling and analysis, and refers to the speed-flow model. EV driving speed v ijThe expression of (t) is:

[0150]

[0151] In the formula: v ij,max is the zero-flow speed of road ij; C ij is the traffic capacity of road ij; Q ij (t) is the traffic flow of road ij at time t; ω is the road traffic level; a, b, and γ are adaptive coefficients for different road levels. The urban road impedance model is expressed as:

[0152]

[0153] In the formula: L ij is the length of road ij; R i is the waiting time for the traffic light at intersection i.

[0154] In the "vehicle-road-network" mode, it is necessary to realize the spatial coupling of the distribution network and the road network. Therefore, when establishing the road network model, it is necessary to establish a suitable distribution network that matches the road network. To obtain the impacts of a large number of EV charging behaviors on the traffic flow of road sections, the spatio-temporal distribution of the power grid load, and the profits of users and charging stations, the present invention conducts two-way power trading between the distribution network as the power supply side and the charging stations. The model expression of the nth distribution network node is:

[0155]

[0156] In the formula: G Y is the position of the source node; C Y is the capacity position of the source node.

[0157] The total sum P n of the total EV charging power connected to the nth distribution network node at time t is:

[0158]

[0159] In the formula: is the charging power of the nth EV in the t time period; N is the total number of EV chargings in the t time period.

[0160] Under normal circumstances, EV users will choose the charging station with the shortest path, but this may lead to load over-limit in a certain time period, and in severe cases, it will affect the operation stability of the power grid. Therefore, the present invention introduces an EV charging electricity price responsiveness model to reasonably regulate EV charging behaviors to balance the EV charging load distribution in each time period.

[0161] EV users will only participate in the regulation process when the price starts to reach the threshold, and there is a price saturation threshold. If the electricity price difference during the regulation period continues to increase after exceeding the saturation threshold, the number of EV users participating in the charging response will not increase. The demand response degree λ of EV users c is expressed as:

[0162]

[0163] In the formula: λ c (t) is the EV demand response degree at time t; △c s,t (t) is the electricity price difference between time k and time t at time t; l s,t is the starting threshold, h s,t is the saturation threshold; λ c,max is the maximum EV demand response degree.

[0164] The trips of EV users determine the initial departure time and return time of EVs. The present invention assumes that the travel behaviors of fuel vehicles and EV users are the same. Therefore, referring to the probability distribution curves of the initial departure time and return time of EVs in the American household travel survey data, the initial departure time of EVs is generated accordingly. It can be seen from the American household travel survey data that the battery capacities Cap of different types of EVs r follow the gamma distribution of Equation (11).

[0165]

[0166] In the formula: Cap r is the EV capacity.

[0167] Assume that the initial state of charge (SOC) is the battery state when the EV has just been fully charged. According to the EV battery capacity and combined with the initial SOC, the initial electricity amount Cap0 is obtained. Since the power consumption of the EV increases linearly with the driving mileage, the remaining electricity amount Cap at time t t is:

[0168] Cap t =η(Cap t-1 -△l·△Cap) (12)

[0169] In the formula: Cap t-1 is the remaining electricity amount of the EV at time t - 1; η is the energy consumption coefficient, indicating the power loss caused by starting and braking during the vehicle's driving process; △l is the driving distance of the vehicle from time t - 1 to time t.

[0170] To analyze the spatio-temporal characteristics of EVs, the OD analysis method is introduced. By consulting the historical data of the transportation department, the traffic volumes of various types of EVs on each section in each period are obtained, and the OD matrix H of each period is inversely deduced from the traffic volumes of the sections according to the OD matrix algorithm of the complex traffic network, and the travel characteristics of different types of EVs are characterized.

[0171] One day is divided into 24 hours, so the OD matrix is divided into 24 parts, and each part is a sub-matrix. m is the number of road nodes in the simulation area, T = 0, 1, …, 23. is the traffic volume of the starting point and destination of vehicles in the time period from T to T + 1. Therefore, in the time period from T to T + 1, the probability that an EV starts from node i and ends at the terminal node j is:

[0172]

[0173] In the formula: represents the number of EVs starting from i and ending at j in the time period from T to T + 1 (1 ≤ i ≤ m, 1 ≤ j ≤ m); represents the probability that an EV stays in place in the time period from T to T + 1; represents the sum of the number of EVs from node i to any node in the time period from T to T + 1.

[0174] EV charging demand prediction is a very important prerequisite for studying the optimal regulation of EV charging and discharging. Under ideal conditions, the starting node and initial travel time t of each EV are assigned by Monte Carlo sampling. s , combined with t s corresponding OD probability matrix, randomly sample to generate the destination node. By repeatedly calling the OD probability matrix of EVs in each time period, the travel trajectory of EVs is depicted. When the EV meets the charging demand, the charging load of EV users can be predicted. In this prediction model, the influence of road impedance on EVs is not considered, and when the EV meets the charging demand, it chooses the nearest charging station to charge. The specific process is as Figure 2 shown. First, obtain the unit mileage △Cap according to the type of each EV. According to the battery capacity of the EV and combined with the initial SOC, obtain the initial power Cap0; where the initial SOC is the battery state when the EV has just been fully charged. Secondly, use Monte Carlo sampling to assign the starting node and initial travel time t of each EV. s , combined with the initial travel time t s corresponding OD probability matrix, randomly sample to generate the destination node; then, depict the travel trajectory of each EV by repeatedly calling the OD probability matrix of EVs in each time period; finally, when the remaining power of the EV is greater than the charging threshold, complete the travel according to the specified route, otherwise, quickly charge nearby and determine the spatio-temporal information of the fast charging load; repeat the above steps to traverse all EVs and count the charging load of each node in the distribution network.

[0175] The present invention guides EV charging in a way that responds to electricity prices. EVs will choose the time period that maximizes their own interests for charging according to different time electricity prices. A charging-discharging-storage integrated station and EVs form a leader-follower game, with the integrated station as the leader and EVs as the followers.

[0176] Maximizing its own profit is the goal of the charging-discharging-storage integrated station. The profit of the integrated station consists of four parts: one is the revenue from selling electricity to the real-time market; the second is its cost of purchasing electricity in the day-ahead market; the third is its cost of purchasing electricity from the real-time power grid; and the fourth is the charging revenue of EVs at the integrated station. Therefore, the objective function expression of the integrated station is:

[0177]

[0178] In the formula: N is the number of EVs charging at the integrated station in a day; △t is the duration spent by the vehicle due to road impedance; δ - (t) is the electricity purchase price of the integrated station from the power grid at time t, and δ + (t) is the electricity selling price of the integrated station to the power grid at time t; E - (t) is the electricity purchase quantity of the integrated station from the real-time market at time t, and E + (t) is the electricity selling quantity of the integrated station to the real-time market at time t; δ d (t) is the day-ahead contract electricity price at time t; E(t) is the contract electricity quantity in the day-ahead market at time t; is the charging electricity price of the nth EV at time t.

[0179] The constraints of the EV integrated station include:

[0180] 1) Charging electricity price constraint:

[0181]

[0182] In the formula: is the highest electricity price at time t, is the lowest electricity price at time t; is the daily average electricity price.

[0183] 2) Day-ahead electricity purchase quantity constraint and real-time electricity purchase and sale quantity constraint of the integrated station:

[0184]

[0185] In the formula: z(t) is a Boolean variable indicating the electricity trading status at time t; M is a positive number; is the energy storage discharge quantity of the integrated station at time t.

[0186] 3) Energy balance constraint:

[0187]

[0188] In the formula: The charging amount of the integrated station energy storage device during period t.

[0189] 4) Charging and discharging constraints of the integrated station energy storage device:

[0190]

[0191] In the formula: u(t) is a Boolean variable representing the state of the integrated station energy storage facility during period t; is the maximum charging power of the integrated station energy storage device, is the maximum discharging power of the integrated station energy storage device.

[0192] 5) Energy storage power constraints of the integrated station:

[0193]

[0194] In the formula: S(t) is the energy storage power of the integrated station energy storage facility during period t; η + is the charging efficiency of the integrated station energy storage device, η - is the discharging efficiency of the integrated station energy storage device; S max is the maximum capacity of the integrated station energy storage device; S(0) is the initial power of the integrated station energy storage device.

[0195] The goal of the EV is to minimize the charging cost, and its objective function is expressed as:

[0196]

[0197] The EV constraint conditions include:

[0198] 1) EV battery state of charge constraint:

[0199] The EV charging amount should make the battery reach the corresponding state of charge, and the constraint conditions are as follows:

[0200]

[0201] In the formula: ξ is the desired battery state of charge of the user; is the battery capacity of the nth EV; is the initial power of the nth EV; μ represents the EV charging efficiency; T a is the EV charging duration.

[0202] 2) Charging power constraint:

[0203]

[0204] In the formula: is the maximum charging power of the nth vehicle.

[0205] The present invention uses the KKT conditions and duality theory to transform the multi-objective master-slave game optimization scheduling model of EVs and integrated stations into a mixed-integer linear programming model, thereby obtaining the optimal solution.

[0206] 1) Equivalent nonlinear programming transformation of the master-slave game model:

[0207] For both sides of the game, the price is determined when the EV makes a decision. Replace the linear programming equations (20)-(22) with the corresponding KKT conditions to eliminate this optimization problem; denote the dual variables as α1, ω it , and the KKT conditions corresponding to the linear programming equations (20)-(22) are:

[0208]

[0209]

[0210]

[0211]

[0212]

[0213] Constraints (25) and (26) are complementary slackness conditions, where x⊥y means that at most one of the scalars x and y can be strictly greater than 0; equations (23)-(27) transform the lower-level optimization into constraints. Therefore, the objective functions (14) and (25), (26) are nonlinear. The following analyzes the linearization of the nonlinear problem.

[0214] 2) Linearization of the complementary slackness conditions:

[0215] Introduce Boolean variables and to transform constraints (25) and (26) into linear inequalities:

[0216]

[0217]

[0218]

[0219]

[0220] 3) Linearization of the integrated station objective function:

[0221] The source of the nonlinearization of the objective function is the product of the electricity price and the charging power The duality theorem of linear programming states that at the optimal solution, the objective function values of the dual problem and the original problem are the same. Transform the linear programming equations (20)-(22) into:

[0222]

[0223] On the premise of satisfying the KKT conditions, the objective function formula (14) is equivalent to:

[0224]

[0225] Formula (33) is linear with respect to the decision variables.

[0226] 4) Equivalent mixed-integer linear programming model of the master-slave game model:

[0227] The electricity price pricing game of the integrated station can be equivalent to the following mixed-integer linear programming:

[0228]

[0229] s.t. formulas (15)-(29), formulas (23)-(24), formulas (27)-(31)

[0230] In the optimal solution of the mixed-integer linear programming formula (34) constitutes the Stackelberg equilibrium of the games (14)-(27), (20)-(22).

[0231] Combined with the EV charging spatio-temporal distribution prediction model and the multi-objective master-slave game optimization scheduling model of electric vehicles and integrated stations, an EV charging and discharging optimization scheduling model is proposed. The specific scheduling process is as Figure 3 shown, and the steps are as follows:

[0232] 1) Obtain the spatio-temporal distribution of the charging demand at each node according to the above prediction results, and obtain its optimized electricity price through the master-slave game;

[0233] 2) Formulate the charging electricity price based on the optimized electricity price according to the load conditions at each node;

[0234] 3) Formulate the optimal charging plan according to the optimal electricity price.

[0235] Case study

[0236] In this case study, a road network in a certain area of a certain city combined with the IEEE 33-node distribution network system is used to conduct a numerical example analysis of the model of the present invention. The road network contains 31 nodes and 52 roads, and the average length of the roads is 1.44 km. The specific road network diagram is as Figure 4 shown.

[0237] Combined with the population quantity and vehicle penetration rate in this area, 1000 EVs are introduced, including 250 private cars, 450 taxis, and 300 other public vehicles. The experimental software environment is MATLAB 2021b, and the CPLEX solver is used for solving.

[0238] The charging demand of EV users needs to be judged according to their travel demand and the current SOC of the vehicle to determine whether they need to charge. Therefore, it is necessary to regulate the charging load of EVs. The basis of load regulation is the prediction of EV charging load, and a prediction model is used to predict the EV load. The spatio-temporal distribution of the charging load at each node is as Figure 5 shown.

[0239] It can be Figure 5 seen that there are 2 peak periods in a day. The first peak period appears in the 08:00-09:00 time period, and the peak value is about 1644 KW; the second peak period appears in the 18:00-19:00 time period, and the peak value is about 4320 KW. In addition, private cars have charging demands in the 02:00-17:00 time period, and reach the peak at 10:00 in the morning; taxis and other public vehicles have charging demands throughout the day, and the charging peak appears at 7:00 in the afternoon.

[0240] The day-ahead market electricity price is shown in Table 2. Generally speaking, the real-time market electricity purchase price will be higher than the day-ahead market electricity purchase price. Therefore, it is assumed that the upper limit of the retail electricity price is 1.2 times the electricity price, the lower limit is 0.8 times the electricity price, and the average electricity price is 0.9 yuan / (kW·h). The experimental parameters are shown in Table 3.

[0241] Table 2 Day-ahead market electricity price

[0242]

[0243] Table 3 Experimental parameters

[0244]

[0245]

[0246] Solving Equation (32) can obtain the maximum benefit of the integrated station of 4823 yuan. The optimal strategy of the charging station is shown in Table 4. According to the comparison between the optimal electricity price and the day-ahead electricity price, the integrated station will always set the upper and lower limits of the electricity price at the EV day-ahead charging and non-charging time periods to meet the average price. Because the real-time market electricity price is higher than the day-ahead market electricity price, there is no need to purchase electricity from the real-time market. This is to cope with some uncertain factors, such as EVs temporarily leaving the charging station; on the other hand, from the EV charge and discharge optimization scheduling model, it can be obtained that the optimal charging time of EVs is in the 01:00-08:00 time period, and the EV charging cost is the lowest in this time period. Rational EV users have no tendency to deviate from this charging time. Therefore, under this optimized electricity price, the charging behavior of EVs is an orderly behavior.

[0247] Table 4 Optimal strategy of the integrated station

[0248]

[0249]

[0250] As can be seen from Table 4, when the distribution network system is operating normally or at low load, the distribution network system gives priority to meeting the charging needs of EVs in the integrated station, ensuring the normal driving of EV users. After meeting the charging needs of EVs in the integrated station, electricity is purchased during the period when the day-ahead market electricity price is relatively low. This not only ensures the normal operation of the distribution network system, but also meets the EV charging needs, reduces the electricity purchase cost, and obtains greater benefits, thus achieving the optimal operation state of this scenario.

[0251] To verify the advantages of the charging, discharging and energy storage integrated station compared with the traditional charging station, S max was set to 0 for simulation, and the optimal strategy of the traditional charging station is shown in Table 5.

[0252] Table 5 Optimal Strategy of Traditional Charging Station

[0253]

[0254]

[0255] As can be seen from Table 5, combined with the day-ahead electricity price, the income of the traditional charging station is 3,191 yuan. It can be known from Section 5.2.2 that the income of the charging, discharging and energy storage integrated station is 4,823 yuan. Therefore, adopting the charging, discharging and energy storage integrated station can obtain greater benefits. Compared with the charging, discharging and energy storage integrated station, the traditional charging station does not have energy storage equipment, while the charging, discharging and energy storage integrated station can purchase electricity from the grid side at a low electricity price and store it in the energy storage equipment, and discharge it to the grid side at the peak of the electricity price, which can not only bring additional income but also reduce the load peak. Therefore, the charging, discharging and energy storage integrated station has more advantages than the traditional charging station.

[0256] On the premise of keeping other parameters unchanged, the pricing lower limit is changed from 0.5 times the day-ahead market electricity price to 0.9 times the day-ahead market electricity price, and the profit of the integrated station and the change of the charging cost of EV users are as Figure 6 、 7 shown.

[0257] From Figure 6 、 7It can be seen that when the lower limit of the charging electricity price is set at 0.5 times the day-ahead market electricity price, the maximum profit of the integrated station is 4,830 yuan. At this time, the charging cost of EV users is 19,172 yuan, which is the highest. This is because the lower limit of the charging electricity price is too low, and the average charging electricity price remains unchanged, resulting in an increase in the charging electricity price during the charging period of EV users, thereby increasing the charging cost of EV users. The integrated station can purchase and store electricity during the period when the day-ahead market electricity price is low and sell it when the charging electricity price is high. Therefore, the profit of the integrated station is the largest at this time. When the lower limit of the charging electricity price is between 0.5 and 0.8 times the day-ahead market electricity price, the profit of the integrated station and the charging cost of EV users will slowly decrease as the lower limit increases. When the lower limit of the charging electricity price exceeds 0.8 times the day-ahead market electricity price, the profit of the integrated station and the charging cost of EV users will decrease rapidly. This is because the increase in the lower limit will increase the price during the EV charging period, but due to the limitation of the average price, it will lead to a decrease in the price during multiple EV charging periods, resulting in a decrease in the profit of the charging station and a reduction in the charging cost of EV users. Therefore, in order to achieve a win-win situation for the charging station and EV users, the lower limit should comprehensively consider external factors.

[0258] Let S max Changes from 4,000 kW·h to 20,000 kW·h, with other parameters remaining unchanged. The change in the profit of the charging station is as Figure 8 shown.

[0259] From Figure 8 it can be seen that when S max changes from 4,000 kW·h to 10,000 kW·h, the profit of the integrated station increases rapidly from 4,830 yuan to 7,302 yuan. However, when the value of S max exceeds 10,000 kW·h, the profit of the integrated station increases slowly and no longer changes after the profit of the power station reaches 7,375 yuan. This is because the larger the capacity of the energy storage device, the more electricity the integrated station can purchase and store at a lower electricity price, saving the purchase cost of electric energy. In addition, as shown in Table 4, the integrated station sells a total of 2,500 kW·h of electricity to the real-time market during the periods of 11:00 - 12:00 and 19:00, further increasing its own profit. However, the charging demand of EV users will not change significantly. Blindly increasing the capacity of the energy storage device of the integrated station will also increase the construction cost of the integrated station.

[0260] To relieve the electricity consumption pressure during the peak period of the power grid, combined with the EV charging electricity price response model, reasonably guide EVs to optimize off-peak charging. In this process, EV users will also save charging costs. From Figure 5 it can be seen that there are a total of 2 peak periods in a day. The first peak period appears in the 08:00 - 09:00 period; the second peak period appears in the 18:00 - 19:00 period. Therefore, these two periods are used as the EV charging regulation periods. The EV charging situation is shown in Table 6.

[0261] Table 6 EV Charging Situation Statistics

[0262]

[0263] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A master-slave game scheduling method considering the interaction between a charging-discharging-storage integrated station and electric vehicles, characterized in that, The steps are as follows: Step 1: Establish a charging, discharging, and energy storage integrated station model, and perform segmented settings for multiple scenarios of the integrated station. Step 2: Establish a dynamic road network model and combine the EV travel characteristics to predict the spatio-temporal distribution of EV charging load under the constraints of the urban regional road network. Step 3: Establish a multi-objective master-slave game optimization scheduling model for EVs and the charging, discharging, and energy storage integrated station, and conduct multi-objective coordination for the benefits of EV users and the integrated station. Step 4: Use the KKT conditions and duality theory to transform the multi-objective master-slave game optimization scheduling model for EVs and the integrated station into a mixed-integer linear programming model, so as to obtain the optimal solution.

2. The master-slave game scheduling method considering the interaction between the charging / discharging / storage integrated station and the electric vehicle according to claim 1, wherein, The multiple scenarios of the charging, discharging, and energy storage integrated station include three behaviors, namely, the energy storage behavior of purchasing electricity in the day-ahead market, the discharging behavior of selling electricity to the grid side, and the charging behavior of EV charging within the integrated station. The setting of multiple scenarios of the charging, discharging, and energy storage integrated station is divided into two stages: 1) When the distribution network system operates normally or at low load, the integrated station gives priority to the charging behavior; when the EV charging demand is met, its energy storage behavior is considered. At this time, the integrated station shows load characteristics. 2) When the distribution network system reaches the peak load operation state, the integrated station gives priority to the discharging behavior; the integrated station sells electricity to the grid side. At this time, the integrated station shows the characteristics of a distributed power source.

3. The master-slave game scheduling method considering the interaction between the charging / discharging / storage integrated station and electric vehicles according to claim 1, wherein The method for establishing the charging, discharging, and energy storage integrated station model is as follows: The charging and discharging adjustment rate ψ(i0) of the integrated station is: wherein, when i0 = 1, it represents the charging and discharging status of the charging system of the EV charging, discharging and energy storage integrated station, and when i0 = -1, it represents the charging and discharging status of the energy storage system of the EV charging, discharging and energy storage integrated station; P I (i0) is the rated charging power of the charging system and the energy storage system; P Ie (i0) is the actual discharge power of the charging system and the energy storage system; when ψ(i0) > 0, it indicates that the EV charging, discharging, and energy storage integrated station is in the charging state, and when ψ(i0) < 0, it indicates that the EV charging, discharging, and energy storage integrated station is in the discharging state; The energy storage system provides electrical energy support for the charging system through a Boost boost circuit, and its operating power is P B (t), P B (t) satisfies the relationship with the remaining power of the charging system and the remaining power of the energy storage system as follows: Where Q C (t) is the remaining power of the charging system; Q F (t) is the remaining power of the energy storage system; is the charging power of the nth EV in the t time period; is the actual charging power of the EV; P C (t) is the power of the charging system at time t; P F (t) is the power of the energy storage system at time t; Due to the limitations of the energy storage device and the converter device of the EV charging, discharging, and energy storage integrated station, the energy storage system and the charging system are restricted by boundary conditions, and their power constraints are as follows: P C,Fmin ≤P C,F ≤P C,Fmax (3) Where: P C,F is the charge-discharge power of the energy storage system of the integrated charging, discharging and storage station; P C,Fmin is the minimum charge-discharge power of the energy storage system of the integrated charging, discharging and storage station, and P C,Fmax is the maximum charge-discharge power of the energy storage system of the integrated charging, discharging and storage station.

4. The master-slave game scheduling method considering the interaction between the charging / discharging / storage integrated station and electric vehicles according to claim 1, characterized in that, The dynamic road network model includes a dynamic traffic road network model, an urban road resistance model, a distribution network model, and an EV charging price response model. The dynamic traffic road network model is expressed as: Where: G is the set of traffic road networks; V is the set of all nodes in the traffic road network; E is the set of all road segments in the traffic road network; H is the set of divided time, that is, the whole day is divided into T time periods; W is the set of road segment weights, representing the travel cost of vehicles passing through this road segment, which can be quantified by travel time, passing speed and cost; the connection relationship between nodes in the traffic road network is described by matrix D; the element d ij (t) is expressed as: Where: v ij is the connection section between the i-th and j-th nodes; w ij (t) is the weight of section v ij at time period t; The urban road resistance model is expressed as: Where: L ij is the length of road ij; R i is the waiting time for traffic lights at intersection i; v ij (t) represents the driving speed of the EV at time t; v ij,max is the zero-flow speed of road ij; C ij is the traffic capacity of road ij; Q ij (t) is the traffic flow of section ij of road at time t; ω is the road traffic level; a, b, γ are adaptive coefficients for different road levels; The distribution network model is expressed as: Wherein: is the nth distribution network node; G Y is the position of the source node; C Y is the position of the source node capacity; The total EV charging power connected to the nth distribution network node at time t is P n which is Wherein: is the charging power of the nth EV in the t period; N is the total number of EV chargings in the t period; The EV charging price response model is expressed as: where: λ c (t) is the EV demand response degree during the t period; △c s,t (t) is the electricity price difference between the kth moment and the tth moment during the t period; l s,t is the starting threshold, h s,t is the saturation threshold; λ c,max is the maximum EV demand response degree.

5. The master-slave game scheduling method considering the interaction between the charging / discharging / storage integrated station and the electric vehicle according to claim 1, characterized in that The method for predicting the spatio-temporal distribution of EV charging load under the constraints of the urban regional road network is as follows: Obtain the unit mileage △Cap according to the type of each EV, and combine the EV battery capacity and the initial SOC to obtain the initial electricity quantity Cap0; among them, the initial SOC is the battery state when the EV has just been fully charged. Use Monte Carlo sampling to allocate the starting node and the initial departure time t for each EV s , combined with the OD probability matrix corresponding to the initial departure time t s , randomly sample to generate the destination node; Depict the travel trajectory of each EV by repeatedly calling the OD probability matrix of each time period for EVs. When the remaining electricity of the EV is greater than the charging threshold, complete the travel according to the specified route; otherwise, quickly charge nearby to determine the spatio-temporal information of the fast charging load. Repeat the above steps to traverse all EVs and count the charging load of each node in the distribution network.

6. The master-slave game scheduling method considering the interaction between the charging / discharging / storage integrated station and electric vehicles according to claim 5, characterized in that, Calculate the remaining battery capacity Cap of the EV at time t according to the battery capacity consumption △Cap per unit mileage t : Cap t = η(Cap t-1 - Δl·ΔCap) (12) Where: Cap t-1 is the remaining power of the EV at time t-1; η is the energy consumption coefficient, representing the power loss caused by starting and braking during vehicle driving; △l is the driving distance of the vehicle from time t-1 to t.

7. The master-slave game scheduling method considering the interaction between the charging / discharging / storage integrated station and electric vehicles according to claim 5, characterized in that, The OD probability matrix is expressed as: In the formula: represents the number of EVs with the starting point at i and the ending point at j during the period from T to T + 1; represents the probability that the EV stays in place during the period from T to T + 1; represents the sum of the number of EVs from node i to any node during the period from T to T + 1.

8. The master-slave game scheduling method considering the interaction between the charging / discharging / storage integrated station and electric vehicles according to claim 4, wherein The multi-objective master-slave game optimization scheduling model for EVs and the charging, discharging, and energy storage integrated station includes the integrated station objective function and EV integrated station constraint conditions, and the EV objective function and EV constraint conditions. The expression of the integrated station objective function is: Where: N is the number of EVs charged at an integrated station in a day; △t is the duration spent by the vehicle due to road impedance; δ - (t) is the electricity purchase price from the power grid by the integrated station at time t, and δ + (t) is the electricity selling price to the power grid by the integrated station at time t; E - (t) is the electricity purchase quantity from the real-time market by the integrated station at time t, and E + (t) is the electricity selling quantity to the real-time market by the integrated station at time t; δ d (t) is the day-ahead contract electricity price at time t; E(t) is the contract electricity quantity in the day-ahead market at time t; is the charging electricity price of the nth EV at time t; The EV integrated station constraint conditions include: Charging price constraint: Wherein: is the highest electricity price during period t, is the lowest electricity price during period t; is the daily average electricity price; Day-ahead power purchase quantity constraint and real-time power purchase and sale quantity constraint of the integrated station: where: z(t) is a Boolean variable representing the electricity trading status during period t; M is a positive number; is the energy storage discharge of the integrated station during period t; Energy balance constraint: Where: is the charging amount of the integrated station energy storage device during the t period; Charging and discharging constraint of the energy storage device of the integrated station: where: u(t) is a Boolean variable representing the state of the integrated station energy storage facility during period t; is the maximum charging power of the integrated station energy storage device, is the maximum discharging power of the integrated station energy storage device; Energy storage electricity quantity constraint of the integrated station: Where: S(t) is the stored energy of the integrated station energy storage facility during the t period; η + is the charging efficiency of the integrated station energy storage device, η - is the discharging efficiency of the integrated station energy storage device; S max is the maximum capacity of the integrated station energy storage device; S(0) is the initial energy of the integrated station energy storage device; The EV objective function is expressed as: The EV constraint conditions include: EV battery state of charge constraint: Where: ξ is the battery charge level expected by the user; is the battery capacity of the nth EV; is the initial power of the nth EV; μ represents the EV charging efficiency; T a is the EV charging duration; Charging power constraint: Where: is the maximum charging power of the nth vehicle.

9. The master-slave game scheduling method considering the interaction between the charging / discharging / storage integrated station and electric vehicles according to claim 8, characterized in that The method of transforming the multi-objective master-slave game optimization scheduling model of EV and integrated station into a mixed-integer linear programming model by using KKT conditions and duality theory is as follows: Equivalent nonlinear programming transformation of the master-slave game model: For both sides of the game, the price is determined when the EV makes a decision. Replace the linear programming equations (20)-(22) with the corresponding KKT conditions to eliminate this optimization problem; Let the dual variables be α1, ω it , and the KKT conditions corresponding to the linear programming equations (20)-(22) are as follows: Constraints (25) and (26) are complementary slackness conditions. x⊥y means that at most one of the scalars x and y can be strictly greater than 0; Equations (23)-(27) transform the lower-level optimization into constraints; Linearization of complementary slackness conditions: Introduce a Boolean variable and Convert constraints (25) and (26) into linear inequalities: Linearization of the integrated station objective function: The source of the non - linearization of the objective function is the product of the electricity price and the charging power According to the duality theorem of linear programming, the linear programming equations (20)-(22) are transformed into: On the premise of satisfying the KKT conditions, the objective function equation (14) is equivalent to: Equivalent mixed-integer linear programming model of the master-slave game model: The electricity price pricing game of the integrated station can be equivalent to the following mixed-integer linear programming: s.t. equations (15)-(29), equations (23)-(24), equations (27)-(31) In the optimal solution of the mixed-integer linear programming equation (34) constitutes the Stackelberg equilibrium of the game equations (14)-(27), (20)-(22).

10. The master-slave game scheduling method considering the interaction between the charging / discharging / storage integrated station and the electric vehicle according to claim 1, wherein, Combining the EV charging spatio-temporal distribution prediction model and the multi-objective master-slave game optimization scheduling model of electric vehicles and integrated stations, an EV charging and discharging optimization scheduling model is proposed. The specific scheduling steps are as follows: Obtain the spatio-temporal distribution of charging demands at each node according to the prediction results in step two, and obtain its optimized electricity price through the master-slave game; Formulate the charging electricity price based on the optimized electricity price according to the load conditions of each node; Formulate the optimal charging plan according to the optimal electricity price.

Images