Electric vehicle cluster multi-market collaborative transaction decision-making method based on double-layer optimization

Abstract

The invention relates to the technical field of electric power systems, and particularly discloses an electric vehicle cluster multi-market collaborative transaction decision-making method based on double-layer optimization, and the method comprises the steps: integrating dispersed EV charging behaviors into a virtual energy storage unit, and firstly building an energy state and a power boundary of a cluster; secondly, describing uncertainty caused by user behaviors through a probability model; and finally, deducing the auxiliary service capacity which can be provided by the cluster on the basis of the deterministic constraint. The step of constructing a double-layer collaborative optimization market transaction decision model comprises the substeps of constructing an upper-layer EV cluster participation multi-level market joint clearing model by taking minimization of system power consumption cost as a target function; constructing a lower-layer EVA multi-level market bidding decision-making model by taking comprehensive income maximization as an objective function; and solving the double-layer collaborative optimization market transaction decision model to obtain a decision scheme. The method has the advantages that the multi-market bidding decision can be optimized, the adjustable capability of the EV cluster is fully played, and the economic benefit is improved.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a multi-market collaborative trading decision-making method for electric vehicle clusters based on two-layer optimization. Background Technology

[0002] Traditional flexible regulation models centered on thermal power units are no longer adequate for scenarios with a high proportion of renewable energy. Electric vehicles (EVs), as a flexible resource on the large-scale load side, play a crucial role in promoting renewable energy consumption and ensuring grid stability through their clustered regulation capabilities. EV clusters face unique challenges in participating in multi-level electricity markets: on the one hand, the objectives of EV users with decentralized decision-making differ from those of operators with centralized control, requiring coordination between wholesale market transactions and physical dispatch; on the other hand, existing trading strategies do not fully consider the spatiotemporal constraints of regulation capabilities under this model, making it difficult to achieve synergistic optimization across multiple markets such as energy, frequency regulation, and reserves.

[0003] To address these challenges, Reference [1] considers a market bidding decision model based on non-cooperative game among charging stations, transforms it into a Nash equilibrium problem for solution, and proposes a two-stage trading model of day-ahead bidding and real-time bidding. It verifies that this strategy can effectively tap the potential of electric vehicle storage and promote its orderly interaction with the power grid as a flexible resource. Reference [2] constructs a dispatchable capacity model for EV cluster storage based on travel uncertainty, comprehensively considers the carbon trading mechanism to establish a master-slave game framework between distribution network operators and electric vehicle aggregators, and takes into account the real-time deviation adjustment mechanism, thus achieving coordinated optimization of low-carbon operation of the system and win-win interests of multiple stakeholders. Reference [3] uses battery swapping price as a demand-guided method and constructs a two-stage optimization bidding model for battery swapping stations to participate in the electricity and frequency regulation market, thus reducing charging costs and maximizing the frequency regulation service potential of battery swapping stations. Reference [4] constructs a decision-making model for intercity passenger transport companies to participate in the day-ahead market, considers the spatiotemporal dynamic transfer of vehicles between multiple cities, and achieves coordinated optimization of intercity schedule planning, vehicle resource allocation and electricity market bidding behavior. Reference [5] proposes an aggregated electric vehicle charging model based on energy boundary and parameterization, which can search for a "good enough" charging strategy with probabilistic performance guarantee within seconds. Reference [6] constructs a wind-storage joint two-layer bidding optimization model that takes into account battery loss and wind power deviation penalty based on the boundary feature analysis of EV full process behavior, and realizes the balance of interests of participants.

[0004] Overall, the aforementioned studies mostly focus on user-side incentive design in decentralized charging scenarios, with few considerations for aggregating charging stations to participate in market transactions. Furthermore, there are few studies on multi-market collaborative trading strategies for electric vehicle aggregators. Finally, the clearing mechanism for vehicle-grid interaction in the joint market of electric energy-frequency regulation-reserve is imperfect, and existing market rules restrict the full realization of the regulatory potential of EVs as a flexible resource.

[0005] References

[0006] [1] Zhan Xiangpeng, Yang Jun, Han Sining, et al. Two-stage market bidding strategy for charging stations considering the dispatchability potential of electric vehicles [J]. Automation of Electric Power Systems, 2021, 45(10):86-96.

[0007] [2] Zhang Panzhao, Xie Lirong, Ma Ruizhen, et al. Multi-entity two-stage low-carbon optimization operation strategy considering the dispatchability of electric vehicle clusters [J]. Power System Technology, 2022, 46(12):4809-4825.

[0008] [3] Li Xianshan, Zhan Ziao, Li Fei, et al. Bidding strategy for battery swapping stations to participate in the power-frequency regulation market considering demand response of battery swapping [J]. Automation of Electric Power Systems, 2024, 48(08):207-215.

[0009] [4] Lu Zhilin, Shang Nan, Chen Zheng, et al. Two-level optimal scheduling algorithm for intercity passenger transport companies to participate in day-ahead electricity market [J]. Automation of Electric Power Systems, 2022, 46(12):220-231.

[0010] [5] LONG Teng, JIA Qingshan, WANG Gongming, et al. Efficient real-time EV charging scheduling via ordinal optimization[J]. IEEE Transactions onSmart Grid, 2021, 12(5): 4029-4038.

[0011] [6] Zhang Qian, Wu Xiaohan, Deng Xiaosong, et al. Bidding strategy for wind power and Large-scale electric vehicles participating in Day-aheadenergy and frequency regulation market[J]. Applied Energy, 2023, 341: 121063. Summary of the Invention

[0012] The purpose of this invention is to overcome the shortcomings of the prior art and provide a multi-market collaborative transaction decision-making method for electric vehicle clusters based on two-layer optimization.

[0013] The objective of this invention is achieved through the following technical solution: a multi-market collaborative transaction decision-making method for electric vehicle clusters based on two-layer optimization, the method comprising,

[0014] Virtual energy storage modeling and uncertainty handling for electric vehicle clusters;

[0015] Based on the charging and discharging characteristics and energy state model of a single EV, an aggregate model describing the overall behavior of the cluster is established.

[0016] The uncertainty brought about by user connectivity status and travel demand is characterized by a probabilistic model, and the schedulable power boundary and energy state feasible region of the cluster are derived based on this.

[0017] Construct a two-tier market transaction decision-making model with upper and lower layers working collaboratively and interactively;

[0018] The upper-level model aims to minimize the system's electricity costs by constructing a joint clearing model for a multi-level market, which transmits the market clearing price and the winning bid volume to the lower-level model.

[0019] The lower-level EVA multi-level market bidding decision model aims to maximize the overall revenue of electric vehicle aggregators and formulates the optimal bidding decision in the multi-level market. Under the guidance of the market signals given by the upper level, the lower-level model also considers the constraints of the aggregation model and the schedulable power boundary and energy state feasible region of the cluster to optimize its bidding quantity and bidding price in each market.

[0020] Solve the aforementioned two-layer collaborative optimization model;

[0021] Through the interaction between upper-level market clearing and lower-level bidding decisions, a Stackelberg equilibrium solution is gradually approximated, so that the upper-level system cost and the lower-level aggregator revenue reach synergistic optimality. Finally, the optimal multi-market bidding strategy of the lower level and the market clearing result of the upper level are output, which together serve as EV clusters to participate in multi-market collaborative transaction decision-making.

[0022] Specifically, the energy state model for a single EV is as follows:

[0023] ;

[0024] In the formula, , These represent the charging and discharging power of the EV at time t, respectively. The charging / discharging state is a 0-1 variable, indicating whether vehicle i is connected to the charging station at time t; , These are the upper limits of the charging / discharging power of the EV; Let t be the state of charge. , These represent charge / discharge efficiencies, respectively.

[0025] Assuming each EV participates in scheduling at its maximum charging and discharging power, the EV cluster aggregation model is as follows:

[0026] ;

[0027] ;

[0028] ;

[0029] In the formula, , These represent the maximum available charging / discharging power of the EV cluster at time t; The total SOC-weighted power consumption of all connected EVs in time period t; , These are the upper and lower limits of the polymerization energy, respectively. , These represent the upper and lower limits of the permissible SOC for a single EV; Let be the available battery energy capacity of the i-th electric vehicle when participating in the aggregation scheduling. A collection of electric vehicles.

[0030] Specifically, the uncertainty arising from the use of probabilistic models to characterize user connectivity status and travel demand includes:

[0031] Modeling the stochastic behavior of EV:

[0032] ;

[0033] In the formula, =12; =18; This refers to the vehicle's arrival time. , These represent the standard deviations of vehicle arrival distribution during the midday and evening peak hours of the day.

[0034] The distance d of a single trip follows a log-normal distribution :

[0035] ;

[0036] In the formula, It is 3.45; It is 0.51.

[0037] Log-normal distribution of driving mileage Mixed Gaussian distribution of arrival time To simulate the random travel behavior of a large number of electric vehicle users;

[0038] The remaining battery power when the vehicle arrives at the station is related to the user's last trip distance d:

[0039] ;

[0040] The time required to fully charge a vehicle depends on the current SOC and charging power:

[0041] ;

[0042] In the formula, This is the initial SOC after the last full charge. This represents the average energy consumption of the vehicle. The time required for charging; This represents the remaining battery power when the vehicle arrives at the station.

[0043] The above aggregation model is updated in real time according to the following steps:

[0044] Vehicle arrival: At time i, vehicle i joins the set and initialize =1, ;

[0045] Vehicle departure: At a user-defined time, the vehicle departs from the collection. Removed from the middle =0;

[0046] SOC Update: Based on the updated vehicle set and status from the above steps, the EV cluster aggregation model is updated synchronously. and Thus, the total power of the EV cluster at time t can be obtained. Maximum charging and discharging power of the cluster , .

[0047] Specifically, the auxiliary service capacity that the cluster can provide is:

[0048] ;

[0049] In the formula, , , These represent the maximum frequency regulation capacity, frequency regulation range, and reserve capacity that the EV cluster can provide under station-controlled charging mode at time t. Frequency modulation capacity-mileage ratio, , These are the upper limits of EV charging and discharging power, respectively. The total SOC-weighted power consumption of all connected EVs during time period t.

[0050] Specifically, the objective function of the joint clearing model is:

[0051] ;

[0052] ;

[0053] ;

[0054] ;

[0055] ;

[0056] In the formula, , The costs of participating in multi-level markets for thermal power units and EV clusters are respectively considered. For the market cost of electricity; To support service market costs; , These are respectively a collection of thermal power units and an EV cluster; , , , The prices in the market for the electrical energy, frequency regulation capacity, frequency regulation mileage, and reserve capacity of thermal power unit i at time t are respectively: , , , Do not specify the amount of electricity, frequency regulation capacity, frequency regulation mileage, and reserve capacity that thermal power unit i won the bid for at time t; , , , , These are the electricity prices for charging, discharging, frequency regulation capacity, mileage, and reserve capacity declared by EV cluster n at time t; , , , , These represent the charging and discharging power, frequency regulation capacity, mileage, and reserve capacity at time t, respectively, when the market is cleared.

[0057] Specifically, the constraints for the EV cluster to participate in the multi-level market joint clearing model include:

[0058] System power balance and DC power flow constraints:

[0059] ;

[0060] ;

[0061] In the formula, Let be the total load demand of node k at time t; , These are the set of regular units and the EV cluster connected to node k, respectively. The set of adjacent nodes directly connected to node k; , Let be the phase angles of nodes k and m at time t, respectively; For the branch road km of line admittance, For line power flow constraints. For the lines between system nodes; , These represent the electricity volumes won by wind and solar renewable energy units in the electricity market.

[0062] Ancillary service demand constraints:

[0063] ;

[0064] In the formula, , , These represent the frequency modulation capacity, frequency modulation mileage, and reserve capacity requirements at time t, respectively. , , Do not specify the frequency regulation capacity, frequency regulation mileage, and reserve capacity of thermal power unit i at time t; , , These represent the frequency regulation capacity, mileage, and reserve capacity that are cleared from the market at time t;

[0065] Market bidding volume constraints;

[0066] ;

[0067] ;

[0068] ;

[0069] ;

[0070] ;

[0071] In the formula, , These are frequency modulation mileage and capacity multiplier, respectively. , , , These refer to the reported charging and discharging capacity of EV clusters in the electric energy market, and the reported capacity limits for the frequency regulation and standby markets. , , These are the reporting limits for conventional generating units in the electricity, frequency regulation, and standby markets, respectively.

[0072] Renewable energy output constraints:

[0073] ;

[0074] In the formula, , These are the predicted renewable energy output values ​​at time t.

[0075] Specifically, the objective function of the EVA multi-level market bidding decision model is:

[0076] ;

[0077] ;

[0078] ;

[0079] ;

[0080] In the formula, , These represent the revenue of EV clusters in the electric energy and ancillary services markets, respectively. Cost of EV battery degradation; , These represent the charging and discharging prices won by the EV cluster in the electric energy market, respectively. , , These are the frequency regulation capacity, frequency regulation mileage, and reserve capacity electricity prices won by the EV cluster; This represents the total purchase cost of the battery pack. This represents the maximum number of cycles within the battery's lifespan.

[0081] Specifically, the constraints of the EVA multi-level market bidding decision model include:

[0082] EVA bidding capacity constraints:

[0083] ;

[0084] ;

[0085] EVA bid price constraints:

[0086] ;

[0087] In the formula, , These represent the upper limits for EVA's charging and discharging bids in the electric energy market. , , , These represent the upper limit of the price quotes for each market.

[0088] The present invention has the following advantages:

[0089] This invention equates the station-controlled battery swapping mode to a "virtual battery" aggregation model, achieving controllable and adjustable capabilities for traditionally dispersed EV resources. It transforms the uncertainties and randomness of EV loads into reliable vehicle-to-grid (V2G) interaction and regulation resources. The proposed multi-market joint clearing and transaction decision-making collaborative optimization method creates diversified revenue streams for EVA (Electric Vehicle Asset Valuation), including peak shaving and valley filling arbitrage in the energy market, ancillary service frequency regulation revenue, and reserve compensation, fully leveraging the regulation capabilities of EV clusters as a flexible load-side resource. Attached Figure Description

[0090] Figure 1 This is a schematic diagram illustrating the solution of the two-layer collaborative optimization market transaction decision model of the present invention;

[0091] Figure 2 This is a topology diagram of the IEEE 30-node network.

[0092] Figure 3 Predict the output diagram for wind turbine units;

[0093] Figure 4 Predicted output diagram for photovoltaic units;

[0094] Figure 5 Boundary diagram of charging and discharging power for EV clusters in station-controlled battery swapping mode;

[0095] Figure 6 Energy boundary diagram of SOC for EV clusters in station-controlled battery swapping mode;

[0096] Figure 7 To clear the electricity price map in the electricity market;

[0097] Figure 8 A price chart for clearing out the ancillary services market;

[0098] Figure 9 Decision graph for EV cluster 1 participating in multi-market trading;

[0099] Figure 10 To enable EV Cluster 2 to participate in multi-market trading decisions. Detailed Implementation

[0100] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0101] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0102] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0103] The present invention will be further described below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0104] like Figures 1 to 10 As shown, a multi-market collaborative transaction decision-making method for electric vehicle clusters based on two-layer optimization is proposed. This method includes:

[0105] Virtual energy storage modeling and uncertainty handling for electric vehicle clusters;

[0106] Based on the charging and discharging characteristics and energy state model of a single EV, an aggregate model describing the overall behavior of the cluster is established.

[0107] The uncertainty brought about by user connectivity status and travel demand is characterized by a probabilistic model, and the schedulable power boundary and energy state feasible region of the cluster are derived based on this.

[0108] The energy state model for a single EV is as follows:

[0109] ;

[0110] In the formula, , These represent the charging and discharging power of the EV at time t, respectively. The charging / discharging state is a 0-1 variable, indicating whether vehicle i is connected to the charging station at time t; , These are the upper limits of the charging / discharging power of the EV; Let t be the state of charge. , These represent charge / discharge efficiencies, respectively.

[0111] Assuming each EV participates in scheduling at its maximum charging and discharging power, the EV cluster aggregation model is as follows:

[0112] ;

[0113] ;

[0114] ;

[0115] In the formula, , These represent the maximum available charging / discharging power of the EV cluster at time t; The total SOC-weighted power consumption of all connected EVs in time period t; , These are the upper and lower limits of the polymerization energy, respectively. , These represent the upper and lower limits of the permissible SOC for a single EV; Let be the available battery energy capacity of the i-th electric vehicle when participating in the aggregation scheduling. A collection of electric vehicles.

[0116] Specifically, the uncertainty arising from the use of probabilistic models to characterize user connectivity status and travel demand includes:

[0117] Modeling the stochastic behavior of EV:

[0118] ;

[0119] In the formula, =12; =18; This refers to the vehicle's arrival time. , These represent the standard deviations of vehicle arrival distribution during the midday and evening peak hours of the day.

[0120] The distance d of a single trip follows a log-normal distribution :

[0121] ;

[0122] In the formula, It is 3.45; It is 0.51.

[0123] Based on a probability model (log-normal distribution of driving mileage) Mixed Gaussian distribution of arrival time This was used to simulate the random travel behavior of a large number of electric vehicle users. These behaviors determined the initial state of each electric vehicle when it connected to a charging station, i.e., its remaining battery power upon arrival. By charging time This determines the "dwell time" of a vehicle when it is connected to a charging station in the power grid. It is directly derived from the SOC equation in the energy state model of a single EV; The size of the cluster is determined. and connection status .

[0124] The remaining battery power when the vehicle arrives at the station is related to the user's last trip distance d:

[0125] ;

[0126] The time required to fully charge a vehicle depends on the current SOC and charging power:

[0127] ;

[0128] In the formula, This is the initial SOC after the last full charge. This represents the average energy consumption of the vehicle. The time required for charging.

[0129] The above aggregation model is updated in real time according to the following steps:

[0130] Vehicle arrival: At time i, vehicle i joins the set and initialize =1, ;

[0131] Vehicle departure: At a user-defined time, the vehicle departs from the collection. Removed from the middle =0;

[0132] SOC Update: Based on the updated vehicle set and status from the above steps, the EV cluster aggregation model is updated synchronously. and Thus, the total power of the EV cluster at time t can be obtained. Maximum charging and discharging power of the cluster , .

[0133] The adjustable capacity of the EV cluster participating in the frequency regulation and backup ancillary service markets determines the actual bidding ceiling of EVA at time t. The schedulable power boundary and energy state feasible region of the cluster are:

[0134] ;

[0135] In the formula, , , These represent the maximum frequency regulation capacity, frequency regulation range, and reserve capacity that the EV cluster can provide under station-controlled charging mode at time t. Frequency modulation capacity-mileage ratio, , These are the upper limits of EV charging and discharging power, respectively.

[0136] A two-layer collaborative optimization market trading decision-making model is constructed, which includes:

[0137] The upper-level model aims to minimize the system's electricity cost and constructs a multi-level market joint clearing model, transmitting the market clearing price and winning bid volume to the lower-level model. The EV cluster participates in the multi-level market joint clearing model, which is led by the distribution system operator (DSO). Based on market participant declarations, wind and solar power output, and load forecasts, it conducts day-ahead market pre-clearing, with the objective function of minimizing the system's electricity cost. It jointly clears the power market and ancillary services market. The DSO must strictly adhere to network security constraints, including system power balance, transmission line power flow restrictions, and ancillary services market demand. After clearing, the trading center releases a unified clearing price and winning bid volume as market signals to the lower-level EVA.

[0138] The objective function of the joint clearing model is:

[0139] ;

[0140] ;

[0141] ;

[0142] ;

[0143] ;

[0144] In the formula, , The costs of participating in multi-level markets for thermal power units and EV clusters are respectively considered. For the market cost of electricity; To support service market costs; , These are respectively a collection of thermal power units and an EV cluster; , , , The prices in the market for the electrical energy, frequency regulation capacity, frequency regulation mileage, and reserve capacity of thermal power unit i at time t are respectively: , , , Do not specify the amount of electricity, frequency regulation capacity, frequency regulation mileage, and reserve capacity that thermal power unit i won the bid for at time t; , , , , These are the electricity prices for charging, discharging, frequency regulation capacity, mileage, and reserve capacity declared by EV cluster n at time t; , , , , These represent the charging and discharging power, frequency regulation capacity, mileage, and reserve capacity at time t, respectively, when the market is cleared.

[0145] The constraints for the EV cluster to participate in the multi-level market joint clearing model include:

[0146] System power balance and DC power flow constraints are derived from Kirchhoff's laws, the most basic laws of power grid operation. They require that the generation and consumption of electricity at each node (including EV charging and discharging and renewable energy injection) must be balanced in real time. In addition, the power flow of each line must not exceed the limit.

[0147] ;

[0148] ;

[0149] In the formula, Let be the total load demand of node k at time t; , These are the set of regular units and the EV cluster connected to node k, respectively. The set of adjacent nodes directly connected to node k; , Let be the phase angles of nodes k and m at time t, respectively; For the branch road km of line admittance, For line power flow constraints. For the lines between system nodes; , These represent the electricity volumes won by wind and solar renewable energy units in the electricity market.

[0150] Ancillary service demand constraints: The power grid dispatch center determines the total frequency regulation / reserve demand based on predicted load fluctuations and potential faults, requiring conventional units and flexible resources such as EV clusters to provide the demand together in order to maintain system frequency and voltage stability;

[0151] ;

[0152] In the formula, , , These represent the frequency modulation capacity, frequency modulation mileage, and reserve capacity requirements at time t, respectively. , , Do not specify the frequency regulation capacity, frequency regulation mileage, and reserve capacity of thermal power unit i at time t; , , These represent the frequency regulation capacity, mileage, and reserve capacity that are cleared from the market at time t;

[0153] Market bidding volume constraints;

[0154] ;

[0155] ;

[0156] ;

[0157] ;

[0158] ;

[0159] In the formula, , These are frequency modulation mileage and capacity multiplier, respectively. , , , These refer to the charging and discharging capacity declared by EV clusters in the electric energy market, as well as the upper limit of declared capacity in the frequency regulation and standby markets. , , These are the reporting limits for conventional generating units in the electricity, frequency regulation, and standby markets, respectively.

[0160] Renewable energy output constraints:

[0161] ;

[0162] In the formula, , These are the predicted renewable energy output values ​​at time t.

[0163] The lower-level EVA multi-level market bidding decision model aims to maximize the overall revenue of electric vehicle aggregators and formulates the optimal bidding decisions in the multi-level market. Guided by market signals from the upper level, the lower-level model simultaneously considers the constraints of the aggregation model and the schedulable power boundary and energy state feasible region of the cluster to optimize its bidding quantity and bidding price in each market. The EVA multi-level market bidding decision model is led by EVA. Its core process is to optimize the bidding and power allocation strategies based on the market clearing signals issued by the trading center, with the objective function of maximizing overall revenue. It needs to take into account the physical characteristics and operating costs of the EV cluster.

[0164] The objective function of the EVA multi-level market bidding decision model is:

[0165] ;

[0166] ;

[0167] ;

[0168] ;

[0169] In the formula, , These represent the revenue of EV clusters in the electric energy and ancillary services markets, respectively. Cost of EV battery degradation; , These represent the charging and discharging prices won by the EV cluster in the electric energy market, respectively. , , These are the frequency regulation capacity, frequency regulation mileage, and reserve capacity electricity prices won by the EV cluster; This represents the total purchase cost of the battery pack. This represents the maximum number of cycles within the battery's lifespan.

[0170] The constraints of the EVA multi-level market bidding decision model include:

[0171] EVA bidding capacity constraints:

[0172] ;

[0173] ;

[0174] EVA bid price constraints:

[0175] ;

[0176] In the formula, , These represent the upper limits for EVA's charging and discharging bids in the electric energy market. , , , These represent the upper limit of the price quotes for each market.

[0177] The EVA is required to declare capacity and price to each market (electric energy, frequency regulation capacity, frequency regulation mileage, and backup) within the bidding limit range stipulated by the trading center; the physical operating limits of the EV cluster are as follows: the total charging and discharging power of the cluster cannot exceed the sum of the rated power of all available vehicles, and the total energy state of the cluster must be maintained within the safe state of charge range after battery aggregation.

[0178] The decision scheme is obtained by solving the two-layer collaborative optimization market transaction decision model.

[0179] Case analysis,

[0180] Based on actual data from a certain region, two EV clusters in that region were selected as research objects. The cluster access nodes and key technologies are shown in Table 1.

[0181] Table 1

[0182] Parameters\Cluster EV Cluster 1 EV Cluster 2 Node 18 24 Capacity-to-mileage ratio 15 13 Maximum power output per EV (kW) 40 40 Maximum capacity of a single EV (kWh) 60 50

[0183] The day is divided into 24 scheduling periods. The case study first conducts a multi-market joint clearing to obtain the time-of-use electricity price and the ancillary service market clearing price.

[0184] Within this framework, we evaluate market participation strategies and EVA economic benefits under different regulatory models. On the load side, we primarily consider the bidirectional charging and discharging response of EV users, assuming that cluster 1 and cluster 2 have sizes of 400 and 300 vehicles respectively, totaling 700 electric vehicles.

[0185] To demonstrate the innovation and superiority of the control strategy of this invention, the following three sets of comparative schemes are set up:

[0186] Option 1 (Invention): Adopt the station-controlled battery swapping mode proposed in this invention to participate in the electricity market and ancillary services market in a coordinated manner.

[0187] Option 2 (Traditional Decentralized Control): Adopt a decentralized charging mode, with user participation.

[0188] Option 3 (Single Market Aggregation): Adopts the station-controlled battery swapping mode, but only participates in the bidding for the electricity market.

[0189] The results of the comparison of benefits of each scheme were obtained through simulation calculations, as shown in Table 2.

[0190] Table 2

[0191] plan Electricity market revenue (RMB) Ancillary services market revenue (RMB) 1 8552.54 3542.11 2 -1317.87 1723.24 3 9788.19 /

[0192] It can be seen that Scheme 1 (the present invention) has the best total revenue, with a revenue of 3542.11 yuan in the ancillary services market; Scheme 2 (decentralized control) incurs a net loss (-1317.87 yuan) due to its inability to respond to electricity price signals; Scheme 3, although having the highest revenue in the electricity market (¥9788.19), has a lower total revenue than the scheme of the present invention because it does not participate in the ancillary services market. The results show that the collaborative optimization strategy proposed in this invention can maximize the flexibility value of EV clusters and achieve maximum revenue.

[0193] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any person skilled in the art can make many possible variations and modifications to the technical solution of the present invention, or modify it into equivalent embodiments, without departing from the scope of the present invention. Therefore, any modifications, equivalent changes, and alterations made to the above embodiments based on the technology of the present invention without departing from the scope of the present invention are within the protection scope of the present invention.

Claims

1. A multi-market collaborative transaction decision-making method for electric vehicle clusters based on two-layer optimization, characterized by: The method includes, Virtual energy storage modeling and uncertainty handling for electric vehicle clusters; Based on the charging and discharging characteristics and energy state model of a single EV, an EV cluster aggregation model describing the overall behavior of the cluster is established. The uncertainty brought about by user connectivity status and travel demand is characterized by a probabilistic model, and the schedulable power boundary and energy state feasible region of the cluster are derived based on this. Construct a two-tier market transaction decision-making model with upper and lower layers working collaboratively and interactively; The upper-level model aims to minimize the system's electricity costs by constructing a joint clearing model for a multi-level market, which transmits the market clearing price and the winning bid volume to the lower-level model. The lower-level EVA multi-level market bidding decision model aims to maximize the overall revenue of electric vehicle aggregators and formulates the optimal bidding decision in the multi-level market. Under the guidance of the market signals given by the upper level, the lower-level model also considers the constraints of the aggregation model and the schedulable power boundary and energy state feasible region of the cluster to optimize its bidding quantity and bidding price in each market. Solve the aforementioned two-layer collaborative optimization model; Through the interaction between upper-level market clearing and lower-level bidding decisions, a Stackelberg equilibrium solution is gradually approximated, so that the upper-level system cost and the lower-level aggregator revenue reach synergistic optimality. Finally, the optimal multi-market bidding strategy of the lower level and the market clearing result of the upper level are output, which together serve as EV clusters to participate in multi-market collaborative transaction decision-making.

2. The electric vehicle cluster multi-market collaborative transaction decision-making method based on two-layer optimization according to claim 1, characterized in that: The energy state model for a single EV is as follows: ； In the formula, , These represent the charging and discharging power of the EV at time t, respectively. The charging / discharging state is a 0-1 variable, indicating whether vehicle i is connected to the charging station at time t; , These are the upper limits of the charging / discharging power of the EV; Let t be the state of charge. , These represent charge / discharge efficiencies, respectively. Assuming each EV participates in scheduling at its maximum charging and discharging power, the EV cluster aggregation model is as follows: ； ； ； In the formula, , These represent the maximum available charging / discharging power of the EV cluster at time t; The total SOC-weighted power consumption of all connected EVs in time period t; , These are the upper and lower limits of the polymerization energy, respectively. , These represent the upper and lower limits of the permissible SOC for a single EV; Let be the available battery energy capacity of the i-th electric vehicle when participating in the aggregation scheduling. A collection of electric vehicles.

3. The electric vehicle cluster multi-market collaborative transaction decision-making method based on two-layer optimization according to claim 2, characterized in that: The uncertainty arising from the use of probabilistic models to characterize user connectivity status and travel demand includes: Modeling the stochastic behavior of EV: ； In the formula, =12; =18; This refers to the vehicle's arrival time. , These represent the standard deviations of vehicle arrival distribution during the midday and evening peak hours within a single day. The distance d of a single trip follows a log-normal distribution : ； In the formula, It is 3.45; It is 0.51; Log-normal distribution of driving mileage Mixed Gaussian distribution of arrival time To simulate the random travel behavior of a large number of electric vehicle users.

4. The electric vehicle cluster multi-market collaborative transaction decision-making method based on two-layer optimization according to claim 3, characterized in that: The remaining battery power when the vehicle arrives at the station is related to the user's last trip distance d: ； The time required to fully charge a vehicle depends on the current SOC and charging power: ； In the formula, This is the initial SOC after the last full charge. This represents the average energy consumption of the vehicle. The time required for charging; This represents the remaining battery power when the vehicle arrives at the station. The EV cluster aggregation model is updated in real time according to the following steps: Vehicle arrival: At time i, vehicle i joins the set and initialize =1, ; Vehicle departure: At a user-defined time, the vehicle departs from the collection. Removed from the middle =0; SOC Update: Based on the updated vehicle set and status from the above steps, the EV cluster aggregation model is updated synchronously. and Thus, the total power of the EV cluster at time t can be obtained. Maximum charging and discharging power of the cluster , .

5. The electric vehicle cluster multi-market collaborative transaction decision-making method based on two-layer optimization according to claim 3, characterized in that: The schedulable power boundary and energy state feasible region of the cluster are: ； In the formula, , , These represent the maximum frequency regulation capacity, frequency regulation range, and reserve capacity that the EV cluster can provide under station-controlled charging mode at time t. Frequency modulation capacity-mileage ratio; , These are the upper limits of EV charging and discharging power, respectively. The total SOC-weighted power consumption of all connected EVs during time period t.

6. The electric vehicle cluster multi-market collaborative transaction decision-making method based on two-layer optimization according to claim 1, characterized in that: The objective function of the joint clearing model is: ； ； ； ； ； In the formula, , The costs of participating in multi-level markets for thermal power units and EV clusters are respectively considered. For the market cost of electricity; To support service market costs; , These are respectively a collection of thermal power units and an EV cluster; , , , The prices in the market for the electrical energy, frequency regulation capacity, frequency regulation mileage, and reserve capacity of thermal power unit i at time t are respectively: , , , Do not specify the amount of electricity, frequency regulation capacity, frequency regulation mileage, and reserve capacity that thermal power unit i won the bid for at time t; , , , , These are the electricity prices for charging, discharging, frequency regulation capacity, mileage, and reserve capacity declared by EV cluster n at time t; , , , , These represent the charging and discharging power, frequency regulation capacity, mileage, and reserve capacity at time t, respectively, when the market is cleared.

7. The electric vehicle cluster multi-market collaborative transaction decision-making method based on two-layer optimization according to claim 6, characterized in that: The constraints for the EV cluster to participate in the multi-level market joint clearing model include: System power balance and DC power flow constraints: ； ； In the formula, Let be the total load demand of node k at time t; , These are the set of regular units and the EV cluster connected to node k, respectively. The set of adjacent nodes directly connected to node k; , Let be the phase angles of nodes k and m at time t, respectively; The line admittance for the branch road is km; For line power flow constraints. For the lines between system nodes; , These represent the electricity volumes won by wind and solar renewable energy units in the electricity market. Ancillary service demand constraints: ； In the formula, , , These represent the frequency modulation capacity, frequency modulation mileage, and reserve capacity requirements at time t, respectively. , , Do not specify the frequency regulation capacity, frequency regulation mileage, and reserve capacity of thermal power unit i at time t; , , These represent the frequency regulation capacity, mileage, and reserve capacity that are cleared from the market at time t; Market bidding volume constraints; ； ； ； ； ； In the formula, , These are frequency modulation mileage and capacity multiplier, respectively. , , , These refer to the charging and discharging capacity declared by EV clusters in the electric energy market, as well as the upper limit of declared capacity in the frequency regulation and standby markets. , , These are the reporting limits for conventional generating units in the electricity, frequency regulation, and standby markets, respectively. Renewable energy output constraints: ； In the formula, , These are the predicted renewable energy output values ​​at time t.

8. The electric vehicle cluster multi-market collaborative transaction decision-making method based on two-layer optimization according to claim 6, characterized in that: The objective function of the EVA multi-level market bidding decision model is: ； ； ； ； In the formula, , These represent the revenue of EV clusters in the electric energy and ancillary services markets, respectively. Cost of EV battery degradation; , These represent the charging and discharging prices won by the EV cluster in the electric energy market, respectively. , , These are the frequency regulation capacity, frequency regulation mileage, and reserve capacity electricity prices won by the EV cluster; This represents the total purchase cost of the battery pack. This represents the maximum number of cycles within the battery's lifespan.

9. The electric vehicle cluster multi-market collaborative transaction decision-making method based on two-layer optimization according to claim 8, characterized in that: The constraints of the EVA multi-level market bidding decision model include: EVA bidding capacity constraints: ； ； EVA bid price constraints: ； In the formula, , These represent the upper limits for EVA's charging and discharging bids in the electric energy market. , , , These represent the upper limit of the price quotes for each market.

Images