A Decision Method for Community Reserve Service Considering the Dual Uncertainties of Electric Vehicle Clusters

Abstract

The present invention discloses a decision-making method for community reserve service considering double uncertainties of an electric vehicle group. The method includes: based on the proposed reserve service architecture, considering factors such as response willingness, operation characteristics, and travel constraints, adopting a combination of basic and call reserve incentives, designing a monthly reserve contract mechanism for the electric vehicle group to establish an uncertainty model of the EV group's grid-connected power; establishing an uncertainty model of the EV group's response rate; based on the monthly reserve contract mechanism for the electric vehicle group, the uncertainty model of the EV group's grid-connected power, and the established uncertainty model of the EV group's response rate, constructing a community reserve service decision-making model with the goal of maximizing revenue; combining the genetic algorithm and the mixed integer programming method to solve the community reserve service decision-making model, and obtaining the optimal declared reserve capacity of the community, the incentive price of the electric vehicle group, and the discharge strategies of the electric vehicle group and energy storage.

Description

Technical Field

[0001] The present invention relates to a decision-making method for community reserve services considering the double uncertainties of electric vehicle groups, belonging to the technical field of power system reserve services. Background Art

[0002] Affected by the characteristics of new energy such as volatility and randomness, the future new power system will face a series of challenges such as power angle voltage stability and complex regulation, and put forward more, faster and more accurate requirements for power auxiliary services. As a public good, the contribution of power auxiliary services is benefited by all links of the power system, and the costs caused by it should also be jointly borne by various market players in the power market. Encouraging third parties such as energy storage facilities and electric vehicles to participate in the auxiliary service market and requiring "introducing more types of market players to improve the competitiveness and activity of market transactions" will further enrich the types of power market players and enhance the flexibility of the auxiliary service market, and deeply promote the rational and optimal allocation of resources.

[0003] At present, communities mainly composed of residential and commercial buildings have become the basic energy-consuming units in cities. They have a large number of flexible resources such as electric vehicles (EVs), electrical energy storage (EES), and smart homes. They are characterized by centralized geographical location, regular demand periods, large individual demand differences, and strong partial load randomness. They are natural subjects to participate in grid reserve services. On the other hand, flexible resources represented by EVs have characteristics such as diverse subjects, inconsistent credit, small single capacity, and random behavior. There is a large uncertainty in both the adjustable time and capacity, the aggregation ability is difficult to accurately evaluate, and the quality of reserve services cannot be guaranteed. Therefore, it is urgent to carry out research on a decision-making method for community reserve services considering the double uncertainties of electric vehicle groups. Summary of the Invention

[0004] The purpose of the present invention is to provide a decision-making method for community reserve services considering the double uncertainties of electric vehicle groups, so as to solve the defects of existing technologies that flexible resources represented by EVs have characteristics such as diverse subjects, inconsistent credit, small single capacity, and random behavior, with large uncertainties in both the adjustable time and capacity, difficult to accurately evaluate the aggregation ability, and unable to guarantee the quality of reserve services.

[0005] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: In the diversified development of the power auxiliary service market, the proposed method can be generally applied to communities with a large number of flexible resources such as electric vehicles and electrical energy storage, giving full play to the ability and advantages of communities in aggregating and managing dispersed and idle flexible resources, being able to further enrich the types of power market players and enhance the flexibility of the auxiliary service market, deeply promoting the rational and optimal allocation of resources, and at the same time helping to establish a normalized trading system for communities to participate in reserve services. Brief Description of the Drawings

[0006] Figure 1 It is the overall flowchart of a decision-making method for community reserve service considering the double uncertainties of an electric vehicle group according to an embodiment of the present invention;

[0007] Figure 2 It is the SOC data of the EV group at 12:00 under ten scenarios in a decision-making method for community reserve service considering the double uncertainties of an electric vehicle group according to an embodiment of the present invention;

[0008] Figure 3 It is the response rate and output of the EV group under ten scenarios in a decision-making method for community reserve service considering the double uncertainties of an electric vehicle group according to an embodiment of the present invention. Detailed implementation manners

[0009] To make the technical means, creative features, achieved purposes and functions of the present invention easy to understand, the present invention will be further described below in conjunction with specific implementation manners.

[0010] As Figures 1 - 3 shown, the purpose of the present invention is to provide a decision-making method for community reserve service considering the double uncertainties of an electric vehicle group, which can be generally applied to communities mainly composed of residential and commercial buildings. The present invention establishes a decision-making model for community reserve service with the goal of maximizing benefits, gives full play to the ability and advantages of the community in aggregating and managing dispersed and idle flexible resources, can further enrich the types of electricity market players and enhance the flexibility of the ancillary service market, deeply promote the rational and optimal allocation of resources, and at the same time help to establish a normalized trading system for the community to participate in reserve services.

[0011] Embodiment 1:

[0012] In this embodiment, a decision-making on the reserve service of flexible resources in a community considering the double uncertainties of electric vehicles by applying the present invention is described. Let the reserve market period be 30 days, the reserve time period be 12:00 - 14:00, the reserve call rate σ be 30%, and the simulation time period be 15 min. Assume that the community has 1000 EVs (parameters: capacity of 64 kWh, maximum charge and discharge power of 10 kW) and 1000 kWh of storage batteries. Considering the limited ability of EVs to continuously provide reserve, for the sake of generality, the single continuous operation time of EVs is set to 1 h, and various technical and economic parameters are shown in Table 1.

[0013] Table 1

[0014]

[0015]

[0016] The specific method includes the following six steps:

[0017] Step 1: Propose a community backup service architecture and assumptions with electric vehicle (EV) group discharging as the main part and energy storage as the auxiliary part.

[0018] As an intermediary between the backup market and user resources, the community integrates the EV groups and energy storage scattered on the user side, interacts with the backup market for demand, conducts market bidding and calls for services, and obtains corresponding backup revenues. Downward, it signs backup contracts with user resources and aggregates and manages flexible resources through an incentive mechanism to support the upward backup services.

[0019] When the community declares the backup capacity, if it is too high, it will face the risk of backup penalties; if it is too low, it will bring risks of redundant flexible resources or excessive backup incentives. Therefore, the community should comprehensively consider a series of factors such as resource requirements and constraint characteristics, and achieve economic optimality by reasonably optimizing the upward declared capacity and the downward incentive price. The following assumptions are made for the community's participation in backup services:

[0020] 1) The community's flexible resources mainly consist of uncertain EV groups and deterministic energy storage. Only considering the community's participation in the upward backup service, the backup capacity is declared monthly.

[0021] 2) The community participates in the backup market as an ordinary entity, and clears and settles according to the backup market rules. Considering the timeliness and enforceability of the contract, the contract period between the community and flexible resources is monthly.

[0022] 3) Considering that the community's backup capacity is limited and has little impact on the clearing result of the backup market, the community is regarded as a price taker to participate in the backup market and adopts the form of reporting quantity without quoting price.

[0023] 4) Assume that the parameters of the EV groups are the same. During the provision of backup services, the control rights are transferred to the community and uniformly called by the community. The aggregation risks caused by uncertain factors such as the online rate, grid-connected power, and response rate of the EV groups are borne by the community.

[0024] 5) When settling the electricity bill, the charging costs of the EV groups and energy storage are calculated according to the difference between their grid-connected power and off-grid power, that is, the EV groups and energy storage will not generate new electricity bill expenditures due to providing backup services, and the charging costs caused by discharging are borne by the community.

[0025] Step 2: Based on the proposed backup service architecture, considering various factors such as response willingness, operation characteristics, and travel constraints, a monthly backup contract mechanism for EV groups that takes into account time, physical, and economic requirements is designed by combining basic and call backup incentives.

[0026] The monthly reserve contract mechanism for an electric vehicle group includes three types of constraints: time, physical, and economic. Time constraints mainly refer to contract cycle and reserve period constraints; physical constraints mainly refer to connection time / disconnection time, connection power / disconnection expected power, EV operation constraints, and EV call constraints; economic constraints mainly include call reserve incentives and basic reserve incentive constraints, as follows:

[0027] 1) Contract cycle M and reserve period [ t s ,t e ]

[0028] Considering the timeliness and enforceability of the contract, let the contract cycle between the community and flexible resources be monthly, and the number of contract days be M (m ∈ M); the reserve period refers to the start time t s to the end time t e .

[0029] 2) Connection time / disconnection time

[0030] To ensure that the k-th (k ∈ K) EV has the call ability during the reserve period, its and are constrained, as shown in Equation (1).

[0031]

[0032] In the formula: Δt gap is the reserve buffer time set for the community to ensure that the EV is online and callable during the reserve period, min.

[0033] 3) Connection power / disconnection expected power

[0034] When an EV participates in reserve, it still needs to meet the travel needs of users, including: ① The power of the k-th EV at the disconnection moment needs to be greater than or equal to the disconnection expected power ; ② The power of the k-th EV needs to be always greater than or equal to the guaranteed power to ensure the indefinite vehicle use needs of users; ③ To ensure the callability of the EV, the connection power of the k-th EV needs to be greater than a certain power threshold, as shown in Equation (2).

[0035]

[0036] In the formula: is the minimum connection power threshold of the EV, kWh.

[0037] 4) EV operation constraints

[0038] The EV should satisfy multiple constraint conditions such as charging and discharging power, storage capacity, etc., as shown in Equation (3).

[0039]

[0040] In the formula: and are the charging and discharging power of the k-th EV at time t, in kW; and are the maximum charging and discharging power of the EV, in kW; σ EV is the self-discharge rate of the EV, in %; η EV,c and η EV,d are the charging efficiency and discharging efficiency of the EV, in kW; and are the charging and discharging state variables of the k-th EV at time t, which are 0-1 variables; Δt is the simulation duration, in h; and are the minimum and maximum storage capacities of the EV, in kWh.

[0041] 5) EV call constraints

[0042] To avoid excessive and frequent calls of EVs, the call behavior of EVs within the contract period is regulated, as shown in Equation (4):

[0043]

[0044] In the formula: N d,k is the call status of the k-th EV on the m-th day, 1 means being called, 0 means not being called; is the maximum allowable call times of the k-th EV within the contract period; is the maximum allowable call duration of the k-th EV during the standby period, in h; E d,max is the maximum daily discharge amount allowed for the EV, in kWh. In Equation (7), the first formula is the cumulative call times constraint of the EV; the second formula is the daily call duration constraint of the EV; the third formula means that the EV can be called at most once during the standby period; the fourth formula is the maximum daily discharge amount constraint of the EV.

[0045] 6) EV standby incentive R EV

[0046] Analogous to the grid-side standby market [4] , the EV standby incentive R EV consists of the basic standby incentive R EV,base and the call standby incentive R EV,act .

[0047]

[0048] In the formula: refers to the basic incentive cost given by the community to encourage the k-th EV to participate in reserve. It is calculated on a daily basis and is independent of whether the k-th EV is called, as shown in Equation (5).

[0049]

[0050] In the formula: γ base is the basic incentive price, yuan / day; D k is the number of reserve days provided for the k-th EV, days.

[0051] Incentive for the k-th EV to be called for reserve is the incentive cost given according to its actual call degree, as shown in Equation (7).

[0052]

[0053] In the formula: γ is the call incentive price, yuan / kWh.

[0054] Step 3: Establish an uncertainty model for the grid-connected power of the EV group, input the historical parameters of the grid-connected power of the EV group, set the number of target scenarios, and use the scenario analysis method to solve the scenario set and its probability of the grid-connected power of the EV group;

[0055] Specifically in this example, Step 3 is refined as follows:

[0056] Scenario generation. Use historical data to generate 1000 scenarios. The data source is the operation monitoring platform of charging facilities in a certain province. The data time is from July 15th to July 18th, 2020, 12:00 - 14:00. The set of EV grid-connected power scenarios monitored by the platform is called scenario set W * ;

[0057] Scenario reduction. Reduce the EV grid-connected power scenario set, merge similar scenarios, and give multiple EV grid-connected power scenarios and corresponding probabilities, so that the reduced typical scenario set is consistent with the original scenario set in trend.

[0058] Let the scenario set be 10. The 10 typical scenario sets of the grid-connected power of the EV group after the scenario reduction method are shown in Table 2. Assume that the average online charging time after the EV is connected to the grid is 2 hours. Based on Table 2 and the incentive calculation formula for the k-th EV to be called for reserve The numerical distribution of the SOC of the EV group at 12:00 when the reserve starts is obtained under the ten scenarios, as Figure 2 shown.

[0059] Table 2

[0060]

[0061] Step 4: Establish an uncertainty model for the EV group response rate. Input the EV group response rate model parameters and their ranges, and use the Latin hypercube sampling method to uniformly sample to generate EV group response rate data under the same call incentive;

[0062] Specifically in this example, Step 4 is refined as follows:

[0063] The uncertainty calculation formula for the EV group response rate Δβ(γ) is: Δβ(γ) = Δβ m (γ) + Δβ err , where γ is the incentive level; Δβ m (γ) is the mean value of the EV group response rate at the incentive level γ, and the calculation formula is: where Δβ up , Δβ down represent the upper and lower limits of the response rate uncertainty in period t; Δβ err is the random response rate of the EV group at the incentive level γ, which is a random variable and is considered to follow a uniform distribution with a mean of Δβ up (λ) within [Δβ down (λ), Δβ m (λ)].

[0064] Step 5: Based on Steps 2, 3, and 4, construct a community backup service decision-making model with the goal of maximizing revenue;

[0065] The calculation formula for the community backup revenue G is:

[0066]

[0067] where W is the number of scenarios; π w is the standardized probability of the occurrence of the w-th type of scenario; is the community backup capacity revenue under the w-th type of scenario; is the revenue of the electric energy actually called when the community provides backup under the w-th type of scenario; is the penalty cost paid by the community for not meeting the backup capacity under the w-th type of scenario; is the backup incentive cost paid by the community to the EV group under the w-th type of scenario; is the EES backup cost under the w-th type of scenario.

[0068] Specifically, the community backup revenue The revenue of the electric energy actually called when the community provides backup The penalty cost paid by the community for not meeting the backup capacity The backup incentive cost paid by the community to the EV group The EES backup cost The calculation formula is as follows:

[0069]

[0070] Where: γ R and γ E are the reserve capacity and the electricity market price respectively; γ c is the average charging price during non-reserve periods; γ P is the penalty price when the reserve capacity is not met; P app is the reserve capacity declared by the community; T R is the reserve duration; D is the cooperation period, d ∈ D; μ w,d is whether the community is called on the d-th day under the w-th scenario, a 0-1 variable, 0 for not called, 1 for called; t s is the start time of the reserve; t e is the end time of the reserve; K is the total number of EVs, k ∈ K; is the discharge power of the k-th EV at the t-th moment on the d-th day under the w-th scenario; is the discharge power of the EES at the t-th moment on the d-th day under the w-th scenario; is the community reserve deficit at the t-th moment on the d-th day under the w-th scenario, and this part will be converted into reserve penalty costs; R EV is the EV reserve incentive; γ EES is the operation and maintenance unit cost; is the stored electricity of the ESS at the start time of the reserve; is the stored electricity of the ESS at the end time of the reserve.

[0071] The model constraint conditions include:

[0072] 1) Reserve balance constraint, the reserve capacity P app is provided by the coordinated operation of EVs and EES. When the community capacity is insufficient, the remaining part is regarded as the reserve deficit.

[0073]

[0074] 2) Maximum reserve deficit rate constraint

[0075]

[0076] Where: θ R,max is the maximum reserve deficit allowed in the reserve market, kW.

[0077] 3) EV group response rate constraint. To avoid excessive response, the actual call rate of the EV group during the reserve period should be less than or equal to the possible maximum response rate.

[0078]

[0079] 4) Spare call rate constraint. The spare call rate σ refers to the probability of the community participating in the reserve market on a monthly basis; this value is the community's predicted value for the reserve market.

[0080]

[0081] Step 6: Solve the community reserve service decision model by combining the genetic algorithm and the mixed-integer programming method to obtain the community's optimal declared reserve capacity, the incentive price for the electric vehicle group, and the discharge strategies for the electric vehicle group and energy storage.

[0082] After optimization, the optimized result of the community's declared reserve capacity is 2087.3 kW, the declared capacity is 2100 kW, the community revenue is 6453.8 yuan, the average response rate of the EV group is 34%, the average output of the EV group is 1480.8 kW, the average output of the EES is 427.5 kW, the reserve shortage is 191.7 kW, and the reserve shortage rate is 9.13%. The community revenues under ten scenarios are shown in Table 3, and the EV group response rates and their outputs under ten scenarios are as Figure 3 shown.

[0083] Table 3

[0084]

[0085] In Scenarios 1 / 2, the state of charge (SOC) of all electric vehicles (EVs) is less than 50%, and none of them meet the discharge conditions. At this time, the EES is responsible for discharging, and the overall reserve shortage is 1672.5 kW. Considering the scenario probability, the community revenues for Scenarios 1 and 2 are -347.6 yuan and -695.3 yuan respectively.

[0086] In Scenario 3, most EVs can execute the discharge strategy, but due to the minimum guaranteed power limit, all EVs cannot continuously discharge for 1 hour. The community optimizes the call incentive price to 0.3 yuan / kWh, the response rate of the EV group reaches the maximum value of 50%, the output of the EES is 427.5 kW, the output of the EV group is 529.9 kW, the reserve shortage is 1142.5 kW, and considering the scenario probability, the community revenue is -842.8 yuan.

[0087] In Scenario 4, all EVs can execute the discharge strategy, 80.5% of the EVs can continuously discharge for 1 hour, the call incentive price is 0.27 yuan / kWh, the response rate reaches 42%, the output of the EV group is 1672.5 kW, and considering the scenario probability, the community revenue is 1375.2 yuan.

[0088] In Scenarios 5 / 6 / 7 / 8 / 9 / 10, all EVs can continuously discharge for 1 hour. At this time, the overall call potential of the EV group increases, the call incentive price is 0.24 yuan / kWh, the EV response rate is 33%, and the output of the EV group is 1672.5 kW.

[0089] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention.

Claims

1. A decision-making method for community reserve service considering the double uncertainties of electric vehicle groups, characterized in that, The method includes: Pre - constructing a monthly reserve contract mechanism for an electric vehicle (EV) group; Inputting historical parameters of the EV group's grid - connected power into a scenario analysis program, setting the number of target scenarios, solving the scenario set and its probability of the EV group's grid - connected power using the scenario analysis method, and establishing an uncertainty model of the EV group's grid - connected power; Inputting the model parameters and range of the EV group's response rate into the scenario analysis program, uniformly sampling using the Latin hypercube sampling method to generate EV group response rate data under the same call incentive, and establishing an uncertainty model of the EV group's response rate; Based on the monthly reserve contract mechanism for the EV group, the uncertainty model of the EV group's grid - connected power, and the uncertainty model of the EV group's response rate, constructing a community reserve service decision - making model with the goal of maximizing revenue; Combining the genetic algorithm and the mixed - integer programming method to solve the community reserve service decision - making model, and obtaining the community's optimal declared reserve capacity, the incentive price for the EV group, and the discharge strategies of the EV group and energy storage; The establishment of the uncertainty model of the EV group's grid - connected power includes: scenario generation and scenario reduction; The above-mentioned scenario generation: Select the historical data of the charging facility operation monitoring platform to establish an EV grid-connected power scenario analysis model. Consider the EV grid-connected power monitored by the platform each time as a scenario w, and the set of EV grid-connected power scenarios monitored by the platform is called the scenario set W * ; The scenario reduction: reducing the EV grid - connected power scenario set to obtain the number of target scenarios \(W\), merging similar scenarios, and giving multiple EV grid - connected power scenarios and their corresponding probabilities; The calculation formula of the uncertainty model \(\Delta\beta(\gamma)\) of the EV group's response rate is: Δβ(γ) = Δβ m (γ) + Δβ err (8) where γ is the excitation level; Δβ m (γ) is the mean value of the EV group response rate at the excitation level γ; Δβ err is the random response rate of the EV group at the excitation level γ; Mean EV group response rate Δβ at excitation level γ m (γ) is given by the formula: where, Δβ up (γ), Δβ down (γ) represent the upper and lower limits of the uncertainty of the EV group response rate at time t under the excitation level γ; Δβ(γ) is a random variable, %, assumed to be uniformly distributed with a mean of Δβ up (γ) within the interval [Δβ down (γ), Δβ m (γ))].

2. The decision-making method for community backup services considering the double uncertainties of electric vehicle clusters according to claim 1, characterized in that The constraints of the monthly reserve contract mechanism for the EV group include: 1) Contract period M and spare period [t s , t e ​ Considering the timeliness and enforceability of the contract, the contract period is M; the standby period refers to the standby start time t s to the standby end time t e ; 2) Internet access time / Disconnection time To enable the k-th EV to be callable during the standby period, where k ∈ K, constraints are imposed on it and as shown in Equation (1): where: Δt gap is the spare buffer time set for the community; 3) Grid-connected electricity quantity / Expected off-grid electricity quantity When EVs participate in reserve, they still need to meet the travel needs of users, including: ① At the off-grid moment of the k-th EV, its battery level needs to be greater than or equal to the expected off-grid battery level. ② The battery level of the k-th EV needs to be always greater than or equal to the guaranteed battery level. To meet the users' unpredictable vehicle usage needs; ③ To achieve the callability of EVs, the on-grid battery level of the k-th EV needs to be greater than a certain battery level threshold, as shown in Equation (2): In the formula: is the minimum grid connection power threshold of EV, kWh; 4) EV operation constraints The EV should satisfy multiple constraint conditions such as charging and discharging power and stored electricity, as shown in Equation (3); In the formula: is the charging power of the k-th EV at time t, is the discharging power of the k-th EV at time t, in kW; is the maximum charging power of the EV, is the maximum discharging power of the EV, in kW; σ EV is the self-discharge rate of the EV, in %; η EV,c is the charging efficiency of the EV, η EV,d is the discharging efficiency of the EV, in kW; is the charging state variable of the k-th EV at time t, is the discharging state variable of the k-th EV at time t, which is a 0-1 variable; Δt is the simulation duration, in h; is the minimum stored electricity of the EV, is the maximum stored electricity of the EV, in kWh; 5) EV call constraints To avoid multiple calls to the EV, the EV call behavior within the contract period is regulated, as shown in Equation (4): where: N m,k is the call status of the k-th EV on the m-th day, 1 means called, 0 means not called; is the maximum allowable call times of the k-th EV within the contract period; is the maximum allowable call duration of the k-th EV during the standby period, h; E d,max is the maximum daily discharge amount allowed for the EV, kWh. In formula (4), the first formula is the constraint on the cumulative call times of the EV; the second formula is the constraint on the daily call duration of the EV; the third formula means that the EV can be called at most once during the standby period; the fourth formula is the constraint on the maximum daily discharge amount of the EV; 6) EV Standby Incentive R EV Analogous to the grid-side reserve market, the EV reserve incentive R EV consists of the basic reserve incentive R EV,base and the reserve call incentive R EV,act ; In the formula: refers to the basic incentive fee given by the community to encourage the k-th EV to participate in reserve, calculated on a daily basis, regardless of whether the k-th EV is called or not, as shown in Equation (6); Wherein: γ base is the basic incentive price, yuan / day; D k is the spare days provided for the k-th EV, day; The k-th EV calls for backup incentives The incentive cost is given according to its actual call level, as shown in Equation (7); where: \(\gamma\) is the call incentive price, yuan / kWh.

3. The decision-making method for community backup service considering the double uncertainties of electric vehicle groups according to claim 1, characterized in that The calculation process of the number of target scenarios \(W\) of the EV grid - connected power includes: 1) Represented by the Xth w class of scenarios, with the corresponding probability being the interval credibility π w , w = 1, 2, …, W * ; 2) Represent the scene X by the absolute value of the difference between the interval means w and X k for the Kantorovich distance between them 3) For each scenario w, find the scenario r that is closest to it and label it as DM w = min(D(X w , X k )) and calculate the product PDM w = DM w × π w ; 4) Search for scenario d among W * scenarios, such that PDM d = min(PDM w ); 5) Add the probability π of scenario d d to the nearest scenario, and at the same time subtract π d from the scenario set W * , that is, the total number of scenarios W * = W * - 1; 6) Determine whether the number of scenarios W that meet the requirements is reached. If so, obtain W and its corresponding probability π w , otherwise repeat the above process.

4. The decision-making method for community backup service considering the double uncertainties of electric vehicle clusters according to claim 1, wherein The calculation formula of the community reserve revenue \(G\) is: where W is the number of scenarios; π w is the standardized probability of the occurrence of the w-th type of scenario; is the community backup capacity revenue under the w-th type of scenario; is the revenue of the electric energy actually called for when the community provides backup under the w-th type of scenario; is the penalty cost paid by the community for the unmet backup capacity under the w-th type of scenario; is the backup incentive cost paid by the community to the EV group under the w-th type of scenario; is the EES backup cost under the w-th type of scenario.

5. The decision-making method for community backup services considering the double uncertainties of an electric vehicle cluster according to claim 4, wherein The community's standby revenue The revenue from the electricity that is actually called upon as standby provided by the community The penalty cost paid by the community for not meeting the standby capacity The standby incentive cost paid by the community to the EV group The standby cost of EES The calculation formula is as follows: Where: γ R is the contract price for reserve capacity, γ E is the electricity market price; γ C is the average charging price during non-reserve periods; γ P is the penalty price when the reserve capacity is not met; P app is the reserve capacity declared by the community; T R is the reserve duration; Let \(M\) be the contract period, \(m\in M\); \(\mu\) w,m is whether the community is called on the \(m\)-th day under the \(w\)-th scenario, a 0-1 variable, 0 for not called, 1 for called; \(K\) is the total number of EVs, \(k\in K\); is the discharge power of the \(k\)-th EV at time \(t\) on the \(m\)-th day under the \(w\)-th scenario; is the EES discharge power at time \(t\) on the \(m\)-th day under the \(w\)-th scenario; is the community reserve deficit at time \(t\) on the \(m\)-th day under the \(w\)-th scenario, converted into reserve penalty cost; \(R\) EV is the EV reserve incentive; \(\gamma\) EES is the operation and maintenance unit cost; is the ESS stored electricity at the start time of reserve; is the ESS stored electricity at the end time of reserve.

6. The decision-making method for community backup services considering the double uncertainties of electric vehicle groups according to claim 1, wherein The constraint conditions of the community reserve service decision - making model: 1) Reserve balance constraint 2) Maximum reserve deficit rate constraint where: θ R,max is the maximum reserve shortage allowed in the reserve market, kW.

7. The decision-making method for community backup services considering the double uncertainties of an electric vehicle cluster according to claim 1, characterized in that, The method of combining the genetic algorithm and the mixed - integer programming method to solve the community reserve service decision - making model includes: Using the backward scenario decomposition method to solve the uncertainty problem of the EV group's grid - connected power, forming \(W\) types of deterministic EV group grid - connected power scenarios; For the \(w\) - th type of scenario, using the genetic algorithm to randomly generate multiple incentive levels \(\gamma\), using the Latin hypercube sampling method to quantify the EV group's response rate under a certain type of incentive level \(\gamma\) into \(X\) deterministic EV group response rates \(\Delta\beta(\gamma)\), and using CPLEX to find the community's optimal solution under the \(x\) - th EV group response rate, including the community's declared capacity and the EV / ESS discharge strategy; Through genetic cross - optimization, obtaining the community's optimal declared reserve capacity, the incentive price for the EV group, and the discharge strategies of the EV group and energy storage.

Images