A Construction Method and Device for a Pricing Model of an Electric Vehicle Virtual Power Plant

Abstract

The present invention provides a method and device for constructing a pricing model of an electric vehicle virtual power plant, which relates to the technical field of power system operation and control. The method includes: determining an uncertainty set composed of typical scenario probability distribution vectors of new energy output prediction errors and corresponding confidence constraints based on the idea of multi-discrete scenario distributionally robust optimization; determining the bidding volume of the electric vehicle virtual power plant in the day-ahead market, and then optimizing based on the uncertainty set and the corresponding confidence constraints to determine the retail price issued to the electric vehicle charging station; adjusting the deviation power of the electric vehicle charging station according to the bidding volume, retail price and real-time market price, and optimizing the charge and discharge power of the energy storage system and the power transaction in the real-time market. The present invention can ensure the maximization of benefits in the worst-case scenario, thus achieving a balance between economy and robustness, and also conforming to the uncertainties of new energy output and the demand response behavior of electric vehicle charging stations in the actual situation.

Description

Technical Field

[0001] The present invention relates to the technical field of power system operation and control, and particularly to a method and device for constructing an electric vehicle virtual power plant pricing model. Background Art

[0002] With the large-scale popularization of new energy vehicles, the problem of disorderly charging brought by them will have a huge impact on the operation of the power grid. A reasonable incentive pricing strategy can achieve "peak shaving and valley filling", which helps the stable operation of the power grid. At the same time, market bidding and incentive pricing are coupled with each other: if the virtual power plant bids too much in the market and offers less incentives to end-users, the virtual power plant will face assessment risks; conversely, if the bidding volume is too small and the incentives are large, the virtual power plant will face the risk of over-response. Both will result in the economic interests of the virtual power plant being damaged and are not conducive to achieving the maximum benefit.

[0003] At present, most relevant research analyzes and models the bidding and pricing of virtual power plants separately. Although it is relatively simple and intuitive and easy to put into practice, due to the lack of full consideration of the relationship between bidding and pricing, there may be a large deviation between the final model and the actual situation. In the context of the increasing electricity demand of electric vehicles, it is necessary to consider the two factors comprehensively to make the model prediction results more immediate and accurate. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and device for constructing an electric vehicle virtual power plant pricing model, which is used to solve the problems that the existing pricing model does not fully consider the relationship between bidding and pricing, and may lead to a large deviation between the final pricing model and the actual situation. It can ensure the maximum benefit in the worst case, thus achieving the balance between economy and robustness, and also conforming to the uncertainty of new energy output and the demand response behavior of electric vehicle charging stations in actual situations.

[0005] To achieve the above purpose, the present invention provides a method for constructing an electric vehicle virtual power plant pricing model, including:

[0006] Step 1: Based on the idea of multi-discrete scenario distribution robust optimization, determine the uncertainty set composed of the typical scenario probability distribution vectors of the new energy output prediction error and the corresponding confidence constraints;

[0007] Step 2: Determine the bidding volume of the electric vehicle virtual power plant in the day-ahead market, and then optimize based on the uncertainty set and the corresponding confidence constraints to determine the retail price issued to the electric vehicle charging station;

[0008] Step 3: According to the bidding volume, retail price and real-time market price, adjust the deviation power of the electric vehicle charging station, and optimize the charging and discharging power of the energy storage system and the power transaction in the real-time market.

[0009] According to a construction method of an electric vehicle virtual power plant pricing model provided by the present invention, in step 1, the earth mover distance method is used to characterize the uncertainty set composed of the typical scenario probability distribution vectors of the new energy output prediction error and the corresponding confidence constraints.

[0010] According to a construction method of an electric vehicle virtual power plant pricing model provided by the present invention, the uncertainty set composed of the typical scenario probability distribution vectors of the new energy output prediction error is:

[0011] (1)

[0012] Wherein, is the uncertainty set; is the typical scenario probability distribution vector of the new energy output prediction error; represents the boundary of the support space of the random variable obtained from the historical sample data of the new energy power generation and the load; and are respectively the power purchase and power sale of the electric vehicle virtual power plant in the real-time market under the typical scenario s at time t; is the confidence level of the fuzzy set of the earth mover distance method;

[0013] The confidence level of the fuzzy set of the earth mover distance method is obtained by the following formula:

[0014] (2)

[0015] (3)

[0016] In the formula, is the reference scenario probability distribution vector; Ns represents the number of all possible scenarios created through the scenario generation process; is a variable, which is determined according to the actual situation; is the diameter of the support space of the random variable; represents the norm.

[0017] According to a construction method of an electric vehicle virtual power plant pricing model provided by the present invention, the confidence constraint is:

[0018]

[0019] In the formula, P represents probability.

[0020] According to a construction method of an electric vehicle virtual power plant pricing model provided by the present invention, step 2 specifically includes:

[0021] Determine the objective function of the retail price based on the settlement revenue of the current market and the revenue from selling electricity to electric vehicle charging stations; determine the constraint conditions of the retail price according to the market rules and the internal supply and demand situation when the electric vehicle virtual power plant participates in the superior market.

[0022] According to a method for constructing an electric vehicle virtual power plant pricing model provided by the present invention, the objective function of the retail price is:

[0023] (5)

[0024] Wherein, ,

[0025] In the formula, is the set of decision variables in the first stage; represents the time length of each decision period in the optimization process; is the total number of time periods; is the total number of electric vehicle charging stations; is the total number of energy storage systems; and are the clearing electricity prices for selling and purchasing electricity in the day-ahead market at time t, respectively; and are the cleared electricity power output volumes sold by the electric vehicle virtual power plant to the market and purchased from the market at time t, respectively; and respectively represent the discharging and charging electricity prices of the energy storage system at time t; and respectively represent the discharging and charging powers of the energy storage system at time t; and are the equivalent charging power and discharging power of the electric vehicle charging station at time t; is the retail price released by the electric vehicle virtual power plant at time t; is a 0-1 variable, represents that the electric vehicle virtual power plant is an electricity producer at time t, represents that the electric vehicle virtual power plant is an electricity consumer at time t; is a 0-1 variable, represents that the energy storage system is in a charging state at time t, represents that the energy storage system is in a discharging state at time t.

[0026] According to a method for constructing an electric vehicle virtual power plant pricing model provided by the present invention, the constraint conditions of the retail price are:

[0027] (6)

[0028] (7)

[0029] (8)

[0030] (9)

[0031] (10)

[0032] (11)

[0033] (12)

[0034] (13)

[0035] In the formula, are respectively the minimum and maximum retail prices allowed by the market rules during the t period; is the daily average retail price threshold; are respectively the maximum cleared power of the electric power sold by the electric vehicle virtual power plant to the market and purchased from the market during the t period; is the maximum charging and discharging power of the energy storage system during the t period; and respectively represent the charging and discharging efficiencies of the energy storage system; represents the state of the energy storage system, is the initial state of the energy storage system.

[0036] According to the construction method of an electric vehicle virtual power plant pricing model provided by the present invention, step 3 specifically includes:

[0037] Establish an expected revenue model for the electric vehicle virtual power plant, and the objective function of the expected revenue model is:

[0038] (14)

[0039] In the formula, and are respectively the purchased power and sold power of the electric vehicle virtual power plant in the real-time market under the typical scenario s during the t period, is the real-time market power purchase price, is the deviation penalty cost coefficient; is the risk preference coefficient; represents the scheduling cost of the energy storage system;

[0040] Characterize the AC power flow in the network within the electric vehicle virtual power plant and determine the constraint conditions of the expected revenue model.

[0041] According to a method for constructing a pricing model of an electric vehicle virtual power plant provided by the present invention, the constraint conditions of the expected revenue model are as follows:

[0042] (15)

[0043] (16)

[0044] (17)

[0045] (18)

[0046] (19)

[0047] (20)

[0048] In the formula, represents the active power of the conventional load in the typical scenario s at time t; is the output of the distributed new energy r in the typical scenario s at time t; is the total number of distributed new energies; is the reactive power injected by the electric vehicle virtual power plant into the power grid at node i, is the reactive power absorbed by the electric vehicle virtual power plant from the power grid at node i; 、 are the voltage amplitudes of nodes i and j respectively, is the node and is the voltage phase difference between them, is the conductance of the line , is the susceptance of the line , represents all the nodes directly connected to node through the line, is the voltage phase angle of node ; are the minimum and maximum values of the node voltage respectively; is the maximum possible output of the distributed new energy r in the typical scenario s at time t.

[0049] In the second aspect, the present invention provides a device for constructing a pricing model of an electric vehicle virtual power plant, including:

[0050] The first determination unit is used to determine the uncertainty set composed of the typical scenario probability distribution vector of the new energy output prediction error and the corresponding confidence constraint based on the idea of multi-discrete scenario distributionally robust optimization;

[0051] A second determination unit, configured to determine the bidding volume of the electric vehicle virtual power plant in the day-ahead market, and then optimize based on the uncertainty set and the corresponding confidence constraint to determine the retail price issued to the electric vehicle charging station;

[0052] An adjustment and optimization unit, configured to adjust the deviation power of the electric vehicle charging station according to the bidding volume, retail price and real-time market price, and optimize the charging and discharging power of the energy storage system and the power transaction in the real-time market.

[0053] A method and device for constructing an electric vehicle virtual power plant pricing model provided by the present invention successfully solve the pricing problem of the electric vehicle virtual power plant under the uncertainty of new energy output and the uncertainty of charging station demand response through a distributed robust optimization method. The present invention performs well in terms of maximizing revenue, scheduling the behavior of electric vehicle charging stations, grid stability and handling the uncertainty of new energy output, and has strong practical application prospects. By comprehensively considering market bidding, retail price, new energy output and electric vehicle charging station demand response, it can help the virtual power plant maximize revenue and minimize risks in a complex market environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0055] In the drawings:

[0056] Figure 1 is a flowchart of the method for constructing the electric vehicle virtual power plant pricing model of the present invention;

[0057] Figure 2 is a comparison diagram of the revenue curves of charging station 1 between the retail price (optimized pricing) given by the pricing model of the present invention and the fixed pricing of the prior art;

[0058] Figure 3 is a comparison diagram of the revenue curves of charging station 2 between the retail price (optimized pricing) given by the pricing model of the present invention and the fixed pricing of the prior art;

[0059] Figure 4 is a comparison diagram of the revenue curves of charging station 3 between the retail price (optimized pricing) given by the pricing model of the present invention and the fixed pricing of the prior art;

[0060] Figure 5 is a comparison diagram of the total revenue curves between the retail price (optimized pricing) given by the pricing model of the present invention and the fixed pricing of the prior art;

[0061] Figure 6 、 Figure 7 、 Figure 8 are respectively the curve graphs of the charging power and discharging power of the real-time market of the charging stations 1, 2, and 3 of the present invention, and the charging power and discharging power of the energy storage system;

[0062] Figure 9 is the curve graph of the total charging power and discharging power of the real-time market of the present invention;

[0063] Figure 10 is the curve graph of the total charging power and discharging power of the energy storage system of the present invention. Detailed implementation manners

[0064] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0065] The following will describe in detail some implementation manners of the present invention in conjunction with the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other. The steps described in the embodiments are only examples if there is no necessary order, and should not be regarded as a limitation. Those of ordinary skill in the art can adjust the step order without destroying the logic.

[0066] The present invention utilizes a two-stage distributionally robust joint optimization strategy, which can take into account the uncertainties of the distributed new energy output in the virtual power plant, the real-time market electricity price, and the demand response behavior of the electric vehicle charging stations, and then ensure that the virtual power plant formulates the optimal incentive electricity price under the worst-case probability distribution of the distributed new energy output: when the uncertainty distribution is unknown, stochastic optimization may lead to large deviations, while the scheduling result of robust optimization is too conservative to meet the actual scheduling requirements. Distributionally robust optimization effectively balances the particularity of stochastic optimization and the conservativeness of robust optimization, and has broad application prospects for the source-load uncertainty problem of the electric vehicle virtual power plant. This method first assumes that the true probability distribution is not known a priori, that is, the exact distributions of the new energy output situation and the demand size of the electric vehicle charging stations are not available in advance, and it belongs to a fuzzy set, which contains all potential probability distributions close to the set probability. Then, by finding the worst-case probability distribution in the possible distribution cluster and optimizing it, the maximum benefit under the worst-case scenario is ensured, thereby achieving the balance between economy and robustness, which also conforms to the uncertainties of the new energy output and the demand response behavior of the electric vehicle charging stations in the actual situation.

[0067] Please refer to Figure 1 , an embodiment of the present invention provides a method for constructing a pricing model of an electric vehicle virtual power plant, including:

[0068] Step 1: Based on the idea of multi-discrete scenario distributionally robust optimization, determine the uncertainty set composed of the typical scenario probability distribution vectors of the new energy output prediction error and the corresponding confidence constraints;

[0069] Specifically, use the Wasserstein distance (earth mover's distance) method to characterize the typical scenario probability distribution vectors constituting the uncertainty set (or called the uncertainty space), that is, the Wasserstein ball is:

[0070] (1)

[0071] Wherein, is the uncertainty set; represents the boundary of the support space of the random variable obtained from the historical sample data of the new energy power generation and the load. and are the purchase / sale electric powers of the VPP in the real-time market under the typical scenario s at time t, respectively. In the constraint, is the radius of the fuzzy set of the Wasserstein distance method, representing the maximum allowable deviation. Through this constraint, the model can handle uncertainties within the fuzzy set range. is obtained by the following formula:

[0072] (2)

[0073] (3)

[0074] In the formula, is the confidence level of the fuzzy set of the Wasserstein distance method, given by the decision maker. is a benchmark scenario probability distribution vector, representing the initial scenario probability distribution generated by a certain method. By introducing the scenario probability distribution , the model can perform robust optimization under uncertainties. Ns represents the number of all possible scenarios created through the scenario generation process. is a variable that can be selected according to the actual situation. is the diameter of the support space of the above-mentioned random variable. A larger means a larger support space range and a more conservative fuzzy set; a smaller means a smaller support space range and a more aggressive fuzzy set. represents the norm.

[0075] Probability distribution vector of typical scenarios of new energy output prediction error The confidence level constraint is as follows:

[0076]

[0077] In the formula, P represents the probability of the inequality in the curly brackets. This confidence level constraint can ensure that in the most adverse situation, the uncertainty set of has a probability deviation floating within a certain limit in the Wasserstein ball.

[0078] It should be noted that in Step 1, the uncertainty set and the corresponding confidence level constraint are determined. By setting the allowable size of the radius of the Wasserstein ball, the error range of possible new energy output prediction is limited, ensuring that the most adverse situation is considered in the subsequent optimization process, and providing an uncertainty set with a probability deviation within a certain range for the subsequent steps.

[0079] Step 2: Determine the bidding volume of the electric vehicle virtual power plant in the day-ahead market, and then optimize based on the uncertainty set and the corresponding confidence level constraint to determine the retail price issued to the electric vehicle charging station;

[0080] In Step 2, according to the uncertainty set and the corresponding confidence level constraint in Step 1, as well as the historical data of the day-ahead market, the bidding volume of the electric vehicle virtual power plant in the day-ahead market can be determined, and the retail price can be issued to the electric vehicle charging station according to the most adverse probability distribution of distributed new energy output.

[0081] In some embodiments, Step 2 specifically includes: determining the objective function of the retail price according to the settlement revenue of the day-ahead market and the revenue from selling electricity to the electric vehicle charging station; determining the constraint conditions of the retail price according to the market rules and the internal supply and demand situation when the electric vehicle virtual power plant participates in the upper-level market. That is, an objective function is established based on the settlement revenue of the day-ahead market and the revenue from selling electricity to the electric vehicle charging station, and at the same time, considering the internal supply and demand situation to determine the role (producer or consumer) of participating in the market, and accordingly formulating the optimal retail price.

[0082] Specifically, the objective function and constraint conditions of the retail price are as follows:

[0083] (5)

[0084] (6)

[0085] (7)

[0086] (8)

[0087] (9)

[0088] (10)

[0089] (11)

[0090] (12)

[0091] (13)

[0092] wherein, is the set of decision variables in the first stage. represents the time length of each decision period in the optimization process; is the total number of time periods; is the total number of electric vehicle charging stations; is the total number of energy storage systems; and are the clearing electricity prices for selling and purchasing electricity in the day-ahead market at time t, respectively; and are the cleared electric power quantities sold by the electric vehicle virtual power plant to the market and purchased from the market at time t, respectively; and respectively represent the discharging and charging electricity prices of the energy storage system at time t; and respectively represent the discharging and charging powers of the energy storage system at time t, and are the equivalent charging power and discharging power of the electric vehicle charging station at time t; is the retail price published by the electric vehicle virtual power plant at time t; are the minimum and maximum retail prices allowed by the market rules at time t, respectively; is the daily average retail price threshold; are the maximum cleared electric power quantities sold by the electric vehicle virtual power plant to the market and purchased from the market at time t, respectively.

[0093] It can be seen that the retail price and its daily average value should be within the range allowed by the market rules, as shown in equations (6) and (7). Equations (8) and (9) illustrate that when the electric vehicle virtual power plant (VPP) participates in the superior power market, it needs to decide whether to participate as an electric energy producer or an electric energy consumer according to the internal supply and demand situation, wherein is a 0-1 variable, Indicates that the VPP is an electric energy producer, and vice versa, it is an electric energy consumer. Equations (10) and (11) show that the charging and discharging power of the VPP and the energy storage system during collaborative operation do not exceed their maximum power limits, and they will not charge and discharge simultaneously. Among them is a 0-1 variable, indicating that the energy storage system is in the charging state, and vice versa, it is in the discharging state. Equation (12) shows that the SOC of the energy storage system changes over time, depending on the charging and discharging power. Among them indicates the state of the energy storage system, and respectively represent the charging and discharging efficiencies of the energy storage system. Equation (13) shows that the energy storage system returns to its initial state after a scheduling period (such as one day)

[0094] Step 3. Adjust the deviation power of the electric vehicle charging station according to the bid volume, retail price, and real-time market price, and optimize the charging and discharging power of the energy storage system and the power trading in the real-time market.

[0095] It should be noted that Step 3 outputs the optimized charging and discharging power of the energy storage system and the power trading strategy in the real-time market according to the optimal decision-making results of Step 2, including the bid volume, retail price, and real-time market price, based on the expected revenue model. On the basis of Step 2, by adjusting the deviation power, the working state of the energy storage system is further optimized, and at the same time, power is bought and sold in the real-time market to maximize the revenue.

[0096] In some embodiments, Step 3 specifically includes:

[0097] Step 3.1. Establish an expected revenue model for the electric vehicle virtual power plant. The objective function of this expected revenue model is:

[0098] (14)

[0099] In the formula, and are the purchased / sold electric power of the VPP in the real-time market under the typical scenario s at time t, respectively, is the real-time market power purchase price, is the deviation penalty cost coefficient. is the risk preference coefficient, which is used to adjust the risk preference of the electric vehicle virtual power plant. represents the scheduling cost of the energy storage system, which can be measured by the charging and discharging efficiency and life loss of the energy storage system.

[0100] Step 3.2: Characterize the AC power flow of the network within the electric vehicle virtual power plant and determine the constraint conditions of the expected revenue model. The constraint conditions of the expected revenue model include power balance constraints, node voltage constraints, charge and discharge power constraints of the energy storage system, and new energy output constraints.

[0101] Specifically, select the AC power flow model to characterize the AC power flow of the network within the VPP. Use the AC power flow model to describe the AC power flow within the VPP to ensure that power transmission conforms to physical laws. This is set based on the topological structure and equipment characteristics of the power system, and its output will serve as the basis for real-time adjustment.

[0102] Equations (15) and (16) are respectively for active / reactive power balance; Equation (17) is used to calculate the node voltage; Equation (18) ensures that the distributed new energy output in each scenario should not exceed the maximum possible output of that scenario; Equation (19) ensures that the node voltage is within the allowable range; Equation (20) ensures that the risk preference coefficient is non-negative.

[0103] The power balance constraint and node voltage constraint are as follows:

[0104] (15)

[0105] (16)

[0106] (17)

[0107] (18)

[0108] (19)

[0109] (20)

[0110] In the equations, represents the active power of the conventional load in the typical scenario s at time t; is the output of the distributed new energy r in the typical scenario s at time t; is the total number of distributed new energies; is the reactive power injected by the electric vehicle virtual power plant into the power grid at node i, is the reactive power absorbed by the electric vehicle virtual power plant from the power grid at node i; 、 are the voltage amplitudes of nodes i and j respectively, is the node and is the voltage phase angle difference between them, is the conductance of the line and is the line susceptance denotes all the nodes directly connected to the node is the voltage phase angle of node ; are the minimum and maximum values of the node voltage respectively; is the maximum possible output of distributed new energy r under the typical scenario s at time t.

[0111] It should be noted that in the process of constructing the pricing model of the electric vehicle virtual power plant of the present invention, step 1 provides the basic framework and parameter setting for the subsequent steps; step 2 depends on the results of step 1 to conduct market bidding and pricing; step 3 is based on the decision results of the previous two steps and makes dynamic adjustments during actual operation. Step 2 determines the bidding strategy of the electric vehicle virtual power plant in the day-ahead market and the retail price issued to the electric vehicle charging station, while step 3 optimizes the power trading and energy storage system operation in the real-time market based on these decisions.

[0112] Although the above description is linear, in practical applications, there may be a feedback mechanism that returns from step 3 to step 2 or even step 1. For example, if the situation in the real-time market significantly deviates from the expectation, it may be necessary to re-evaluate the bidding strategy or adjust the pricing model. The whole process can be regarded as an iterative optimization process. With the acquisition of more data and the change of market conditions, the pricing model can continuously update and improve its decision-making strategy.

[0113] Generally speaking, the pricing model of the electric vehicle virtual power plant constructed by the present invention is a power system pricing model specifically for the charging demand of electric vehicles and the complexity and uncertainty of the operation of the electric vehicle virtual power plant. This model is based on the idea of distributed robust optimization and aims to influence the charging behavior of electric vehicle users through a reasonable pricing mechanism, so as to effectively manage the grid load, maximize the economic benefits of the electric vehicle virtual power plant operator while ensuring the stable operation of the grid.

[0114] Based on the same inventive concept, another embodiment of the present invention provides a device for constructing a pricing model of an electric vehicle virtual power plant. This device corresponds to the method of the foregoing embodiment. The device includes:

[0115] A first determination unit, configured to determine an uncertainty set composed of a typical scenario probability distribution vector of the new energy output prediction error and the corresponding confidence constraint based on the idea of multi-discrete scenario distribution robust optimization;

[0116] A second determination unit, configured to determine the bidding volume of the electric vehicle virtual power plant in the day-ahead market, and then optimize based on the uncertainty set and the corresponding confidence constraint to determine the retail price issued to the electric vehicle charging station;

[0117] An adjustment and optimization unit is configured to adjust the deviation power of an electric vehicle charging station, optimize the charging and discharging power of an energy storage system, and optimize power trading in the real-time market according to the bidding volume, retail price, and real-time market price.

[0118] The following is a specific embodiment of the present invention.

[0119] According to the method for constructing the pricing model of the electric vehicle virtual power plant of the present invention, Python is used for simulation. By optimizing the bidding decisions and retail prices of the electric vehicle virtual power plant in the day-ahead market, corresponding formulas are formulated. Three electric vehicle charging stations are set up, denoted as charging station 1, charging station 2, and charging station 3 respectively. The initial parameters are set as follows: the total number of time periods N T = 24 hours, the number of charging stations N CS = 3, the total number of generated scenarios is 10, the deviation penalty cost coefficient = 0.1, the risk preference coefficient , , the confidence level , the charging and discharging efficiency is 95% for both. Finally, the revenue and charging and discharging power conditions of each charging station are obtained.

[0120] Figure 2 It is a comparison chart of the revenue curves of charging station 1 between the retail price (optimized pricing) given by the pricing model of the present invention and the fixed pricing of the prior art. Figure 3 It is a comparison chart of the revenue curves of charging station 2 between the retail price (optimized pricing) given by the pricing model of the present invention and the fixed pricing of the prior art. Figure 4 It is a comparison chart of the revenue curves of charging station 3 between the retail price (optimized pricing) given by the pricing model of the present invention and the fixed pricing of the prior art. Figure 5 It is a comparison chart of the total revenue curves between the retail price (optimized pricing) given by the pricing model of the present invention and the fixed pricing of the prior art. From Figure 2 , Figure 3 , Figure 4 and Figure 5 it can be concluded that the retail price obtained in the method for constructing the pricing model proposed by the present invention results in a greater revenue for the charging stations compared to the fixed pricing of the prior art, verifying that the pricing model provided by the present invention has the function of real-time response and effectiveness in terms of revenue optimization.

[0121] Figure 6 , Figure 7 , Figure 8 respectively show the changes in the charging power and discharging power in the real-time market and the charging power and discharging power of the energy storage system of charging station 1, charging station 2, and charging station 3 of the present invention within 24 hours, Figure 9Shows the variation of the total charging power and discharging power in the real-time market. Figure 10 Shows the variation of the total charging power and discharging power of the energy storage system. It can be learned therefrom that the charging and discharging behaviors of the electric vehicle charging station between different time periods and the real-time market and the energy storage system have obvious fluctuations. The changes in the charging power and discharging power reflect the response ability of the electric vehicle charging station to the power grid, verifying the rationality of the pricing model of the present invention in scheduling the behavior of the electric vehicle charging station. Moreover, the electric vehicle charging station tends to discharge to support the power grid during peak hours, while it tends to charge to store electric energy during off-peak hours. This behavior pattern helps to achieve "peak shaving and valley filling" of the power grid, improving the stability and economy of the power grid.

[0122] In summary, the method and device for constructing the pricing model of the electric vehicle virtual power plant provided by the present invention effectively handle the uncertainty of new energy output through the multi-discrete scenario distributionally robust optimization method. Even in the worst probability distribution case, it can still formulate the optimal incentive electricity price to ensure the economy and robustness of the virtual power plant. By constructing the uncertainty set and confidence constraints, it can still maintain a high revenue and power grid stability under the fluctuation of new energy output. It has strong practical application prospects and can provide a scientific pricing strategy and scheduling plan for the operation of the electric vehicle virtual power plant. By comprehensively considering market bidding, retail price, new energy output and demand response of the electric vehicle charging station, it can help the virtual power plant to maximize revenue and minimize risks in a complex market environment, enabling it to maximize revenue even in the worst case.

[0123] After considering the specification and practicing the embodiments disclosed herein, those skilled in the art will readily conceive of other embodiments of the present invention. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include known common knowledge or conventional technical means in the technical field not disclosed by the present invention. It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A method for constructing a pricing model of an electric vehicle virtual power plant, characterized in that Including: Step 1: Based on the idea of multi-discrete scenario distributionally robust optimization, determine the uncertainty set composed of the typical scenario probability distribution vectors of new energy output prediction errors and the corresponding confidence constraints; Step 2: Determine the bidding volume of the electric vehicle virtual power plant in the day-ahead market, and then optimize based on the uncertainty set and the corresponding confidence constraints to determine the retail price issued to the electric vehicle charging station; Step 3: According to the bidding volume, retail price and real-time market price, adjust the deviation power of the electric vehicle charging station, and optimize the charging and discharging power of the energy storage system and the power transaction in the real-time market; The uncertainty set composed of the typical scenario probability distribution vectors of the new energy output prediction errors is: (1) Among them, is the uncertainty set; is the probability distribution vector of typical scenarios of new energy output prediction error; represents the boundary of the support space of random variables obtained from historical sample data of new energy power generation and load; and are the power purchase and power sales of the electric vehicle virtual power plant in the real-time market under the typical scenario s at time t, respectively; is the confidence level of the fuzzy set of the bulldozer distance method; The confidence level of the fuzzy set of the earth mover's distance method is obtained by the following formula: (2) (3) In the formula, is the probability distribution vector of the reference scenario; Ns is the number of all possible scenarios created through the scenario generation process; is a variable, depending on the actual situation; is the diameter of the support space of the random variable; represents the norm; The confidence constraint is: In the formula, P represents probability.

2. The method for constructing an electric vehicle virtual power plant pricing model according to claim 1, characterized in that In Step 1, use the earth mover's distance method to characterize the uncertainty set composed of the typical scenario probability distribution vectors of the new energy output prediction errors and the corresponding confidence constraints.

3. The method for constructing the pricing model of the electric vehicle virtual power plant according to claim 1, characterized in that Step 2 specifically includes: Determine the objective function of the retail price according to the settlement revenue of the day-ahead market and the revenue from selling electricity to the electric vehicle charging station; determine the constraint conditions of the retail price according to the market rules and the internal supply and demand situation when the electric vehicle virtual power plant participates in the upper-level market.

4. The method for constructing an electric vehicle virtual power plant pricing model according to claim 3, characterized in that, The objective function of the retail price is: (5) Among them, , In the formula, is the set of decision variables in the first stage; represents the time length of each decision period in the optimization process; is the total number of time periods; is the total number of electric vehicle charging stations; is the total number of energy storage systems; and are the clearing electricity prices for selling and purchasing electricity in the day-ahead market at time t, respectively; and are the cleared electric power quantities sold by the electric vehicle virtual power plant to the market and purchased from the market at time t, respectively; and respectively represent the discharge and charge electricity prices of the energy storage system at time t; and respectively represent the discharge and charge powers of the energy storage system at time t; and are the equivalent charging power and discharging power of the electric vehicle charging station at time t; is the retail price released by the electric vehicle virtual power plant at time t; is a 0-1 variable, indicating that the electric vehicle virtual power plant is an electricity producer at time t, indicating that the electric vehicle virtual power plant is an electricity consumer at time t; is a 0-1 variable, indicating that the energy storage system is in a charging state at time t, indicating that the energy storage system is in a discharging state at time t.

5. The method for constructing the pricing model of the electric vehicle virtual power plant according to claim 4, wherein The constraint conditions of the retail price are: (6) (7) (8) (9) (10) (11) (12) (13) In the formula, are respectively the minimum and maximum retail prices allowed by the market rules during the t period; is the daily average retail price threshold; are respectively the maximum cleared power output of the electric vehicle virtual power plant selling to and purchasing from the market during the t period; is the maximum charge-discharge power of the energy storage system during the t period; and respectively represent the charging and discharging efficiencies of the energy storage system; represents the state of the energy storage system, is the initial state of the energy storage system.

6. The method for constructing an electric vehicle virtual power plant pricing model according to claim 5, wherein Step 3 specifically includes: Establish an expected revenue model for the electric vehicle virtual power plant, and the objective function of the expected revenue model is: (14) wherein and are the electricity purchase power and electricity selling power of the electric vehicle virtual power plant in the real-time market under the typical scenario s at time t, respectively, is the electricity purchase price in the real-time market, is the deviation penalty cost coefficient; is the risk preference coefficient; represents the scheduling cost of the energy storage system; Characterize the AC power flow of the network within the electric vehicle virtual power plant, and determine the constraint conditions of the expected revenue model.

7. The method for constructing the pricing model of the electric vehicle virtual power plant according to claim 6, characterized in that The constraint conditions of the expected revenue model are: (15) (16) (17) （18） （19） (20) Wherein, represents the active power of the conventional load under the typical scenario s at time t; is the output of the distributed new energy r under the typical scenario s at time t; is the total number of distributed new energies; is the reactive power injected into the power grid by the electric vehicle virtual power plant at node i, is the reactive power absorbed from the power grid by the electric vehicle virtual power plant at node i; and are the voltage amplitudes of nodes i and j respectively, is the node and is the voltage phase difference between them, is the conductance of the line ; is the susceptance of the line ; represents all the nodes directly connected to node through the line, is the voltage phase angle of node ; are the minimum and maximum values of the node voltage respectively; is the maximum possible output of the distributed new energy r under the typical scenario s at time t.

8. An apparatus for constructing a pricing model of an electric vehicle virtual power plant, characterized in that, Including: The first determination unit is used to determine the uncertainty set composed of the typical scenario probability distribution vectors of the new energy output prediction errors and the corresponding confidence constraints based on the idea of multi-discrete scenario distributionally robust optimization; The second determination unit is used to determine the bidding volume of the electric vehicle virtual power plant in the day-ahead market, and then optimize based on the uncertainty set and the corresponding confidence constraints to determine the retail price issued to the electric vehicle charging station; The adjustment and optimization unit is used to adjust the deviation power of the electric vehicle charging station according to the bidding volume, retail price and real-time market price, and optimize the charging and discharging power of the energy storage system and the power transaction in the real-time market; The uncertainty set composed of the typical scenario probability distribution vectors of the new energy output prediction errors is: (1) Among them, is the uncertainty set; is the probability distribution vector of the typical scenarios of the new energy output prediction error; represents the boundary of the support space of the random variable obtained from the historical sample data of the new energy power generation and the load; and are respectively the electricity purchase power and the electricity sale power of the electric vehicle virtual power plant in the real-time market under the typical scenario s at time t; is the confidence level of the fuzzy set of the bulldozer distance method; The confidence level of the fuzzy set of the earth mover's distance method is obtained by the following formula: (2) (3) In the formula, is the probability distribution vector of the reference scenario; Ns represents the number of all possible scenarios created through the scenario generation process; is a variable, which depends on the actual situation; is the diameter of the support space of the random variable; represents the norm; The confidence constraint is: In the formula, P represents probability.

Images