Multi-interest subject day-ahead electric energy transaction method based on hybrid game theory

Abstract

The invention provides a multi-benefit subject day-ahead electric energy transaction method based on a mixed game theory, and the method comprises the steps: firstly constructing a revenue function and a game model of different benefit subjects based on the mixed game theory; based on this, designing an electricity buyer electricity buying strategy and an electricity seller bidding strategy which consider reference information delay and intelligence thereof in a strategy updating process; furthermore, a centralized electricity market clearing scheme considering electricity price willingness of bottom-layer market participants is designed. The method can effectively improve the income of the electricity selling main body in the market bidding process, prevents an electricity market transaction center from occupying a monopoly position in the electricity price making process, and further enhances the enthusiasm of each benefit main body at the bottom layer to participate in the day-ahead electricity market.

Description

Technical Field

[0001] The present invention relates to the technical field of power market, and in particular to a multi-stakeholder day-ahead power trading method based on hybrid game theory. Background Art

[0002] With the widespread adoption of distributed energy resources such as photovoltaics, wind turbines, and electric vehicles, distribution networks have evolved from traditional passive to active distribution networks, making it possible to achieve localized self-sufficiency in electricity supply and demand. Furthermore, the development of advanced aggregation technologies such as virtual power plants (VPPs) has enabled geographically dispersed, small-unit distributed energy resources to participate in the dispatch of larger power grids as clusters, making localized power self-sufficiency technically feasible. However, due to the limited aggregation capacity of a single VPP, it often struggles to achieve internal supply and demand balance, making power surpluses and shortages more likely. Therefore, a comprehensive power trading mechanism between VPPs is urgently needed to facilitate power trading among multiple entities, achieve regional power self-balancing, and enhance the overall benefits of all participants. Furthermore, studying the impact of competition and cooperation between VPPs on the benefits of all parties involved has important guiding significance for the future development of small-scale power trading markets.

[0003] The mature development of game theory has comprehensively addressed the challenge of depicting the complex relationships between real-world electricity market participants. Non-cooperative game models primarily depict competing interests among several entities; cooperative game models primarily describe cooperative relationships between non-competitive entities that tend to collaborate to accomplish a common task; and Stackelberg games primarily describe master-slave relationships between stakeholders with distinct hierarchical or superior relationships.

[0004] In the existing multi-participant day-ahead electricity trading market system, the market clearing price setting method is mostly determined by the power market trading center alone, without considering the electricity price willingness of other stakeholders, and without fully integrating the principles of market economics into the electricity price setting process. It has monopoly characteristics and is not conducive to the healthy development of electricity market transactions. On the other hand, the existing optimal strategy decision-making method for the underlying participating units does not take into account the complex game situation they face in a competitive environment and the influence of competitors on their own decisions. At the same time, the competitor information used in the strategy update process is the relevant information of the previous iteration round, and there is a time lag. This restricts the efficient and healthy development of distribution network side electricity market transactions and reduces the enthusiasm of the underlying participating units. Summary of the Invention

[0005] In order to effectively improve the enthusiasm of the underlying participants and enhance the fairness of the power market transactions on the distribution network side, the present invention proposes a multi-stakeholder day-ahead power trading method based on hybrid game theory to enhance the fairness of the power trading market on the distribution network side, while increasing the purchase and sale income of the participants, enhancing their enthusiasm for participating in power market transactions, and promoting the healthy development of the small-scale and fragmented power market on the distribution network side. The present invention provides the following technical solutions:

[0006] A multi-stakeholder day-ahead electricity trading method based on hybrid game theory comprises the following steps:

[0007] Step 1: Divide the participants in the day-ahead electricity trading market into electricity buyers and electricity sellers. Consider all electricity buyers as a power purchasing alliance. Use a cooperative game model to construct the power purchasing alliance's profit function. With the goal of maximizing profits, uniformly adjust the elastic load resources with different adjustment costs within the alliance and determine the power purchasing strategy.

[0008] Step 2: The electricity seller uses a non-cooperative game model to construct the electricity seller's profit function, which consists of the following four parts: the profit obtained by the electricity seller from the successful bidding part in the electricity market transaction, the profit obtained from the unsuccessful bidding part absorbed by the power grid company, the power generation cost of the electricity seller, and the grid connection fee paid by the electricity seller to the power grid company for using the transmission line on the grid side; the profit obtained by the electricity seller from the successful bidding part in the electricity market transaction is determined by the expectation of the next round of clearing price under the game situation, and the optimal electricity sales strategy is obtained by solving the profit function.

[0009] Preferably, the profit function of the power purchasing alliance is as follows:

[0010]

[0011] in represents the overall revenue of the power purchasing alliance during period t, and They represent the fixed load and elastically adjustable load during period t, and Indicates the boundary value of the elastic load adjustment amount, represents the initial value of the elastically adjustable load, M* represents the set of electricity buyers, and the variable j is an element in the set M*; represents the adjustment cost of the elastic load, represents the load revenue of electricity buyer j in period t, represents the reward function of the power market trading center to the power buyer, and the variable It indicates the prediction of the future clearing price by the power purchasing alliance and reflects the impact of the change of the power purchasing alliance strategy on future prices.

[0012] Preferably, the prediction of the future clearing price by the power purchasing coalition is obtained by the following iterative process:

[0013] Let denote the impact of the power purchasing coalition's anticipated strategy change on the clearing price:

[0014]

[0015] where the superscript k denotes the iteration number, and μ denotes the adjustment factor. After obtaining the information of the last round clearing price communicated by the power market trading center, the power purchasing coalition will determine its own power purchasing scheme in combination with its own benefit function.

[0016] Preferably, the SCA algorithm is used to convexly approximate the benefit function of the power purchasing coalition, and to solve the optimal power purchasing scheme of the current iteration round.

[0017] Preferably, each power seller adopts a segmented bidding strategy to bid for a total amount of power corresponding to the marginal cost centered on the marginal cost , the bids in the interval are divided into n equal parts, i.e., the bid corresponds to the amount of power , is the bidding strategy of the power seller i at time period t, where is a known quantity, and N* denotes the set of power sellers, and the benefit function of the power seller is:

[0018]

[0019] where α denotes the number of overall competitive situations that the power seller i can face; denotes the probability of occurrence of a certain game situation l, and satisfies , denotes the benefit of the power seller i at time period t under the game situation l, which is composed of the following four parts: and denote the benefits of the power seller i in the power market trading, which are the successfully bid part and the unsuccessfully bid part absorbed by the grid company, respectively; denotes the power generation cost of the power seller; denotes the over-grid fee paid by the power seller i to the grid company due to the use of the transmission line on the grid side; denotes the absorption price of the grid company under the game situation l; denotes the expectation of the power seller under the game situation l for the next round clearing price, denotes the power generation cost function of the power seller; denotes the over-grid fee function submitted by the power seller; and It is used to distinguish the expected winning and unwinning electricity in the bidding strategy of the electricity seller.

[0020] For the constraints, and Indicates the power limit for the corresponding bidding electricity quantity under the segmented electricity price in the power sales side bidding strategy; Indicates the restrictions on the electricity sales quotation.

[0021] Preferably, the electricity seller's expectation of the next round of clearing price under the game situation is obtained through the following iterative process:

[0022] Under different game situations l, the expected clearing price of the next round obtained by the electricity seller i by changing its bidding strategy is:

[0023]

[0024] in It represents the elasticity coefficient, which varies among the participants according to the degree of risk aversion; represents the weighted average electricity price of electricity seller i in the kth and k+1th rounds of bidding; Indicates the degree of dependence of electricity seller i on the change in clearing price caused by other participants; It represents the predicted value of the impact of other competitors' strategies on the change in clearing price in the lth final game scenario faced by electricity seller i.

[0025] Preferably, a composite differential evolution algorithm is used to solve the optimal electricity sales strategy of electricity seller i during the iteration process.

[0026] Secondly, a multi-stakeholder day-ahead electricity trading market clearing method based on hybrid game theory is used to formulate the clearing price in the aforementioned method. The Stackelberg game is used to describe the naturally existing hierarchical relationship between the electricity market trading center and several virtual power plants participating in the electricity trading market. By constructing the profit function of the virtual power plant, a centralized market clearing plan that takes into account the electricity price willingness of electricity market participants is formulated.

[0027] Preferably, the revenue function of the virtual power plant is:

[0028]

[0029]

[0030] The revenue function consists of four parts: the profit from investing the power generated by the power plant into the electricity market, the profit from the power recovered from the electricity market at a penalty price, the profit from the self-sufficiency of each virtual power plant to the grid company, and the profit from collecting the grid access fees from each virtual power plant.

[0031] It is used to quantify the profits brought to the power market trading center by reducing the dispatching burden of the power grid company due to the self-sufficiency of power among virtual power plants. Indicators The definition method is as follows, where Represents the weighting coefficient:

[0032]

[0033] is the clearing price, The unit power generation cost of the power grid company in different time periods; The electricity sales price for the power grid company; It is the penalty electricity price for the electricity seller i that has unsuccessful bids under the clearing price decision.

[0034] Preferably, As the domain of the revenue function of the virtual power plant, The composite differential evolution algorithm is used to calculate the clearing price of the virtual power plant. Solve it.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] The market clearing scheme proposed in the present invention takes into account the electricity price willingness of trading units in the trading market, which can fully consider the electricity price willingness of market participants and disperse the clearing price decision-making power to all market participants, thereby greatly improving the fairness of the electricity trading market.

[0037] The market participation scheme for electricity buyers and sellers proposed in the present invention takes into account factors such as the intelligence of the electricity market participants, the complex game situation faced in the actual bidding process for electricity purchases, and the time lag of information used in the strategic decision-making process in the existing technology. It can adapt to the complex competition and cooperation relationships in reality, improve their own profits in the process of electricity trading, and thus enhance the enthusiasm of various entities to participate in the electricity market transactions on the distribution network side. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0039] Figure 1 It is the main flow chart of the present invention. DETAILED DESCRIPTION

[0040] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0041] To make the above-mentioned objects, features and effects of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments. To avoid confusion, the following explanations are first given of the various terms mentioned in the present invention.

[0042] Day-ahead electricity trading market: referred to as the electricity market in this invention, refers to a centralized electricity market trading scenario organized by the electricity market trading center and including several virtual power plants. This electricity market achieves the supply and demand balance of electricity traded the next day by organizing related transactions one day in advance.

[0043] Grid company: The organizer of the electricity market and the guarantor of electricity transactions, responsible for absorbing excess electricity during the transaction process or making up for the electricity shortage during the transaction process.

[0044] Power Market Trading Center: A department of the power grid company, responsible for organizing the specific trading behaviors of the power market entities. In this invention, the power market and the power grid company are regarded as a community of interests.

[0045] Virtual power plant: An electricity market entity, divided into power purchasing virtual power plants (referred to as power buyers) and power selling virtual power plants (referred to as power sellers) based on the relationship between its internal load demand and total power generation.

[0046] Example 1: A multi-stakeholder day-ahead energy trading method based on hybrid game theory, such as Figure 1 As shown, the following steps are included.

[0047] Step 1: Construct a transaction model and game model for different stakeholders based on hybrid game theory, which includes the classification of participants in the day-ahead electricity trading market, the construction of a game model and load adjustment model for a power-purchasing virtual power plant, and the construction of a game model and bidding model for a power-selling virtual power plant.

[0048] Step 1.1, Classification of the participants in the day-ahead electricity trading market: In the day-ahead market, consider a centralized electricity market trading scenario organized by the electricity market trading center and including several virtual power plants. Based on the power load and power generation unit power forecast, and according to the relationship between internal load demand D and total power generation P, virtual power plants are divided into power purchasing virtual power plants (D>P) and power selling virtual power plants (D<P), with the number of M and N respectively. At the same time, the day-ahead trading cycle is divided into T=24 small cycles, and clearing transactions are organized in each of the 24 time periods. The set of trading periods is .

[0049] Step 1.2, construct the game model and load adjustment model of power-purchasing virtual power plants: consider all power-purchasing virtual power plants as an alliance, and use the cooperative game method to uniformly adjust the elastic load resources with different adjustment costs within the alliance, thereby affecting the market supply and demand relationship.

[0050] After obtaining the total power supply from the power market trading center Afterwards, the power purchasing alliance will coordinate and adjust the elastic load within the alliance based on the overall profit function. The profit function of the power purchasing alliance is as follows:

[0051] (1)

[0052] in represents the overall revenue of the power purchasing alliance during period t, and They represent the fixed load and elastically adjustable load during period t, and Indicates the boundary value of the elastic load adjustment amount, represents the initial value of the elastically adjustable load, M* represents the set of power-purchasing virtual power plants, and the variable j is an element in the set M*.

[0053] represents the adjustment cost of the elastic load, as shown in formula (2), where It represents the benchmark value of the power shortage fee for flexible load in period t, which may vary between different power purchasing virtual power plants. represents the penalty factor for period t; It represents the power shortage ratio of the elastic load in period t, as shown in formula (3).

[0054] (2)

[0055] (3)

[0056] represents the load revenue of the power purchasing virtual power plant j in period t, as shown in formula (4), and is a deterministic parameter.

[0057] (4)

[0058] It represents the reward function of the power market trading center for promoting the consumption of new energy to the power purchaser, as shown in formula (5), where is a deterministic parameter. During peak trading hours, the day-ahead market primarily aims to maintain a balance between supply and demand, so incentives are lower. During off-peak trading hours, however, the power trading center will increase incentives to a certain extent to increase the absorption rate of renewable energy.

[0059] (5)

[0060] variable It is an important part of the load revenue function of the power purchasing alliance, representing the prediction of the power purchasing alliance for the future clearing price and reflecting the impact of the change of the power purchasing alliance strategy on the future price, which will be specifically reflected in the subsequent step 2.

[0061] Step 1.3, construct the game model and bidding model of the power sales virtual power plant: In order to increase the amount of electricity won in the bidding game and reduce the risk of not winning the bid during the transaction process, each power sales virtual power plant (hereinafter referred to as the power seller) adopts a segmented bidding strategy. The corresponding marginal cost Centered and extending vertically , the interval The quotes in the table are divided into n equal parts, that is, the quotes The corresponding power is , is the bidding strategy of electricity seller i in period t, where is a known quantity, and N* represents the set of electricity-selling virtual power plants.

[0062] A non-cooperative game model is used to describe the competitive relationship between electricity sales virtual power plants. Based on the prescribed segmented bidding format and incorporating the core ideas of game theory, a payoff function that comprehensively considers all game situations is described as:

[0063] (6)

[0064] (7)

[0065] (8)

[0066] (9)

[0067] Where α represents the number of overall competitive situations that electricity seller i may face; Represents the probability of a certain game situation l occurring, and satisfies , It represents the revenue of electricity seller i under the game situation l in period t, and consists of the following four parts: and represents the revenue obtained by electricity seller i in the electricity market transaction, which is the successful bidding part in the electricity market transaction and the unsuccessful bidding part absorbed by the grid company; represents the electricity generation cost of the electricity seller; = represents the transmission fee paid by the electricity seller i to the grid company for using the transmission line on the grid side. Formula (10) represents the power generation cost function of the electricity seller, where represents the cost coefficient of the electricity seller; Formula (11) represents the transmission fee function submitted by the electricity seller, and z represents the transmission fee to be paid per unit of electricity.

[0068] (10)

[0069] (11)

[0070] represents the absorption price of the power grid company under the game situation l, and the calculation method is similar to the formula (21) in the subsequent step 3; It is the most important component in formula (6), which represents the electricity seller’s expectation of the next round of clearing price under the game situation. Its iterative formula is shown in formula (16) in the subsequent step 2.2.

[0071] As for the constraints, Equations (7) and (8) represent the power limits of the corresponding bidding electricity under the segmented electricity price in the power sales side bidding strategy; Equation (9) represents the constraints on the power sales side bidding.

[0072] function and It is used to distinguish the expected winning and unwinning electricity in the bidding strategy of the electricity seller, as shown in Equations (12) and (13), where the part of the segmented electricity price that is higher than the predicted clearing price is the unwinning part; conversely, the part that is lower than the clearing price is the winning part.

[0073] (12)

[0074] (13)

[0075] Step 2: Design the purchasing strategy of the power buyer and the bidding strategy of the power seller that take into account the reference information delay and their own intelligence during the strategy update process, including designing the purchasing strategy of the power buyer that takes into account the delay of the clearing electricity price information during the strategy update process, and designing the bidding strategy of the power seller that takes into account the delay of the competitor's bidding strategy information during the strategy update process.

[0076] Step 2.1: Consider the power purchaser's power purchase strategy with the delay of clearing price information during the strategy update process:

[0077] use represents the impact of the power purchasing alliance's expected change in its own strategy on the clearing price, as shown in Equation (14), where the superscript k represents the number of iterations and μ represents the adjustment factor.

[0078] (14)

[0079] After receiving the previous round's clearing price information from the electricity market trading center, the power purchasing alliance will determine its own power purchasing plan based on its own profit function, namely, the alliance's internal flexible load adjustment. Because the power purchasing alliance's objective function is a non-convex quadratic form, the SCA algorithm is used to convexly approximate it to solve the optimal power purchasing plan for the current iteration.

[0080] Step 2.2: Consider the power seller's bidding strategy when competitor bidding strategy information is delayed during the strategy update process: In each round of bidding, the power seller will face different winning bids and clearing prices. Use δ to determine the relationship between the clearing price and the most recent tiered bid, using a three-stage bidding strategy as an example. 、 and It represents the difference between the clearing price and the first, second and third (highest) tiered bidding prices. 、 and The electricity price is declared in three sections, representing the lowest price section, the middle price section and the highest price section.

[0081] (1) When the position relationship between the previous round’s clearing price and the three-stage tiered electricity price satisfies When the three bids are all in a risk-free winning state, the electricity seller will increase the amount of electricity it bids in the high price segment, thereby increasing the next round of clearing prices and thus increasing its bidding revenue. It will adopt a relatively conservative bidding strategy of raising electricity prices, so that the expected value of the next round of market clearing prices after the modified strategy is around In the equation, τ is a deterministic parameter that represents the parameter for dividing the predicted bid price into specific intervals. At the same time, under the condition of the same profit, the scheme with the largest bidding volume in the middle price range will be selected as the optimal scheme.

[0082] (2) When the position relationship between the previous round’s clearing price and the three-stage tiered electricity price satisfies When the bid for the third segment is in a risky winning state, the bids for the first and second segments are in a risk-free winning state. When considering its own profit function, the electricity seller only considers the profit brought by 0.8 times the third segment bidding electricity, and the rest is calculated based on the unsuccessful bidding electricity. It will implement a relatively conservative bidding strategy, so that the predicted next round of market clearing price is Inside.

[0083] (3) When the position relationship between the previous round’s clearing price and the three-stage tiered electricity price satisfies and When the bid for the third segment fails to win the bid, the bidder for the first and second segments will be in a risk-free bidding state. The bidder will not consider the benefits brought by the winning bid for the third segment, but will only increase the bid for the high-priced segment accordingly, increase the next round of clearing prices, and thus increase the winning benefits of the bid for the first and second segments. It will implement a relatively conservative bidding strategy, so that the predicted next round of market clearing prices will be around At the same time, under the same profit conditions, the plan with the largest amount of electricity in the middle price range will be selected as the optimal plan.

[0084] (4) When the position relationship between the previous round’s clearing price and the three-stage tiered electricity price satisfies At this time, the third bid cannot win the bid, the second bid is in a risky bid state, and the first bid is in a risk-free bid state. The bidder will not consider the benefits brought by the third bid electricity, and only consider the profit brought by 0.8 times the second bid electricity. The remaining second bid electricity will be considered as unsuccessful bid electricity. It will implement a relatively conservative bidding strategy, so that the predicted next round of market clearing price is Inside.

[0085] (5) When the relationship between the previous round’s clearing price and the three-stage tiered electricity price position satisfies and At this time, the second and third bids cannot win the bid, and the first bid is in a risk-free state. The bidder will not consider the benefits brought by the second and third bids, but will only increase the bids for the high-priced segments accordingly, increase the next round of clearing prices, and thus increase the benefits of winning the first bid. It will implement a relatively conservative bidding strategy, so that the predicted next round of market clearing prices will be around At the same time, under the condition of the same profit, the plan with the largest amount of electricity in the middle price range will be selected as the optimal plan.

[0086] (6) When the position relationship between the previous round’s clearing price and the three-stage tiered electricity price satisfies At this time, the second and third segments cannot be successful, and the first segment is in a risky bidding state. The bidding party will increase the bidding capacity of the high price segment, increase the clearing price in the next round, and thus increase the probability of winning the first segment. It will no longer consider the benefits of winning the second and third segments, and will implement a relatively aggressive bidding strategy, so that the predicted clearing price in the next round is within , and at the same profit, the scheme with the largest bidding capacity in the middle price segment will be selected as the optimal scheme.

[0087] The above guidelines for developing bidding strategies are public knowledge for each power seller participating in the bidding game. Therefore, when predicting the actions of other participants in the next round of bidding, the previous clearing price transmitted by the power market trading center and the information of the bidding strategy of each party are combined to determine which guideline the other power sellers will adopt, and on this basis, the value of the change in the clearing price caused by the bidding strategy of the other power sellers in the next round is defined. The interval of the predicted clearing price in the next round in the above case is uniformly formatted as , then the predicted value of the impact of the bidding action of the competitor i' on the future clearing price by the power seller i is The calculation method is as follows:

[0088] (15)

[0089] s represents the number of predicted scenarios for the bidding actions of the competitor i', and u represents the total number of these predicted scenarios. For , the probability is , which is determined by the experience value of the power seller i itself. It mainly combines the amount of electricity corresponding to the lowest price segment in the previous bidding strategy of the competitor i' to assess its ability to affect the clearing price in the next round, and thus determine the probability size. When the proportion of the lowest price segment is relatively low, the competitor i' is not capable of increasing the clearing price in the next round, so the probability of reaching a higher level is small.

[0090] After evaluating all participants except itself, a complete estimation space is obtained; then, with the help of probability theory and mathematical statistics, the power seller i faces , a total of kinds of game situations, each of which contains the total impact of the strategies adopted by the remaining power sellers on the clearing price in the next round and the corresponding probability. For a specific scenario, the total impact on the future clearing price and the corresponding probability are calculated. The former is calculated by summing up the impact of all competitors on the future clearing price in this scenario The latter is obtained by adding the probability of each competitor's price impact in this scenario. Finally, through the clustering method, we can get the 10 possible game situations that the electricity seller i may face in the next round of bidding. , and its corresponding probability satisfies , l represents the number of the final game scenario.

[0091] Therefore, taking into account the impact of other participants, the expected clearing price of the next round obtained by the electricity seller i by changing its bidding strategy under different game situations l is:

[0092] (16)

[0093] in It represents the elasticity coefficient, which varies among the participants according to the degree of risk aversion; represents the weighted average electricity price of electricity seller i in the kth and k+1th rounds of bidding; Indicates the degree of dependence of electricity seller i on the change in clearing price caused by other participants; It represents the predicted value of the impact of other competitors' strategies on the change in clearing price in the lth final game scenario faced by electricity seller i.

[0094] Combining the above content, the optimal electricity sales strategy of electricity seller i during the iterative process is solved. Since the objective function is a high-order, nonlinear, and piecewise complex function, the composite differential evolution algorithm (CoDE) is used to solve it.

[0095] Example 2: The dual role of the power market trading center should be further considered in the process of setting clearing prices. It is not only the maintainer of the supply and demand balance of the entire system, but also a stakeholder similar to the power sales virtual power plant. It can also obtain profits by selling electricity directly to power buyers, but the amount of electricity sold is subject to certain restrictions. The Stackelberg game is used to describe the natural upper and lower-level relationship between the power market trading center and several virtual power plants participating in the underlying power trading market. A centralized power market clearing scheme that takes into account the price willingness of the underlying market participants is designed, as shown in Equations (17) and (18).

[0096] (17)

[0097] (18)

[0098] The profit function consists of four parts: the profit obtained by investing one's own electricity into the electricity market, the profit brought by the electricity recovered from the electricity market at a penalty price, the profit brought by the self-sufficiency of the underlying virtual power plants to the power grid company (which forms a community of interests with the electricity market trading center), and the profit brought by collecting grid access fees from each virtual power plant.

[0099] It is used to quantify the profit brought to the power market trading center by reducing the dispatch burden of the power grid company due to the self-sufficiency of electricity among virtual power plants. As shown in formula (19), it is a quadratic function, and its index is The definition of is shown in formula (20). Represents the weighting coefficient.

[0100] (19)

[0101] (20)

[0102] The unit power generation cost of the power grid company in different time periods; The electricity sales price for the power grid company; is the penalty price for the electricity seller i that has unsuccessful bids under the clearing price decision, is the penalty factor, as shown in formula (21). By introducing the penalty price mechanism, the bad behavior of electricity sellers to drive up electricity prices can be effectively curbed, and the healthy development of the electricity market can be promoted.

[0103] (twenty one)

[0104] In order to avoid monopoly and unfair trading problems caused by the market clearing price being determined by the electricity market trading center, part of the market transaction pricing power will be transferred to each virtual power plant to increase the enthusiasm of all participants and promote the market-oriented development of electricity trading.

[0105] Regarding the formation process of the electricity price willingness of the electricity seller: first, the electricity seller's quotations are sorted in order from low to high and redefined as , and at the same time summarize the electricity quantities under the same electricity price; then, remove the price segment corresponding to the lowest price segment and the largest part of the bidding electricity quantity; finally, calculate the weighted average electricity price of the remaining part. The specific expression is shown in formula (22), where m is the subscript of the electricity price corresponding to the highest electricity value.

[0106] (twenty two)

[0107] As for the formation process of the electricity price willingness on the power purchaser side: first calculate the supply and demand ratio of the power purchaser and seller in a certain iteration, and then refer to the macro electricity price law under different supply and demand ratios formed by the power market trading center to calculate the electricity price willingness formed on the power purchaser side .

[0108] (twenty three)

[0109] (twenty four)

[0110] By integrating Equation (22) and Equation (24), the comprehensive electricity price willingness of the underlying virtual power plant can be derived, as shown in Equation (25).

[0111] (25)

[0112] (26)

[0113] Adjustment Factor Will have an impact on price willingness, the larger Favors the seller. Usually, The decrease means that the supply is relatively tight. According to the macro-control mechanism, this will cause the price willingness to shift to the seller side.

[0114] Considering the non-convexity and discontinuity of the objective function, the optimal solution in the iterative process is related to the definition domain of its optimization variables; therefore, As its definition domain, the underlying electricity price intention can be incorporated into the process of setting the market clearing price. Similar to the calculation method on the electricity seller side, the compound differential evolution algorithm (CoDE) is used to solve the clearing price.

[0115] After receiving the market-clearing price and related bidding information from the power market trading center, both buyers and sellers adjust their strategies and upload them to the power market trading center again. This leads to repeated bidding and negotiation between the upper and lower levels until an equilibrium solution is reached. When the final equilibrium solution is reached, all participants can obtain acceptable and substantial benefits.

[0116] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein with equivalents. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A multi-stakeholder day-ahead electricity trading method based on hybrid game theory, characterized in that: The following steps are involved: Step 1: Divide the participants in the day-ahead electricity trading market into electricity buyers and electricity sellers. Consider all electricity buyers as a power purchasing alliance. Use a cooperative game model to construct the power purchasing alliance's profit function. With the goal of maximizing profits, uniformly adjust the elastic load resources with different adjustment costs within the alliance and determine the power purchasing strategy. Step 2: The electricity seller uses a non-cooperative game model to construct the electricity seller's profit function, which consists of the following four parts: the profit obtained by the electricity seller from the successful bidding part in the electricity market transaction, the profit obtained from the unsuccessful bidding part absorbed by the power grid company, the power generation cost of the electricity seller, and the grid connection fee paid by the electricity seller to the power grid company for using the transmission line on the grid side; the profit obtained by the electricity seller from the successful bidding part in the electricity market transaction is determined by the expectation of the next round of clearing price under the game situation, and the optimal electricity sales strategy is obtained by solving the profit function.

2. The multi-stakeholder day-ahead power trading method based on hybrid game theory according to claim 1, characterized in that: The revenue function of the power purchasing alliance is as follows: ； in represents the overall revenue of the power purchasing alliance during period t, and They represent the fixed load and elastically adjustable load during period t, and Indicates the boundary value of the elastic load adjustment amount, represents the initial value of the elastically adjustable load, M* represents the set of electricity buyers, and the variable j is an element in the set M*; represents the adjustment cost of the elastic load, represents the load revenue of electricity buyer j in period t, represents the reward function of the power market trading center to the power buyer, and the variable It indicates the prediction of the future clearing price by the power purchasing alliance, and reflects the impact of the change of the power purchasing alliance strategy on future prices.

3. The multi-stakeholder day-ahead power trading method based on hybrid game theory according to claim 2, characterized in that: The PPA's forecast of future clearing prices is obtained through the following iterative process: use The impact of the PPC's expected changes in strategy on the clearing price is shown below: ； The superscript k represents the number of iterations, and μ represents the adjustment factor. After obtaining the previous round of clearing price information conveyed by the electricity market trading center, the power purchasing alliance will determine its own power purchasing plan based on its own profit function.

4. The multi-stakeholder day-ahead power trading method based on hybrid game theory according to claim 3, characterized in that: The SCA algorithm is used to perform convex approximation on the profit function of the power purchasing alliance and solve the optimal power purchasing plan for the current iteration round.

5. The multi-stakeholder day-ahead power trading method based on hybrid game theory according to claim 1, characterized in that: Each electricity seller adopts a segmented bidding strategy to bid for the total electricity The corresponding marginal cost Centered and extending vertically , the interval The quotes in the table are divided into n equal parts, that is, the quotes The corresponding power is , is the bidding strategy of electricity seller i in period t, where is a known quantity, N* represents the set of electricity sellers, and the revenue function of the electricity sellers is: ； Where α represents the number of overall competitive situations that electricity seller i may face; Represents the probability of a certain game situation l occurring, and satisfies , It represents the revenue of electricity seller i under the game situation l in period t, and consists of the following four parts: and represents the revenue obtained by electricity seller i in the electricity market transaction, which is the successful bidding part in the electricity market transaction and the unsuccessful bidding part absorbed by the grid company; represents the electricity generation cost of the electricity seller; It represents the transmission fee paid by the electricity seller i to the grid company for using the transmission line on the grid side; represents the absorption price of the power grid company under the game situation l; It indicates the electricity seller’s expectation of the next round of clearing price under the game situation. represents the electricity generation cost function of the electricity seller; Function representing the transmission fee paid by the electricity seller; function and Used to distinguish the expected winning and unwinning electricity in the electricity seller's bidding strategy; For the constraints, and Indicates the power limit for the corresponding bidding electricity quantity under the segmented electricity price in the power sales side bidding strategy; Indicates the restrictions on the electricity sales quotation.

6. The multi-stakeholder day-ahead power trading method based on hybrid game theory according to claim 5, characterized in that: The electricity seller's expectation of the next round of clearing price under the game situation is obtained through the following iterative process: Under different game situations l, the expected clearing price of the next round obtained by the electricity seller i by changing its bidding strategy is: ； in It represents the elasticity coefficient, which varies among the participants according to the degree of risk aversion; represents the weighted average electricity price of electricity seller i in the kth and k+1th rounds of bidding; Indicates the degree of dependence of electricity seller i on the change in clearing price caused by other participants; It represents the predicted value of the impact of other competitors' strategies on the change in clearing price in the lth final game scenario faced by electricity seller i.

7. The multi-stakeholder day-ahead power trading method based on hybrid game theory according to claim 6, characterized in that: The composite differential evolution algorithm is used to solve the optimal electricity sales strategy of electricity seller i in the iterative process.

8. A multi-stakeholder day-ahead electricity trading market clearing method based on hybrid game theory, used to formulate the clearing price in any of the methods described in claims 1-7, characterized in that: The Stackelberg game is used to describe the naturally existing hierarchical relationship between the electricity market trading center and several virtual power plants participating in the electricity trading market. By constructing the profit function of the virtual power plants, a centralized market clearing plan that takes into account the electricity price willingness of electricity market participants is formulated.

9. The method for clearing a multi-stakeholder day-ahead power trading market based on hybrid game theory according to claim 8, characterized in that: The revenue function of the virtual power plant is: ； ； The revenue function consists of four parts: the profit from investing the power generated by the power plant into the electricity market, the profit from the power recovered from the electricity market at a penalty price, the profit from the self-sufficiency of each virtual power plant to the grid company, and the profit from collecting the grid access fees from each virtual power plant. It is used to quantify the profits brought to the power market trading center by reducing the dispatching burden of the power grid company due to the self-sufficiency of power among virtual power plants. Indicators The definition method is as follows, where Represents the weighting coefficient: ； is the clearing price, The unit power generation cost of the power grid company in different time periods; The electricity sales price for the power grid company; It is the penalty electricity price for the electricity seller i that has unsuccessful bids under the clearing price decision.

10. The method for clearing a multi-stakeholder day-ahead power trading market based on hybrid game theory according to claim 9, characterized in that: by As the domain of the revenue function of the virtual power plant, The composite differential evolution algorithm is used to calculate the clearing price of the virtual power plant. Solve it.

Images