A shared energy storage service game pricing method

Abstract

The application provides a shared energy storage service game pricing method, first, a P2P energy transaction pricing mechanism is established based on a shadow price; second, a double-layer game model is constructed with the P2P transaction price as prior information, and the double-layer game model takes the maximum expected income of an upper operator and the minimum comprehensive power consumption cost of a lower producer and consumer as targets; finally, the double-layer model is simplified into a single-layer MIQP problem for efficient solution through a scenario method, a strong duality theory, KKT conditions and a large M method.

Description

Technical Field

[0001] This invention relates to the field of new energy technology, specifically to a game-theoretic pricing method for shared energy storage services. Background Technology

[0002] In a producer-consumer interactive energy market, shared energy storage capacity leasing is a crucial link in balancing energy supply and demand and improving the utilization rate of energy storage resources. Its pricing rationality directly determines the efficiency of energy storage resource allocation and the balance of interests between the transacting parties. However, current shared energy storage capacity leasing pricing typically does not incorporate peer-to-peer (P2P) energy trading factors among producers and consumers. This leads to pricing ignoring the actual supply and demand interaction in the market, potentially resulting in pricing deviating from a reasonable range. Furthermore, shared energy storage operators struggle to obtain accurate data on individual producer-consumers' power generation, consumption, and regulation capabilities, facing pricing dilemmas in incomplete information scenarios. This can easily lead to pricing deviating from actual market supply and demand, making it difficult to effectively guarantee the profitability of shared energy storage. Existing shared energy storage capacity leasing pricing models based on two-layer game theory have complex solution processes, and their solutions usually require the aggregation of private information such as individual producer-consumers' power generation and consumption, which can easily lead to information leakage and is difficult to adapt to distributed producer-consumer interactive market scenarios. Summary of the Invention

[0003] This invention addresses the shortcomings of existing technologies by providing a game-theoretic pricing method for shared energy storage services. It includes a producer-consumer energy trading pricing mechanism based on shadow prices, characterizing the P2P energy trading price mechanism; constructing a two-layer game model for shared energy storage service pricing by combining P2P energy trading prices; and finally simplifying the two-layer game model into a single-layer optimization problem for solution, thus balancing economic efficiency and practicality.

[0004] The technical solution adopted by this invention to address the shortcomings of the prior art is as follows:

[0005] A game-theoretic pricing method for shared energy storage services, comprising the following steps:

[0006] Step 1: Establishing a P2P energy trading pricing mechanism:

[0007] Based on the KKT conditional dual variable (shadow price) of the producer-consumer optimization model, the price of time-sharing P2P energy trading for producers-consumers is calculated, which serves as the quantitative basis for pricing shared energy storage capacity leasing.

[0008] Step 2: Construct a two-layer optimization model for pricing shared energy storage capacity leasing:

[0009] For scenarios with incomplete information between shared energy storage operators and prosumers, based on Bayesian sequential game theory, and using the P2P energy trading price as prior information, a two-layer game model is constructed with the goal of maximizing the expected revenue of the upper-layer shared energy storage operators and minimizing the comprehensive electricity cost of the lower-layer prosumer group. The power capacity leasing price curve and storage capacity leasing price curve of shared energy storage are determined by solving the game equilibrium.

[0010] Step 3: Simplification and solution of the two-layer model:

[0011] The two-layer game model is transformed into a single-layer MIQP problem by employing the scenario method, KKT conditions, strong duality theory, and the Big M method, thus completing the model solution.

[0012] The beneficial effects that can be achieved by adopting the above-mentioned technical solution in this invention are:

[0013] A two-layer pricing game model that comprehensively considers the P2P energy trading mechanism and does not rely on the privacy information of producers and consumers is proposed, and an efficient engineering solution method adapted to it is formed. Attached Figure Description

[0014] Figure 1 This is a market architecture diagram of shared energy storage, prosumers, and the power grid proposed in this invention;

[0015] Figure 2 This is a framework diagram of the Bayesian sequential game bi-level optimization model proposed in this invention. Detailed Implementation

[0016] 1. Establish an operational model for shared energy storage, a market between producers / consumers and the power grid.

[0017] The research scenario architecture diagram is as follows: Figure 1 As shown, the entire market consists of a centralized shared energy storage system and N prosumers. The centralized shared energy storage system is invested in, constructed, and operated by a third party. Its profit model is as follows: the system leases its capacity to prosumers, who can then utilize it within the leased capacity. The centralized shared energy storage system charges a rental fee. When a prosumer leases shared energy storage, it essentially owns a "virtual energy storage" unit. During the lease period, the prosumer can send charge and discharge commands to the shared energy storage, which then performs the charging and discharging operations, considered as the prosumer's own charging and discharging. Prosumers have built distributed photovoltaic and wind power systems on top of their existing electricity loads. Some prosumers also have a small amount of their own energy storage and interact with the grid, purchasing or selling electricity to the grid. Prosumers engage in P2P energy trading, selling excess electricity to other prosumers, storing it in their own energy storage, or storing it in leased energy storage, thereby generating more revenue.

[0018] 2. P2P Energy Trading Pricing Mechanism

[0019] This invention proposes a P2P energy trading pricing mechanism based on shadow prices. To implement this pricing mechanism, it is first necessary to establish an optimization model for prosumers and consumers, and determine the transaction price by analyzing the dual variable (shadow price) in its KKT conditions.

[0020] The P2P energy trading price mentioned above serves only as a quantitative basis for pricing shared energy storage capacity leasing. The core pricing objects are the power capacity leasing price and storage capacity leasing price of shared energy storage.

[0021] 2.1 Construction of the Prosumer Optimization Model

[0022] Let S be the set of producers and consumers. Let i represent the i-th producer-consumer; T be the set of transaction periods. This represents the t-th time period; For scene collection, Indicates the first There are 3 scenarios, and the probability of each scenario occurring is... ,satisfy All producers and consumers simultaneously make decisions with the goal of minimizing their own total cost, forming a generalized Nash game. Through KKT conditional transformation, this generalized Nash game problem can be transformed into an equivalent quadratic programming problem:

[0023]

[0024] in, For producers and consumers The decision variable vector; Let i be the vector of decision variables for all prosumers except prosumer i. For producers and consumers The decision variables that do not affect other producers and consumers (including their own energy storage charging and discharging power, transferable load, etc.); For producers and consumers The decision variables that affect other prosumers (including P2P transaction power, shared energy storage usage power, etc.); For producers and consumers The cost function does not affect other prosumers; For producers and consumers The impact on the cost function of other prosumers.

[0025] The constraints for this optimization problem are:

[0026]

[0027]

[0028] in, For producers and consumers The independent constraint coefficient matrix; For producers and consumers The coupling constraint coefficient matrix; For producers and consumers The constraint right-hand side vector; For producers and consumers The shared constraint independent term coefficient matrix; For producers and consumers is the coefficient matrix of the shared constraint coupling terms; f is the vector of the right-hand side of the shared constraint terms.

[0029] 2.2 Pricing Mechanism Based on Shadow Prices

[0030] Since the buyer's expenditure is equal to the seller's income in any P2P energy transaction, the P2P transaction price does not affect the final generalized Nash equilibrium solution (the optimal decisions of each participant and the energy supply and demand balance). However, a reasonable pricing mechanism can ensure that each producer and consumer is willing to participate in the transaction and that the transaction price reflects the market supply and demand relationship, embodying the time-sharing value of energy. This invention borrows the meaning of shadow price of even variables from economics for pricing.

[0031] Select the dual variable of the shared constraint (3) in problem (1). Reflecting the value of energy. If And producers and consumers For the energy buyer in a P2P energy transaction, the absolute value of the dual variable means: the producer and consumer at time t. From producers and consumers Purchase more units of energy, producers and consumers Costs that can be saved; at this time, the overall market supply is less than demand, and the P2P energy trading market is in a seller's market state. If And producers and consumers For the energy seller in a P2P energy transaction, the dual variable is defined as: the producer and consumer at time t. To consumers Multiple sales unit energy, producers and consumers This can increase profits; at this time, the P2P energy trading market is in a state of oversupply and is a buyer's market.

[0032] in, The dual variable of the shared constraint (3) is also called the shadow price; Indicates producer-consumer and consumers .

[0033] When the market is a seller's market, energy sellers should reap higher profits; therefore, energy trading centers should raise market prices. To adjust the energy trading price by an amount, we have:

[0034]

[0035] in, This is for calculating the maximum value. This indicates that all producers and consumers are... dual variables Take the maximum value; Indicates in 0 and Take the larger one. hour, ;when hour, .

[0036] When the market is a buyer's market, energy buyers should pay less, therefore energy trading centers should suppress market prices. To adjust the energy trading price by a certain amount, we have:

[0037]

[0038] in, Indicates in 0 and Take the larger one. hour, ;when hour, .

[0039] Ultimately, the energy trading center adjusted its pricing as follows:

[0040]

[0041] in, The P2P transaction price for time period t; The P2P transaction price before adjustment for time period t; The price increase amount for time period t; The amount of price reduction for time period t; The price adjustment factor satisfies Equation (6) indicates that the final P2P transaction price equals the initial price plus the adjustment value after subtracting the downward adjustment. Each producer and consumer settles payment according to the new pricing and the optimized P2P energy transaction volume. By utilizing dual variables, on the one hand, it ensures that all producers and consumers can profit from P2P energy transactions, and on the other hand, it better reflects the time value of energy, using price to reflect the supply and demand relationship at different times.

[0042] 3. Two-tier optimization model for pricing shared energy storage capacity leasing

[0043] Building upon the established P2P energy trading pricing mechanism, this invention further constructs a two-layer optimization model for shared energy storage capacity leasing pricing. This model addresses scenarios where operators lack access to accurate data on individual producers and consumers, employing a Bayesian sequential game framework.

[0044] The incomplete information scenario specifically refers to the fact that shared energy storage operators only obtain probability distribution information on the electricity generation and consumption, regulation capacity, and usage costs of the prosumer group, and cannot obtain accurate data on individual prosumers.

[0045] 3.1 Incomplete Information Assumption

[0046] While shared energy storage cannot obtain information on the power generation, consumption, and regulation capabilities of every individual producer and consumer, it does possess incomplete information about the overall market's power generation, consumption, and regulation capabilities, as well as the probability distribution of shared energy storage usage costs. This information can be mathematically expressed as follows:

[0047]

[0048]

[0049] in, In the scene The total renewable energy power generation of the downstream consumer group during time period t; In the scene Lower-level consumers The spontaneous power generation during time period t; In the scene The total baseline load power of the consumer group in time period t; In the scene Lower-level consumers The electrical load during time period t.

[0050] The estimated unit cost of shared energy storage for all producers and consumers using shared energy storage is: The estimated value of the overall transferable load cost coefficient for producers and consumers is These parameters all follow a certain probability distribution.

[0051] 3.2 Upper-level model: Maximizing the expected revenue of shared energy storage operators

[0052] The upper-level model, aiming to maximize the expected revenue of shared energy storage operators, can be expressed as:

[0053]

[0054] in, To share the benefits of energy storage; The price parameter for storage capacity (upper-level decision variable); The price parameter for power capacity (upper-level decision variable); Shared energy storage capacity leased by the total prosumer group; The total shared energy storage capacity leased by the prosumer group; Let be the expectation operator, representing a weighted average over all scenarios.

[0055] The constraints of the upper-level model are:

[0056]

[0057] in, This indicates finding the minimum value of the decision variable x; x is a vector containing the decision variables of all producers and consumers. Let be the comprehensive electricity cost function for the prosumer group. Equation (10) represents the capacity decision for shared energy storage. and It must be within the set of optimal solutions that minimize the costs of prosumers and consumers, which reflects the master-slave relationship in a two-layer game.

[0058] 3.3 Lower-level model: Minimizing the overall electricity cost for the prosumer group

[0059] The lower-level model, aiming to minimize the overall electricity cost for all producers and consumers, can be expressed as:

[0060]

[0061] in, The overall electricity cost for the prosumer group; The grid interaction cost for all producers and consumers; Shared energy storage costs for all producers and consumers; For the transferable load cost for all producers and consumers; The cost of energy storage for all producers and consumers.

[0062] The detailed expressions for each cost item are as follows:

[0063] (1) Grid interaction cost:

[0064]

[0065] in, The total power purchased by the consumer group from the grid during time period t; The total power sold to the grid by the prosumer group during time period t; The grid purchase price per unit of electricity during time period t; Let t be the unit price of electricity sold by the power grid during the time period t.

[0066] (2) Cost of shared energy storage:

[0067]

[0068] in, Shared energy storage capacity leased to prosumer groups; Shared energy storage capacity leased to prosumer groups; Estimate the unit power cost of shared energy storage for all producers and consumers using shared energy storage; The charging power of shared energy storage for the consumer group during time period t; The discharge power of the consumer group on the shared energy storage during time period t. To share the cost of energy storage capacity leasing, Cost of using shared energy storage power.

[0069] (3) Cost of transferable load:

[0070]

[0071] in, An estimate of the cost coefficient of shared energy storage for the overall transferable load of producers and consumers; Let be the transferable load power of the prosumer group in time period t. This formula is derived through the KKT conditions, compressing the transferable load decision of each prosumer into the decision of the prosumer group as a whole.

[0072] (4) Cost of self-owned energy storage:

[0073]

[0074] in, The unit power cost for producers and consumers to use their own energy storage as a whole; The charging power of the consumer group for its own energy storage during time period t; The discharge power of the consumer group relative to its own energy storage during time period t.

[0075] 3.4 Constraints

[0076] The lower-level model must satisfy the following constraints:

[0077] (1) Energy balance constraint:

[0078]

[0079]

[0080] in, The net load power of the prosumer group during time period t; The total baseline load power of the producer-consumer group in time period t; The charging power of the consumer group for its own energy storage during time period t; The discharge power of the consumer group to its own energy storage during time period t; The total renewable energy power generation of the prosumer group during time period t; The transferable load power of the consumer group during time period t.

[0081] (2) Shared energy storage constraints:

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089] in, Shared energy storage capacity for rent by consumer groups during a given time period (t); Shared energy storage capacity leased by prosumer groups during time period t-1; For the charging and discharging efficiency of shared energy storage; The charging power of shared energy storage for the consumer group during time period t; The discharge power of the consumer group on the shared energy storage during time period t; The duration of the time period; Shared energy storage capacity leased to prosumer groups; For the shared energy storage charging status of the consumer group during the time period t; Shared energy storage capacity leased to prosumer groups; This is the lower limit of the state of charge of shared energy storage; This represents the upper limit of the shared energy storage state of charge. For shared initial energy storage capacity; For producers and consumers Leased shared energy storage capacity; For producers and consumers Leased shared energy storage capacity; The total storage capacity of the shared energy storage system; The total power capacity of the shared energy storage system.

[0090] (3) Transferable load constraints:

[0091]

[0092] in, Let be the transferable load power of the producer-consumer group in time period t. This constraint indicates that the sum of transferable load in all time periods is 0, that is, the total load transfer is balanced.

[0093] (4) Constraints on self-owned energy storage:

[0094]

[0095]

[0096]

[0097]

[0098]

[0099] in, The self-owned energy storage capacity of the prosumer group during time period t; The energy storage capacity owned by the prosumer group during time period t-1; The charging and discharging efficiency of its own energy storage; The charging power of the consumer group for its own energy storage during time period t; The discharge power of the consumer group to its own energy storage during time period t; The duration of the time period; For the self-owned energy storage capacity of the prosumer group; The state of charge of the consumer group's own energy storage during time period t; For the self-owned energy storage capacity of the prosumer group; This is the lower limit of the state of charge of the self-storage energy; This represents the upper limit of the state of charge of the self-owned energy storage.

[0100] 4. Simplification and solution of the two-layer model

[0101] 4.1 Problem Abstraction

[0102] To facilitate the solution, the Bayesian sequential game problem is first abstracted into a generalized mathematical problem. Let... For the decision variables of the prosumers as a whole, For decision variables of shared energy storage. The decision variable represents the cost of the total cost to producers and consumers as a whole, which is a quadratic function. Let $\mathbf$ represent the decision variables that are linearly related to the costs of the total producers and consumers. This Bayesian sequential game problem can be abstracted as follows:

[0103]

[0104]

[0105]

[0106] in, Both are parameter matrices; For parameter vectors; For the dual variables of the overall constraint on prosumers; These represent the estimated parameters related to the overall prosumer-consumer relationship of shared energy storage. Equation (31) is the upper-level objective function, Equation (32) is the optimality condition of the lower-level problem, and Equation (33) is the constraint condition of the lower-level problem. For the dual variable of constraint (33).

[0107] 4.2 Duality Problem and Strong Duality Theory

[0108] In the overall optimization problem that prosumers need to consider, y is already given by the shared energy storage, and therefore is a constant for the prosumers as a whole. The prosumer problem is a quadratic programming (QP) problem. Its dual problem is:

[0109]

[0110]

[0111]

[0112] Equation (34) is the dual objective function, Equation (35) is the dual feasibility constraint, and Equation (36) is the condition that the gradient is zero. Representation matrix The inverse matrix; Representing vectors The transpose of .

[0113] The original problem is a convex optimization problem. According to strong duality theory, at the optimal solution, the duality gap between the minimum value of the original problem and the maximum value of the dual problem is 0. Therefore, we can deduce that:

[0114]

[0115]

[0116] This allows us to identify the non-convex bilinear term in the shared energy storage objective function. Transform it into a quadratic problem and eliminate the bilinear terms.

[0117] 4.3 Scenario-based transformation

[0118] For prosumer models containing boundary information about probability distributions, a scenario-based approach can be used. Shared energy storage samples based on its available probability distribution information to generate several scenarios, and then optimizes the calculation of capacity rental unit prices suitable for these scenarios. Assuming the number of scenarios is K, the optimization problem can be written as:

[0119]

[0120]

[0121]

[0122] Where K is the total number of scenes; For the k-th scene; We assign weights to each scenario (assuming equal probability for each scenario). Through the scenario-based transformation, the original Bayesian problem is transformed into a deterministic problem.

[0123] 4.4 KKT Conditions and Big M Method Conversion

[0124] Considering that the producer-consumer problem is a QP problem, the KKT conditions can be used to transform it into constraints for a shared energy storage problem, thus converting the sequential game problem into a single-layer model. For the complementary relaxation constraints in the KKT conditions, the Big M method can be used for linearization, further simplifying the solution.

[0125] The final problem is transformed into an MIQP problem:

[0126]

[0127]

[0128]

[0129]

[0130]

[0131] in, Let M be the coefficient matrix of the quadratic term; M is a sufficiently large constant introduced by the Big M method; The 0-1 variable vector is introduced. Equation (43) is the gradient zero condition (the derivative of the objective function with respect to the decision variables is 0); Equation (44) is the dual feasibility condition; Equation (45) is the dual variable nonnegativity condition and the upper bound of the Big M method; Equation (46) is the complementary relaxation condition after linearization of the Big M method, which includes the product constraint. Linearization.

[0132] 4.5 Parameter Adjustment for Variable Pricing Rules

[0133] The solution for shared energy storage yields the capacity leasing unit price under a fixed pricing rule. Subsequently, it is possible to estimate the overall capacity of shared energy storage leased by producers and consumers. and The expected value. Finally, shared energy storage is priced according to a pre-defined pricing rule slope. Adjust the parameters as follows:

[0134]

[0135]

[0136] in, The adjusted storage capacity price parameters; The storage capacity and price parameters before adjustment; The slope of the price of storage capacity as a function of the amount of capacity leased. The expected value of shared energy storage capacity for the entire production and consumption process; The adjusted power capacity price parameters; The power capacity and price parameters before adjustment; The slope of the price of power capacity as a function of the amount of capacity leased. The expected value of shared energy storage capacity for producers and consumers as a whole.

[0137] This pricing rule implements "variable pricing": when the shared energy storage capacity leased by the prosumer group is large, the price is lowered accordingly to promote leasing; when the leased capacity is small, the price is raised accordingly to ensure returns. This mechanism can reduce pricing risk, especially the risk of reduced leasing and lower returns when the price is set high.

[0138] The game-theoretic pricing method for shared energy storage services proposed in this invention includes the following steps:

[0139] 1. Establishment of a P2P energy trading pricing mechanism. In the context of an energy market where producers and consumers interact, a shadow price is introduced to depict the scarcity of energy supply and demand at different times, forming a benchmark price for P2P transactions and ensuring that the transaction price matches the actual value of energy.

[0140] 2. Construct a two-layer optimization model for shared energy storage capacity leasing pricing. In the case of incomplete information where operators cannot obtain accurate data on individual producers and consumers, and in conjunction with the P2P energy trading price in step (1), construct a Bayesian sequential game two-layer optimization model with the goal of maximizing the expected revenue of the upper-layer shared energy storage operators and minimizing the comprehensive electricity cost of all producers and consumers in the lower layer.

[0141] 3. Simplification and Solution of the Two-Layer Model. Based on the probability distribution of producer-consumer group data, deterministic scenarios are generated. Through dual transformation, scenario method transformation, and KKT conditions (optimal solution decision conditions in convex optimization problems), the two-layer Bayesian sequential game model is simplified into a single-layer mixed integer quadratic programming (MIQP) problem. A commercial solver is used to complete the solution, so as to simultaneously determine the optimal P2P transaction price and the shared energy storage leasing price curve, taking into account both economic efficiency and practicality.

Claims

1. A game-theoretic pricing method for shared energy storage services, characterized in that, The steps are as follows: Step 1: Establishing a P2P energy trading pricing mechanism: Based on the KKT conditional dual variable (shadow price) of the producer-consumer optimization model, the price of time-sharing P2P energy trading for producers-consumers is calculated, which serves as the quantitative basis for pricing shared energy storage capacity leasing. Step 2: Construct a two-layer optimization model for pricing shared energy storage capacity leasing: For scenarios with incomplete information between shared energy storage operators and prosumers, based on Bayesian sequential game theory, and using the P2P energy trading price as prior information, a two-layer game model is constructed with the goal of maximizing the expected revenue of the upper-layer shared energy storage operators and minimizing the comprehensive electricity cost of the lower-layer prosumer group. The power capacity leasing price curve and storage capacity leasing price curve of shared energy storage are determined by solving the game equilibrium. Step 3: Simplification and solution of the two-layer model: The two-layer game model is transformed into a single-layer MIQP problem by employing the scenario method, KKT conditions, strong duality theory, and the Big M method, thus completing the model solution.

2. The game-theoretic pricing method for shared energy storage services according to claim 1, characterized in that, In step 1, to implement this pricing mechanism, it is first necessary to establish an optimization model for prosumers. The transaction price is determined by analyzing the dual variable (shadow price) in the KKT conditions. The specific method is as follows: 1.1 Construction of the Prosumer Optimization Model Let S be the set of producers and consumers. Represents the i-th producer-consumer; T is the set of trading periods. This represents the t-th time period; For scene collection, Indicates the first There are 3 scenarios, and the probability of each scenario occurring is... ,satisfy All producers and consumers simultaneously make decisions with the goal of minimizing their own total cost, forming a generalized Nash game. Through KKT conditional transformation, this generalized Nash game problem can be transformed into an equivalent quadratic programming problem: in, For producers and consumers The decision variable vector; Let i be the vector of decision variables for all prosumers except prosumer i. For producers and consumers The decision variables that do not affect other prosumers' decisions; For producers and consumers Influencing decision-making variables of other prosumers; For producers and consumers The cost function does not affect other prosumers; For producers and consumers The cost function that affects other prosumers, The constraints for this optimization problem are: in, For producers and consumers The independent constraint coefficient matrix; For producers and consumers The coupling constraint coefficient matrix; For producers and consumers The constraint right-hand side vector; For producers and consumers The shared constraint independent term coefficient matrix; For producers and consumers The coefficient matrix of the shared constraint coupling terms; f is the vector of the right-hand side terms of the shared constraints. 1.2 Pricing Mechanism Based on Shadow Prices Select the dual variable of the shared constraint (3) in problem (1). Reflecting the value of energy, if And producers and consumers For the energy buyer in a P2P energy transaction, the absolute value of the dual variable means: the producer and consumer at time t. From producers and consumers Purchase more units of energy, producers and consumers Cost savings are possible; at this time, the overall market supply is insufficient to meet demand, and the P2P energy trading market is in a seller's market state. And producers and consumers For the energy seller in a P2P energy transaction, the dual variable is defined as: the producer and consumer at time t. To consumers Multiple sales unit energy, producers and consumers This could increase profits; at this point, the P2P energy trading market is experiencing a supply exceeding demand, and the market is in a buyer's market state. in, The dual variable of the shared constraint (3) is also called the shadow price; Indicates producer-consumer and consumers , When the market is a seller's market, energy sellers should reap higher profits; therefore, energy trading centers should raise market prices. To adjust the energy trading price by an amount, we have: in, This is for calculating the maximum value. This indicates that all producers and consumers are... dual variables Take the maximum value; Indicates in 0 and Take the larger one. hour, ;when hour, , When the market is a buyer's market, energy buyers should pay less, therefore energy trading centers should suppress market prices. To adjust the energy trading price by a certain amount, we have: in, Indicates in 0 and Take the larger one. hour, ;when hour, , Ultimately, the energy trading center adjusted its pricing as follows: in, The P2P transaction price for time period t; The P2P transaction price before time period t is adjusted; The price increase amount for time period t; The amount of price reduction for time period t; The price adjustment factor satisfies Equation (6) indicates that the final P2P transaction price is equal to the initial price plus the adjustment value after the upward adjustment minus the downward adjustment. Each producer and consumer makes payment and settlement according to the new pricing and the optimized P2P energy transaction volume.

3. The game-theoretic pricing method for shared energy storage services according to claim 1, characterized in that, In step 2, the two-layer optimization model addresses the scenario where operators cannot obtain accurate data on individual prosumers due to incomplete information. It employs a Bayesian sequential game framework, with the specific method as follows: 2.1 Incomplete Information Assumption While shared energy storage cannot obtain information on the power generation, consumption, and regulation capabilities of every individual producer and consumer, it does possess incomplete information on the overall market's power generation, consumption, and regulation capabilities, as well as the probability distribution of shared energy storage usage costs. This information can be mathematically expressed as follows: in, In the scene The total renewable energy power generation of the downstream consumer group during time period t; In the scene Lower-level consumers The spontaneous power generation during time period t; In the scene The total baseline load power of the consumer group in time period t; In the scene Lower-level consumers The electrical load during time period t The estimated unit cost of shared energy storage for all producers and consumers using shared energy storage is: The estimated value of the overall transferable load cost coefficient for producers and consumers is These parameters all follow a certain probability distribution. 2.2 Upper-level model: Maximizing the expected revenue of shared energy storage operators The upper-level model, aiming to maximize the expected revenue of shared energy storage operators, can be expressed as: in, To share the benefits of energy storage; Price parameters for storage capacity; Price parameters for power capacity; Shared energy storage capacity leased by the total prosumer group; The total shared energy storage capacity leased by the prosumer group; Let be the expectation operator, representing a weighted average over all scenarios. The constraints of the upper-level model are: in, This indicates finding the minimum value of the decision variable x; x is a vector containing the decision variables of all producers and consumers. Equation (10) represents the capacity decision of shared energy storage, which is a comprehensive electricity cost function for the prosumer group. and The optimal solution must be found within the set of solutions that minimize the costs for both producers and consumers; this reflects the master-slave relationship in a two-layer game. 2.3 Lower-level model: Minimizing the overall electricity cost for the prosumer group The lower-level model, aiming to minimize the overall electricity cost for all producers and consumers, can be expressed as: in, The overall electricity cost for the prosumer group; The grid interaction cost for all producers and consumers; Shared energy storage costs for all producers and consumers; For the transferable load cost of all prosumers; For the energy storage costs of all producers and consumers, The detailed expressions for each cost item are as follows: (1) Grid interaction cost: in, The total power purchased by the consumer group from the grid during time period t; The total power sold to the grid by the prosumer group during time period t; The grid purchase price per unit of electricity during time period t; The unit price of electricity sold by the power grid during time period t. (2) Cost of shared energy storage: in, Shared energy storage capacity leased to prosumer groups; Shared energy storage capacity leased to prosumer groups; Estimate the unit power cost of shared energy storage for all producers and consumers using shared energy storage; The charging power of shared energy storage for the consumer group during time period t; For the discharge power of the consumer group on the shared energy storage during time period t, To share the cost of energy storage capacity leasing, To share the cost of energy storage power usage, (3) Cost of transferable load: in, An estimate of the cost coefficient of shared energy storage for the overall transferable load of producers and consumers; Let be the transferable load power of the prosumer group in time period t. This formula is derived through the KKT conditions, compressing the transferable load decision of each prosumer into the decision of the prosumer group as a whole. (4) Cost of self-owned energy storage: in, The unit power cost for producers and consumers to use their own energy storage as a whole; The charging power of the consumer group for its own energy storage during time period t; The discharge power of the consumer group relative to its own energy storage during time period t. 2.4 Constraints The lower-level model must satisfy the following constraints: (1) Energy balance constraint: in, The net load power of the prosumer group during time period t; The total baseline load power of the producer-consumer group in time period t; The charging power of the consumer group for its own energy storage during time period t; The discharge power of the consumer group to its own energy storage during time period t; The total renewable energy power generation of the prosumer group during time period t; The transferable load power of the consumer group in time period t. (2) Shared energy storage constraints: in, Shared energy storage capacity for rent by consumer groups during time period t; Shared energy storage capacity leased by prosumer groups during time period t-1; For the charging and discharging efficiency of shared energy storage; The charging power of shared energy storage for the consumer group during time period t; The discharge power of the consumer group on the shared energy storage during time period t; The duration of the time period; Shared energy storage capacity leased to prosumer groups; For the shared energy storage charging status of the consumer group during the time period t; Shared energy storage capacity leased to prosumer groups; This is the lower limit of the state of charge of shared energy storage; This represents the upper limit of the shared energy storage state of charge. For shared initial storage capacity; For producers and consumers Leased shared energy storage capacity; For producers and consumers Leased shared energy storage capacity; The total storage capacity of the shared energy storage system; To share the total power capacity of the energy storage system, (3) Transferable load constraints: in, Let be the transferable load power of the producer-consumer group in time period t. This constraint indicates that the sum of the transferable load in all time periods is 0, that is, the total load transfer is balanced. (4) Constraints on self-owned energy storage: in, The self-owned energy storage capacity of the prosumer group during time period t; The energy storage capacity owned by the prosumer group during time period t-1; The charging and discharging efficiency of its own energy storage; The charging power of the consumer group for its own energy storage during time period t; The discharge power of the consumer group to its own energy storage during time period t; The duration of the time period; For the self-owned energy storage capacity of the prosumer group; The state of charge of the consumer group's own energy storage during time period t; For the self-owned energy storage capacity of the prosumer group; This is the lower limit of the state of charge of the self-storage energy; This represents the upper limit of the state of charge of the self-owned energy storage.

4. The game-theoretic pricing method for shared energy storage services according to claim 1, characterized in that, The specific method for step 3 is as follows: 3.1 Problem Abstraction set up For the decision variables of the prosumers as a whole, For decision variables of shared energy storage, The decision variable represents the cost of the total cost to producers and consumers as a whole, which is a quadratic function. Let $\mathbf$ represent the decision variables that are linearly related to the costs of the total producers and consumers. This Bayesian sequential game problem can be abstracted as follows: in, Both are parameter matrices; For parameter vectors; For the dual variables of the overall constraint on prosumers; The parameters representing the estimated parameters of shared energy storage on the overall parameters related to producers and consumers are: Equation (31) is the upper-level objective function, Equation (32) is the optimality condition of the lower-level problem, and Equation (33) is the constraint condition of the lower-level problem. For the dual variable of constraint (33), 3.2 Duality Problem and Strong Duality Theory In the optimization problem that the prosumer-consumer group needs to consider, y is already given by the shared energy storage, so it is a constant for the prosumer-consumer group. The prosumer-consumer group problem is a QP problem, and its dual problem is: in, Equation (34) is the dual objective function, Equation (35) is the dual feasibility constraint, and Equation (36) is the condition that the gradient is zero. Representation matrix The inverse matrix; Representing vectors transpose, The original problem is a convex optimization problem. According to strong duality theory, at the optimal solution, the duality gap between the minimum value of the original problem and the maximum value of the dual problem is 0. Therefore, we can deduce that: This allows us to identify the non-convex bilinear term in the shared energy storage objective function. Transform it into a quadratic problem, eliminate bilinear terms, 3.3 Scenario-based transformation For prosumer models containing boundary information of probability distributions, a scenario-based approach can be used. Shared energy storage samples based on its available probability distribution information to generate several scenarios. The optimal capacity rental price for these scenarios is then calculated. Assuming the number of scenarios is K, the optimization problem can be written as: Where K is the total number of scenes; For the k-th scene; By assigning weights to each scenario and assuming equal probability across scenarios, the scenario-based transformation converts the original Bayesian problem into a deterministic problem. 3.4 KKT conditions and Big M method conversion Considering that the producer-consumer problem is a QP problem, the KKT conditions are used to transform it into constraints for a shared energy storage problem, thus converting the sequential game problem into a single-layer model. For the complementary relaxation constraints in the KKT conditions, the Big M method can be used for linearization, further simplifying the problem solution. The final problem is transformed into an MIQP problem: in, Let M be the coefficient matrix of the quadratic term; M is a sufficiently large constant introduced by the Big M method; For the introduced 0-1 variable vector, equation (43) is the gradient zero condition, and the derivative of the objective function with respect to the decision variables is 0; equation (44) is the dual feasibility condition; equation (45) is the dual variable nonnegativity condition and the upper bound of the Big M method; equation (46) is the complementary relaxation condition after linearization of the Big M method, which includes the product constraint. Linearization, 3.5 Parameter Adjustment for Variable Pricing Rules The solution for shared energy storage yields the capacity leasing unit price under a fixed pricing rule. Subsequently, it is possible to estimate the overall capacity of shared energy storage leased by producers and consumers. and The expected value, and finally, the shared energy storage is based on the preset pricing rule slope. Adjust the parameters as follows: in, The adjusted storage capacity price parameters; The storage capacity and price parameters before adjustment; The slope of the price of storage capacity as a function of the amount of capacity leased. The expected value of shared energy storage capacity for the entire production and consumption process; The adjusted power capacity price parameters; The power capacity and price parameters before adjustment; The slope of the price of power capacity as a function of the amount of capacity leased. The expected value of shared energy storage capacity for producers and consumers as a whole.

Images